Multislice CT [4 ed.] 9783319425863, 3319425862

Table of contents :
In Memoriam
Contents
Part I: Techniques
Multi-slice CT: Current Technology and Future Developments
1 Introduction
2 System Design
2.1 Gantry
2.2 X-Ray Tube and Generator
2.3 MDCT Detector Design and Slice Collimation
2.4 Dual Source CT
3 Measurement Techniques
3.1 MDCT Sequential (Axial) Scanning
3.2 MDCT Spiral (Helical) Scanning
3.2.1 Pitch
3.2.2 Collimated and Effective Slice Width
3.2.3 Image Reconstruction
3.2.4 Double z-Sampling
3.3 ECG-Synchronized Cardiovascular CT
3.3.1 ECG-Gated Spiral CT
3.3.2 ECG-Triggered Sequential (Axial) CT
3.3.3 ECG-Triggered High-Pitch Spiral CT
3.4 Dual Energy Computed Tomography
4 Radiation Dose Reduction
4.1 Anatomical Tube Current Modulation
4.2 Adaptation of the X-Ray Tube Voltage
4.3 Spectral Shaping
4.4 Iterative Reconstruction
5 Future Developments
References
Radiation Dose Optimization in CT
1 Introduction
2 Dose Metrics
3 Justification for CT
4 Effects of Scan Factors on Radiation Dose and Image Quality
4.1 Reconstruction Techniques
4.2 Body Size-Adapted CT Protocols
5 Image Quality and Radiation Dose
6 Dose Optimization Based on Specific Body Regions
References
Spectral CT/Dual-Energy CT
1 Introduction
2 Concept of Dual Energy
3 Design of Devices
3.1 Single-Source Dual-Energy CT (ssDECT)
3.1.1 Sequential Acquisition
3.1.2 Rapid kV Switching
3.1.3 Split-Filter Technology
3.1.4 Sandwich or Layer Detectors
3.1.5 Photon-Counting Detectors
3.2 Dual-Source Dual-Energy CT (dsDECT)
4 Image Post-Processing and Clinical Applications
4.1 “Morphological” Images
4.2 VME Images
4.2.1 Clinical Applications
Artifact Reduction
Optimizing Contrast
Reducing the Intravenous Contrast Media
4.3 Material-Specific Images
4.3.1 Clinical Applications of Different Materials
Water
Iodine
Calcium
Urate
Fat
Radiation Dose Considerations
Challenges
Future Opportunities
References
Contrast Enhancement at CT
1 Pharmacokinetics of the Iodine Contrast Material
1.1 Distribution of Contrast Material (CM) in the Body (Fig. 1)
1.2 Phases of Contrast Enhancement
1.3 Time-Density Curve
1.4 Simulation of CM Distribution in the Body
2 Factors Influencing the Enhancement of Blood Vessels and Organs
2.1 Patient Factors
2.1.1 Body Size
2.1.2 Cardiac Function
2.1.3 Renal Function
2.1.4 Hepatic Function
2.1.5 Other Patient Factors
2.2 Contrast Material Factors
2.2.1 Iodine Dose
2.2.2 Injection Duration (Injection Rate)
2.2.3 Iodine Concentration
2.2.4 CM Osmolality and Viscosity
2.2.5 Saline Chaser
2.3 Scan Factors
3 Optimization of Scan Protocols Taking into Account CM Pharmacokinetics
References
Image Processing from 2D to 3D
1 Introduction
2 Principles of Rendering Techniques
3 Surface Rendering
4 Maximum Intensity Projection
5 Volume Rendering
6 Cinematic Rendering
References
Perfusion CT: Technical Aspects
1 Introduction
2 Principles of Perfusion CT
2.1 Contrast Agent Kinetics
2.2 Biological Correlates of Perfusion CT Parameters
3 Acquisition Protocols
3.1 Patient Preparation
3.2 Contrast Agent Administration
3.3 Data Acquisition
3.4 Motion Correction and Image Registration
3.5 Arterial Input Function and Venous Output
4 Clinical Implementation
4.1 Quality Assurance and Quality Control
4.2 Accuracy and Precision
4.3 Radiation Dose
5 Summary
References
Part II: Neuro/ENT
Stroke/Cerebral Perfusion CT: Technique and Clinical Applications
1 Introduction
2 Perfusion CT Technique
2.1 Image Acquisition
2.2 Image Processing
3 Clinical Applications
3.1 Stroke
3.2 Vasospasm
3.3 Moyamoya
3.4 Traumatic Brain Injury
3.5 Brain Tumors
4 Summary
References
CT Diagnostics in Brain Tumors
1 Introduction
2 Specific Role of CT Imaging
2.1 Diagnostic and Differential Diagnostic
2.2 Therapy Planning
2.3 Treatment Follow-Up
3 CT Characteristics of Brain Tumors
3.1 WHO Classification of Brain Tumors
3.2 Contrast Enhancement of Brain Tumors
3.2.1 Mechanisms of Contrast Enhancement in Brain Tumors
4 Primary Intraaxial Gliomatous Tumors
4.1 WHO Grade I Tumors
4.2 WHO Grade II Tumors
4.3 WHO Grade III
4.4 Glioblastoma
4.5 Gliomatosis Cerebri
4.6 Oligodendroglioma
4.7 Mixed Glioma
5 Lymphoma
6 Metastasis
7 Others Intraaxial Tumors
7.1 Ependymoma
7.2 Subependymoma
7.3 Anaplastic Ependymoma
7.4 Medulloblastoma
8 Extraaxial
8.1 Meningioma
8.2 Hemangiopericytoma
8.3 Melanocytoma
8.4 Choroid Plexus Tumors
9 Tumors of the Sellar Region
9.1 Pituitary Tumors
9.2 Craniopharyngioma
9.3 Rathke’s Cleft Cyst
10 Tumors of Peripheral Nerves that Affect the CNS
10.1 Schwannoma
11 Non-tumorous Changes
11.1 Arachnoid Cysts
11.2 Epidermoid Cysts
11.3 Dermoid Cysts
12 Extraaxial Metastasis
References
MDCT in Neurovascular Imaging
1 Background
2 Protocol Parameters for Neurovascular MDCT
3 Clinical Applications of Neurovascular MDCT
3.1 Comprehensive Stroke Imaging
3.2 Carotid Artery Stenosis
3.3 Basilar Artery Occlusions
3.4 Intracranial Aneurysms
3.5 Cerebral Venous Thrombosis
3.6 Cervical Artery Dissection
References
Anatomy and Pathology of the Temporal Bone
1 Introduction
2 Anatomy
3 Multi-slice CT Technique
4 Pathology
4.1 Inflammation
4.2 Trauma
4.3 Tumour
4.4 Malformation
4.5 Otosclerosis
4.6 Postoperative CT
5 Take-Home Points
Further Reading
Dental CT: Pathologic Findings in the Teeth and Jaws
1 Systematic Approach for Characterizing Pathologic Findings
1.1 Radiodensity
1.2 Transition Zone
1.3 Bone Remodeling
1.4 Location
1.5 Relation to Teeth
1.6 Clinical Symptoms and History
2 Clinical Examination and Imaging
2.1 Dental CT Examination Protocols
2.2 Anatomy
2.3 Tooth
3 Radiolucent Lesions with Well-Defined Borders
3.1 Radicular Cyst
3.2 Residual Cyst
3.3 Follicular Cyst
3.4 Keratocystic Odontogenic Tumor
3.5 Ameloblastoma
3.6 Stafne Cyst
3.7 Simple Bone Cyst
3.8 Primordial Cyst
4 Sclerotic, Tooth-Related Jaw Lesions
4.1 Cementoblastoma
4.2 Cemento-Osseous Dysplasia (COD)
4.3 Condensing Osteitis
4.4 Odontoma
4.5 Idiopathic Osteosclerosis
5 Sclerotic Not Tooth-Related Jaw Lesions
5.1 Osteoma
5.2 Tori and Exostoses
6 Mixed Lytic and Sclerotic Jaw Lesions
6.1 Osteomyelitis
6.1.1 AO and SCO
6.1.2 PCO
6.1.3 Proliferative Periostitis
6.2 Osteoradionecrosis of the Jaw
6.2.1 Osteoradionecrosis (ORN)
6.2.2 Medication-Related Osteonecrosis of the Jaw
References
Anatomy and Corresponding Oncological Imaging of the Head and Neck
1 Introduction
2 Paranasal Sinuses and Nasopharynx
2.1 Anatomy
2.2 Oncological Imaging of Sinonasal Masses
3 Oral Cavity and Oropharynx
3.1 Anatomy of Oral Cavity
3.1.1 Anatomy of Oropharynx
3.2 Oncological Imaging of Oral Masses
4 Hypopharynx and Larynx
4.1 Anatomy of Hypopharynx
4.1.1 Anatomy of Larynx
4.2 Imaging of Hypopharyngeal and Laryngeal Masses
4.2.1 Hypopharynx
4.2.2 Larynx
References
Part III: Chest
Interstitial Lung Diseases
1 Introduction
2 Anatomic and Technical Considerations
2.1 Normal Lung Anatomy
2.1.1 The Peribronchovascular Interstitium
2.1.2 The Secondary Pulmonary Lobule
2.1.3 Interlobular Septa
2.1.4 Centrilobular Region
2.1.5 Lobular Parenchyma
2.2 CT-Technique
2.2.1 Spaced High-Resolution CT
2.2.2 Volumetric High-Resolution CT
2.2.3 Other Parameters Influencing Radiation Dose
2.2.4 Reconstruction Kernels
2.2.5 Tube Voltage Selection and Tube Current Modulation
2.2.6 Spectral Shaping
2.2.7 Low Dose CT
2.2.8 Protocol Decision
3 HRCT Pattern
3.1 Linear and Reticular Pattern
3.2 Nodular Pattern
3.2.1 Perilymphatic Nodules
3.2.2 Centrilobular Nodules
3.2.3 Random Nodules
3.3 High Attenuation Pattern
3.4 Low Attenuation Pattern
4 Interstitial Lung Diseases That Have No Known Cause
4.1 Idiopathic Interstitial Pneumonias (IIP)
4.1.1 Idiopathic Pulmonary Fibrosis (IPF)
4.1.2 Nonspecific Interstitial Pneumonia (NSIP)
4.1.3 Cryptogenic Organizing Pneumonia (COP)
4.1.4 Respiratory Bronchiolitis-Associated Interstitial Lung Disease (RB-ILD)
4.1.5 Desquamative Interstitial Pneumonia (DIP)
4.1.6 Lymphoid Interstitial Pneumonia (LIP)
4.1.7 Acute Interstitial Pneumonia (AIP)
4.2 Sarcoidosis
4.3 Miscellaneous Rare Forms of Interstitial Lung Disease of Unknown Etiology
4.3.1 Pulmonary Langerhans Cell Histiocytosis
4.3.2 Lymphangioleiomyomatosis
4.3.3 Eosinophilic Pneumonia
4.3.4 Pulmonary Alveolar Proteinosis
4.3.5 Pulmonary Microlithiasis
5 Interstitial Lung Diseases of Known Cause
5.1 Occupational and Environmental Lung Disease
5.1.1 Hypersensitivity Pneumonitis
5.1.2 Pneumoconiosis
5.1.3 Drug-Induced Lung Disease
5.2 Radiation-Induced Lung Injury
5.3 Collagen Vascular Lung Disease
5.4 Diffuse Pulmonary Hemorrhage
References
Pneumonia
1 Introduction
1.1 General Recommendations for Imaging
1.2 Methods and Techniques
1.2.1 Chest X-Ray
1.2.2 Computed Tomography (CT)
1.2.3 Magnetic Resonance Imaging (MRI)
1.2.4 Positron Emission Tomography
2 Forms of Pneumonia
2.1 Bacterial Pneumonia
2.2 Mycobacterial Pneumonia
2.3 Viral Pneumonia
2.4 Pneumocystis Jirovecii Pneumonia
2.5 Fungal Pneumonia
2.6 Parasitic Lung Infection
3 Noninfectious Infiltrates
3.1 Organizing Pneumonia
3.1.1 Drug Toxicity
3.1.2 Radiation Pneumonia
3.2 Graft Versus Host Disease
3.3 Other Diseases
4 Special Forms and Complications of Pneumonia
4.1 Aspiration Pneumonia
4.2 Retention Pneumonia
4.3 Post Infarction Pneumonia
4.4 Septic Embolism
4.5 Lung Abscess
4.6 Pleural Empyema
5 Patients with Underlying Disorders
5.1 Pneumonia in Immunocompromised Patients
5.2 Pneumonia in Patients with Altered Airways
References
CT of the Airways
1 Introduction
1.1 Radiologic Anatomy of the Airways and Physiology
2 Technical Aspects
2.1 Scanning Protocol
2.2 Multiplanar Reformats, MIP, MinIP, and Image Processing
2.3 Virtual Bronchoscopy and 3D Views
3 Evaluation
3.1 Visual Evaluation
3.2 Quantitative Evaluation
4 Elementary CT Findings in Airway Diseases
4.1 Bronchiectasis and Bronchiolectasis
4.2 Bronchial Wall Thickening
4.3 Mucus Plugging
4.4 Tree-in-Bud Pattern
4.5 Mosaic Attenuation Pattern and Air Trapping
5 Diseases
5.1 Tracheobronchomalacia
5.2 Tracheobronchomegaly (Mounier-Kuhn Syndrome)
5.3 Tracheal Stenosis
5.4 Saber Sheath Trachea
5.5 Primary Ciliary Dyskinesia (Immotile Cilia Syndrome)
5.6 Cystic Fibrosis
5.7 Tracheobronchial Amyloidosis
5.8 Sarcoidosis of the Airways
5.9 Relapsing Polychondritis
5.10 Granulomatosis with Polyangiitis
5.11 Tracheopathia Osteoplastica
5.12 Asthma
5.13 Bronchiolitis
5.14 Bronchiolitis Obliterans
5.15 Swyer-James Syndrome
References
Lung Cancer Screening
1 Text
1.1 NLST: First Study to Demonstrate a Mortality Benefit of LDCT-Screening
1.2 Screening Programs and Reimbursement in the USA
2 European Studies and Recommendations
2.1 High False-Positive Rate
2.2 Overdiagnosis
2.3 Radiation Dose
2.4 Interval Cancers
2.5 Management of New Solid Nodules
2.6 Management of Nonsolid Nodules
2.7 Smoking Cessation
2.8 Biomarkers
2.9 Cost-Effectiveness
2.10 CT Can Do More Than Identifying Lung Nodule: Screening for “Big-Three”
References
Chest Neoplasias
1 Lung Cancer
1.1 CT Manifestations of Lung Cancer
1.1.1 Pulmonary Nodule or Mass
1.1.2 Hilar or Mediastinal Enlargement
1.1.3 Airway Abnormalities
1.1.4 Pulmonary Consolidation
1.1.5 Lymphangitic Tumor Spread
1.2 Lung Cancer by Cell Types
1.2.1 Adenocarcinoma
1.2.2 Squamous Cell Carcinoma
1.2.3 Large Cell Carcinoma
1.2.4 Neuroendocrine Tumors
1.3 Lung Cancer Staging
1.3.1 Primary Tumor (T Descriptor)
1.3.2 Regional Lymph Node Status (N Descriptor)
1.3.3 Distant Metastatic Disease (M Descriptor)
2 Malignant Pleural Mesothelioma (MGM)
3 Primary Neoplasms of the Mediastinum
3.1 Thymic Neoplasms
3.1.1 Thymic Epithelial Tumors
3.1.2 Rare Thymic Tumors
3.2 Germ Cell Tumor
3.2.1 Teratoma
3.2.2 Seminoma
3.2.3 NSGCT
3.3 Lymphoma
3.3.1 HL
3.3.2 NHL
3.4 Neurogenic Tumors
References
CT of Pulmonary Embolism: Imaging Update
1 Introduction
2 New Options in the Diagnostic Approach
2.1 Clinical Decision Support
2.2 Optimization of the Radiation Dose
2.3 Optimization of the Iodine Load
2.3.1 Low Contrast Medium Volume
2.3.2 Low-Concentrated Contrast Agents
2.4 Current Role of CAD Systems
2.5 Is There Always a High Level of Confidence in Diagnosing Acute PE?
3 New Options in the Prognostic Assessment
3.1 Clot Burden
3.2 Right Ventricular Dysfunction
3.3 Novel Approaches
4 The Radiologist’s Report: Which Information Is Particularly Relevant for Clinicians?
5 Pulmonary Embolism from Pregnancy to Young Adults
5.1 Pulmonary Embolism in Pregnancy
5.2 Pulmonary Embolism in Children
6 When Acute PE Evolves Toward Chronic PE
References
COPD
1 Definition and Clinical Features of COPD
2 Imaging
2.1 Identification and Phenotyping
2.1.1 Chest Radiography
2.1.2 Computed Tomography
Technical Considerations
CT Appearance
2.2 Quantification
2.3 Role of Quantitative Imaging in Clinical Practice: The Present and the Future
3 Exacerbation and Comorbidities
References
Part IV: Abdomen
Focal Lesions in Non-cirrhotic Liver
1 Introduction
2 Protocols for MDCT of the Liver
3 Radiological Appearance of Different Focal Liver Lesions
3.1 Solid Benign Lesions
3.2 Metastatic Liver Lesions
3.3 Primary Liver Tumors
3.4 Other Lesions
4 Role of MDCT in Emergencies and Follow-Up After Radiological Interventions
5 Summary
Literature
Cirrhotic Liver
1 Introduction
2 CT Scan Protocols
2.1 Amount of Intravenous Contrast Material
2.2 Flow Rate of Contrast Material
2.3 Scan Timing
2.4 Radiation Issues
3 Cirrhosis
3.1 Epidemiology
3.2 Pathophysiology
3.3 Features of Liver Cirrhosis on CT
4 Cirrhotic Nodules
4.1 Multistep Carcinogenesis of Liver Cirrhosis
4.2 Multistep Changes in Intranodular Blood Flow
4.3 Regenerative Nodules
4.4 Dysplastic Nodules
5 CT Findings for HCC
5.1 Progressed HCC
5.2 Major Imaging Features
5.2.1 Arterial Phase Hyperenhancement
5.2.2 Washout Appearance
5.2.3 Capsular Enhancement
5.3 Other Imaging Features
5.3.1 Vascular Invasion
5.3.2 Fatty Metamorphosis
5.3.3 Mosaic Architecture
5.3.4 Nodule-In-Nodule Appearance
6 Early HCC
7 HCC Variants
7.1 Diffuse or Infiltrative HCC
7.2 Scirrhous HCC
7.3 Sarcomatous HCC
7.4 Fibrolamellar HCC
7.5 Combined HCC-Cholangiocarcinoma
8 Advanced CT Techniques for HCC
8.1 Four-Dimensional CT
8.2 Low-Kilovoltage CT
8.3 Perfusion CT
8.4 Dual-Energy CT
References
Pancreatic Tumors
1 Introduction
2 Standard CT for Pancreatic Tumor Evaluation
3 Pancreatic Ductal Adenocarcinoma (PDAC)
3.1 Morphologic Evaluation of Main Tumor and Secondary Signs
3.2 Vascular Evaluation
3.3 Extrapancreatic Evaluation
3.4 Performance of CT for Staging and Resectability of Pancreatic Cancer
4 Pancreatic Neuroendocrine Tumors
4.1 Nonfunctioning Pancreatic Neuroendocrine Tumors
4.2 Insulinoma
4.3 Gastrinoma
5 Cystic Neoplasms
5.1 Serous Cystadenoma (SCA)
5.2 Mucinous Cystic Neoplasm (MCN)
5.3 Intraductal Papillary Mucinous Neoplasm (IPMN)
5.4 Solid Pseudopapillary Epithelial Neoplasm (SPEN)
5.5 Cystic Change of Solid Tumors
6 Secondary Pancreatic Neoplasms
7 Nonneoplastic Solid Lesions Mimicking Pancreatic Solid Tumors
7.1 Focal Pancreatitis
7.2 Fatty Infiltration-Replacement
7.3 Intrapancreatic Accessory Spleen
References
Acute and Chronic Pancreatitis
1 Acute Pancreatitis
1.1 Introduction
1.2 Diagnostic Criteria
1.3 Phases of Acute Pancreatitis
1.4 Severity of Acute Pancreatitis
1.5 Imaging
1.6 Morphological Classification
1.7 Pancreatic and Peripancreatic Collections
1.8 Acute Peripancreatic Fluid Collection and Pseudocyst
1.9 Acute Necrotic Collection and Walled-Off Necrosis
1.10 Infection
1.11 Vascular Complications
1.12 Treatment
2 Chronic Pancreatitis
2.1 Introduction
2.2 Imaging
2.3 Treatment
2.4 Other Forms of Chronic Pancreatitis
2.4.1 Paraduodenal Pancreatitis
2.4.2 Autoimmune Pancreatitis
2.4.3 Hereditary Chronic Pancreatitis
References
Spleen
1 Imaging Anatomy and Function of the Spleen
2 Trauma
2.1 Introduction
2.2 Classification
2.3 Imaging Characteristics
2.4 Image Examples (Figs. 6–10)
2.5 Treatment Options
3 Benign Lesions
3.1 General Principles
3.2 Cysts
3.3 Epidermoid Cyst
3.4 Hemangioma
3.5 Lymphangioma
3.6 Hamartoma
3.7 Sclerosing Angiomatoid Nodular Transformation
3.8 Angiomyolipoma
4 Semi-malignant Lesions
4.1 Littoral Cell Angioma
4.2 Hemangioendothelioma
5 Malignant Lesions
5.1 Lymphoma
5.2 Splenic Metastases
5.3 Angiosarcoma
5.4 Hemangiopericytoma
6 Infection
6.1 Abscess
6.2 Hydatid Infection
6.3 Tuberculosis
6.4 Malaria
7 Miscellaneous
7.1 Splenic Infarction
7.2 Extramedullary Hematopoiesis
7.3 Splenic Sarcoidosis
7.4 Splenic Amyloidosis
8 Differential Diagnosis Based on Imaging Finding
8.1 Splenomegaly
8.2 Splenic Infarction
8.3 Splenic Rupture
References
Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques
1 Introduction
2 CT and PET-CT Technique
2.1 CT Technique
2.2 PET-CT Technique
3 Esophageal Cancer
3.1 CT, PET-CT Imaging of the Esophagus
3.2 Tumor Detection and Classification
3.2.1 N Staging
3.2.2 M Staging
3.3 Follow-Up After Esophagectomy
3.3.1 CT Findings
3.3.2 CT and PET-CT for Predicting Survival
4 Other Esophageal Malignancies
4.1 Esophageal Lymphoma
4.1.1 CT and PET/CT Findings
4.2 Leiomyoma and GIST
4.2.1 CT Findings
4.3 Fibrovascular Polyps
4.3.1 CT Findings
4.4 Esophageal Fistula
4.4.1 CT Findings
4.5 Achalasia
4.5.1 CT Findings
4.6 Dysphagia Lusoria
4.6.1 CT Findings
5 Gastroesophageal Diseases
5.1 Gastro- and Esophageal Perforation
5.1.1 CT Findings
5.2 Diverticula
5.2.1 CT Findings
5.3 Duplication Cyst
5.3.1 CT Findings
5.4 Hiatal Hernia
5.4.1 CT Findings
5.5 Esophagitis, Gastritis, and Ulcer
5.5.1 CT Findings
5.6 Gastroesophageal Varices
5.6.1 CT Findings
6 Gastric Adenocarcinoma
6.1 Tumor Detection and Classification
6.2 T Staging
6.3 N Staging
6.4 M Staging
6.5 Follow-Up After Partial Gastrectomy
7 Other Gastric Malignancies
7.1 Gastric Lymphoma
7.1.1 CT Findings
7.2 Gastro-Entero-Pancreatic Neuroendocrine Tumors (GEPNETs)
7.2.1 CT Findings
7.3 Gastrointestinal Stromal Tumors (GISTs)
7.4 Neural Tumors
7.5 Gastric Outlet Obstruction
References
Small Bowel MDCT
1 Introduction
2 General Recommendations for Imaging
2.1 Patient Preparation
2.2 Acquisition Protocol
2.3 Enteral Contrast Medium
3 Small Bowel Diseases
3.1 Tumours
3.1.1 CT Findings of Benign Tumours
3.1.2 CT Findings of Malignant Tumours
3.2 Crohn’s Disease
3.2.1 CT Findings
3.3 Intestinal Ischemia and Infarction
3.3.1 CT Findings
3.4 Obstructions and Perforations
3.4.1 CT Findings
3.5 Other Entities
3.5.1 Graft-Versus-Host Disease (GVHD)
3.5.2 Coeliac Disease
References
Imaging of Large Bowel with Multidetector Row CT
1 Colorectal Cancer
1.1 Colorectal Cancer Pathophysiology
1.2 Colorectal Staging
1.3 Colorectal Screening
1.4 CT Colonography
1.5 CTC Technique
1.6 CTC: Polyps and Cancer
1.7 CTC Reporting
1.8 CTC Screening
1.9 CT Colonography Indications/Contraindications
2 Colonic Lymphoma
3 Colitis
3.1 Inflammatory Bowel Disease: Ulcerative Colitis and Crohn’s Disease
3.2 Infectious Colitis
3.3 Pseudomembranous Colitis
3.4 Ischemic Colitis
3.5 Typhlitis
3.6 Stercoral Colitis
4 Acute Diverticulitis
5 Appendix
5.1 Appendicitis
5.2 Primary Neoplasms of the Appendix
5.2.1 Mucinous Epithelial Neoplasm: Mucocele of the Appendix
5.2.2 Nonmucinous Epithelial Neoplasm
5.2.3 Carcinoid Tumor
5.2.4 Other Neoplasms of the Appendix
6 Epiploic Appendagitis
7 Colonic Volvulus
7.1 Cecal Volvulus
7.2 Sigmoid Volvulus
8 Lower Gastrointestinal Bleeding: Role of CTA
References
Peritoneal Surface Malignancy
1 Introduction
2 Definition and Clinical Features of PC
3 CT Imaging
3.1 Technical Consideration
3.2 CT Appearance with Pathological Correlation
3.3 Quantification
4 Role of Laparoscopy, US, MR, and PET
References
Multislice PET/CT in Neuroendocrine Tumors
1 Introduction
2 Diagnostic Algorithm in NET
3 Somatostatin Receptor Imaging with PET
3.1 Somatostatin Receptor Agonists
3.1.1 Detection of Primary
3.1.2 Suspicion of Tumor Recurrence
3.1.3 Theranostics
3.1.4 Response Prediction to PRRT
3.1.5 Therapy Monitoring
3.2 Somatostatin Receptor Antagonists
4 PET Using Fluorodeoxyglucose (FDG)
5 PET Using Fluorodihydroxyphenylalanine (FDOPA)
6 Glucagon-like Peptide-1 Receptor (GLP-1R) Imaging
References
Adrenals
1 Introduction
2 Morphology
3 CT Densitometry
4 Contrast Media Kinetics
5 Multi-energy CT
References
Kidneys, Ureters, and Bladder
1 Kidneys
1.1 Anatomy
1.2 Renal Imaging
1.3 Incidental Renal Lesions
1.4 Multi-energy Imaging of Incidental Renal Lesions
1.5 Cystic Renal Masses
1.6 Solid Renal Masses
1.7 Renal Cell Carcinoma
1.8 Pseudotumor
1.9 Renal Infections
1.10 Traumatic Renal Injury
2 Ureters
3 Bladder
References
Part V: Cardiovascular
Technical Innovations and Concepts in Coronary CT
1 Introduction
2 Coronary CT Angiography: Current Status
2.1 Technical Principle
2.2 Acquisition Techniques
3 Tube Voltage
3.1 Technical Background
3.2 Impact on Image Luminal Contrast
3.3 Impact on Radiation Exposure
4 Iterative Reconstruction
4.1 Technical Principles
4.2 Technical Evolution
4.3 Impact on Image Appearance/Quality
4.4 Impact on Radiation Exposure
4.5 Limitations
5 New Concepts in Cardiac CT
5.1 Assessment of Cardiac Function
5.2 Techniques for the Evaluation of Hemodynamic Significance
5.2.1 Cardiac Perfusion
5.2.2 Transluminal Attenuation Gradient (TAG)
5.2.3 CT Fractional Flow Reserve (CT-FFR)
References
Noninvasive Coronary Artery Imaging
1 Introduction
2 Technical Issues
2.1 Radiation Dose
2.1.1 ECG Gating
2.1.2 ECG Pulsing
2.1.3 Tube Current Modulation and Tube Voltage Adaptation
2.2 Image Reconstruction
2.2.1 Data Reconstruction
2.3 Reconstruction Phases
2.3.1 Filtered Back Projection
2.3.2 Iterative Reconstruction
2.3.3 Post-processing
3 Accuracy
4 Prognosis
5 Indications
5.1 Coronary Artery Disease
5.2 Coronary Artery Anomalies
5.3 Coronary Stents and Bypass Grafts
6 Summary
References
Pre- and Postinterventional/Surgical Evaluation by CT
1 Preprocedural Imaging
2 CCTA After Percutaneous Coronary Intervention
3 CT Imaging of Coronary Artery Stents
4 CT Imaging Features Affecting Image Quality
4.1 Temporal Resolution
4.2 Contrast Enhancement
4.3 Image Reconstruction and Scan Parameters
4.4 Display Techniques and Window Settings
4.5 Stent Properties Affecting Image Quality
4.6 Stent Location and Size
5 Results of Stent Imaging with Different CT Scanners
6 CT-Imaging of Coronary Bypass Grafts
7 CT Imaging Features Affecting Image Quality
8 Results of Graft Imaging with CT
References
Computed Tomography in the Management of Electrophysiology Procedures
1 Atrial Fibrillation
1.1 Atrial Fibrillation and Pulmonary Vein Isolation
1.2 Technical Considerations for Computed Tomography Prior to Pulmonary Vein Isolation
1.2.1 Radiation Dose Concerns
1.3 Pulmonary Vein and Left Atrial Imaging in Atrial Fibrillation Ablation
1.3.1 Pulmonary Vein Anatomy
1.3.2 Left Atrial Anatomy
1.4 Electro-anatomic Mapping System Fusion with Computed Tomography Imaging
1.5 Pulmonary Vein Stenosis
1.6 Left Atrial Appendage Thrombus
1.7 Left Atrial Appendage Morphology Assessment
1.8 Left Atrial Appendage Occlusion Devices
1.8.1 LARIAT
1.8.2 WATCHMAN
2 Cardiac Resynchronization Therapy
2.1 Cardiac Resynchronization Therapy and Coronary Venous Imaging
2.2 Technical Considerations for Computed Tomography Coronary Venous Imaging
2.3 Noninvasive Coronary Venous Mapping Prior to Cardiac Resynchronization Therapy
2.4 Coronary Venous Imaging for Cardiac Resynchronization Therapy Nonresponders
3 Ventricular Tachycardia
3.1 Ventricular Tachycardia and Myocardial Scar
3.2 Technical Considerations for Late-Enhancement Computed Tomography for Scar Evaluation
3.3 Late-Enhancement Computed Tomography for Myocardial Scar Assessment
3.4 Computed Tomography Integration into Electro-anatomic Mapping Prior to Ventricular Tachycardia Ablation
4 Conclusions
References
Functional Cardiac CT Angiography
1 Introduction
2 CT Myocardial Perfusion Imaging
2.1 General Consideration
2.2 Static CTMPI: Single Energy Technique
2.2.1 Acquisition Technique and Image Analysis
2.2.2 Clinical Results
2.3 Static CTMPI: Dual-Energy Technique
2.3.1 Acquisition Technique and Image Analysis
2.3.2 Clinical Results
2.3.3 Limitations
2.4 Dynamic CTMPI
2.4.1 Acquisition Technique and Image Analysis
2.4.2 Clinical Results
2.4.3 Limitations
2.5 Pharmacological Stress Agents
3 Radiation Dose Considerations
4 Computed Tomography-Derived FFR
4.1 CT-FFR Basics
4.2 Clinical Results
4.3 Limitations
5 Summary
References
CT Angiography of the Peripheral Arteries
1 Introduction
1.1 Extremity CTA Technique
1.2 Patient Preparation
1.3 Image Acquisition
1.3.1 Protocol Series
1.4 Contrast Medium Administration
1.5 Strategies for Contrast Medium Administration
1.5.1 Synchronization
1.5.2 Scan Duration
1.5.3 Injection Parameters
1.5.4 Saline Flush
1.6 Image Display
1.6.1 Source Images
1.6.2 Visualization Techniques
1.6.3 Overview of Relevant Anatomy
1.6.4 Vessel Analysis
1.7 Clinical Applications
1.7.1 Peripheral Arterial Disease
Acute Ischemia
1.7.2 Aneurysms
1.7.3 Vasculitis
1.7.4 Trauma
1.7.5 Compression and Entrapment Syndromes
1.7.6 Arteriovenous Fistulae, Vascular Malformations, and Vascular Masses
1.7.7 Reconstruction Surgery
References
Acute Aortic Syndromes
1 Introduction
2 Aortic Imaging Technique
2.1 Selection of Modality
2.2 CTA: Scan Range and Phases
2.3 Scan Parameters
2.4 Contrast Material
2.5 Radiation Dose
2.6 Reconstruction
3 Role of Other Imaging Modalities
4 Aortic Dissection and Subtypes
4.1 Definition and Clinical Background
4.2 Classification
4.3 Classical Aortic Dissection
4.3.1 CT Morphology
4.4 Intramural Hematoma (IMH)
4.4.1 CT Morphology
4.5 Penetrating Atherosclerotic Ulcer (PAU)
4.5.1 CT Morphology
5 Symptomatic Aortic Aneurysms
5.1 CT Morphology
References
CT Venography
1 Introduction
2 Technical Considerations
3 CT Venography: Lower Extremity, Thromboembolic Disease
3.1 Indications
3.2 Technique
3.3 Imaging Findings
3.4 Advantages and Limitations
4 CT Venography: Lower Extremity, Other
4.1 Indications
4.2 Technique
4.3 Imaging Findings
4.4 Advantages and Limitations
5 CT Venography: Upper Extremity
5.1 Indications
5.2 Technique
5.3 Imaging Findings
5.4 Advantages and Limitations
6 CT Venography: Pulmonary
6.1 Indications
6.2 Technique
6.3 Imaging Findings
6.4 Advantages and Limitations
7 CT Venography: Portomesenteric
7.1 Indications
7.2 Technique
7.3 Imaging Findings
7.4 Advantages and Limitations
8 CT Venography: Iliocaval and Pelvic Veins
8.1 Indications
8.2 Technique
8.3 Imaging Findings
8.4 Advantages and Limitations
References
Aortic Aneurysm and Stent Graft Assessment
1 Aortic Aneurysm
1.1 Thoracic Aortic Aneurysm
1.2 Abdominal Aortic Aneurysm
2 Technical Considerations in CT Evaluation
2.1 Data Acquisition and Contrast Administration
2.2 Postprocessing Reconstruction
2.3 Technical Advances
3 Preoperative Evaluation for Abdominal Endovascular Aortic Repair (EVAR)
3.1 Vascular Access
4 Postoperative Evaluation of Abdominal Endovascular Aortic Repair (EVAR)
4.1 Other Complications
5 Pre- and Postoperative Evaluation of Thoracic Aortic Aneurysm by Stent Graft (TEVAR)
References
Part VI: Interventions
CT-Guided Biopsy and Drainage
1 CT-Guided Biopsy
1.1 Introduction
1.2 Patient Preparation and Aftercare
1.3 Sequential and CTF Guidance
1.4 CT-Guided Aspiration Biopsy
1.4.1 Indications
1.4.2 Materials
1.4.3 Technique
General Considerations
Special Considerations
1.4.4 Results
1.4.5 Complications
1.4.6 Key Points
1.5 CT-Guided Punch Biopsy
1.5.1 Indications
1.5.2 Materials
1.5.3 Technique
1.5.4 Results
1.5.5 Complications
1.5.6 Key Points
1.6 CT-Guided Drill Biopsy
1.6.1 Indications
1.6.2 Materials
1.6.3 Technique
1.6.4 Results
1.6.5 Complications
1.6.6 Key Points
2 CT-Guided Drainage
2.1 Introduction
2.2 Indications
2.3 Patient Preparation and Aftercare
2.4 Materials
2.5 Technique
2.5.1 Special Considerations
Abdomen
Chest
2.6 Results
2.6.1 Abdomen
2.6.2 Chest
2.7 Complications
2.8 Key Points
References
CT-Guided Spinal Interventions: Vertebroplasty/Kyphoplasty
1 Introduction
1.1 Osteoporotic Vertebral Body Fracture
1.2 Tumoral Osteolysis
2 Patient Selection
2.1 Osteoporotic Vertebral Body Fracture
2.2 Tumoral Osteolysis
3 Indications and Contraindications
3.1 Indications
3.2 Contraindications
4 Bone Cement
5 Preprocedural Evaluation
6 Technique
6.1 Biplane Fluoroscopy Guidance
6.2 Dual Guidance
6.3 CT-Fluoroscopy Guidance
7 Cement Application
8 Complications
9 Post-procedural Care
10 Results
10.1 Osteoporotic Vertebral Fractures
10.1.1 Benefits of Vertebroplasty Versus Sham Procedure (Placebo)
10.1.2 Benefits of Vertebroplasty Versus Optimal Medical Management
10.1.3 Vertebroplasty Versus Kyphoplasty
10.2 Harms of Vertebroplasty/Kyphoplasty
10.2.1 New Apparent Vertebral Fractures
10.3 Tumoral Osteolysis
10.3.1 Benefits for Analgesic Effects
10.3.2 Biomechanical Stabilization
References
CT-Guided Tumor Ablation
1 Introduction
2 Technique
2.1 Radiofrequency Ablation
2.2 Microwave Ablation (MWA)
2.3 Irreversible Electroporation
2.4 Laser
2.5 Kryoablation
3 Clinical Applications
3.1 Primary and Secondary Liver Tumors
3.1.1 Indications and Contraindications
3.1.2 Results in Liver Metastases
3.1.3 Results in HCC
3.1.4 Complications and Side Effects
3.2 Renal Cell Carcinoma
3.3 Lung Tumors
3.4 Bone and Soft Tissue Tumors
References
Functional CT for Image-Guided Personalized Tumor Interventions
1 Transarterial Embolization (TAE) and Transarterial Chemoembolisation (TACE)
2 Radiofrequency Ablation (RFA), Microwave- and Cryoablation
3 Intraarterial Radioembolization (Selective Internal Radiation Therapy—SIRT)
4 Systemic Therapy
References
Part VII: Pediatrics
Dedicated CT Protocols for Children
1 Introduction
2 General Optimization of Scanning Protocols
2.1 Prescanning Preparation
2.2 Implementation of Actual Technologic Developments
2.2.1 Automated Tube Current Modulation and Tube Voltage Adjustment
2.2.2 Iterative Reconstruction Techniques
2.2.3 New Scan Modes
2.2.3.1 High-Pitch Scanning
2.2.3.2 Volume Scanning
2.2.3.3 Organ-Based Tube Current Modulation
2.3 Contrast Media
3 Typical Indications for Pediatric CT and Sample Protocols
3.1 Head
3.2 Sinuses and Neck
3.3 Chest
3.4 Abdomen and Pelvis
3.5 Whole-Body
References
Congenital Heart Disease in Children
1 Introduction
2 Technical Tips and Tricks
2.1 Coronary Anomalies
2.2 Tetralogy of Fallot
2.3 Transposition of the Great Arteries
2.4 Aortic Coarctation
2.5 Anomalous Pulmonary Venous Return
2.6 Fontan
References
Chest CT Imaging in Children
1 Tracheobronchial Anomalies
1.1 Tracheobronchial Branching Anomalies
1.2 Congenital Tracheal Stenosis
1.3 Vascular Congenital Large Airway Disorders
1.3.1 Innominate Artery Compression Syndrome
1.3.2 Double Aortic Arch
1.3.3 Pulmonary Artery Sling
1.4 Bronchial Atresia
1.5 Bronchogenic Cyst
2 Lung Anomalies
2.1 Pulmonary Underdevelopment
2.1.1 Pulmonary Agenesis
2.1.2 Pulmonary Hypoplasia
2.2 Congenital Lobar Emphysema
2.3 Congenital Pulmonary Airway Malformation (CPAM)
2.4 Pulmonary Sequestration
3 Childhood Interstitial Lung Diseases (ChILD)
3.1 Diffuse Developmental Disorders
3.2 Growth Abnormalities
3.2.1 Bronchopulmonary Dysplasia
3.2.2 Filamin A Mutations
3.2.3 Pulmonary Hypoplasia
3.3 Surfactant Disorders and Pulmonary Alveolar Proteinosis
3.4 Specific Conditions of Unknown or Poorly Understood Pathology
3.4.1 Neuroendocrine Cell Hyperplasia of Infancy
3.4.2 Pulmonary Interstitial Glycogenosis (PIG)
References
CT of the Pediatric Abdomen
1 Introduction
2 Technical Considerations
3 Indications, Imaging Findings, and Examples
4 Summary
References
Part VIII: Miscellaneous Topics
Emergency CT
1 Introduction
2 Trauma Imaging Technique
2.1 Use of Intravenous Contrast
2.2 Use of Oral Contrast
2.3 Radiation Dose
2.4 Total-Body Protocol
2.5 Role of Computed Tomography Angiography (CTA)
2.6 Total-Body CT Versus Standard Workup
2.7 Split-Bolus Single-Pass CT
2.8 Dual-Energy CT (DECT)
3 Role of Other Modalities
3.1 Conventional Radiographs
3.2 Ultrasound (US)
3.3 MRI
4 Neurological Trauma
4.1 Skull Fracture
4.2 Temporal Bone Fracture
4.3 Facial Fractures
4.3.1 Nasal Bone
4.3.2 Le Fort Fractures
4.3.3 Zygomaticomaxillary Fracture Complex
4.3.4 Orbital Blowout Fracture
4.3.5 Management
4.4 Extra-axial Hemorrhage
4.4.1 Epidural Hemorrhage (EDH)
4.4.2 Subdural Hemorrhage (SDH)
4.4.3 Subarachnoid Hemorrhage (SAH)
4.5 Intra-axial Injuries
4.5.1 Cerebral Contusion
4.5.2 Intraparenchymal Hematoma (IPH)
4.5.3 Diffuse Axonal Injury (DAI)
4.6 Secondary Traumatic Injury
4.6.1 Cerebral Edema
4.6.2 Cerebral Herniation
4.7 Cervical Trauma
4.7.1 Vascular Trauma
4.7.2 Vertebral Trauma
5 Chest Trauma
5.1 Pneumothorax (PTX)
5.1.1 Tension Pneumothorax (TPTX)
5.1.2 Occult Pneumothorax
5.2 Hemothorax
5.3 Pulmonary Contusion
5.3.1 Occult Pulmonary Contusion
5.4 Traumatic Aortic Injury
6 Body Trauma
6.1 Spleen
6.2 Liver
6.3 Pancreas
7 Genitourinary Trauma
7.1 Renal
7.2 Bladder
References
Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery
1 Introduction
2 Trauma
2.1 Polytrauma
2.1.1 Acquisition Technique
2.1.2 Clinical Applications
2.2 Acute Trauma (Preoperative Assessment)
2.2.1 Acquisition Technique
2.2.2 Clinical Application
2.3 Subacute or Chronic Injuries
2.3.1 Acquisition Technique
2.3.2 Clinical Application
3 Tumors
3.1 Acquisition Technique
3.2 Clinical Applications
4 Postsurgery
4.1 Acquisition Techniques
4.1.1 Metal Artifact Reduction
Conventional Technique
Iterative Reconstruction
Monochromatic Reconstruction of Dual-Energy CT Scans
Metal Artifact Reduction Algorithms
4.1.2 Intravenous Injection of Contrast Medium and Subtraction Techniques
4.1.3 Visualization and Post-processing
4.2 Clinical Applications
4.2.1 Bone Healing
4.2.2 Prosthesis Complications
4.2.3 Follow-Up of Limb Salvage Surgery or Pelvic Reconstruction
Bibliography
Incidental Findings in Multislice CT of the Body
1 Introduction
2 Definition and Misunderstandings Regarding Incidental Findings
3 Frequency and Spectrum of Incidental Findings on Multislice CT of the Abdomen
3.1 Incidental Renal Tumors and Cysts
3.1.1 Solid Renal Tumors
3.1.2 Simple Cysts
3.1.3 Complex Cysts
3.2 Incidental Adrenal Lesions
3.2.1 Management of Adrenal Lesions
Patients Without Known Extra-Adrenal Malignancy
Patients with a History of Extra-Adrenal Malignancy
Young Patients with Adrenal Incidentaloma
3.3 Incidental Liver Lesions
3.3.1 Approach to an Incidental Liver Mass Detected on CT
3.3.2 Steatosis
3.4 Incidental Lesions of the Gallbladder and Biliary Tree
3.5 Incidental Lymphadenopathy
3.6 Incidental Pancreatic Lesions
3.6.1 Solid Tumors
3.6.2 Cystic Lesions
3.7 Incidental Abdominal Vascular Findings
3.8 Incidental Adnexal and Uterine Lesions
3.8.1 Adnexal Cysts
3.8.2 Uterine Lesions
4 Frequency and Spectrum of Incidental Findings on Multislice CT of the Chest
4.1 Incidental Pulmonary Nodules
4.2 Incidental Thoracic Vascular Calcifications
4.3 Incidental Thyroid Lesions
5 How Extensively Should Incidental Findings Be Searched for on Multislice CT?
6 Technical Factors Limiting Detection and Characterization of Incidental CT Findings
7 Reporting of Incidental Findings
7.1 Incidental Findings Are Not Always Reported
7.2 Reasons for Reporting or Not Reporting Incidental CT Findings
8 The Problem of False-Positive Findings, Overdiagnosis, and Indolent Cancers
9 Do the Patients Want to Know About Incidental Findings?
9.1 Who Should Decide Which Information to Convey to the Referring Physician and to the Patient?
9.2 Potential Impact of e-Medicine
References
Correction to: Chest Neoplasias
Correction to: Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques
Correction to: Chapter “Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques” in: Med Radiol Diagn Imaging (2018), https://doi.org/10.1007/174_2018_183

Citation preview

Medical Radiology · Diagnostic Imaging Series Editors: H.-U. Kauczor · P. M. Parizel · W. C. G. Peh

Konstantin Nikolaou · Fabian Bamberg  Andrea Laghi · Geoffrey D. Rubin Editors

Multislice CT Fourth Edition

Medical Radiology Diagnostic Imaging Series Editors Hans-Ulrich Kauczor Paul M. Parizel Wilfred C. G. Peh

For further volumes: http://www.springer.com/series/4354

Konstantin Nikolaou • Fabian Bamberg Andrea Laghi • Geoffrey D. Rubin Editors

Multislice CT Fourth Edition

Editors Konstantin Nikolaou Department of Radiology University Hospitals Tübingen Tübingen, Baden-Württemberg Germany Andrea Laghi Department of Surgical and Medical Sciences and Translational Medicine “Sapienza” – University of Rome Rome Italy

Fabian Bamberg Department of Diagnostic and Interventional Radiology University of Freiburg Freiburg Germany Geoffrey D. Rubin Department of Radiology Duke University School of Medicine Durham, NC USA

ISSN 0942-5373           ISSN 2197-4187 (electronic) Medical Radiology ISBN 978-3-319-42585-6    ISBN 978-3-319-42586-3 (eBook) https://doi.org/10.1007/978-3-319-42586-3 Library of Congress Control Number: 2018966548 © Springer International Publishing AG, part of Springer Nature 2001, 2004, 2009, 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

In Memoriam

This book is dedicated to our friend, colleague, and brilliant radiologist, who left us much too early. Gary Glazer was a CT pioneer, contributing some of our earliest understandings to the value of CT in the assessment of neoplasia, particularly metastatic involvement of mediastinal lymph nodes. He is, however, best remembered and revered for his leadership in building groundbreaking research programs rooted in a culture of collegiality and teamwork that was ahead of its time, serving as a model for innovation in a world of academia that is often hindered by silos. His passion for international collaboration in education and research was visionary and provided a basis for many meaningful collaborations and lasting friendships, including those that bind us and our diverse contributors from around the world. The editors

v

Contents

Part I Techniques Multi-slice CT: Current Technology and Future Developments . . 3 Stefan Ulzheimer, Malte Bongers, and Thomas Flohr Radiation Dose Optimization in CT. . . . . . . . . . . . . . . . . . . . . . . . . 35 Shaunagh McDermott, Alexi Otrakji, and Mannudeep K. Kalra Spectral CT/Dual-Energy CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Anushri Parakh, Manuel Patino, and Dushyant V. Sahani Contrast Enhancement at CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Kazuo Awai, Toru Higaki, and Fuminari Tatsugami Image Processing from 2D to 3D. . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Steven P. Rowe and Elliot K. Fishman Perfusion CT: Technical Aspects. . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Vicky Goh and Davide Prezzi Part II Neuro/ENT Stroke/Cerebral Perfusion CT: Technique and Clinical Applications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 Adrienne Moraff, Jeremy Heit, and Max Wintermark CT Diagnostics in Brain Tumors. . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Marco Essig MDCT in Neurovascular Imaging. . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Giovanna Negrao de Figueiredo and Birgit Ertl-Wagner Anatomy and Pathology of the Temporal Bone. . . . . . . . . . . . . . . . 207 Sabrina Kösling Dental CT: Pathologic Findings in the Teeth and Jaws. . . . . . . . . . 217 Wolfgang Wuest Anatomy and Corresponding Oncological Imaging of the Head and Neck. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Christian Czerny and Juergen Lutz

vii

viii

Part III Chest Interstitial Lung Diseases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 Christina Mueller-Mang, Helmut Ringl, and Christian Herold Pneumonia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289 Sabine Dettmer and Jens Vogel-Claussen CT of the Airways. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 Michael Trojan, Hans-Ulrich Kauczor, Claus Peter Heußel, and Mark Oliver Wielpütz Lung Cancer Screening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331 Oyunbileg von Stackelberg and Hans-Ulrich Kauczor Chest Neoplasias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345 Tina D. Tailor CT of Pulmonary Embolism: Imaging Update . . . . . . . . . . . . . . . . 395 Antoine Hutt, Paul Felloni, Jacques Remy, and Martine Remy-Jardin COPD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409 Anna Rita Larici, Paola Franchi, Giuseppe Cicchetti, and Lorenzo Bonomo Part IV Abdomen Focal Lesions in Non-cirrhotic Liver. . . . . . . . . . . . . . . . . . . . . . . . . 433 Christoph J. Zech Cirrhotic Liver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447 Keitaro Sofue, Masakatsu Tsurusaki, and Takamichi Murakami Pancreatic Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491 Jeong Min Lee and Hyo-Jin Kang Acute and Chronic Pancreatitis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 527 G. Zamboni, M. Chincarini, R. Negrelli, and R. Pozzi Mucelli Spleen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555 Andre Euler and Sebastian T. Schindera Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 579 Ahmed Ba-Ssalamah, Sarah Poetter-Lang, Nina Bastati, Jacqueline C. Hodge, Helmut Ringl, and Richard M. Gore Small Bowel MDCT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 619 Marco Rengo, Simona Picchia, and Andrea Laghi Imaging of Large Bowel with Multidetector Row CT. . . . . . . . . . . 641 Jay D. Patel, Heather I. Gale, and Kevin J. Chang

Contents

Contents

ix

Peritoneal Surface Malignancy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 667 Davide Bellini, Paolo Sammartino, and Andrea Laghi Multislice PET/CT in Neuroendocrine Tumors. . . . . . . . . . . . . . . . 675 Gabriele Pöpperl and Clemens Cyran Adrenals. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 691 Christoph Schabel and Daniele Marin Kidneys, Ureters, and Bladder. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 697 Christoph Schabel and Daniele Marin Part V Cardiovascular Technical Innovations and Concepts in Coronary CT . . . . . . . . . . 713 Nils Vogler, Mathias Meyer, and Thomas Henzler Noninvasive Coronary Artery Imaging. . . . . . . . . . . . . . . . . . . . . . . 729 Manoj Mannil and Hatem Alkadhi Pre- and Postinterventional/Surgical Evaluation by CT. . . . . . . . . 743 Harald Seifarth and David Maintz Computed Tomography in the Management of Electrophysiology Procedures. . . . . . . . . . . . . . . . . . . . . . . . . . . . 755 Joseph Negusei, Ian R. Drexler, Jim Cheung, and Quynh A. Truong Functional Cardiac CT Angiography. . . . . . . . . . . . . . . . . . . . . . . . 777 Domenico De Santis, Marwen Eid, Taylor M. Duguay, U. Joseph Schoepf, and Carlo N. De Cecco CT Angiography of the Peripheral Arteries. . . . . . . . . . . . . . . . . . . 805 Newton B. Neidert, Nikkole M. Weber, Jeffrey C. Hellinger, and Eric E. Williamson Acute Aortic Syndromes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 825 Christian Loewe CT Venography. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 855 Simer Grewal, Patrick Sutphin, and Sanjeeva P. Kalva Aortic Aneurysm and Stent Graft Assessment. . . . . . . . . . . . . . . . . 869 Ilya Livshits, Safet Lekperic, and Robert Lookstein Part VI Interventions CT-Guided Biopsy and Drainage. . . . . . . . . . . . . . . . . . . . . . . . . . . . 893 Giovanna Negrão de Figueiredo and Christoph G. Trumm CT-Guided Spinal Interventions: Vertebroplasty/Kyphoplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 925 Tobias F. Jakobs and Stefanie C. Surwald

x

CT-Guided Tumor Ablation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 945 Ralf-Thorsten Hoffmann Functional CT for Image-Guided Personalized Tumor Interventions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 957 Horger Marius Part VII Pediatrics Dedicated CT Protocols for Children. . . . . . . . . . . . . . . . . . . . . . . . 969 Ilias Tsiflikas Congenital Heart Disease in Children. . . . . . . . . . . . . . . . . . . . . . . . 987 Aurelio Secinaro and Davide Curione Chest CT Imaging in Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1011 Sebastian Ley and Julia Ley-Zaporozhan CT of the Pediatric Abdomen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1037 Michael Riccabona and Alexander Pilhatsch Part VIII Miscellaneous Topics Emergency CT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1051 Samad Shah, Sunil Jeph, and Savvas Nicolaou Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery. . . . . . . . . . . . . . . . . . . . . . . . . 1079 Pedro Augusto Gondim Teixeira and Alain Blum Incidental Findings in Multislice CT of the Body . . . . . . . . . . . . . . 1107 Mikael Hellström Correction to: Chest Neoplasias . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1139 Tina D. Tailor Correction to: Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques. . . . . . . . . . . . . . . . . . . . . . . . . . 1141 Ahmed Ba-Ssalamah, Sarah Poetter-Lang, Nina Bastati, Jacqueline C. Hodge, Helmut Ringl, and Richard M. Gore

Contents

Part I Techniques

Multi-slice CT: Current Technology and Future Developments Stefan Ulzheimer, Malte Bongers, and Thomas Flohr

Contents

Abstract

1    Introduction

 3

2    System Design 2.1  Gantry 2.2  X-Ray Tube and Generator 2.3  MDCT Detector Design and Slice Collimation 2.4  Dual Source CT

 6  6  7  8  10

3    Measurement Techniques 3.1  MDCT Sequential (Axial) Scanning 3.2  MDCT Spiral (Helical) Scanning 3.3  ECG-Synchronized Cardiovascular CT 3.4  Dual Energy Computed Tomography

 11  11  12  17  21

4    Radiation Dose Reduction 4.1  Anatomical Tube Current Modulation 4.2  Adaptation of the X-Ray Tube Voltage 4.3  Spectral Shaping 4.4  Iterative Reconstruction

 23  23  24  25  26

5    Future Developments

 27

References

 30

S. Ulzheimer Siemens Healthcare GmbH, Computed Tomography, Forchheim, Germany e-mail: [email protected] M. Bongers (*) Institute for Diagnostic and Interventional Radiology, Eberhard-Karls-University, Tübingen, Germany e-mail: [email protected] T. Flohr Siemens Healthcare GmbH, Computed Tomography, Forchheim, Germany Institute for Diagnostic and Interventional Radiology, Eberhard-Karls-University, Tübingen, Germany e-mail: [email protected]

Since its introduction in the early seventies of the past century, computed tomography (CT) has undergone tremendous improvements in terms of technology, performance, and clinical applications. Based on the historic evolution of CT and basic CT physics this chapter describes the status quo of the technology and tries to anticipate future developments. Besides the description of key components of CT systems, a special focus is laid on breakthrough developments such as multi-slice CT and dedicated scan modes for cardiac imaging.

1

Introduction

In 1972, the English engineer G.N.  Hounsfield built the first commercial medical X-ray computed tomography (CT) scanner for the company EMI Ltd. as a pure head scanner with an X-ray tube and two detector elements moving incrementally around the patient. It was able to acquire twelve slices with 13  mm slice thickness each and reconstruct the images with a matrix of 80 × 80 pixels (Fig. 1a) in approximately 35 min. Even though the performance of CT scanners increased dramatically over time there were no principally new developments in conventional CT until 1989. By then the acquisition time for one image decreased from 300 s in 1972 to 1–2 s, thin slices of down to 1 mm became possible and the in-plane resolution increased from 3 line pairs

Med Radiol Diagn Imaging (2018) https://doi.org/10.1007/174_2018_187, © Springer International Publishing AG Published Online: 07 July 2018

3

4

S. Ulzheimer et al.

Fig. 1  Development of computed tomography over time. sectional slices in the year 2007 (Image courtesy of Mayo (a) Cross-sectional image of a brain in the year 1971 and Clinic Rochester) (b) the whole brain with sagittal, coronal, and cross-­

per cm (lp/cm) to 15  lp/cm with typically 5122 matrices. As acquisition times of mechanical CT scanners were expected to be far too long for high quality cardiac imaging for the next years or even decades to come, a completely new technical concept for a CT scanner without moving parts for extremely fast data acquisition within 50 ms was suggested and promoted as cardiovascular CT (CVCT) scanner. These scanners were also called “Ultrafast CT” scanners or “Electron Beam CT” (EBT or EBCT) scanners. High cost and limited image quality combined with low volume coverage prevented the wide propagation of the modality, and the production and distribution of these scanners were soon discontinued. Based on the introduction of slip ring technology to get power to and data off the rotating gantry, continuous rotation of the X-ray tube and the detector became possible. The ability of continuous rotation led to the development of spiral CT scanners in the early nineties (Kalender et  al. 1990; Crawford and King 1990). Volume data could be acquired without the danger of mis- or double-registration of anatomical details. Images could be reconstructed at any position along the patient axis (through-plane axis, z-axis), and

overlapping image reconstruction could be used to improve through-plane resolution. Volume data became the very basis for applications such as CT angiography (CTA) (Rubin et  al. 1995), which has revolutionized noninvasive assessment of vascular disease. The ability to acquire volume data was the prerequisite for the development of three-dimensional image processing techniques such as multiplanar reformations (MPR), maximum intensity projections (MIP), surface shaded displays (SSD), or volume rendering techniques (VRT), which have become a vital component of medical imaging today. Main drawbacks of single-slice spiral CT are either insufficient volume coverage within one breath-hold time of the patient or missing spatial resolution in z-axis due to wide collimation. With single-slice spiral CT the ideal isotropic resolution, i.e., of equal resolution in all three spatial axes, can only be achieved for very limited scan ranges (Kalender 1995). Larger volume coverage in shorter scan times and improved through-plane resolution became feasible after the broad introduction of 4-slice CT systems by all major CT manufacturers in 1998 (Klingenbeck-Regn et al. 1999; Mccollough and Zink 1999; Hu et al. 2000). The increased perfor-

Multi-slice CT: Current Technology and Future Developments

mance allowed for the optimization of a variety of clinically relevant scan parameters. Examination times for standard protocols could be significantly reduced; alternatively, scan ranges could be significantly extended. Furthermore, a given anatomic volume could be scanned within a given scan time with substantially reduced slice width. This way, for many clinical applications the goal of isotropic resolution was within reach with 4-slice CT systems. Multi-detector row CT (MDCT) also dramatically expanded into areas previously considered beyond the scope of third-generation CT scanners based on the mechanical rotation of X-ray tube and detector, such as cardiac imaging with the addition of ECG gating capability enabled by gantry rotation times down to 0.5  s (Ohnesorge et al. 2000; Kachelriess et al. 2000). Despite all these promising advances, clinical challenges and limitations remained for 4-slice CT systems. True isotropic had not yet been achieved for many routine applications requiring extended scan ranges, since wider collimated slices (4  ×  2.5  mm or 4 × 3.75 mm) had to be chosen to complete the scan within a breath-hold time of the patient. For ECG-gated coronary CTA, stents or severely calcified arteries constituted a diagnostic dilemma, mainly due to partial volume artifacts as a consequence of insufficient through-­ plane resolution (Nieman et  al. 2001), and reliable imaging of patients with higher heart rates was not possible due to limited temporal resolution. As a next step, the introduction of an 8-slice CT system in 2000 enabled shorter scan times, but did not yet provide improved through-plane resolution (thinnest collimation 8  ×  1.25  mm). This was achieved with the introduction of 16-slice CT (Flohr et al. 2002a, b), which made it possible to routinely acquire substantial anatomic volumes with isotropic submillimeter spatial resolution. ECG-gated cardiac scanning was enhanced by both improved temporal resolution achieved by gantry rotation time down to 0.375 s and improved spatial resolution (Nieman et  al. 2002; Ropers et al. 2003). In 2004, all major CT manufacturers introduced MDCT systems with simultaneous acquisition of 64-slices at 0.5, 0.6 mm, or 0.625 mm

5

collimated slice width, and further reduced rotation times (down to 0.33  s). GE, Philips, and Toshiba aimed at an increase in volume coverage speed by using detectors with 64 rows instead of 16, thus providing 32–40  mm z-coverage. Siemens used 32 physical detector rows in combination with double z-sampling, a refined z-sampling technique enabled by a z-flying focal spot (see Sect. 3.2.4), to simultaneously acquire 64 overlapping 0.6  mm slices with the goal of pitch-independent increase of through-plane resolution and reduction of spiral artifacts (Flohr et al. 2004, 2005). With 64-slice CT systems, CT scans with submillimeter resolution became feasible even for extended anatomical ranges. The improved temporal resolution due to faster gantry rotation increased clinical robustness of ECG-­ gated scanning, thereby facilitating the successful integration of CT coronary angiography into routine clinical algorithms (Leber et  al. 2005; Leschka et al. 2005), although higher and irregular heart rates were still problematic. In 2007, one vendor introduced a MDCT system with 128 simultaneously acquired slices, based on a 64-row detector with 0.6  mm collimated slice width (38.4 mm z-axis coverage) and double z-sampling by means of a z-flying focal spot. Later, simultaneous acquisition of 256 slices became available with a CT system equipped with a 128-row detector (0.625  mm collimated slice width, 80  mm z-axis coverage) and double z-sampling. Clinical experience with 64-, 128-, or 256-­ slice CT indicated that adding even more detector rows would not by itself translate into increased clinical benefit. Instead, new CT concepts were introduced to solve remaining limitations of MDCT. One remaining challenge for MDCT is the visualization of dynamic processes in extended anatomical ranges, e.g., to characterize the inflow and outflow of contrast agent in the arterial and venous system in dynamic CTAs, or to determine the enhancement characteristics of the contrast agent in volume perfusion studies. Dynamic CT examinations are enabled by CT systems with area detectors large enough to cover entire organs, such as the heart, the kidneys, or the

S. Ulzheimer et al.

6

brain, in one axial scan (requiring 120 mm volume coverage or more). Another way to acquire dynamic volume data is the introduction of “shuttle modes” with periodic table movement between two z-positions (e.g., Goetti et al. 2010). In 2007, a CT scanner with 16 cm z-axis coverage at isocenter was introduced which has the potential to provide dynamic volume data without table movement. Currently, two CT scanners following this design principle are commercially available by different vendors, one with 320 × 0.5 mm collimation and 0.27 s gantry rotation time, and the other with 256 × 0.625 mm collimation and 0.28  s gantry rotation time. CT systems with 16  cm detector coverage can also scan the entire heart in one axial scan, thereby avoiding the typical stair-step artifacts in cardiac CT images which are a potential problem in CT systems with smaller detectors (Rybicki et  al. 2008; Steigner et al. 2009; Dewey et al. 2009). Motion artifacts due to insufficient temporal resolution remain an important challenge for cardiothoracic imaging and coronary CTA even with the latest generation of MDCT. Image temporal resolution of less than 100 ms at all heart rates can be achieved with dual source CT (DSCT) systems, i.e., CT scanners with two X-ray tubes and two corresponding detectors offset by 90° (Flohr et  al. 2006). Meanwhile, three generations of DSCT systems have been commercially introduced, and clinical studies have demonstrated the potential of DSCT to reliably perform coronary CTA with little or no dependence on the patient’s heart rate (Scheffel et al. 2006; Matt et al. 2007; Ropers et al. 2007; Weustink et al. 2009). DSCT scanners also show promising properties for general radiology applications. Both X-ray tubes can be operated simultaneously in standard acquisitions to provide high power reserves when necessary. Additionally, both X-ray tubes can be operated at different kV settings and/or different pre-filtrations, enabling dual energy CT.  Since the introduction of DSCT, tissue characterization, calcium quantification, and quantification of the local blood volume in contrast-enhanced scans have been investigated as potential dual energy applications (Johnson et al. 2007).

2

System Design

The overall performance of a MDCT system depends on several key components. These components include the gantry, X-ray source, a high-­ powered generator, detector and detector electronics, data transmission systems (slip-­ rings), and the computer system for image reconstruction and manipulation.

2.1

Gantry

Third-generation CT scanners employ the so-­called “rotate/rotate” geometry, in which both X-ray tube and detector are mounted onto a rotating gantry and rotate around the patient (Fig. 2). A CT detector has 700 or more individual detector elements in the fan angle direction (see Fig. 2) which cover a scan

Fig. 2  Basic system components of a modern “third-­ generation” CT system. First-generation systems used a collimated pencil beam and therefore required a translation of the pencil beam and the single detector element before each rotational step to scan the whole object. Second-generation scanner used a small fan beam but still required translational and rotational movement patterns of the X-ray source and the small detector array. The fan beam of third-generation scanners covers a SFOV of typically 50  cm diameter and allows for a pure rotational motion of the tube and the detector around the patient. This was the key to reduce scan times per image from minutes to less than a second. All medical CT scanners today are third-generation scanners

Multi-slice CT: Current Technology and Future Developments

field of view (SFOV) of usually 50 cm diameter. In a MDCT system, the detector comprises several detector rows in the z-axis direction (patient direction). The X-ray attenuation of the object is measured by the individual detector elements. All measurement values acquired at the same angular position of the measurement system form a “projection” or “view.” About 1000 projections are measured during each 360° rotation. Key requirement for the mechanical design of the gantry is the stability of both focal spot and detector position during rotation, in particular with regard to the rapidly increasing rotational speeds of modern CT systems (from 0.75  s in 1994 to 0.25 s in 2017). Hence, the gantry as well as all the

7

components on it, such as X-ray tube, tube collimator, and data measurement system (DMS), have to be designed to withstand the high gravitational forces associated with fast gantry rotation.

2.2

X-Ray Tube and Generator

State-of-the-art X-ray tube/generator combinations provide a peak power of 60–120 kW, usually at various, user-selectable voltages, e.g., 70–140 kV in steps of 10 kV. In a conventional tube design, an anode plate of typically 160– 220  mm diameter rotates in a vacuum housing (Fig.  3). The heat storage capacity of anode

Anode Cooling oil

X-rays

Anode Deflection unit

e-beam Cathode

X-rays

Fig. 3  Schematic drawings and pictures of a conventional X-ray tube (top) and a rotating envelope tube (bottom). The electrons emitted by the cathode are represented by green lines, the X-rays generated in the anode are depicted as purple arrows. In a conventional X-ray tube the anode plate rotates in a vacuum housing. Heat is mainly dissipated via thermal radiation. In a rotating

envelope tube, the anode plate constitutes an outer wall of the tube housing and is in direct contact with the cooling oil. Heat is more efficiently dissipated via thermal conduction, and the cooling rate is significantly increased. Rotating envelope tubes have no moving parts and no bearings in the vacuum. (Images not to scale)

S. Ulzheimer et al.

8

plate and tube housing—measured in Mega Heat Units (MHU)—determines the performance level: the bigger the anode plate is, the larger is the heat storage capacity, and the more scan-seconds can be delivered until the anode plate reaches its temperature limit. An alternative design is the rotating envelope tube (Schardt et al. 2004). The anode plate constitutes an outer wall of the rotating tube housing; it is therefore in direct contact with the cooling oil and can be efficiently cooled via thermal conduction (Fig. 3). This way, a very high heat dissipation rate and fast anode cooling is achieved, enabling high power scans in rapid succession. Due to the central rotating cathode, permanent electromagnetic deflection of the electron beam is needed to position and shape the focal spot on the anode. The electromagnetic deflection is also used for the double z-sampling technology of high-end scanners of one vendor (Flohr et  al. 2004, 2005). Different clinical applications require different X-ray spectra and hence different kV settings for optimum image quality and/or best possible signal-to-noise ratio at lowest radiation dose. Contrast-enhanced CT scans using iodinated contrast agent, in particular CT angiographic examinations, benefit from low kV settings. Because of the increased iodine contrast at lower kV, the contrast-to-noise ratio (CNR) in contrast-­ enhanced images increases at low kV if the radiation dose is kept constant (McCollough et  al. 2009). Vice versa, lower radiation dose is sufficient at low kV to maintain a desired CNR (Schaller et al. 2001a; Wintersperger et al. 2005; McCollough et al. 2009). Because of limitations of the X-ray tube current low kV protocols have so far been limited to small patients and children. Recent progress in X-ray tube design has led to the introduction of X-ray tubes capable of providing high power reserves at 70, 80 and 90 kV. They have the potential to enable contrast-­ enhanced low kV scans in adult and in obese patients without compromising CNR.  In coronary CTA, as an example, a radiation dose reduction by 49–68% even in obese patients has been reported when using 70 and 80  kV protocols (Meinel et al. 2014).

2.3

 DCT Detector Design M and Slice Collimation

All commercially available MDCT systems to-­ date are equipped with solid-state scintillation detectors. Each detector element consists of a radiation-sensitive solid-state material (such as cadmium tungstate, gadolinium-oxide, or gadolinium oxi-sulfide) with suitable dopings, which converts the absorbed X-rays into visible light. The light is then detected by a Si photodiode attached to the backside of the scintillator. The resulting electrical current is amplified and converted into a digital signal. Key requirements for a suitable detector material are good detection efficiency, i.e., high atomic number, and very short afterglow time to enable the high gantry rotation speeds that are essential for the imaging of moving organs and for fast volume coverage. CT detectors must provide different slice widths to adjust the scan speed, through-plane resolution, and image noise for each application. With a single-slice CT detector, different collimated slice widths are adjusted by pre-patient collimation of the X-ray beam.1 For a 2-slice CT detector, different slice widths can be obtained by pre-patient collimation if the detector is separated midway along the z-width of the X-ray beam. To simultaneously acquire more than 2 slices at different slice widths, detectors with a larger number of detector rows than finally read-out slices have to be used. The required total beam width in the z-direction is adjusted by prepatient collimation, and the signals of every two (or more) detectors along the z-axis are electronically combined to thicker slices. The detector of a 16-slice CT (Siemens SOMATOM Emotion 16) as an example consists of 16 central rows, each with 0.6  mm collimated slice width at isocenter, and 4 outer rows on either side, each with 1.2 mm collimated slice width— in total, 24 rows with a z-width of 19.2 mm at isocenter (Fig. 4). In a 16 × 0.6 mm acquisition Note that the slice width is always measured at the isocenter of the CT system. 1 

Multi-slice CT: Current Technology and Future Developments

9

Fig. 4  Principle of MDCT slice collimation. Example of a 16-slice detector, which consists of 24 detector rows and provides either 16 collimated 0.6 mm slices (top) or—by

combination of the signals of every 2 central rows—16 collimated 1.2 mm slices (bottom)

mode the X-ray beam width is adjusted such that only the central 16 rows are irradiated which are read-out individually (Fig. 4, top). To obtain 16 collimated 1.2  mm slices, the prepatient collimator is opened. The full z-width of the detector is irradiated, and the signals of every 2 central rows are electronically combined. This results in 8 central 1.2  mm slices plus 4 outer 1.2 mm slices on either side of the

detector, in total 16 collimated 1.2  mm slices (Fig 5, bottom). The 16-slice detectors of other vendors are similarly designed, providing, e.g., 16 collimated 0.625 mm slices or 16 collimated 1.25 mm slices. MDCT detectors with 64 detector rows provide 64 collimated 0.5, 0.6  mm, or 0.625  mm slices, depending on the manufacturer. They allow acquisition of thicker slices by electronic

S. Ulzheimer et al.

10

90°

33 cm

Detector A

26 cm

Detector A

Detector B

95°

Detector B

Fig. 5  Dual source CT scanner with two independent measurement systems. A scanner of this type provides temporal resolution equivalent to a quarter of the gantry rotation time, independent of the patient’s heart rate. First generation (center). The measurement systems are at an

angle of 90°. One detector (a) covers the entire SFOV with a diameter of 50 cm, while the other detector (b) is restricted to a smaller, central field of view. Second generation (right). To enlarge the SFOV of detector B, the system angle was increased to 95°

combination of every two detector rows. This results in 32 collimated 1.0, 1.2 mm, or 1.25 mm slices. One CT system has a detector with 128 collimated 0.625  mm slices (z-width 80  mm at isocenter). Meanwhile, CT systems with 80 or 160 collimated 0.5  mm slices have been introduced. Third-generation DSCT systems provide 96 collimated 0.6 mm slices with both detectors (z-width 57.6  mm). The widest commercially available CT detectors cover 16 cm at isocenter, either by acquiring 320 collimated 0.5 mm slices or 256 collimated 0.625 mm slices, depending on the vendor.

0.6 mm slices per rotation. The shortest gantry rotation time is 0.33 s. Meanwhile, newer generations of DSCT systems have been introduced, which are equipped with 128- and 196-slice detectors, respectively, and provide gantry rotation times down to 0.25 s. To enlarge the SFOV of detector B, the system angle was increased to 95°. One key benefit of DSCT is improved temporal resolution for the examination of moving organs, such as the heart, the lung, and the thoracic vessels. The shortest data interval needed for image reconstruction at the isocenter is half a rotation of scan data—a so-called half scan sinogram. In a DSCT scanner, the halfscan sinogram can be split up into two quarter scan sinograms which are simultaneously acquired by the two measurement systems in the same relative phase of the patient’s cardiac cycle and at the same anatomical level due to the 90° angle between both detectors. With this approach, constant temporal resolution equivalent to a quarter of the gantry rotation time trot/4 is achieved in a centered region of the SFOV— 83 ms for the first-generation DSCT, 75 ms for the second-generation DSCT, and 66  ms for the third-generation DSCT, independent of the patient’s heart rate. DSCT is sufficiently accurate to diagnose coronary artery disease in patients with high and even irregular heart rates, and in difficult-to-image patients (e.g., Sun et  al. 2011; Lee et  al. 2012; Paul et  al. 2013; Westwood et al. 2013). The good tempo-

2.4

Dual Source CT

A dual source CT (DSCT) is a CT system with two X-ray tubes and two detectors, see Fig.  5. Both measurement systems operate simultaneously and acquire CT scan data at the same anatomical level of the patient (same z-position). In 2005, the first DSCT system was commercially introduced (Flohr et  al. 2006). The two acquisition systems are mounted onto the rotating gantry with an angular offset of 90°. One detector (A) covers the entire SFOV with a diameter of 50 cm, while the other detector (B) is restricted to a smaller, central field of view because of space limitations on the gantry. Figure  5 illustrates the principle. Using the z-­flying focal spot technique (Flohr et al. 2004, 2005), each detector acquires 64 overlapping

Multi-slice CT: Current Technology and Future Developments

ral resolution is also beneficial to reduce motion artifacts in cardiothoracic studies (e.g., Hutt et al. 2016). In addition to improving temporal resolution, the dual source principle can be exploited favorably in other clinical situations. If both X-ray tubes are simultaneously operated at the same tube potential (kV), up to 240  kW peak power is available with the third-generation DSCT. These power reserves are not only beneficial for the examination of morbidly obese patients, whose number is dramatically growing in western societies, but also to maintain adequate X-ray photon flux when high volume coverage speed is needed. Additionally, both X-ray tubes can be operated at different kV and mA settings, enabling the acquisition of dual energy data. While dual energy CT was already evaluated 30  years ago (Kalender et  al. 1986; Vetter et al. 1986), technical limitations of the CT scanners at those times prevented the development of routine clinical applications. On the DSCT system dual energy data can be acquired nearly simultaneously with sub-second gantry rotation times and fast volume coverage. The use of dual energy CT data can in principle add functional information to the morphological information based on X-ray attenuation coefficients that is usually obtained in a CT examination. Meanwhile, a variety of different applications of dual energy CT scans have been evaluated, with some of them on their way to clinical routine (see the reviews in Lu et  al. 2012; Remy-Jardin et  al. 2014; Marin et  al. 2014; Agrawal et al. 2014; Patino et al. 2016). By simultaneously operating both measurement systems in a spiral scan mode, DSCT systems provide very high scan speeds up to 737 mm/s, see the clinical example in Fig. 6. As a drawback of DSCT systems, the SFOV of the second detector cross-scattered radiation, i.e., scattered radiation originating from tube A and detected by detector B (at an angle of 90°) and vice versa, has to be carefully corrected for to avoid distortions of CT numbers by cupping or streaking artifacts. This can be done either by measurement of cross-scattered radiation or by model-based approaches (Petersilka et al. 2010).

11

Fig. 6  Case study illustrating the clinical performance of third-generation DSCT.  ECG-triggered high-pitch spiral scan of the aorta and the iliac arteries in a patient with aortic dissection. Scan parameters: 0.25  s rotation time, pitch 3.2, scan speed 738 mm/s, 90 kV, DLP = 177 mGy cm. Total scan time 0.8 s. Note the clear visualization of the right coronary artery (arrow). Courtesy of Klinikum Großhadern, Munich, Germany

3

Measurement Techniques

The two basic modes of MDCT data acquisition are sequential (axial) and spiral (helical) scanning.

3.1

 DCT Sequential (Axial) M Scanning

Using sequential (axial) scanning, the scan volume is covered by consecutive axial scans in a

S. Ulzheimer et al.

12 Fig. 7 Schematic illustration of axial CT scanning with a CT system with an area detector wide enough to cover entire organs such as the heart. Two commercially available CT systems provide 16 cm z-coverage at isocenter. Their SFOV is cone-shaped

“step-and-shoot” technique. Between the scans the table is moved to the next z-position. The number of images per scan corresponds to the number of active detector slices. By adding the detector signals of the active slices during image reconstruction, the number of images per scan can be reduced, and the image slice width can be increased. Modern MDCT scanners offer retrospective reconstruction of axial scan data with different user-selectable slice widths by means of refined data interpolation approaches. The option to realize wider sections is beneficial for examinations that require narrow collimation to avoid partial volume artifacts and low image noise to detect low-contrast details, such as examinations of the posterior fossa of the skull or the cervical spine. With the advent of MDCT, axial scanning has remained in use for only few clinical applications, such as head scanning, high-resolution lung scanning, perfusion CT, and interventional applications. Furthermore, most cardiac CT examinations with MDCT systems with 64 and more slices are performed with ECG-triggered axial scanning, because of the significantly reduced radiation dose to the patient compared to ECG-gated spiral scans (Earls et  al. 2008; Scheffel et al. 2008). Axial “step-and-shoot” scanning plays a bigger role for CT systems with area detectors with 16  cm z-axis coverage at isocenter. These systems can cover selected organs, such as the heart, the kidneys, or the brain, in one axial scan without table movement. CT scanners with area detectors show advantages in cardiac scanning and in the acquisition of dynamic CT data. They can avoid the typical stair-step artifacts in ECG-­ synchronized examinations of the heart, and they

can acquire dynamic volume data with high temporal resolution. Successful use of the commercially available CT systems with 16  cm z-axis coverage for coronary CTA has been demonstrated (Rybicki et al. 2008; Steigner et al. 2009; Dewey et al. 2009). Meanwhile, the application spectrum has been extended to scanning of patients with atrial fibrillation (Kondo 2013). In an axial scan the reconstructed SFOV (SFOV) is cone-shaped, see Fig. 7. While a SFOV of, e.g., 16  cm z-width can be reconstructed at isocenter, only 11.7 cm z-coverage is feasible at a distance of 160  mm from the isocenter. Larger scan volumes in the z-direction have to be covered by “stitching,” i.e., by appending axial scans shifted in the z-direction. With increasing SFOV, more overlap in the z-direction is required for gapless volume coverage (see Fig.  7), which results in correspondingly increased radiation dose.

3.2

 DCT Spiral (Helical) M Scanning

Spiral/ helical scanning is characterized by continuous gantry rotation and continuous data acquisition while the patient table is moving at constant speed, see Fig. 8.

3.2.1 Pitch An important parameter to characterize a spiral/ helical scan is the pitch p. According to IEC specifications (International Electrotechnical ­ Commission 2002) p is given by: p  =  table feed  per  rotation/total width of the collimated beam.

Multi-slice CT: Current Technology and Future Developments

z-axis

Fig. 8 Principle of spiral/ helical CT scanning: the patient table is continuously translated while multiple rotations of scan data are acquired. The path of X-ray tube and detector relative to the patient is a helix. An interpolation of the acquired measurement data has to be performed in the z-direction to estimate a complete CT data set at the desired image position

This definition holds for single-slice CT as well as for MDCT. It shows whether data acquisition occurs with gaps (p  >  1) or with overlap (p  1.5:1) nonoverlapping pitch, the tube current does not increase sufficiently to offset the effect of increased pitch on radiation dose and as a result this scanning mode is associated with substantially lower radiation dose and higher image noise. For high-contrast CT examinations such as CT angiography or whole spine CT for assessing spinal deformities, such high pitch enables faster scanning at substantially lower radiation dose without any noticeable effect of increased image noise on the diagnostic confidence (Kalra et al. 2013a; Schulz et al. 2012). Detector configuration: This refers to the number of z-axis data channels being used and the effective detector-row thickness of each data channel. The product of the number of data

41

c­hannels and the effective detector-row thickness determines the X-ray beam collimation. On some older MDCT scanners, the X-ray beam used to fall beyond the extent of the detector (referred to as overbeaming) leading to lower dose efficiency since X-rays falling beyond the detectors did not contribute to image formation but did lead to additional radiation dose (Kalra et al. 2004d). Subsequently, online collimator cams were introduced to curtail the X-ray beam and focus it on the detector assembly. Different MDCT scanners offer different types of detector geometries. Some scanners have constant width (matrix-array) configuration where all detector rows have equal width whereas others had variable detector row widths with central rows being thinner and peripheral rows being thicker. The variable detector arrays systems implied that thinner images required use of thinner X-ray beam width (less dose efficient) compared to thicker beam width which could not generate the thinner images. On such variable detector array scanners, choice of detector width depends on the clinical need for thinner or thicker images. Most recent MDCT scanners have matrix array detectors so that thinner images can be acquired over the entire detector width. Another aspect of beam collimation is over-­ ranging which pertains to scanning with helical mode. Over-ranging refers to half of the collimated X-ray beam falling beyond the beginning of prescribed scan range and half falling beyond the end of the scan range. The over-ranging X-ray beam does not contribute in image formation but leads to additional radiation dose. With wider detector area MDCT scanners, over-ranging can lead to substantial radiation penalty particularly for short scan range. To address this issue, some CT vendors have introduced adaptive X-ray shielding which use a filter to curtail the over-­ ranging X-ray beam at the beginning and at the end of the prescribed scan range. Scan modes: Axial scan mode (also called sequential or step and shoot) refers to scan data acquisition with stationary patient table. For longer scan range, the table incrementally moves along the z-axis and then data acquisition resumes. Helical or spiral scan mode involves

42

scan data acquisition while the patient table continuously moves along the z-axis. The latter enables faster data acquisition over longer body regions (longer than the combined width of the detector rows) as compared to axial scan mode. MDCT scanners with wider detector configuration (16 cm coverage with 256-detector rows from GE and 320–detector rows from Toshiba) enable scanning of smaller organs such as head, heart, and other small body regions (1 cm short-axis diameter) from nonenlarged nodes (≤1 cm short-axis diameter). However, it is important to note that nodal metastases are variable in size. While large nodes (>2 cm) are almost always malignant, a lymph node that measures 2 mm), pericardial nodularity, pericardial mass, or visualization of direct

Fig. 32  Malignant pleural effusion. Axial CT in a patient with lung adenocarcinoma shows a loculated right pleural effusion (asterisks). There is associated nodular thickening of the pleura (arrowheads), as well as a pleural nodule along the right lateral pleura (arrow). The presence of a malignant pleural effusion indicates M1a disease

Fig. 34  Pericardial metastasis. Axial CT shows circumferential thickening of the pleura caused by pericardial metastases from mesothelioma. Also present is a left pleural effusion. Metastases to the pericardium may manifest as a pericardial effusion, pericardial thickening or nodularity, or a pericardial mass

a

Fig. 33 Pericardial metastases from lung cancer. A 54-year-old male with a history of right upper lobe adenocarcinoma. (a) Axial CT in soft tissue window performed after a right upper lobectomy shows a normal pericardium. The normal pericardium is barely perceptible and should be a thin band of soft tissue measuring 7 cm), and, less reliably, infiltration of the surrounding mediastinal fat are suggestive of with invasive thymoma. Other more direct signs of invasion include vascular encasement or obliteration, endoluminal soft tissue within a vessel, pericardial thickening continuous with the tumor, and pleural dissemination, the latter of which can manifest pleural thickening, nodules, or masses, occurring almost always ipsilateral to tumor site (so-called drop metastases to the pleura) (Webb and Higgins 2010; Benveniste et al. 2011) (Fig. 37b). Pleural effusion is uncommon, even in the setting of extensive pleural metastases (Benveniste et al. 2011). Thymic Carcinoma Thymic carcinoma is rare, comprising 20% of thymic epithelial tumors. Several histological subtypes of thymic carcinoma are recognized by the WHO (Marx et al. 2015) (Table 11). Thymic carcinomas typically lack a well-defined capsule, and unlike thymomas, which show morphologic features of thymic epithelial cells at pathology, the cells of thymic carcinoma show overt atypia (Nasseri and Eftekhari 2010). Thymic carcinoma portends a poorer prognosis than thymoma (5-year survival of 40% versus 78% for thymic carcinoma and thymoma, respectively) (Rashid et al. 2013). The distinction between thymoma and thymic carcinoma is not possible by imaging alone. Nevertheless, thymic carcinoma has a greater propensity for mediastinal invasion and metastases than thymoma; whereas 5% of patients with invasive thymoma have distant metastases at the time of diagnosis, up to 50–65% of patients with thymic carcinoma have distant metastases at the time of diagnosis (Nasseri and Eftekhari 2010). CT features that are more frequently observed in thymic carcinoma compared to thymoma include irregular margins, necrotic or cystic components, great vessel invasion, heterogeneous contrast enhancement, and lymphadenopathy (Sadohara

381

et al. 2006) (Fig. 38). If any of these features are present, the possibility of thymic carcinoma should be raised.

3.1.2 Rare Thymic Tumors While thymic epithelial tumors comprise the vast majority of thymic tumors, there are a number of other thymic neoplasias recognized by the WHO. Many of these lesions are indistinguishable from thymic epithelial lesions at imaging. Although a complete review of these thymic lesions is beyond the scope of this text, salient features of some of these neoplasms are discussed below. Thymic Neuroendocrine Tumors Neuroendocrine tumors of the thymus include carcinoid tumors, large cell tumors, and small cell carcinomas. While the behavior of these lesions depends upon the histological subtype and grade, neuroendocrine tumors of the thymus carry a poorer prognosis than thymoma, with a tendency for recurrence and metastases (Webb and Higgins 2010). There is an association between neuroendocrine tumors of thymus and endocrine syndromes such as Cushing’s syndrome and multiple endocrine neoplasia types I and II. The appearance at CT is nonspecific, but these neoplasms are typically large anterior mediastinal masses with irregular margins, internal necrosis, hemorrhage, and occasional calcification (Nasseri and Eftekhari 2010). Thymic Lymphoma Lymphoma may involve the thymus. Most commonly, lymphoma of the thymus occurs in the setting of disseminated disease, most frequently in the setting of Hodkin’s lymphoma (Nasseri and Eftekhari 2010). Although rare, isolated lymphoma of the thymoma may also occur. At CT, lymphoma of the thymus may mimic thymoma; however, thymic lymphoma typically occurs in a younger age group. Suggestive features include homogenous thymic enlargement, lobular margins, homogenous contrast enhancement, and lymphadenopathy in the mediastinum (Nishino et al. 2006). Necrosis and calcification are rare.

T.D. Tailor

382

a

b

c

d

Fig. 38 Metastatic thymic carcinoma. A 71-year-old male with history of cough. (a, b) Axial chest CT shows an anterior mediastinal mass (arrow) with a punctate focus of internal calcification. There is a mildly enlarged right internal mammary chain lymph node, as well as an 8-mm right paratracheal lymph node (arrowheads). (c) Axial fused PET-CT shows FDG avidity in both the right internal mammary node and right paratracheal node

(arrows). In this patient with pathology-proven thymic carcinoma, these FDG-avid nodes were presumed malignant. (d) Follow-up CT performed four years after initial diagnosis shows a new nodal metastasis in the right paracardial fat (arrow) and a small right pleural effusion. Although distinction of thymoma from thymic carcinoma is not possible by CT alone, lymphadenopathy is more commonly observed in thymic carcinoma than thymoma

Thymolipoma Thymolipomas are rare, well-encapsulated benign tumors of the thymus. The mean age of presentation is 33 years, and there is no gender predilection (Moran et al. 1995). They are usually asymptomatic, despite their often large size at the time of diagnosis (average diameter 20 cm). Typical CT findings include a large fat-­containing

mass with linear bands of soft tissue. Lesions arise within the thymus or can be connected to the thymus via a pedicle. However, owing to its pliable nature and large size, thymomas may drape over the heart, extend inferiorly into the cardiophrenic or costophrenic angles, or may occupy an entire hemithorax (Nasseri and Eftekhari 2010; Moran et al. 1995).

Chest Neoplasias

3.2

383

Germ Cell Tumor

The mediastinum is the most common extragonadal site for GCT. GCTs account for 15% of all mediastinal tumors in adults and approximately 24% mediastinal tumors in children (Strollo et al. 1997a). The classification of mediastinal GCTs is identical to gonadal GCTs; that is, GCTs are characterized as teratomas (mature and immature subtypes), pure seminomas, non-seminomatous GCT (NSGCT, which comprises a variety of entities, including embryonal carcinoma, yolk sac tumor, choriocarcinoma, and mixed GCTs). Mature teratomas are the most common mediastinal GCT, followed by seminoma. Mediastinal GCTs can occur at any age; however, there is a bimodal age distribution with a peak during infancy and a second peak in young adulthood (28–33 years of age) (Travis et al. 2004). The vast majority of mediastinal GCT occur in the anterior mediastinum, most commonly within or adjacent to the thymus. Rarely, mediastinal GCTs (termatomas and yolk sac tumors) may occur in the posterior myocardium. GCTs are thought to occur as a result of arrested primordial germ cell migration (Travis et al. 2004; Drevelegas et al. 2001). While most mediastinal GCTs are benign, malignant GCTs can occur in both children and adults.

3.2.1 Teratoma Teratomas, the most common type of mediastinal GCT, are composed of several tissues derived from two to three germinal layers (ectoderm, endoderm, and mesoderm). Teratomas may be mature, meaning they contain exclusively mature, adult-type tissues, or immature, the latter implying some component of immature embryonic or fetal tissue. Mature teratomas are more common than immature teratomas; the former is benign; however, the latter may be malignant. Although rare, teratomas may rupture into the lung, pleural space, or pericardium. Unless large, teratomas are usually asymptomatic or incidentally found. The most common CT appearance of a teratoma is that of a heterogeneous anterior mediastinal mass containing cystic and solid components (Fig.  39). The combination of fluid, soft tissue,

Fig. 39  Teratoma. A 40-year-old with mature teratoma. Axial CT with contrast shows a well-circumscribed anterior mediastinal mass with multiple tissue densities. The lesion causes mild mass effect on the right ventricle and right atrium. There is mild rim calcification along the medial margin of the tumor (arrowhead). There is a focal nodule of fat (−50 HU) in the right posterolateral aspect of the lesion (arrow). Up to 70% of teratomas contain visible fat

calcification, and fat in an anterior mediastinal mass is diagnostic of a teratoma (Silverman 2012). However, not at teratomas contain all of these tissue types. Fluid components or cysts are observed in 85–95% of teratomas (Patel et al. 2013). As many as 50–70% of teratomas contain fat, which can occur in variable amounts, ranging from a small fatty nodule, to almost complete fatty mass, or to a fat-fluid level (Drevelegas et al. 2001). Calcification is reported in approximately 50% of cases and may be focal, punctate, or rimlike; rarely, bones or teeth may be seen. With the administration of intravenous contrast, teratomas demonstrate varying degrees of contrast enhancement, including enhancing soft tissue nodules, enhancing internal septations, and/or rim enhancement (Travis et al. 2004; Drevelegas et al. 2001). Various CT findings can be helpful in differentiating mature (benign) from immature (potentially malignant) teratomas. Mature teratomas typically have smooth or lobular margins, do not invade adjacent mediastinal structures, and are more likely to have fat and fluid components. In contrast, immature teratomas often have irregular margins, are more likely to contain solid tissue

T.D. Tailor

384

components, less likely to contain fat, may have necrosis, and may compress or invade nearby structures (Webb and Higgins 2010).

Pericardial or pleural effusions are frequently observed (Webb and Higgins 2010; Drevelegas et al. 2001; Juanpere et al. 2013) (Fig. 40b).

3.2.2 Seminoma Seminoma is the most common primary malignant GCT of the mediastinum and occurs almost exclusively in men. Unlike teratomas, which are usually asymptomatic, patients with seminoma often experience symptoms, which can range from chest pain, respiratory difficulty, dysphagia, or fever (Drevelegas et al. 2001). At CT, the appearance of seminoma often resembles lymphoma (Travis et al. 2004). Seminomas are usually large anterior mediastinal masses with smooth or lobular margins. Lesions are typically homogenous with mild contrast enhancement (Fig. 40). Although uncommon, low-density cystic spaces or necrosis may occur. Calcification is rare. The mass often obliterates mediastinal fat planes; however, invasion into mediastinal structures or the chest wall is rare.

3.2.3 NSGCT While NSGCTs comprise a variety of different entities, from an imaging and classification standpoint, these tumors are often grouped together due to similarity in imaging appearances and aggressive behavior (Travis et al. 2004). At CT, NSGCTs are commonly heterogeneous anterior mediastinal masses with irregular or spiculated margins. Unlike seminomas, which tend to be homogenous in attenuation, these lesions are invariably heterogeneous, showing both low-density and high-density areas, as a result of cystic change, necrosis, or hemorrhage (Fig.  41). Calcification is more common in NSGCT than in seminoma. Unlike seminomas, NSGCTs may invade adjacent structures, including the chest wall, pleura, great vessels, pericardium, and the heart (Drevelegas et al. 2001).

a

b

* *

Fig. 40  Seminoma. A 20-year-old male who presented with fever and chest discomfort. (a) Contrast-enhanced CT shows a large mass in the anterior mediastinum with mild, relatively homogenous, contrast enhancement. A few foci of low attenuation are present in the mass (arrows), which may suggest necrosis or cystic change. The fat plane between the main pulmonary artery (aster-

isk) and the mass is obliterated, and the main pulmonary artery is also mildly displaced in the posterior direction. (b) Noncontrast CT scan, below the level of the anterior mediastinal mass, performed within 2 weeks of (a), shows a small pericardial effusion (asterisk) and a small left pleural effusion (arrow)

385

Chest Neoplasias

a

b

Fig. 41  Nonseminomatous germ cell tumor. Contrastenhanced axial CT in a 21-year-old male. (a) A heterogeneously enhancing mass in the anterior mediastinum. (b) More inferiorly, the mass displaces the heart mildly to the right side. Within the mass, there is at least one thin

3.3

Lymphoma

enhancing septation (arrow), as well as a course calcification. In contrast to seminomas, which tend to be relatively homogenous, nonseminomatous germ cell tumors are often heterogeneous in appearance

or NK cells or their precursors may give rise to NHL (Campo et al. 2011). Lymphomas are primary neoplasms of the lymWhile the HL and NHL cannot be definitively phoreticular system. Lymphomas account for distinguished by CT appearance, certain CT fea15% of all primary mediastinal masses in adults tures are helpful in differentiating HL and and 45% of anterior mediastinal masses in chil- NHL. For both HL and NHL, mediastinal lymphdren. Lymphomas are classified into two main adenopathy is the most common manifestation in categories: Hodkin’s lymphoma (HL) and non-­ the chest. Enlarged lymph nodes are usually Hodkin’s lymphoma (NHL). HL is the most com- homogenous with attenuation similar to that of mon lymphoma to occur in the mediastinum; muscle. With intravenous contrast delivery, mild however overall, NHL is more common than HL diffuse contrast enhancement is usually observed. (Shields et al. 2009). Rarely (up to 20% of cases), low attenuation HL may occur at any age; however, there is a areas of necrosis may be present. The presence of bimodal age distribution with incidence peaking necrosis does not affect therapeutic response and in the third decade of life and after age 50 (Tecce overall patient survival (Fishman et al. 1991). In et al. 1994). NHL is predominantly a disease of the setting of bulky disease, blood vessels are older adults with a median age of diagnosis of often displaced and/or mildly narrowed. However, 55 years (Au and Leung 1997). Histologically, severe vascular narrowing and vascular invasion HL is comprised of Reed-Sternberg cells. A vari- are rare. Enlarged lymph nodes in HL and NHL ety of histological subtypes are recognized, may assume various morphologies, such as disincluding nodular sclerosing, mixed cellularity, crete nodal clusters with preserved mediastinal lymphocyte rich, and lymphocyte depleted, and fat planes, matted lymph nodes with poor visualnodular lymphocyte predominant HL. In devel- ization of mediastinal fat planes, or diffuse medioped countries, nodular sclerosing HL is the most astinal infiltration with loss of individual nodal common subtype (Jaffe et al. 1999). Regarding margins. Nodal calcification, in the absence of NHL, the majority arise from B cells, although T treatment, is rare (Au and Leung 1997).

386

T.D. Tailor

a

In most cases of HL, multiple nodal groups are involved. Less commonly, a single nodal group is enlarged (up to 40% of cases); in these cases, the anterior mediastinum (prevascular space) is most common site of involvement. Unlike NHL, HL spreads via continuous lymph node involvement. Hence, it is unusual for HL to skip nodal groups (Tecce et al. 1994; Au and Leung 1997). HD also has a predilection to involve the thymus, which is enlarged in 30–40% of cases. Often, owing to anterior mediastinal lymphadenopathy, visualization of an enlarged thymus is obscured unless the normal thymic shape is preserved (Webb and Higgins 2010). Pulmonary parenchymal involvement by lymphoma is reported in 38% of patients with HL (Berkman et al. 1996). Pulmonary parenchymal involvement by HL almost always is associated with intrathoracic lymphadenopathy (Hare et al. 2012). Three patterns of parenchymal HL are described: (1) multiple pulmonary nodules or masses (the most common manifestation), (2) lobar consolidation (which is indistinguishable from pneumonia), and (3) interlobular septal thickening, which can be a sequela of lymphatic and/or venous obstruction by mediastinal lymphadenopathy or lymphangitic tumor spread (Hare et al. 2012; Au and Leung, 1997; Berkman et al. 1996). Pleural effusions are reported in approximately 15% of patients with HL and are usually due to lymphatic and/or venous obstruction by lymphadenopathy (Castellino et al. 1986).

b

Fig. 42  Hodgkin’s lymphoma. Contrast-enhanced axial CT of the chest in a 21-year old male. (a) Superior mediastinal lymphadenopathy with mild homogenous enhancement encases the proximal left common carotid artery and the left internal mammary artery (arrow). (b) More inferiorly, there is bulky lymphadenopathy in the anterior mediastinum and the right paratracheal regions. These lymph nodes were continuous with the superior mediastinal lymphadenopathy shown in (a). The lymphadenopathy encases and mildly displaces the left internal mammary artery. While lymphoma may encase and narrow blood vessels, vascular invasion is rare

3.3.1 HL The most common CT manifestation of HL is anterior or superior mediastinal (prevascular, paratracheal, aorticopulmonary) lymphadenopathy (Fig.  42). These lymph nodes are involved in 90–100% of patients with HL. Hilar lymph nodes (unilateral or bilateral) are enlarged in approximately 30% of cases, and the subcarinal lymph nodes are involved in approximately 20% of cases (Castellino et al. 1986). The presence of hilar lymphadenopathy without mediastinal lymphadenopathy is unusual. Other less common lymph node chains to be affected by HL include the paracardiac (cardiophrenic angle), paraesophageal, internal mammary, and posterior mediastinal lymph nodes (Au and Leung 1997; Castellino et al. 1986).

3.3.2 NHL Similar to HL, mediastinal lymphadenopathy is the most common thoracic manifestation on NHL (Fig. 43a). However, unlike in patients HL, the majority of whom have mediastinal nodal involvement, less than half of patients with NHL have mediastinal nodal involvement (Au and Leung 1997). While anterior and superior mediastinal nodes are most commonly affected by NHL, involvement of posterior mediastinal, retrocrural, and cardiophrenic nodes is more common in NHL than in HL, with a reported prevalence of 7–10% (Castellino et al. 1996). Involvement of a single node group is more common in patients with NHL than HL. When a single node group is involved, the

387

Chest Neoplasias

a

b

Fig. 43  Non-Hodgkin’s lymphoma. A 57-year-old male with diffuse large B-cell lymphoma. (a) Axial noncontrast chest CT shows prevascular and right paratracheal lymphadenopathy (arrows). There is a smoothly marginated nodule in the right lung (arrowhead). Surgical biopsy of this nodule revealed pulmonary involvement by diffuse large

B-cell lymphoma. The high-density focus posterolateral to lung nodule is suture material from the biopsy. (b) Coronal CT shows bilateral pulmonary masses with lobular margins, compatible with pulmonary involvement by lymphoma. Pulmonary lymphoma may present as nodules, masses, consolidation, or interlobular septal thickening

posterior mediastinal is the most common site of involvement. Isolated cardiophrenic angle lymphadenopathy may also occur, which is atypical for HL (Tecce et al. 1994). Although uncommon, bulky mediastinal lymphadenopathy with compression of adjacent structures may occur, particularly for high-grade subtypes (Au and Leung 1997). Pulmonary parenchymal involvement is less common in NHL than in HL (24% versus 38 for NHL and HL, respectively). Similar to HL, parenchymal involvement of the lungs usually manifests as pulmonary masses or nodules (Fig. 43b); less commonly, interlobular septal thickening or consolidation may occur. In contrast to HL, where pulmonary involvement is almost always associated with mediastinal nodal enlargement, this is not the case for NHL; pulmonary involvement by NHL frequently occurs in the absence of mediastinal lymphadenopathy (Hare et al. 2012). Other extranodal findings of NHL include pleural effusion, pleural mass, pericardial effusion, or pericardial mass (Castellino et al. 1996). High-grade lesions may invade the chest wall (Au and Leung 1997).

3.4

Neurogenic Tumors

Neurogenic tumors are the most common cause of a posterior mediastinal mass, comprising 20% of mediastinal neoplasms in adults and approximately 35% of mediastinal neoplasms in children (Strollo et al. 1997b). Neurogenic tumors are grouped into three categories based upon their neural tissue of origin: (1) peripheral nerve sheath tumors (which include schwannoma, neurofibroma, and malignant nerve sheath tumor), (2) sympathetic ganglia tumors (which include ganglioneuroma, ganglioneuroblastoma, and neuroblastoma), and (3) parasympathic tumors (paraganglioma). Peripheral nerve sheath tumors are most common in adults, whereas sympathetic ganglia tumors are most common in children (Kawashima et al. 1991). CT cannot reliably differentiate between the various neurogenic tumor types; however, patient age and history may help guide specific differential considerations. Up to 45% of neurofibromas occur in patients with neurofibromatosis. Neurofibromas are the most common peripheral nerve sheath tumors in patients with neurofibromatosis 1, and multiple neurofibromas or a single

T.D. Tailor

388

plexiform neurofibroma are diagnostic of neurofibromatosis type 1. Schwannomas are the most common type of peripheral nerve sheath tumor in patients with neurofibromatosis 2. While rare, up to 50% of malignant peripheral nerve sheath tumors occur in patients with neurofibromatosis (Strollo et al. 1997b). At CT, most neurogenic tumors are paraspinal in location, although may also occur in the intercostal spaces, following the course of intercostal a

nerves. Neurogenic tumors are commonly sharply marinated and spherical or oblong in shape (Fig.  44). Punctate or stippled calcifications are occasionally detected. Up to 70% of peripheral nerve sheath tumors are low in attenuation (with respect to muscle). The low attenuation is attributable to lipid associated with neurogenic tissue, interstitial fluid, or cystic degeneration (Kawashima et al. 1991). Neurogenic tumors demonstrate variable enhancement following b

c

Fig. 44  Schwannoma in the posterior mediastinum. A 30-year-old male with a history of neurofibromatosis type 2 and a paraspinal schwannoma. (a) Coronal noncontrast CT shows a smoothly marinated low-attenuation oblong paraspinal mass adjacent to the mid-thoracic spine. (b) With the administration of intravenous contrast, the mass

demonstrates minimal diffuse contrast enhancement (Hounsfield unit increase by 5–7 HU following contrast administration). (c) Axial CT in bone window shows no evidence of bone erosion or periosteal reaction. Neurogenic tumors are the most common tumors of the posterior mediastinum and are usually paraspinal in location

Chest Neoplasias

389

pulmonary neuroendocrine proliferations and neoplasms. Radiographics 33(6):1631–1649. doi:10.1148/ rg.336135506. PubMed PMID: 24108555 Benveniste MF, Rosado-de-Christenson ML, Sabloff BS, Moran CA, Swisher SG, Marom EM (2011) Role of imaging in the diagnosis, staging, and treatment of thymoma. Radiographics 31(7):1847–61; discussion 61–3. doi:10.1148/rg.317115505. PubMed PMID: 22084174 Berkman N, Breuer R, Kramer MR, Polliack A (1996) Pulmonary involvement in lymphoma. Leuk Lymphoma 20(3-4):229–237. doi:10.3109/10428199609051612. PubMed PMID: 8624461 Brooks DR, Austin JH, Heelan RT, Ginsberg MS, Shin V, Olson SH et al (2005) Influence of type of cigarette on peripheral versus central lung cancer. Cancer Epidemiol Biomarkers Prev 14(3):576–581. doi:10.1158/10559965.EPI-04-0468. PubMed PMID: 15767332 Byrd RB, Carr DT, Miller WE, Payne WS, Woolner LB (1969) Radiographic abnormalities in carcinoma of the lung as related to histological cell type. Thorax 24(5):573–575. PubMed PMID: 5348324; PubMed Central PMCID: PMCPMC472052 Campo E, Swerdlow SH, Harris NL, Pileri S, Stein H, Jaffe References ES (2011) The 2008 WHO classification of lymphoid neoplasms and beyond: evolving concepts and practical applications. Blood 117(19):5019–5032. doi:10.1182/ Abidin AZ, Garassino MC, Califano R, Harle A, Blackhall blood-2011-01-293050. PubMed PMID: 21300984; F (2010) Targeted therapies in small cell lung cancer: PubMed Central PMCID: PMCPMC3109529 a review. Ther Adv Med Oncol 2(1):25–37. doi:10.1177/1758834009356014. PubMed PMID: Carbone M, Pass HI, Rizzo P, Marinetti M, Di Muzio M, Mew DJ et al (1994) Simian virus 40-like DNA 21789124; PubMed Central PMCID: PMCPMC sequences in human pleural mesothelioma. Oncogene 3126006 9(6):1781–1790. PubMed PMID: 8183577 Aguayo SM, Miller YE, Waldron JA Jr, Bogin RM, Sunday ME, Staton GW Jr et al (1992) Brief report: idiopathic Castellino RA, Blank N, Hoppe RT, Cho C (1986) Hodgkin disease: contributions of chest CT in the initial staging diffuse hyperplasia of pulmonary neuroendocrine cells evaluation. Radiology 160(3):603–605. doi:10.1148/ and airways disease. N Engl J Med 327(18):1285–1288. radiology.160.3.3737899. PubMed PMID: 3737899 doi:10.1056/NEJM199210293271806. PubMed PMID: Castellino RA, Hilton S, O'Brien JP, Portlock CS (1996) 1406819 Non-Hodgkin lymphoma: contribution of chest CT in Alberg AJ, Ford JG, Samet JM, American College of the initial staging evaluation. Radiology 199(1):129– Chest P (2007) Epidemiology of lung cancer: ACCP 132. doi:10.1148/radiology.199.1.8633134. PubMed evidence-based clinical practice guidelines (2nd ediPMID: 8633134 tion). Chest 132(3 Suppl):29S–55S. doi:10.1378/ Chiles C, Woodard PK, Gutierrez FR, Link KM (2001) chest.07-1347. PubMed PMID: 17873159 Metastatic involvement of the heart and pericardium: Arenas-Jimenez J, Alonso-Charterina S, Sanchez-Paya J, CT and MR imaging. Radiographics 21(2):439–449. Fernandez-Latorre F, Gil-Sanchez S, Lloret-Llorens M doi:10.1148/radiographics.21.2.g01mr15439. (2000) Evaluation of CT findings for diagnosis of pleuPubMed PMID: 11259706 ral effusions. Eur Radiol 10(4):681–690. doi:10.1007/ Choi JB, Park CK, Park DW, Kim Y, Kim YS, Choi YW s003300050984. PubMed PMID: 10795555 et al (2002) Does contrast enhancement on CT suggest Au V, Leung AN (1997) Radiologic manifestations of lymtumor response for chemotherapy in small cell carciphoma in the thorax. AJR Am J Roentgenol 168(1):93–98. noma of the lung? J Comput Assist Tomogr 26(5):797– doi:10.2214/ajr.168.1.8976927. PubMed PMID: 8976927 800. PubMed PMID: 12439316 Austin JH, Garg K, Aberle D, Yankelevitz D, Kuriyama K, Lee HJ et al (2013) Radiologic implications of the 2011 Chong S, Lee KS, Chung MJ, Han J, Kwon OJ, Kim TS (2006) Neuroendocrine tumors of the lung: clinical, classification of adenocarcinoma of the lung. Radiology pathologic, and imaging findings. Radiographics 266(1):62–71. doi:10.1148/radiol.12120240. PubMed 26(1):41–57. doi:10.1148/rg.261055057. discussion PMID: 23070271 -8. PubMed PMID: 16418242 Bach PB, Kattan MW, Thornquist MD, Kris MG, Tate RC, Barnett MJ et al (2003) Variations in lung cancer risk Clinical Lung Cancer Genome P, Network Genomic M (2013) A genomics-based classification of human among smokers. J Natl Cancer Inst 95(6):470–478. lung tumors. Sci Transl Med 5(209):209ra153. PubMed PMID: 12644540 doi:10.1126/scitranslmed.3006802. PubMed PMID: Benson RE, Rosado-de-Christenson ML, Martinez-­ 24174329; PubMed Central PMCID: PMCPMC4006630 Jimenez S, Kunin JR, Pettavel PP (2013) Spectrum of

intravenous contrast delivery. Up to 50% of will produce pressure erosion or deformity of the ribs, vertebral bodies, or neural foramina. A finding that is highly suggestive CT features of neurogenic tumors is extension into the neural foramina, which is observed in up to 10% of peripheral nerve sheath tumors. In such cases, the paraspinal mass often assumes a “dumbbell” configuration (Strollo et al. 1997b; Kang et al. 2010). While not definitive, sympathetic ganglia tumors tend to have an elongated craniocaudal dimension compared to peripheral nerve sheath tumors; the former often spans three to five vertebral bodies in length, whereas the l­ atter typically spans one to two vertebral bodies in length (Strollo et al. 1997b).

390 Dales RE, Stark RM, Raman S (1990) Computed tomography to stage lung cancer. Approaching a controversy using meta-analysis. Am Rev Respir Dis 141(5 Pt 1):1096–1101. doi:10.1164/ajrccm/141.5_Pt_1.1096. PubMed PMID: 2160211 Drevelegas A, Palladas P, Scordalaki A (2001) Mediastinal germ cell tumors: a radiologic-pathologic review. Eur Radiol 11(10):1925–1932. doi:10.1007/s003300000725. PubMed PMID: 11702124 Dwamena BA, Sonnad SS, Angobaldo JO, Wahl RL (1999) Metastases from non-small cell lung cancer: mediastinal staging in the 1990s–meta-analytic comparison of PET and CT. Radiology 213(2):530–536. doi:10.1148/radiology.213.2.r99nv46530. PubMed PMID: 10551237 Dynes MC, White EM, Fry WA, Ghahremani GG (1992) Imaging manifestations of pleural tumors. Radiographics 12(6):1191–1201. doi:10.1148/radiographics.12.6.1439021. PubMed PMID: 1439021 Edge SB, American Joint Committee on Cancer (2010). xiv, 648 p. p. AJCC cancer staging manual, 7th edn. Springer, New York/London Engels EA (2010) Epidemiology of thymoma and associated malignancies. J Thorac Oncol 5(10 Suppl 4):S260–S265. doi:10.1097/JTO.0b013e3181f1f62d. PubMed PMID: 20859116; PubMed Central PMCID: PMCPMC2951303 Fishman EK, Kuhlman JE, Jones RJ (1991) CT of lymphoma: spectrum of disease. Radiographics 11(4):647– 669. doi:10.1148/radiographics.11.4.1887120. PubMed PMID: 1887120 Folkert MR, Timmerman R (2015) Review of treatment options for oligometastatic non-small cell lung cancer. Clin Adv Hematol Oncol 13(3):186–193. PubMed PMID: 26352427 Funai K, Yokose T, Ishii G, Araki K, Yoshida J, Nishimura M et al (2003) Clinicopathologic characteristics of peripheral squamous cell carcinoma of the lung. Am J Surg Pathol 27(7):978–984. PubMed PMID: 12826890 Gaerte SC, Meyer CA, Winer-Muram HT, Tarver RD, Conces DJ Jr (2002) Fat-containing lesions of the chest. Radiographics 22 Spec No:S61–S78. doi:10.1148/radiographics.22.suppl_1.g02oc08s61. PubMed PMID: 12376601 Gaikwad A, Souza CA, Inacio JR, Gupta A, Sekhon HS, Seely JM et al (2014) Aerogenous metastases: a potential game changer in the diagnosis and management of primary lung adenocarcinoma. AJR Am J Roentgenol 203(6):W570–W582. doi:10.2214/ AJR.13.12088. PubMed PMID: 25415722 Glazer HS, Duncan-Meyer J, Aronberg DJ, Moran JF, Levitt RG, Sagel SS (1985) Pleural and chest wall invasion in bronchogenic carcinoma: CT evaluation. Radiology 157(1):191–194. doi:10.1148/radiology.157.1.4034965. PubMed PMID: 4034965 Glazer HS, Kaiser LR, Anderson DJ, Molina PL, Emami B, Roper CL et al (1989) Indeterminate mediastinal invasion in bronchogenic carcinoma: CT evaluation. Radiology 173(1):37–42. doi:10.1148/radiology.173.1.2781028. PubMed PMID: 2781028

T.D. Tailor Goldstraw P, Crowley J, Chansky K, Giroux DJ, Groome PA, Rami-Porta R et al (2007) The IASLC lung cancer staging project: proposals for the revision of the TNM stage groupings in the forthcoming (seventh) edition of the TNM Classification of malignant tumours. J Thorac Oncol 2(8):706–714. doi:10.1097/JTO.0b013e31812f3c1a. PubMed PMID: 17762336 Govindan R, Page N, Morgensztern D, Read W, Tierney R, Vlahiotis A et al (2006) Changing epidemiology of small-cell lung cancer in the United States over the last 30 years: analysis of the surveillance, epidemiologic, and end results database. J Clin Oncol 24(28):4539–4544. doi:10.1200/JCO.2005.04.4859. PubMed PMID: 17008692 Hansell DM, Bankier AA, MacMahon H, McLoud TC, Muller NL, Remy J (2008) Fleischner Society: glossary of terms for thoracic imaging. Radiology 246(3):697–722. doi:10.1148/radiol.2462070712. PubMed PMID: 18195376 Hare SS, Souza CA, Bain G, Seely JM, Frcpc, Gomes MM et al (2012) The radiological spectrum of pulmonary lymphoproliferative disease. Br J Radiol 85(1015):848–864. doi:10.1259/bjr/16420165. PubMedPMID:22745203.; PubMed Central PMCID: PMCPMC3474050 Hecht SS (1999) Tobacco smoke carcinogens and lung cancer. J Natl Cancer Inst 91(14):1194–1210. PubMed PMID: 10413421 Heelan RT, Rusch VW, Begg CB, Panicek DM, Caravelli JF, Eisen C (1999) Staging of malignant pleural mesothelioma: comparison of CT and MR imaging. AJR Am J Roentgenol 172(4):1039–1047. doi:10.2214/ ajr.172.4.10587144. PubMed PMID: 10587144 Henschke CI, Yankelevitz DF, Mirtcheva R, McGuinness G, McCauley D, Miettinen OS et al (2002) CT screening for lung cancer: frequency and significance of part-solid and nonsolid nodules. AJR Am J Roentgenol 178(5):1053–1057. doi:10.2214/ajr.178.5.1781053. PubMed PMID: 11959700 Henschke CI, Yankelevitz DF, Yip R, Reeves AP, Farooqi A, Xu D et al (2012) Lung cancers diagnosed at annual CT screening: volume doubling times. Radiology 263(2):578–583. doi:10.1148/radiol.12102489. PubMed PMID: 22454506; PubMed Central PMCID: PMCPMC3329268 Herman SJ, Winton TL, Weisbrod GL, Towers MJ, Mentzer SJ (1994) Mediastinal invasion by bronchogenic carcinoma: CT signs. Radiology 190(3):841– 846. doi:10.1148/radiology.190.3.8115637. PubMed PMID: 8115637 Hollings N, Shaw P (2002) Diagnostic imaging of lung cancer. Eur Respir J 19(4):722–742. PubMed PMID: 11999004 Honda O, Tsubamoto M, Inoue A, Johkoh T, Tomiyama N, Hamada S et al (2007) Pulmonary cavitary nodules on computed tomography: differentiation of malignancy and benignancy. J Comput Assist Tomogr 31(6): 943–949. doi:10.1097/RCT.0b013e3180415e20. PubMed PMID: 18043361 Hsu-Kim C, Hoag JB, Cheng GS, Lund ME (2013) The microbiology of postobstructive pneumonia in lung

Chest Neoplasias cancer patients. J Bronchology Interv Pulmonol 20(3):266–270. doi:10.1097/LBR.0b013e31829ddf01. PubMed PMID: 23857204 Hussain SP, Harris CC (1998) Molecular epidemiology of human cancer: contribution of mutation spectra studies of tumor suppressor genes. Cancer Res 58(18):4023–4037. PubMed PMID: 9751603 Ikezoe J, Kadowaki K, Morimoto S, Takashima S, Kozuka T, Nakahara K et al (1990) Mediastinal lymph node metastases from nonsmall cell bronchogenic carcinoma: reevaluation with CT. J Comput Assist Tomogr 14(3):340–344. PubMed PMID: 2159492 Jaffe ES, Harris NL, Diebold J, Muller-Hermelink HK (1999) World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues. A progress report. Am J Clin Pathol 111(1 Suppl 1):S8–12. PubMed PMID: 9894466 Juanpere S, Canete N, Ortuno P, Martinez S, Sanchez G, Bernado L (2013) A diagnostic approach to the mediastinal masses. Insights Imaging 4(1):29–52. doi:10.1007/ s13244-012-0201-0. PubMed PMID: 23225215; PubMed Central PMCID: PMCPMC3579993 Kang J, Litmanovich D, Bankier AA, Boiselle PM, Eisenberg RL (2010) Manifestations of systemic diseases on thoracic imaging. Curr Probl Diagn Radiol 39(6):247–261. doi:10.1067/j.cpradiol.2009.07.008. PubMed PMID: 20875612 Kawashima A, Libshitz HI (1990) Malignant pleural mesothelioma: CT manifestations in 50 cases. AJR Am J Roentgenol 155(5):965–969. doi:10.2214/ ajr.155.5.2120965. PubMed PMID: 2120965 Kawashima A, Fishman EK, Kuhlman JE, Nixon MS (1991) CT of posterior mediastinal masses. Radiographics 11(6):1045–1067. doi:10.1148/radiographics.11.6.1749849. PubMed PMID: 1749849 Lewis ER, Caskey CI, Fishman EK (1991) Lymphoma of the lung: CT findings in 31 patients. AJR Am J Roentgenol 156(4):711–714. doi:10.2214/ ajr.156.4.2003430. PubMed PMID: 2003430 Lindeman NI, Cagle PT, Beasley MB, Chitale DA, Dacic S, Giaccone G et al (2013) Molecular testing guideline for selection of lung cancer patients for EGFR and ALK tyrosine kinase inhibitors: guideline from the College of American Pathologists, International Association for the Study of Lung Cancer, and Association for Molecular Pathology. J Thorac Oncol 8(7):823–859. doi:10.1097/ JTO.0b013e318290868f. PubMed PMID: 23552377; PubMed Central PMCID: PMCPMC4159960 Lubin JH, Wang ZY, Boice JD Jr, Xu ZY, Blot WJ, De Wang L et al (2004) Risk of lung cancer and residential radon in China: pooled results of two studies. Int J Cancer 109(1):132–137. doi:10.1002/ijc.11683. PubMed PMID: 14735479 Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan BW et al (2004) Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med 350(21):2129–2139. doi:10.1056/ NEJMoa040938. PubMed PMID: 15118073 MacMahon H, Austin JH, Gamsu G, Herold CJ, Jett JR, Naidich DP et al (2005) Guidelines for management of

391 small pulmonary nodules detected on CT scans: a statement from the Fleischner Society. Radiology 237(2):395–400. doi:10.1148/radiol.2372041887. PubMed PMID: 16244247 Maldonado RL (1999) The CT angiogram sign. Radiology 210(2):323–324. doi:10.1148/radiology.210.2.r99fe 14323. PubMed PMID: 10207409 Marom EM, Milito MA, Moran CA, Liu P, Correa AM, Kim ES et al (2011) Computed tomography findings predicting invasiveness of thymoma. J Thorac Oncol 6(7):1274–1281. doi:10.1097/JTO.0b013e31821c4203. PubMed PMID: 21623235 Marx A, Chan JK, Coindre JM, Detterbeck F, Girard N, Harris NL et al (2015) The 2015 World Health Organization classification of tumors of the thymus: continuity and changes. J Thorac Oncol 10(10):1383– 1395. doi:10.1097/JTO.0000000000000654. PubMed PMID: 26295375; PubMed Central PMCID: PMCPMC4581965 Masaoka A, Monden Y, Nakahara K, Tanioka T (1981) Follow-up study of thymomas with special reference to their clinical stages. Cancer 48(11):2485–2492. PubMed PMID: 7296496 McDonald AD, McDonald JC (1980) Malignant mesothelioma in North America. Cancer 46(7):1650–1656. PubMed PMID: 7417959 McLoud TC (2003) The role of PET scanning in the evaluation of lung carcinoma. Cancer Imaging 3(2):83–84. doi:10.1102/1470-7330.2003.0003. PubMed PMID: PMC4448639 Meisinger QC, Klein JS, Butnor KJ, Gentchos G, Leavitt BJ (2011) CT features of peripheral pulmonary carcinoid tumors. AJR Am J Roentgenol 197(5):1073– 1080. doi:10.2214/AJR.10.5954. PubMed PMID: 22021498 Modlin IM, Lye KD, Kidd M (2003) A 5-decade analysis of 13,715 carcinoid tumors. Cancer 97(4):934–959. doi:10.1002/cncr.11105. PubMed PMID: 12569593 Moran CA, Rosado-de-Christenson M, Suster S (1995) Thymolipoma: clinicopathologic review of 33 cases. Mod Pathol 8(7):741–744. PubMed PMID: 8539231 Munden RF, Swisher SS, Stevens CW, Stewart DJ (2005) Imaging of the patient with non-small cell lung cancer. Radiology 237(3):803–818. doi:10.1148/ radiol.2373040966. PubMed PMID: 16251391 Murata K, Takahashi M, Mori M, Shimoyama K, Mishina A, Fujino S et al (1994) Chest wall and mediastinal invasion by lung cancer: evaluation with multisection expiratory dynamic CT. Radiology 191(1):251–255. doi:10.1148/radiology.191.1.8134583. PubMed PMID: 8134583 Naidich DP, McCauley DI, Khouri NF, Leitman BS, Hulnick DH, Siegelman SS (1983) Computed tomography of lobar collapse: 1. Endobronchial obstruction. J Comput Assist Tomogr 7(5):745–757. PubMed PMID: 6309926 Naidich DP, Bankier AA, MacMahon H, Schaefer-Prokop CM, Pistolesi M, Goo JM et al (2013) Recommendations for the management of subsolid pulmonary nodules detected at CT: a statement from the Fleischner

392 Society. Radiology 266(1):304–317. doi:10.1148/ radiol.12120628. PubMed PMID: 23070270 Nakazawa K, Kurishima K, Tamura T, Kagohashi K, Ishikawa H, Satoh H et al (2012) Specific organ metastases and survival in small cell lung cancer. Oncol Lett 4(4):617–620. doi:10.3892/ol.2012.792. PubMed PMID: 23205072; PubMed Central PMCID: PMCPMC3506697 Nasseri F, Eftekhari F (2010) Clinical and radiologic review of the normal and abnormal thymus: pearls and pitfalls. Radiographics 30(2):413–428. doi:10.1148/ rg.302095131. PubMed PMID: 20228326 National Cancer Institute. SEER stat fact sheets: lung and bronchus cancer Bethesday, MD [October 10, 2016 ]. Available from: http://www.ugpti.org/pubs/pdf/DP236. pdf Ngo AV, Walker CM, Chung JH, Takasugi JE, Stern EJ, Kanne JP et al (2013) Tumors and tumorlike conditions of the large airways. AJR Am J Roentgenol 201(2):301–313. doi:10.2214/AJR.12.9043. PubMed PMID: 23883210 Nishino M, Ashiku SK, Kocher ON, Thurer RL, Boiselle PM, Hatabu H (2006) The thymus: a comprehensive review. Radiographics 26(2):335–348. doi:10.1148/ rg.262045213. PubMed PMID: 16549602 Okumura M, Ohta M, Tateyama H, Nakagawa K, Matsumura A, Maeda H et al (2002) The World Health Organization histologic classification system reflects the oncologic behavior of thymoma: a clinical study of 273 patients. Cancer 94(3):624–632. PubMed PMID: 11857293 Park CM, Goo JM, Lee HJ, Kim MA, Lee CH, Kang MJ (2009) Tumors in the tracheobronchial tree: CT and FDG PET features. Radiographics 29(1):55–71. doi:10.1148/rg.291085126. PubMed PMID: 19168836 Patel IJ, Hsiao E, Ahmad AH, Schroeder C, Gilkeson RC (2013) AIRP best cases in radiologic-pathologic correlation: mediastinal mature cystic teratoma. Radiographics 33(3):797–801. doi:10.1148/rg.333125088. PubMed PMID: 23674775 Patz EF Jr, Shaffer K, Piwnica-Worms DR, Jochelson M, Sarin M, Sugarbaker DJ et al (1992) Malignant pleural mesothelioma: value of CT and MR imaging in predicting resectability. AJR Am J Roentgenol 159(5):961–966. doi:10.2214/ajr.159.5.1414807. PubMed PMID: 1414807 Pennes DR, Glazer GM, Wimbish KJ, Gross BH, Long RW, Orringer MB (1985) Chest wall invasion by lung cancer: limitations of CT evaluation. AJR Am J Roentgenol 144(3):507–511. doi:10.2214/ ajr.144.3.507. PubMed PMID: 3871560 Peto R (1986) Influence of dose and duration of smoking on lung cancer rates. IARC Sci Publ 74:23–33. PubMed PMID: 3623669 Pfister DG, Johnson DH, Azzoli CG, Sause W, Smith TJ, Baker S Jr et al (2004) American Society of Clinical Oncology treatment of unresectable non-small-cell lung cancer guideline: update 2003. J Clin Oncol 22(2):330–353. doi:10.1200/ JCO.2004.09.053. PubMed PMID: 14691125

T.D. Tailor Postmus PE, Brambilla E, Chansky K, Crowley J, Goldstraw P, Patz EF Jr et al (2007) The IASLC lung cancer staging project: proposals for revision of the M descriptors in the forthcoming (seventh) edition of the TNM classification of lung cancer. J Thorac Oncol 2(8):686–693. doi:10.1097/JTO.0b013e31811f4703. PubMed PMID: 17762334 Prakash P, Kalra MK, Sharma A, Shepard JA, Digumarthy SR (2010) FDG PET/CT in assessment of pulmonary lymphangitic carcinomatosis. AJR Am J Roentgenol 194(1):231–236. doi:10.2214/AJR.09.3059. PubMed PMID: 20028927 Priola AM, Priola SM, Di Franco M, Cataldi A, Durando S, Fava C (2010) Computed tomography and thymoma: distinctive findings in invasive and noninvasive thymoma and predictive features of recurrence. Radiol Med 115(1):1–21. doi:10.1007/s11547-009-0478-3. PubMed PMID: 20017005 Quint LE (2004) Lung cancer: assessing resectability. Cancer Imaging 4(1):15–18. doi:10.1102/1470­7330.2003.0028. PubMed PMID: 18211855; PubMed Central PMCID: PMCPMC1434522 Rashid OM, Cassano AD, Takabe K (2013) Thymic neoplasm: a rare disease with a complex clinical presentation. J Thorac Dis 5(2):173–183. doi:10.3978/j. issn.2072-1439.2013.01.12. PubMed PMID: 23585946; PubMed Central PMCID: PMCPMC3621939 Ratto GB, Piacenza G, Frola C, Musante F, Serrano I, Giua R et al (1991) Chest wall involvement by lung cancer: computed tomographic detection and results of operation. Ann Thorac Surg 51(2):182–188. PubMed PMID: 1846522 Reck M, Heigener DF, Mok T, Soria JC, Rabe KF (2013) Management of non-small-cell lung cancer: recent developments. Lancet 382(9893):709–719. doi:10.1016/ S0140-6736(13)61502-0. PubMed PMID: 23972814 Rusch VW (1995) A proposed new international TNM staging system for malignant pleural mesothelioma. From the International Mesothelioma Interest Group. Chest 108(4):1122–1128. PubMed PMID: 7555126 Ryu JS, Ryu HJ, Lee SN, Memon A, Lee SK, Nam HS et al (2014) Prognostic impact of minimal pleural effusion in non-small-cell lung cancer. J Clin Oncol 32(9):960– 967. doi:10.1200/JCO.2013.50.5453. PubMed PMID: 24550423 Sadohara J, Fujimoto K, Muller NL, Kato S, Takamori S, Ohkuma K et al (2006) Thymic epithelial tumors: comparison of CT and MR imaging findings of ­low-­risk thymomas, high-risk thymomas, and thymic carcinomas. Eur J Radiol 60(1):70–79. doi:10.1016/j. ejrad.2006.05.003. PubMed PMID: 16766154 Salahudeen HM, Hoey ET, Robertson RJ, Darby MJ (2009) CT appearances of pleural tumours. Clin Radiol 64(9):918–930. doi:10.1016/j.crad.2009.03.010. PubMed PMID: 19664483 Sheth S, Ebert MD, Fishman EK (2010) Superior vena cava obstruction evaluation with MDCT. AJR Am J Roentgenol 194(4):W336–W346. doi:10.2214/ AJR.09.2894. PubMed PMID: 20308479

Chest Neoplasias Shields TW, Altorki NK, Disa JJ, Downey RJ, Flores R, Ilson DH et al (2009) General thoracic surgery, 7th edn. Wolters Kluwer Health/Lippincott Williams & Wilkins, Philadelphia Shim SS, Lee KS, Kim BT, Chung MJ, Lee EJ, Han J et al (2005) Non-small cell lung cancer: prospective comparison of integrated FDG PET/CT and CT alone for preoperative staging. Radiology 236(3):1011–1019. doi:10.1148/radiol.2363041310. PubMed PMID: 16014441 Siegelman SS, Khouri NF, Leo FP, Fishman EK, Braverman RM, Zerhouni EA (1986) Solitary pulmonary nodules: CT assessment. Radiology 160(2):307– 312. doi:10.1148/radiology.160.2.3726105. PubMed PMID: 3726105 Silverman PM (2012) Oncologic imaging a multidisciplinary approach. Elsevier/Saunders, Philadelphia. Available from: http://getitatduke.library.duke. edu/?sid=sersol&SS_jc=TC0000614957&title=Oncol ogic imaging %3A a multidisciplinary approach Strollo DC, Rosado de Christenson ML, Jett JR (1997a) Primary mediastinal tumors. Part 1: tumors of the anterior mediastinum. Chest 112(2):511–522. PubMed PMID: 9266892 Strollo DC, Rosado-de-Christenson ML, Jett JR (1997b) Primary mediastinal tumors: part II. Tumors of the middle and posterior mediastinum. Chest 112(5):1344– 1357. PubMed PMID: 9367479 Swensen SJ, Viggiano RW, Midthun DE, Muller NL, Sherrick A, Yamashita K et al (2000) Lung nodule enhancement at CT: multicenter study. Radiology 214(1):73–80. doi:10.1148/radiology.214.1.r00ja1473. PubMed PMID: 10644104 Tailor TD, Schmidt RA, Eaton KD, Wood DE, Pipavath SN (2015) The pseudocavitation sign of lung adenocarcinoma: a distinguishing feature and imaging biomarker of lepidic growth. J Thorac Imaging 30(5):308–313. doi:10.1097/RTI.0000000000000168. PubMed PMID: 26258598 Tecce PM, Fishman EK, Kuhlman JE (1994) CT evaluation of the anterior mediastinum: spectrum of disease. Radiographics 14(5):973–990. doi:10.1148/radiographics.14.5.7991827. PubMed PMID: 7991827 Tomiyama N, Muller NL, Ellis SJ, Cleverley JR, Okumura M, Miyoshi S et al (2001) Invasive and noninvasive thymoma: distinctive CT features. J Comput Assist Tomogr 25(3):388–393. PubMed PMID: 11351188 Travis WD, World Health Organization, International Agency for Research on Cancer, International Association for the Study of Lung Cancer (2004) International Academy of Pathology. Pathology and genetics of tumours of the lung, pleura, thymus and heart. IARC Press Oxford University Press (distributor), Lyon/Oxford, p 344 Travis WD, Brambilla E, Noguchi M, Nicholson AG, Geisinger KR, Yatabe Y et al (2011) International association for the study of lung cancer/american thoracic society/european respiratory society international multidisciplinary classification of lung adenocarcinoma.

393 J Thorac Oncol 6(2):244–285. doi:10.1097/ JTO.0b013e318206a221. PubMed PMID: 21252716; PubMed Central PMCID: PMCPMC4513953 Travis WD, Brambilla E, Nicholson AG, Yatabe Y, Austin JH, Beasley MB et al (2015) The 2015 World Health Organization classification of lung tumors: impact of genetic, clinical and radiologic advances since the 2004 classification. J Thorac Oncol 10(9):1243–1260. doi:10.1097/JTO.0000000000000630. PubMed PMID: 26291008 Trigaux JP, Gevenois PA, Goncette L, Gouat F, Schumaker A, Weynants P (1996) Bronchioloalveolar carcinoma: computed tomography findings. Eur Respir J 9(1):11– 16. PubMed PMID: 8834327 Truong MT, Ko JP, Rossi SE, Rossi I, Viswanathan C, Bruzzi JF et al (2014) Update in the evaluation of the solitary pulmonary nodule. Radiographics 34(6):1658– 1679. doi:10.1148/rg.346130092. PubMed PMID: 25310422 Tsao AS, Wistuba I, Roth JA, Kindler HL (2009) Malignant pleural mesothelioma. J Clin Oncol 27(12):2081–2090. doi:10.1200/JCO.2008.19.8523. PubMed PMID: 19255316; PubMed Central PMCID: PMCPMC4881753 US Department of Health and Human Services (2014) The health consequences of smoking—50 years of progress: a report of the surgeon general. https://www. surgeongeneral.gov/library/reports/50-years-ofprogress/50-years-of-progress-by-section.html UyBico SJ, Wu CC, Suh RD, Le NH, Brown K, Krishnam MS (2010) Lung cancer staging essentials: the new TNM staging system and potential imaging pitfalls. Radiographics 30(5):1163–1181. doi:10.1148/ rg.305095166. PubMed PMID: 20833843 Valli M, Fabris GA, Dewar A, Hornall D, Sheppard MN (1994) Atypical carcinoid tumour of the lung: a study of 33 cases with prognostic features. Histopathology 24(4):363–369. PubMed PMID: 8045525 Vallieres E, Shepherd FA, Crowley J, Van Houtte P, Postmus PE, Carney D et al (2009) The IASLC lung cancer staging project: proposals regarding the relevance of TNM in the pathologic staging of small cell lung cancer in the forthcoming (seventh) edition of the TNM classification for lung cancer. J Thorac Oncol 4(9):1049–1059. doi:10.1097/ JTO.0b013e3181b27799. PubMed PMID: 19652623 Vansteenkiste JF, Stroobants SG, De Leyn PR, Dupont PJ, Bogaert J, Maes A et al (1998) Lymph node staging in non-small-cell lung cancer with FDG-PET scan: a prospective study on 690 lymph node stations from 68 patients. J Clin Oncol 16(6):2142–2149. PubMed PMID: 9626214 Wang ZJ, Reddy GP, Gotway MB, Yeh BM, Hetts SW, Higgins CB (2003) CT and MR imaging of pericardial disease. Radiographics 23 Spec No:S167–S180. doi:10.1148/rg.23si035504. PubMed PMID: 14557510 Webb WR, Higgins CB (2010) Thoracic imaging, 2nd edn. Lippincott Williams & Wilkins, Philadelphia

394 Webb WR, Gamsu G, Glazer G (1981) Computed tomography of the abnormal pulmonary hilum. J Comput Assist Tomogr 5(4):485–490. PubMed PMID: 7263986 Winer-Muram HT (2006) The solitary pulmonary nodule. Radiology 239(1):34–49. doi:10.1148/ radiol.2391050343. PubMed PMID: 16567482 Wislez M, Antoine M, Rabbe N, Gounant V, Poulot V, Lavole A et al (2007) Neutrophils promote aerogenous spread of lung adenocarcinoma with bronchioloalveolar

T.D. Tailor carcinoma features. Clin Cancer Res 13(12):3518– 3527. doi:10.1158/1078-0432.CCR-06-2558. PubMed PMID: 17575214 Yokota J, Shiraishi K, Kohno T (2010) Genetic basis for susceptibility to lung cancer: recent progress and future directions. Adv Cancer Res 109:51–72. doi:10.1016/B978-0-12-380890-5.00002-8. PubMed PMID: 21070914

CT of Pulmonary Embolism: Imaging Update Antoine Hutt, Paul Felloni, Jacques Remy, and Martine Remy-Jardin

Contents 1    Introduction 2    New Options in the Diagnostic Approach 2.1  Clinical Decision Support 2.2  Optimization of the Radiation Dose 2.3  Optimization of the Iodine Load 2.4  Current Role of CAD Systems 2.5  Is There Always a High Level of Confidence in Diagnosing Acute PE?

 395  396  396  396  397  399

6    When Acute PE Evolves Toward Chronic PE

 405

Conclusion

 406

References

 406

Abstract

This article provides an update of the current trends in the management of acute thromboembolic disease, aimed at emphasizing new options applicable in daily practice. CT is no longer exclusively dedicated to the diagnosis of PE but also participates in the prognostic approach of this disease focusing on the detection of CT features of right ventricular dysfunction. The availability of perfusion CT allows evaluation of the extent of perfusion impairment, thus introducing a more reliable prognostic parameter than the estimation of the clot burden. Radiologists can provide clinicians with relevant information regarding the early risk stratification that is worth describing in daily practice. Optimized management of young patients, pregnant women, and evolution toward chronic pulmonary embolism will also be covered in this chapter.

 400

3    New Options in the Prognostic Assessment 3.1  Clot Burden 3.2  Right Ventricular Dysfunction 3.3  Novel Approaches

 400  400  401  402

4    The Radiologist’s Report: Which Information Is Particularly Relevant for Clinicians?

 402

5    Pulmonary Embolism from Pregnancy to Young Adults 5.1  Pulmonary Embolism in Pregnancy 5.2  Pulmonary Embolism in Children

 403  403  404

A. Hutt, MD • P. Felloni, MD • J. Remy, MD Department of Thoracic Imaging, Hospital Calmette (EA 2694), Univ Lille Nord de France, F-59000 Lille, France M. Remy-Jardin, MD, PhD (*) Department of Thoracic Imaging, Hospital Calmette (EA 2694), Univ Lille Nord de France, F-59000 Lille, France

1

Department of Thoracic Imaging, Hospital Calmette, Boulevard Jules Leclercq - 59037LILLE cedex, Lille, France e-mail: [email protected]

Multidetector-row computed tomographic angiography (CTA) has become the first-line modality for imaging the pulmonary vasculature in patients

Med Radiol Diagn Imaging (2017) DOI 10.1007/174_2017_19, © Springer International Publishing AG Published Online: 17 March 2017

Introduction

395

A. Hutt et al.

396

with suspected acute pulmonary embolism (PE) (Remy-Jardin et al. 2007). This disease is the third most common acute cardiovascular disease after myocardial infarction and stroke in the Unites States and results in many deaths each year. This data fully justifies a regular update of the radiologists’ role in the management of patients suspected of acute pulmonary embolism. Some aspects are currently well established such as the CT features of endovascular clots whose description has not changed over time. However, their detection has greatly benefited from the improvement of crosssectional imaging that has almost completely suppressed technically related artifacts. Some clinically relevant aspects have to be considered by radiologists in their CT reports owing to their impact in early risk stratification. The purpose of this chapter is to provide an insight into new options applicable in the daily practice in the clinical context of acute pulmonary embolism.

department was found to be associated with a significant decrease in the use and significant increase in the yield of CT pulmonary angiography for the evaluation of acute PE during a 2-year period (Raja et al. 2012). These conclusions are shared by Dunne et al. who have recently reported that implementation of evidence-­ based clinical decision support was associated with a 12.3% immediate and sustained decrease in use of CT pulmonary angiographic (CTPA) examinations in the evaluation of inpatients for acute PE (Dunne et al. 2015). However, one should underline that a negative chest CTPA for acute PE can help recognize alternative diagnoses for the patient’s symptoms in 25–67% of cases. These cases usually include congestive heart failure, pneumonia, pleural effusion, or atelectasis. Green et al. have recently proposed a structured approach for recognition of alternative explanations approachable on CT images (Green et al. 2015).

2

New Options in the Diagnostic Approach

2.2

2.1

Clinical Decision Support

In parallel to the abovementioned efforts to decrease the number of unnecessary examinations, the radiological community is directly involved in the optimization of the radiation dose delivered to patients. As recently underlined in a panel discussion (Araoz et al. 2012), the radiation risk from pulmonary CT angiography (CTA) is strongly dependent on age, sex, and pulmonary CTA acquisition parameters (Woo et al. 2012). The radiological community demonstrated the usefulness of several practical methods, the most frequently employed relying on individual adjustment of milliamperage to patient weight, alone or in association with automated tube current modulation systems. This approach achieves an average dose reduction of 20% (Kubo et al. 2008; Christner et al. 2010). However, lowering the kilovoltage has a greater effect on patient dose than reducing the tube current, and the current trend is to perform chest CTAs with a kilovoltage adapted to the patient’s body weight (Matsuoka et al. 2009) (Tables 1a and 1b) (Fig. 1). This

Over the last decade, numerous articles have underlined the increase in CT examinations indicated for clinical suspicion of PE with a parallel description of a high rate of negative examinations (Mamlouk et al. 2010). Mamlouk et al. showed that a CT angiogram positive for pulmonary embolism was extremely unlikely (0.95% chance) if patients had none of the studied thromboembolic risk factors, raising questions on the appropriate indication of the CT examination. In order to improve patient selection, clinicians can utilize risk factor assessment. Special attention should be directed to the patient’s age and immobilization, both risk factors of most concern for a CT angiogram positive for PE. Another option has been proposed, based on a computerized clinical decision support (CDS). At each stage of the proposed decision tree, clinicians could either cancel the imaging or ignore the advice. The implementation of evidence-based CDS in the emergency

Optimization of the Radiation Dose

397

CT of Pulmonary Embolism: Imaging Update Table 1a  Selection of kilovoltage and milliamperage according to the patient’s body weight in routine clinical practice. Scanning parameters for CT examinations reconstructed with filtered-back projection Patient’s weight (kg) < 50 50–80 81–100 ≥ 100

Kilovoltage (kV) 80 100 120 140

Reference (mAs) 120 90 90 90–140

From de Broucker et al. (2012) Abbreviations: kg kilogram, kV kilovoltage, mAs milliampere-second Table 1b  Selection of kilovoltage and milliamperage according to the patient’s body weight in routine clinical practice. Scanning parameters for low-kilovoltage examinations reconstructed with raw-data based iterative reconstructions Patient’s weight (kg) < 50 50–80 81–100 ≥ 100

Kilovoltage (kV) 80 100 100 120

Reference (mAs) 120 65 90 90–140

From de Broucker et al. (2012) Abbreviations: kg kilogram, kV kilovoltage, mAs milliampere-second

approach was investigated in several studies in adult populations, from which several conclusions were drawn (Sigal-Cinqualbre et al. 2004; Schueller-Weidekamm et al. 2006; Szucs-Farkas et al. 2008; Gorgos et al. 2009). First, substantial dose reduction can be achieved, with an average dose reduction of 40% when lowering the setting from 120 to 80 kVp. Second, we improve vascular enhancement, as the attenuation of iodinated contrast media increases at low tube voltage. This was found to improve the analyzability of central and peripheral pulmonary arteries, even when reducing the volume of contrast material (Szucs-­ Farkas et al. 2011). However, radiologists may be reluctant to apply low kVp protocols in daily clinical routine because of the lack of standardized guidelines not only for the tube potential selection but also for the adjustment of the tube current to avoid grainy images. This difficulty can be overcome by the use of automated ­systems

which can determine the most appropriate kV settings in relation to the patient’s attenuation (Niemann et al. 2013). Lastly, the availability of iterative reconstructions offers a unique means of combining dose reduction (up to 50%, averaged DLP less than 80 mGy.cm), excellent vascular enhancement, and better image quality than that obtained at standard dose (Pontana et al. 2013; Kaul et al. 2014; McLaughlin et al. 2015). With such possibilities of dose reduction, it is no longer necessary to try to save dose by reducing the scan length. Important additional or alternative diagnosis may be excluded from the limited imaging volume and therefore go undetected.

2.3

Optimization of the Iodine Load

2.3.1 Low Contrast Medium Volume Although iodinated contrast medium is relatively safe, some adverse reactions such as contrast-­ induced nephropathy may occur. The risk of developing contrast-induced nephrotoxicity is increased with higher doses of iodinated contrast medium. Subsequently, one can decrease the risks by reducing the volume of highly concentrated contrast agents administered during a chest CT angiographic examination. This approach can be facilitated when combining low tube voltage and high-pitch techniques together with iterative reconstruction (Sodickson and Weiss 2012; Lu et al. 2014). The volumes administered varied between 50 and 20 mL of contrast material. 2.3.2 Low-Concentrated Contrast Agents With the advent of fast CT scanning modes, administration of contrast media with high iodine concentration (i.e., 300–370 mg of iodine per milliliter) has become routine clinical practice to maximize the arterial enhancement of systemic and pulmonary arterial circulation. However, the inflow of highly concentrated contrast material to systemic veins can generate streak artifacts between the concentrated agent and the surrounding structures that may obscure

A. Hutt et al.

398

a

b

c

d

e

Fig. 1  Chest CT angiography obtained in a 50-year-old male patient referred to the emergency department with a clinical suspicion of acute pulmonary embolism (182 cm, 69 kg). The examination was obtained at 100 kVp and 90 ref mAs (35% iodinated contrast agent, flow rate: 4 mL/s).

The dose-length product was 128 mGy.cm. Sharp delineation of numerous endoluminal clots identified on both sides (a–d). (e) Transverse CT section obtained at the level of cardiac cavities showing a RV/LV ratio greater than 1, suggestive of right ventricular dysfunction

mediastinal and pathological hilar and right upper lobe pulmonary arterial abnormalities. Moreover, the ­application of CTPA is limited in patients with impaired renal function because of the risk of developing contrast-medium induced

nephropathy. Many patients at risk of developing pulmonary embolism are elderly and have comorbid conditions that increase the risk for renal injury. Therefore, it could be interesting to perform chest CT angiographic examinations

CT of Pulmonary Embolism: Imaging Update

with reduced iodine load. To date, this approach has not been found to be clinically acceptable owing to the poor level of arterial enhancement on images acquired at high kilovoltages. This limitation has recently been overcome by acquiring data sets with dual-energy CT. This new scanning mode offers the possibility of generating virtual monochromatic spectral (VMS) imaging with a wide range of energy levels accessible from a single data set. Using single-source, dual-energy CT, Yuan et al. were the first authors to d­ emonstrate that these acquisitions facilitated iodine load reduction at CT pulmonary angiography (Yuan et al. 2012). Delesalle et al. investigated the energy levels providing optimal imaging of the thoracic circulation at dual-source, dual-­energy CT angiography with reduced iodine load (Delesalle et al. 2013). These authors found that the use of low-concentration contrast media enabled suppression of streak artifacts around systemic veins on high-energy images while providing satisfactory central arterial enhancement on low-energy images. The additional advantage of this scanning mode was the considerable reduction in the amount of iodine administered to patients (Fig. 2). These preliminary studies confirm that dual-energy CT has the potential to represent a new option for greater use of

a

Fig. 2  Chest CT angiography obtained in a 52-year-old female patient (164 cm, 60 kg). The examination was acquired with dual source, dual energy after administration of a low-concentrated contrast agent (i.e., 125 mg I/ mL). (a) CT section obtained at the level of the right bron-

399

low-concentration contrast agents for chest CT examinations in routine clinical practice.

2.4

Current Role of CAD Systems

Whereas the diagnosis of acute PE remains based on the visual depiction of endoluminal filling defects, identification of peripheral clots remains a difficult task and more than 30% can be missed on initial review (Ritchie et al. 2007). This limitation can be solved by the use of computer-aided diagnostic (CAD) systems which have been developed to aid radiologists with the depiction of endovascular clots that requires careful analysis of hundreds of pulmonary vascular branches. Used as a second reader, these systems can help detect small clots initially missed (Kligerman et al. 2014), increasing reader sensitivity for the detection of peripheral emboli (Dewailly et al. 2010; Wittenberg and Berger 2012). In addition, the high negative predictive value of these tools is helpful to reassure inexperienced readers (Blackmon et al. 2013). However, these results are obtained at the expense of an increased reading time due to the presence of numerous false-­ negative and false-positive findings, as recently demonstrated by Wittenberg et al. (Wittenberg and Berger 2012). These authors also

b

chus intermedius illustrating the excellent quality of attenuation within all thoracic circulations on this image generated from both tubes. (b) CT section obtained at the level of the lower lung zones illustrating the additional excellent opacification of the cardiac cavities

A. Hutt et al.

400

a

b

Fig. 3  Dual-energy chest CT angiography obtained in a 30-year-old male suspected of acute pulmonary embolism (185 cm, 73 kg). The examination was obtained with dual-­ source, dual-energy CT (tube A: 80 kV; tube B: 140 kVp; 35% iodinated contrast agent; flow rate: 4 mL/s). The dose-length product was 260 mGy.cm. (a) The presence

of a small-sized, peripheral clot in a subsegmental pulmonary artery of the anterior segment of the left upper lobe (arrow). (b) Whereas the small clot is difficult to visualize, the corresponding perfusion defect (arrowheads) is easily depicted

d­ emonstrated a strong association between CT image quality and the number of false-positive findings indicated by the CAD system which is a current limitation of CAD in clinical practice. An alternative to CAD for the detection of small-sized clots can theoretically be found with dual-energy CT which can provide perfusion imaging in addition to cross-sectional imaging of the pulmonary circulation (Fig. 3). Abnormal pulmonary blood distribution shown at dual-­ source CT improves detection of acute PE, particularly by emphasizing the presence of subsegmental pulmonary iodine mapping defects. However, acute PE cannot be assessed on the sole finding of perfusion defects, even if observed as triangular-shaped defects known to be suggestive of acute PE. Small-airways disease can lead to similar filling defects, depicted in 30% of COPD patients (Pontana et al. 2012). Moreover, the presence of an underlying lung disease altering lung perfusion makes it more difficult to detect PE-related filling defects. Therefore, CT detection of small-sized clots remains a difficult task. In daily practice, this does not represent a major clinical limitation except for the subset of patients with isolated subsegmental PE in whom cross-­sectional imaging may fail to depict them. The role of perfusion imaging in the clinical context of acute PE is more generally considered as a helpful tool for identifying the presence or absence of PE (Cai et al. 2015).

2.5

I s There Always a High Level of Confidence in Diagnosing Acute PE?

From the radiologist perspective, the confidence in identifying endovascular clots has dramatically increased with CT technological improvement over years. In a recent study, Bedayat et al. have evaluated the clinical characteristics associated with low confidence in diagnosis of acute PE as expressed in CTPA reports and evaluated the effect of confidence level in PE diagnosis on patient clinical outcomes (Bedayat et al. 2015). They observed that roughly 10% of positive CTPA reports had uncertainties in PE findings; patients with reports categorized as low confidence had smaller emboli and more comorbidities. Although the low-confidence group was less likely to receive PE-related therapies, patients in this group were not associated with higher probability of short-term mortality.

3

New Options in the Prognostic Assessment

3.1

Clot Burden

The degree of vascular obstruction on CT images can be estimated with semiquantitative scores (Quanadli et al. 2001; Mastora et al. 2003) with

401

CT of Pulmonary Embolism: Imaging Update

recent introduction of quantitative estimates of blood clot volume (Furlan et al. 2011; Hariharan et al. 2016). There is a controversy on the association between mortality and clot burden indexes, mainly related to the lack of correlation between these results and the presence of a pre-­ existing pulmonary disease. Whereas an isolated subsegmental clot may have no clinical consequence in an otherwise healthy patient, the same clot may lead to respiratory failure in a patient with poor respiratory condition due to previously altered pulmonary perfusion. Therefore, it appears interesting to switch from clot burden estimates to the analysis of the extent of perfusion impairment on dual-energy CT perfusion images (Chae et al. 2010; Bauer and Frellesen 2011; Thieme et al. 2012). New scores have been defined, grading the degree of perfusion defect per lobe (Chae et al. 2010; Thieme et al. 2012), estimating the volume of perfusion defects relatively to the total lung volume (Bauer and Frellesen 2011). Good correlations were found between perfusion impairment and CT features of right ventricular dysfunction, suggesting that perfusion defects could be a predictor of patient outcome. To our knowledge, a single study used a model adjusted for age, gender, and prior history of COPD and heart failure (Thieme et al. 2012). Good correlations were found between the proposed dual-energy based PE score and a number of parameters of PE severity. These authors concluded that this approach was easier and faster to perform than the traditional CT scoring methods for vascular obstruction with better prognostic implications.

3.2

Right Ventricular Dysfunction

As previously underlined, the presence of right ventricular dysfunction in hemodynamically stable patients is the major determinant of patient’s outcome. Recent updates in the literature should help radiologists gather accurate information. Regarding the CT features of right ventricular dysfunction, the description published by Reid and Murchison in 1998 remains of major clinical usefulness (Reid and Murchison 1998) (Table 2). In the list of CT features of right ventricular

Table 2  CT features of right ventricular dysfunction 1. Main indicators of right ventricular dysfunction  Dilatation of the right ventricle  Interventricular septal shift (displaced to the left)  Compression of the left ventricle 2. Secondary effects of right ventricular dilatation and dyskinesia  Tricuspid regurgitation  Right atrial enlargement  Reflux of contrast material within the inferior vena cava and hepatic veins  Dilatation of the coronary sinus, superior vena cava and azygos vein From Reid and Murchison (1998)

d­ysfunction, the RV/LV diameter ratio is the most important parameter to consider, as recently confirmed by Becattini et al. in a meta-analysis (Becattini et al. 2014). Several practical approaches have been proposed to determine the RV/LV diameter ratio, including measurements on transverse CT sections, short-axis images and four-chamber views of the cardiac cavities. Kamel et al. were the first authors to suggest that measurements of ventricular diameters on transverse CT sections were accurate enough to estimate the RV/LV ratio (Kamel et al. 2008). This has been recently confirmed by Lu et al. (2012) who have reported that the axial RV/LV diameter ratio is no less accurate than the reformatted four-chamber RV/LV diameter ratio for predicting 30-day mortality after PE (Fig. 1). A step further in the simplification of the estimation of the RV/LV diameter ratio has been proposed by Kumamaru et al. (2012). These authors have recently demonstrated that complex measurements of RV/LV diameter ratios can be replaced by subjective determination of right ventricular enlargement. When the right ventricle appeared larger than the left ventricle, it provided prognostic information that did not significantly differ from that of more traditional, quantitative RV/LV diameter ratios. The authors concluded that a right ventricle that appears larger than the left ventricle should be reported by the radiologist and interpolated into clinical risk stratification. This information can be reinforced when a prior CT examination negative for PE indications is available. In such circumstances, Lu et al. have

A. Hutt et al.

402

shown that the interval increase in four-chamber Table 3  Risk stratification of patients with acute PE: RV/LV diameter ratio is more accurate than the standard approaches and new horizons diameter ratio of the CT examination with posi- 1. High-risk PE (i.e., “massive acute PE”) tive findings for PE alone for mortality prediction  (a) P E-associated arterial hypotension or shock at the time of presentation after acute PE (Lu et al. 2012). Kim et al. have  (b) Short-term mortality of at least 15% recently completed the list of conventional chest  (c) Consensus on thrombolytic therapy CT-derived hemodynamic findings with the left-­ 2. Intermediate-risk PE (i.e., “submassive acute PE”) bulging atrial septum, an abnormal sign indicat (a) Hemodynamic stability but the presence of right ing hemodynamic overloading of the right heart ventricular dysfunction at the time of presentation (Kim et al. 2014).  (b) Mortality risk similar to that of massive PE

3.3

Novel Approaches

In addition to the estimation of the clot burden and features of RV dysfunction, CT angiography can also reveal important ancillary findings in the scanned field, particularly malignancy and abnormal lung parenchymal features (e.g., infarction, interstitial lung disease, or effusions) that affect the patient’s prognosis. Based on these assumptions, Kumamaru et al. have developed a comprehensive risk-scoring system based on CTPA findings for predicting mortality within 30 days after acute PE (Kumamaru et al. 2016). Their model was found to be superior to PESI (PE severity index) in predicting mortality, leading the authors to suggest that the incorporation of this scoring system into image interpretation workflows might help clinicians to select the most appropriate management approach for individual patients.

4

 he Radiologist’s Report: T Which Information Is Particularly Relevant for Clinicians?

The various mortality rates reported among studies illustrate the heterogeneous clinical and prognostic spectrum in acute PE (Table 3). This situation has raised debates on the most appropriate therapeutic options for the various PE-related risk categories. Regarding the prognostic parameters, it is well established that the hemodynamic status at the time of presentation has the strongest prognostic implication

 (c) Thrombolytic therapy or anticoagulation alone 3. Low-risk PE (new category upon discussion – from Lankeit and Konstantinides (2012))  (a) Possible criteria for early discharge and home treatment:    (i) Absence of overt right heart failure    (ii) Absence of right ventricular dysfunction    (iii) Absence of serious comorbidity    (iv) Low risk of early recurrence    (v) Exclusion of a patent foramen ovale  (b) Outpatient treatment (awaiting for results of ongoing trials) 4. No-risk PE (new category upon discussion – from Stein et al. (2012))  (a) Criteria for considering to withhold treatment    (i) Good pulmonary-respiratory reserve    (ii) No evidence of deep venous thrombosis with serial leg tests    (iii) Transient major risk factor for PE that is no longer present    (iv) No history of central venous catheterization or atrial fibrillation    (v) Patient’s willingness to return for serial venous ultrasound  (b) Full patients’ information

for short-term mortality. Therefore, the highest risk is that of “massive acute PE,” characterized by the presence of PE-associated arterial hypotension or shock. Accounting for 5% of all cases of PE, consensus guidelines recommend treatment with thrombolysis. These patients are not referred to the CT room. In the remaining majority of PE patients who present without hypotension, there is a subgroup of patients with “submassive acute PE,” characterized by the presence of right ventricular dysfunction at the time of diagnosis. These patients are considered as having a higher risk of clinical deterioration than those with preserved systemic arterial blood

403

CT of Pulmonary Embolism: Imaging Update

pressure and normal right ventricular function. They correspond to the “intermediate risk” category. Current debates question the recommendations for thrombolytic therapy in this subset of patients (Todd and Tapson 2009; Piazza and Goldhaber 2010; Jimenez et al. 2013). It was only recently that improved risk assessment strategies permitted advances in the identification of another category, i.e., the “low-risk PE.” The current state of knowledge of this category has been recently summarized by Lankeit and Konstantinides (2012). Patients presenting without hemodynamic instability and without elevated biomarker levels or imaging findings indicating right ventricular dysfunction or myocardial injury may constitute this low-risk group (Torbicki et al. 2008). In this category, selected patients might be considered for early discharge and treatment at home. Lastly, the high quality of chest CT examinations currently achievable enables depiction of small-sized pulmonary embolism, sometimes incidentally diagnosed, which could represent a “no-risk category.” These situations have raised debates in the literature on the clinical significance of such clots, described as “isolated subsegmental PE,” comprising a spectrum ranging from a single subsegmental clot to multiple clots exclusively confined to the subsegmental arterial bed. Eyer et al. were the first authors to report the clinicians’ decision to withhold anticoagulation in this category of patients, without adverse effects of this clinical decision (Eyer et al. 2005). Later, Anderson et al. suggested that some pulmonary emboli detected by CT might be clinically unimportant, the equivalent of deep vein thrombosis isolated to the calf veins that might not require anticoagulant therapy (Anderson et al. 2007). In a recent review article, Stein et al. summarized the conditions which should be fulfilled for such therapeutic decisions (Stein et al. 2012). From this description of current trends in PE-risk stratification (Table 4), one can deduce that the radiologist’s report should provide clinicians with relevant information for early risk stratification. Consequently, particular attention should be directed toward analysis of the amount and location of clots in the pulmonary

Table 4  New messages for radiologists Some clots may not require anticoagulant therapy Selected patients might be treated as outpatients Some stable patients might receive thrombolytic therapy

arterial tree, description of cardiac cavity morphology with special attention to CT features suggestive of right ventricular dysfunction as well as abnormalities suggesting pre-existing pulmonary and/or cardiac disease.

5

Pulmonary Embolism from Pregnancy to Young Adults

5.1

Pulmonary Embolism in Pregnancy

To prepare the mother for the blood losses associated with delivery, a state of hypercoagulability develops during pregnancy that explains the increased risk for venous thromboembolism reported during pregnancy. Because clinical symptoms are nonspecific, reliable diagnostic tests are needed but the most adapted diagnostic strategy remained a matter of discussion until the publication of clinical practice guidelines in 2011 (Leung et al. 2011). A multidisciplinary panel developed evidence-based guidelines using the Grades of Recommendation, Assessment, Development and Evaluation (GRADE) system. Strong recommendations were made for three specific scenarios: (a) performance of chest radiography as the first radiation-associated procedure, (b) use of lung scintigraphy as the preferred test in the setting of a normal chest radiograph, and (c) performance of CT pulmonary angiography rather than digital substraction angiography in a pregnant woman with a nondiagnostic ventilation-­ perfusion result. In addition to general recommendations for radiation dose savings for the fetus and the maternal breast, the scanning protocol for a chest CT pulmonary angiography in a pregnant patient should be adapted to the hemodynamic effects of pregnancy. These

A. Hutt et al.

404

a

b

Fig. 4  Chest CT angiography obtained in a 29-year-old pregnant patient at 21 weeks gestation (163 cm, 54 kg). The examination was obtained at 100 kVp and 90 ref mAs (30% iodinated contrast agent, 4 mL/s). The dose-length product was 83 mGy.cm. Images were reconstructed with

raw-data-based iterative reconstruction. Transverse CT section obtained at the level of the right upper lobe bronchus (a) and right middle lobe bronchus (b) illustrating the good level of vascular enhancement and the possibility to depict large endoluminal clots on both sides

effects combine an increase in cardiac output, heart rate, and plasma volume leading to dilution of the ­contrast bolus (Ridge et al. 2011). Moreover, suboptimal opacification can also be due to the increased venous return of nonopacified blood to the right atrium during inspiration. Consequently, several technical adjustments have been proposed, including the use of automated bolus triggering, a high contrast medium flow rate, a high concentration of contrast medium and acquisition during quiet or suspended respiration rather than at deep inspiration (Fig. 4). Regarding potential harmful effects of a chest CT examination to the fetus, it is important to be aware that the radiation dose delivered is in the range of that absorbed by the fetus from naturally occurring background radiation during the 9-month gestational period. In a series of 343 neonates exposed to an iodinated contrast agent at various stages of gestation, all had a normal tyroxine level at birth. In the 85 neonates tested for thyroid-stimulating hormone, only one (with comorbid conditions) had a transiently abnormal level that reverted to normal by day 6 (Bourjeily et al. 2010). From this study, it was concluded that a single, high-dose in utero exposure to water-soluble, low-osmolar iodinated intravenous products, such as iohexol, is unlikely to have a clinically important effect on thyroid function at birth.

5.2

Pulmonary Embolism in Children

As recently reported by Lee et al. (Lee et al. 2012a), the incidence of PE ranges from 0.73% to 4.2% in the pediatric population with concerns about potential overutilization of CT pulmonary angiography in children suspected of having PE. From their study, it was concluded that risk factor assessment should be a primary tool for guiding when to perform CT pulmonary angiography in this population. With such an approach, CT pulmonary angiography can be targeted more appropriately, with the potential to substantially reduce costs and radiation exposure. Five independent risk factors were found to be significantly associated with a positive CT pulmonary angiography result, namely immobilization, hypercoagulable state, excess estrogen state, indwelling central venous line, and prior PE and/ or deep venous thrombosis. Lastly, they observed that the D-dimer test was of little value in screening for PE among children with a high clinical probability of PE. From the same group, it appears that similar conclusions can be drawn in older children and young adults (Lee et al. 2012b). In a recent study, Hennelly et al. have proposed a pediatric PE clinical decision rule that was derived from commonly used adult-based PE algorithms (Hennelly et al. 2016).

CT of Pulmonary Embolism: Imaging Update

a

405

b

c

Fig. 5  CT angiography obtained in a 38-year-old female patient evaluated for chronic thromboembolic pulmonary hypertension. The examination was obtained with dual-­ source, dual-energy CT (tube A: 80 kV; tube B: 140 kVp; 35% iodinated contrast agent; flow rate: 4 mL/s). The doselength product was 378 mGy.cm. (a) CT section obtained at the level of the right pulmonary artery showing an intra-

luminal web at the level of the left interlobar pulmonary artery. (b) CT section obtained at the level of the origin of the right middle lobe artery showing a mural defect (arrow) at the level of the right interlobar pulmonary artery. (c) CT section obtained at the level of the carina showing tiny arterial sections in the posterior segment of the right upper lobe and mosaic perfusion on both sides

6

prevalence occurring in 0.1–4.0% of patients in retrospective studies (Fedullo et al. 2001; Pengo et al. 2004). In a recent prospective study, the incidence of symptomatic CTEPH was found to represent a relatively common complication after an episode of acute pulmonary embolism, reported in 5.4% of the studied population (Guerin et al. 2014). The high rate of complete thrombus resolution in the majority of patients combined with the lack of precise knowledge of the risk factors for CTEPH do not justify the routine use of follow-up CTPA imaging in patients treated for acute PE (Den Exter et al. 2015). The interpretation of residual defects after acute PE is

 hen Acute PE Evolves W Toward Chronic PE

Whereas complete resolution of endoluminal clots represents the most frequent outcome of acute pulmonary embolism, some patients can develop chronic obstruction of the pulmonary circulation. Dependent on the severity of chronic obstruction, pulmonary hypertension can subsequently develop, related to incomplete resolution of clots or recurrent pulmonary embolism (Fig.  5). Chronic thromboembolic pulmonary hypertension (CTEPH) represents one of the leading causes of pulmonary hypertension, with a

406

not straightforward. Whereas it may be related to a slow rate of resolution of endoluminal clots, one should integrate the possibility of acute PE in the context of unknown chronic PE. Guerin et al. have recently underlined that a majority of patients with CTEPH had previously unknown pulmonary hypertension at the time of the acute event (Guerin et al. 2014). Six months after diagnosis of first or recurrent PE, residual pulmonary perfusion defects encountered on V/Q-SPECT were found to correspond with hypoperfusion on iodine map dual-energy CT in the majority of patients with chronic thromboembolic disease seen on dual-energy CT (de Broucker et al. 2012). Conclusion

CT pulmonary angiography is a well-recognized diagnostic tool but is also a unique means of providing prognostic information from the same examination as that used for diagnostic purposes. Radiologists should be aware of current trends in the management of patients with acute PE as this knowledge has direct influence on the content of their daily reports. RV function and information on the likelihood of an underlying cardiopulmonary disease is of the utmost importance for risk stratification. Lastly, CT is a rapidly evolving technology, and radiologists should regularly adapt their single energy scanning protocols and consider new options with dual-energy CT whenever available.

References Anderson DR, Kahn SR, Rodgers MA et al (2007) Computed tomographic pulmonary angiography vs ventilation-perfusion lung scanning in patients with suspected pulmonary embolism. JAMA 298:2743–2753 Araoz PA, Haramati LB, Mayo JR, Mortani Barbosa EJ Jr, Rybicki FJ, Colletti PM (2012) Panel discussion : pulmonary embolism imaging and outcomes. AJR Am J Roentgenol 198:1313–1319 Bauer RW, Frellesen C, Renker M et al (2011) Dual energy CT pulmonary blood volume assessment in acute pulmonary embolism – correlation with D-dimer level, right heart strain and clinical outcome. Eur Radiol 21:1914–1921 Becattini C, Agnelli G, Germini F, Vedovati MC (2014) Computed tomography to assess risk of death in acute

A. Hutt et al. pulmonary embolism: a meta-analysis. Eur Respir J 43:1678–1690 Bedayat A, Sewatkar R, Cai T et al (2015) Association between confidence level of acute pulmonary embolism diagnosis on CTPA images and clinical outcome. Acad Radiol 22:1555–1561 Blackmon KN, Florin C, Bogoni L, McCain JW, Koonce JD, Bastarrika G, Thilo C, Costello P, Salganicoff M, Schoepf UJ (2013) Computer-aided detection of pulmonary embolism at CT pulmonary angiography: can it improve performance of inexperienced readers? Eur Radiol 21:1214–1223 Bourjeily G, Chalhoub M, Phornphutkul C, Alleyne TC, Woodfield CA, Chen KK (2010) Neonatal thyroid function: effect of a single exposure to iodinated contrast medium in utero. Radiology 256:744–750 Cai XR, Feng YZ, Qiu L et al (2015) Iodine distribution map in dual-energy computed tomography pulmonary artery imaging with rapid kVp switching for the diagnosis analysis and quantitative evaluation of acute pulmonary embolism. Acad Radiol 22:743–751 Chae EJ, Seo JB, Jang MY et al (2010) Dual-energy CT for assessment of the severity of acute pulmonary embolism: pulmonary perfusion defect score compared with CT angiographic obstruction score and right ventricular/left ventricular diameter ratio. AJR Am J Roentgenol 194:604–610 Christner JA, Zavaletta VA, Eusemann CD, Walz-­Flannigan AI, McCollough CH (2010) Dose reduction in helical CT: dynamically adjustable z-axis X-ray beam collimation. AJR Am J Roentgenol 194:W49–W55 de Broucker T, Pontana F, Santangelo T, Faivre JB, Tacelli N, Delannoy-Deken V, Duhamel A, Remy J, Remy-­ Jardin M (2012) Single- and dual-source chest CT protocols: levels of radiation dose in routine clinical practice. Diagn Interv Imaging 93:852–858 Delesalle MA, Pontana F, Duhamel A, Faivre JB, Flohr T, Tacelli N, Remy J, Remy-Jardin M (2013) Spectral optimization of chest CT angiography with reduced iodine load: experience in 80 patients evaluated with dual-source, dual-energy CT. Radiology 267:256–266 Den Exter PL, Van Es J, LJM K et al (2015) Thromboembolic reoslution assessed by CT pulmonary angiography after treatment of acute pulmonary embolism. Thromb Haemost 114:26–34 Dewailly M, Remy-jardin M, Duhamel A et al (2010) Computer-aided detection of acute pulmonary embolism with 64-slice multidetector row computed tomography: impact of the scanning conditions and overall image quality in the detection of peripheral clots. J Comput Assist Tomogr 34:23–30 Dunne RM, Ip IK, Abbett S et al (2015) Effect of evidence-­ based clinical decision support on the use and yield of CT pulmonary angiographic imaging in hospitalized patients. Radiology 276:167–174 Eyer BA, Goodman LR, Washington L (2005) Clinicians’ response to radiologists’ reports of isolated subsegmental pulmonary embolism or inconclusive ­ interpretation of pulmonary embolism using MDCT. AJR Am J Roentgenol 184:623–628

CT of Pulmonary Embolism: Imaging Update Fedullo PF, Auger WR, Kerr KM et al (2001) Chronic thromboembolic pulmonary hypertension. N Engl J Med 345:1465–1472 Furlan A, Patil A, Park B et al (2011) Accuracy and reproducibility of blood clot burden quantification with pulmonary CT angiography. AJR Am J Roentgenol 196:516–523 Gorgos A, Remy-Jardin M, Duhamel A, Faivre JB, Tacelli N, Delannoy V, Remy J (2009) Evaluation of peripheral pulmonary arteries at 80 kV and 140 kV: dual-­energy computed tomography assessment in 51 patients. J Comput Assist Tomogr 33:981–986 Green DB, Raptis CA, Garin IAH, Bhalla S (2015) Negative computed tomography for acute pulmonary embolism: important differential diagnosis considerations for acute dyspnea. Radiol Clin North Am 53:789–799 Guerin L, Coutiraud F, Parent F et al (2014) Prevalence of chronic thromboembolic pulmonary hypertension after acute pulmonary embolism. Thromb Haemost 112:598–605 Hariharan P, Dudzinski DM, Rosovosky R et al (2016) Relation among clot burden, right-sided heart strain and adverse events after acute pulmonary embolism. Am J Cardiol 118:1568–1573 Hennelly KE, Baskin MN, Monuteuax MC et al (2016) Detection of pulmonary embolism in high-risk children. J Pediatr 178:214–218 Jimenez D, Billello KL, Murin S (2013) Point/ Counterpoint editorials. Should systemic lytic therapy be used for submassive pulmonary embolism? Yes. Chest 143:296–299 Kamel EM, Schmidt S, Doenz F, Adler-Etechami G, Schnyder P, Quanadli SD (2008) Computed tomographic angiography in acute pulmonary embolism: do we need multiplanar reconstructions to evaluate the right ventricular dysfunction? J Comput Assist Tomogr 32:438–443 Kaul D, Grupp U, Kahn J, Ghadjar P, Wiener E, Hamm B, Streitparth F (2014) Reducing radiation dose in the diagnosis of pulmonary embolism using adaptive statistical iterative reconstruction and lower tube potential in computed tomography. Eur Radiol 24:2685–2692 Kim MJ, Jung HO, Jung JI, Kim KJ, Jeon DS, Youn HJ (2014) CT-derived atrial and ventricular septal signs for risk stratification of patients with acute pulmonary embolism: clinical associations of CT-derived signs for prediction of short-term mortality. Int J Cardiovasc Imaging 30:25–32 Kligerman SJ, Lahiji K, Galvin JR et al (2014) Missed pulmonary emboli on CT angiography/ assessment with pulmonary embolism – computer-aided detection. AJR Am J Roentgenol 202:65–73 Kubo T, Lin PJP, Stiller W, Takahashi M, Kauczor HU, Ohno Y, Hatabu H (2008) Radiation dose reduction in chest CT: a review. AJR Am J Roentgenol 190:335–343 Kumamaru KK, Hunsaker AR, Bedayat A, Soga S, Signorelli J, Adams K, Wake N, Lu MT, Rybicki FJ (2012) Subjective assessment of right ventricle

407 enlargement from computed tomography pulmonary angiography images. Int J Cardiovasc Imaging 28:965–973 Kumamaru KK, Saboo SS, Aghayev A et al (2016) CT pulmonary angiography-based scoring system to predict the prognosis of acute pulmonary embolism. J Cardiovasc Comput Tomogr 10(6):473–479 Lankeit M, Konstantinides S (2012) Is it time for home treatment of pulmonary embolism? Eur Respir J 40:742–749 Lee EY, Tse SKS, Zurakowski D, Johnson VM, Lee NJ, Tracy DA, Boiselle PM (2012a) Children suspected of having pulmonary embolism: multidetector CT pulmonary angiography – thromboembolic risk factors and implications for appropriate use. Radiology 262:242–251 Lee EY, Neuman MI, Lee NJ, Johnson VM, Zurakowski D, Tracy DA, Boiselle PM (2012b) Pulmonary embolism detected by pulmonary MDCT angiography in older children and young adults: risk factor assessment. AJR Am J Roentgenol 198:1431–1437 Leung AN, Bull TM, Jaeschke R et al (2011) An official American Thoracic Society/Society of Thoracic radiology clinical practice guideline: evaluation of suspected pulmonary embolism in pregnancy. Am J Respir Crit Care Med 184:1200–1208 Lu MT, Demehri S, Cai T, Parast L, Hunsaker AR, Goldhaber SZ, Rybicki FJ (2012) Axial and reformatted four-chamber right ventricle-to-left ventricle diameter ratios on pulmonary CT angiography as predictors of death after acute pulmonary embolism. AJR Am J Roentgenol 198:1353–1360 Lu GM, Luo S, Meinel FG, McQuiston AD, Zhou CS et al (2014) High-pitch computed tomography pulmonary angiography with iterative reconstruction at 80 kVp and 20 mL contrast agent volume. Eur Radiol 24:3260–3268 Mamlouk MD, vanSonnenberg E, Gosalia R et al (2010) Pulmonary embolism at CT angiography: implications for appropriateness, cost and radiation exposure in 2003 patients. Radiology 256:625–632 Mastora I, Remy-Jardin M, Masson P, Galland E, Delannoy V, Bauchart JJ, Remy J (2003) Severity of acute pulmonary embolism: evaluation of a new spiral CT angiographic score in correlation with echocardiographic data. Eur Radiol 13:29–35 Matsuoka S, Hunsaker AR, Gill RR et al (2009) Vascular enhancement and image quality of MDCT pulmonary angiography in 400 cases: comparison of standard and low kilovoltage settings. AJR Am J Roentgenol 192:1651–1656 McLaughlin PD, Liang T, Homiedan M, Louis LJ, O’Connel TW, Krzymyk K, Nicolaou S, Mayo MR (2015) High pitch, low voltage dual source CT ­pulmonary angiography: assessment of image quality and diagnostic acceptability with hybrid iterative reconstruction. Emerg Radiol 22:117–123 Niemann T, Simon H, Faivre JB, Yasunaga K, Bendaoud S, Simeone A, Remy J, Duhamel A, Flohr T, Remy-­ Jardin M (2013) Clinical evaluation of automatic tube

408 voltage selection in chest CT angiography. Eur Radiol 23:2643–2651 Pengo V, Lensing AW, Prins MH et al (2004) Incidence of thromboembolic pulmonary hypertension after pulmonary embolism. N Engl J Med 350:2257–2264 Piazza G, Goldhaber SZ (2010) Management of submassive pulmonary embolism. Circulation 122:1124–1129 Pontana F, Chalayer C, Faivre JB, Murphy C, Remy-­ Jardin M, Remy J (2012) Pseudo-embolic perfusion defects in COPD: evaluation with dual-energy CT angiography (DECT) in 170 patients. Abstract B 0272, European Congress of Radiology, SS504 – CTTA - Dual energy and dose reduction Pontana F, Pagniez J, Duhamel A et al (2013) Reduced-­ dose low-voltage chest CT angiography with sinogram-­ affirmed iterative reconstruction versus standard-dose filtered back projection. Radiology 267:609–618 Quanadli SD, EI Hajjam M, Vieillard-Baron A, Joseph T, Mesurolle B, Oliva VL, Barré O, Bruckert F, Dubourg O, Lacombe P (2001) New CT index to quantify arterial obstruction in pulmonary embolism: comparison with angiographic index and echocardiography. AJR Am J Roentgenol 176:1415–1420 Raja AS, Ip IK, Prevedello LM et al (2012) Effect of computerized clinical decision support on the use and yield of CT pulmonary angiography in the emergency department. Radiology 262:468–474 Reid JH, Murchison JT (1998) Acute right ventricular dilatation: a new helical CT sign of massive pulmonary embolism. Clin Radiol 53:694–698 Remy-Jardin M, Pistolesi M, Goodman LR, Gefter WB, Gottschalk A, Mayo JR, Sostman HD (2007) Management of suspected acute pulmonary embolism in the era of CT angiography: a statement from the Fleischner Society. Radiology 245:315–329 Ridge CA, Mhuircheartaigh JN, Dodd JD, Skehan SJ (2011) Pulmonary CT angiography protocol adapted to the hemodynamic effects of pregnancy. AJR Am J Roentgenol 197:1058–1063 Ritchie G, McGurk S, Mc Creath C, Graham C, Murchison JT (2007) Prospective evaluation of unsuspected pulmonary embolism on contrast multidetector CT (MDCT) scanning. Thorax 62:536–540 Schueller-Weidekamm C, Schaefer-Prokop CM, Weber M, Herold CJ, Prokop M (2006) CT angiography of pulmonary arteries to detect pulmonary embolism: improvement of vascular enhancement with low kilovoltage settings. Radiology 241:899–907 Sigal-Cinqualbre AB, Hennequin R, Abada HT, Chen X, Paul JF (2004) Low-kilovoltage multi-detector row

A. Hutt et al. chest CT in adults: feasibility and effect on image quality and iodine dose. Radiology 231:169–174 Sodickson A, Weiss M (2012) Effect of patient size on radiation dose reduction and image quality in low-kVp CT pulmonary angiography performed with reduced IV contrast dose. Emerg Radiol 19:437–445 Stein PD, Goodman LR, Hull RD et al (2012) Diagnosis and management of isolated subsegmental pulmonary embolism: review and assessment of the options. Clin Appl Thromb Hemost 18:20–26 Szucs-Farkas Z, Kurmann L, Strautz T, Patak MA, Vock P, Schindera ST (2008) Patient exposure and image quality of low-dose pulmonary computed tomography angiography: comparison of 100- and 80-kVp protocols. Invest Radiol 43:871–876 Szucs-Farkas Z, Schibler F, Cullmann J et al (2011) Diagnostic accuracy of pulmonary CT angiography at low tube voltage: intraindividual comparison of a normal-­dose protocol at 120 kVp and a low-dose protocol at 80 kVp using reduced amount of contrast medium in a simulation study. AJR Am J Roentgenol 197:W852–W859 Thieme SF, Ashoori N, Bamberg F et al (2012) Severity assessment of pulmonary embolism using dual energy CT – correlation of a perfusion defect score with clinical and morphological parameters of blood oxygenation and right ventricular failure. Eur Radiol 22:269–278 Todd JL, Tapson VF (2009) Thrombolytic therapy for acute pulmonary embolism: a critical appraisal. Chest 135:1321–1329 Torbicki A, Perrier A, Konstantinides SV et al (2008) Guidelines on the diagnosis and management of acute pulmonary embolism: the task force for the diagnosis and management of acute pulmonary embolism of the European Society of Cardiology (ESC). Eur Heart J 29:2276–2315 Wittenberg R, Berger FH, Peters JF et al (2012) Acute pulmonary embolism: effect of a computer-assisted detection prototype on diagnosis – an observer study. Radiology 262:305–313 Woo JKH, Chiu RYW, Thakur Y, Mayo JR (2012) Risk-­ benefit analysis of pulmonary CT angiography in patients with suspected pulmonary embolus. AJR Am J Roentgenol 198:1332–1339 Yuan R, Shuman WP, Earls JP, Hague CJ, Mumtaz HA, Scott-Moncrieff A, Ellis JD, Mayo JR, Leipsic JA (2012) Reduced iodine load at CT pulmonary angiography with dual-energy monochromatic imaging: comparison with standard CT pulmonary angiography – a prospective randomized trial. Radiology 262:290–297

COPD Anna Rita Larici, Paola Franchi, Giuseppe Cicchetti, and Lorenzo Bonomo

Contents 1    Definition and Clinical Features of COPD

Abstract  410

2    Imaging 2.1  Identification and Phenotyping 2.2  Quantification 2.3  Role of Quantitative Imaging in Clinical Practice: The Present and the Future

 411  411  422

3    Exacerbation and Comorbidities

 426

References

 427

A.R. Larici, MD (*) • P. Franchi • G. Cicchetti • L. Bonomo Diagnostic Imaging, Catholic University of the Sacred Heart, Policlinico A. Gemelli Foundation, Largo A. Gemelli, 8, 00168 Rome, Italy e-mail: [email protected]; [email protected]

 425

Chronic obstructive pulmonary disease (COPD) is a common preventable and treatable disease characterized by persistent airflow limitation associated with an enhanced chronic inflammatory response to noxious particles or gases in the lungs and airways. Airflow limitation that characterizes COPD is caused by a combination of emphysema-induced loss of elastic recoil and small airway chronic inflammation and remodeling. Their contribution to the disease largely varies among patients, and current spirometric criteria are inadequate to assess the predominant contributor. In this context imaging plays an essential role. Chest radiography is a simple tool for diagnosing moderate-to-severe emphysema, and it can be suitable for phenotyping COPD in severe cases. Multidetector CT (MDCT) is currently the most widely available and precise imaging method for diagnosing and characterizing the morphological phenotypes of COPD, also in the early stage. MDCT allows the automatic quantification of the presence and percentage of emphysematous lung, the lobar and zonal distribution of the low-attenuation areas, the changes in airway wall and luminal caliber, and the severity of air trapping due to small airway involvement. MDCT provides information on significant concomitant pulmonary and extrapulmonary diseases and can be employed to plan interventional procedures in patients with severe disease. Imaging quantitative mea-

Med Radiol Diagn Imaging (2017) DOI 10.1007/174_2017_10, © Springer International Publishing AG Published Online: 04 March 2017

409

A.R. Larici et al.

410

surements of COPD phenotype have also been associated with exacerbation frequency and risk of lung cancer. In the future imaging quantitative analysis will enable the identification and stratification of patients with certain disease features for clinical trials focused on targeted treatments, thus improving patient management.

1

 efinition and Clinical D Features of COPD

Chronic obstructive pulmonary disease (COPD) is a major cause of morbidity and mortality throughout the world. According to the World Health Organization, by the year 2030, COPD will be the third leading cause of death worldwide (Organization WH 2010). COPD is a common preventable and treatable disease characterized by persistent airflow limitation that is usually progressive and associated with an enhanced chronic inflammatory response to noxious particles or gases in the airways and the lungs. Spirometry is required to make a clinical diagnosis of COPD. Usually, a value of forced expiratory volume in 1 s (FEV1) to forced vital capacity (FVC) ratio 1 and PAD >30 mm, even though CT cannot exclude reliably a PH if the PAD is less than 30 mm, particularly in patients with end-stage COPD (Hoesein et al. 2016). Enlargement of the pulmonary artery at CT scans was shown to be an independent risk factor for exacerbations in patients with COPD (Wells et al. 2012).

2.2

Quantification

Recent advances in MDCT technology and software implementation permit automated detection and quantification of pulmonary emphysema and airway disease in COPD patients. The goals of quantitative CT (QCT) in COPD are to automatically measure the presence and percentage of emphysema-like lung (expresses as low-attenuation areas), the lobar and zonal distribution of the low-attenuation areas, the changes in airway wall and luminal caliber, and the severity of gas trapping at expiratory CT (Lynch et al. 2015). Emphysematous lung destruction results in replacement of normal lung (which has a typical attenuation about −850 HU on inspiratory CT scans) by air-containing spaces, having a CT attenuation close to −1,000 HU. Since the early assessment of the lung density at CT, it was evident that the measurement of attenuation values could help in quantification of the emphysema extent (Lynch and Al-Quaisi 2013). The main technique to quantify emphysema, termed CT densitometry, consists in applying a density mask to the lung parenchyma and setting a threshold below which all voxels are assumed to be emphysema, giving a low-attenuation areas percentage (%LAA) (Fig. 10.10d–f). Madani et al. demonstrated that the highest correlation between quantitative CT metrics and histology is

found when the CT threshold is set at −960 or −970 HU (Madani et al. 2006). However, in the interests of balancing sensitivity and specificity, the threshold of −950 HU is now most commonly used (Coxson et al. 2013) (Fig. 10.10). An alternative approach to emphysema quantification, termed percentile densitometry, implies choosing a threshold percentile in the attenuation distribution curve, which provides the density value (in HU) under which a percentage of the voxels are distributed. A number of different thresholds between 1% and 18% have been used and correlate strongly with microscopic emphysema on histological specimens (Madani et al. 2006). The most commonly used threshold is 15% (Heussel et al. 2009). There is still no definite consensus about which of these methods is best, and, even now, different studies use different thresholds. There is some evidence that the percentile approach is more robust for longitudinal evaluation of emphysema and less sensitive to change in lung volume. However, given the fact that emphysema is an all-or-nothing phenomenon, it seems more intuitive to use CT densitometry, which uses an absolute cutoff to quantify it (Ostridge et al. 2016). Since emphysema is a regionally distributed disease, it makes sense to determine the zonal or lobar distribution of emphysema. Most available QCT methods can divide each lung into upper, mid, and lower zones of equal height or volume, and ratios between upper and lower lung LAA measurements can be computed. Newer methods also permit automatic or semiautomatic segmentation of the lung lobes and to perform a per-lobe and all-lung quantification of volumes and extent of the LAA with respect to the normal lung parenchyma (by subtracting the volume of emphysema from the total volume) (Lynch and Al-Quaisi 2013). Thanks to recent development of dedicated software, it is possible to generate two-dimensional multiplanar reconstructions and three-dimensional volume rendering (VR) reconstructions used for quantification (Chen-Yoshikawa and Date 2016) (Fig. 10.10a–c). The application of this imaging procedure in the preoperative evaluation for lung resection in

COPD

423

a

b

d

c

e

Interval -950 HU > -950 HU

Values [–1024 / –950] [–950 / –200] Totale

f

Right Lung

Left Lung

Total Lung Volume

50,4074 %

46,2724 %

48,5263 % / 3,0836 L

49,5926 %

53,7276 %

51,4737 % / 3,2708 L

3,4637 L

2,8907 L

6, 3544 L

Fig. 10.10  CT densitometry in a patient with advanced destructive centrilobular emphysema. By setting a threshold value of −950 HU in inspiratory scans, the software automatically assesses and quantifies the emphysema extent in the middle-upper portion of both lungs (blue areas) (d–f). Software semiautomatically performs a per-­

lobe analysis and provides a quantification of the lung volume and the emphysema percentage. Lung lobes are visually represented with different colors in different planes on MPR and VR reconstructions (a–c). Volumetry of one lobe (right lower lobe) is showed in the coronal VR reconstruction (c)

COPD candidates allows a “virtual lobectomy” similar to the surgical one and provides quantitative data (volumes) necessary to derive parameters predictive of postoperative lung reserve and function (MDCT-virtual lobectomy). Furthermore, there is convincing evidence that CT densitometry has a strong association with airflow obstruction, and analysis of over 4,000 CT scans from the COPDGene study showed a strong negative correlation between %LAA under −950 HU (%LAA−950) and FEV1, demonstrating an increase in the emphysema index with the worsening of GOLD severity (Schroeder et al. 2013). CT densitometry also

has strong associations with gas transfer (transfer factor of the lung for carbon monoxide, DLCO) (Diaz et al. 2010), 6-min walk distance test (Diaz et al. 2010), BODE score (body mass index, obstruction, dyspnea, exercise capacity) (Martinez et al. 2012), and body composition (Rutten et al. 2011). CT densitometry has also been found to correlate with mortality in patients with COPD (Haruna et al. 2010), with %LAA having a better predictive value for respiratory and cardiac mortality than GOLD staging (Johannessen et al. 2013). Emphysematous changes on CT have been associated with higher exacerbation rate

A.R. Larici et al.

424

and increased mortality from exacerbations (Jairam et al. 2015). A potentially important application of CT imaging is to track the changes in emphysema over time and see how it contributes to the lung functional decline. Until recently, the majority of these studies were conducted in patients with α1-antitrypsin deficiency and have shown that CT densitometry can accurately detect progression of emphysema and correlates with functional decline (Dirksen et al. 2009). The percentage of emphysema on the follow-up scans may be corrected for lung volume using the volume achieved on the baseline scan (Bakker et al. 2005). End-expiratory CT, as already said, is an excellent way to assess air trapping in COPD that is a key finding for depicting small airway obstruction on CT scans (Sverzellati et al. 2007). Most studies have evaluated the presence of gas trapping by calculating the % of low-attenuation areas at a threshold of −856 or −850 HU (LAAexp-856 or LAAexp-850) (Murphy et al. 2012). This value is chosen because it is the attenuation of normally inflated inspiratory lung, so that it is assumed that expiratory lung should have higher attenuation than this. Other authors have used different indices of gas trapping, including the ratio of inspiratory to a

b

expiratory lung volume, expiratory to inspiratory lung attenuation ratio (E/I-ratioMLD), and the expiratory to inspiratory relative volume change of voxels with attenuation values between −860 and −950 (RVC-860 to −950) (Mets et al. 2012). Quantitative CT evaluation of severity of emphysema and expiratory gas trapping provides a simple way to assign individual COPD subjects to subgroups characterized by predominant emphysema, mixed emphysema and air trapping, or predominant air trapping. Airway wall remodeling is an important feature in COPD, and histological specimens confirm airway wall thickening throughout the bronchial tree. The limited resolution of CT means only large- and intermediate-sized airways can be visualized directly. Early measurements of the airway relied on manual tracing; however, a number of automated methods have been developed. Dedicated software let automated extraction of bronchial tree, beginning from the trachea and extending into the sixth to eighth bronchial generation, and allow measurements of airway lumen area, inner and outer diameters, wall thickness, and wall area (Mayer et al. 2004; Tschirren et al. 2005) (Fig. 10.11). Some studies demonstrated that MDCT with automatic measurement of the airway c

d

Fig. 10.11 Automated software airway extraction. Automated detection of the bronchial tree (a) and representation of the right lower lobe bronchus in an oblique coronal view (b). (d) Image showing the right lower lobe

bronchus stretched along its course. A true axial image (c) derived from the stretched reconstruction of the bronchus is used to measure airway parameters

COPD

parameters may have a role in quantification of airflow obstruction in COPD. In particular Hasegawa et al. (2006) demonstrated high correlations between the airway luminal area and wall thickness with FEV1. Matsuoka et al. (2008) analyzed the relationship between airflow limitation and airway dimensions from the third to the fifth generation of bronchi in COPD patients by using inspiratory and expiratory MDCT scans and demonstrated a good correlation between dynamic modification of distal airway luminal area (expressed by the ratio of expiratory to inspiratory airway luminal area) and FEV1. Recent studies have shown that both airway wall and lumen size are reduced in COPD, although proportionally resulting in a larger wall area percentage (wall area/(lumen + wall area) (Smith et al. 2014; Washko et al. 2014). Bronchial wall markers also correlate with functional markers in the form of the BODE index (Martinez et al. 2012), exercise capacity (Rambod et al. 2012), and body composition (Rutten et al. 2011). On the other hand, simply measuring markers of bronchial wall thickness may be insufficient to describe the airway remodeling that occurs in COPD. In theory, CT should be a useful tool for assessing the morphology of the large- and intermediate-­sized airways. However, there are still many uncertainties regarding this technology and measurements, and no definite evidence has shown that this is a particularly useful tool in COPD (Ostridge et al. 2016). There is increasing interest in using more sophisticated textural analysis to evaluate smoking-­related lung injury, including emphysema. Ginsburg et al. (2012) showed that a texture-­based approach could discriminate quite effectively between the lungs of never-smokers, smokers without emphysema, and smokers with emphysema. This suggests that textural analysis may be able to identify the early phase of smoking-­related lung injury, prior to the development of emphysema. Conventional anatomical CT cannot provide direct functional information regarding which parts of the lung receive ventilation or perfusion. Other techniques such as dual-energy CT (DECT)

425

have been evaluated in this context. A recent study investigating DECT with combined xenon-­ enhanced ventilation and iodine contrast-­ enhanced perfusion imaging demonstrated the feasibility of obtaining regional and quantitative ventilation and perfusion measurements and, importantly, the measurements of the ­ventilation-­perfusion relationship on a voxel-byvoxel basis (Hwang et al. 2016). In patients with COPD, DECT ventilation-perfusion measurements were shown to be significantly associated with measures of pulmonary function (Hwang et al. 2016). Although quantitative CT provides useful information regarding emphysema, airways, and air trapping and provides a means of objectively characterizing and following these pathologic processes, visual assessment of CT scans remains important to describe patterns of altered lung structure in COPD and provides distinct phenotypes not currently identified with quantitative CT (Lynch et al. 2015).

2.3

 ole of Quantitative Imaging R in Clinical Practice: The Present and the Future

Imaging has an important and rapidly developing role to play in the investigation of COPD. MDCT aids diagnosis and provides information on significant concomitant disease and can be employed to plan interventional procedures for those patients with severe disease. Furthermore, quantitative analysis techniques permit objective measurements not only of pulmonary manifestations of the disease, as mentioned above, but also of extrapulmonary manifestations. Indeed COPD has multiple systemic manifestations, and CT of the chest is able to capture some of this information. Cachexia and skeletal muscle wasting are a significant problem in the clinical context of COPD. A technique to assess pectoralis muscle area on CT showed a significantly reduction of muscle area in COPD patients versus healthy controls, and this data has been associated with GOLD staging (McDonald et al. 2014). Various fat compartments can also be

A.R. Larici et al.

426

assessed by CT, including anterior chest wall subcutaneous fat and intra-abdominal fat, even if these methods are not standardized yet. Osteoporosis is also common in COPD. Vertebral body attenuation values can be measured by CT and is associated with bone mineral density on dual-energy X-ray absorptiometry (Romme et al. 2012). These techniques can give important insights into COPD and help explore the heterogeneity and underlying mechanisms of this condition. In time, it is hoped that these techniques can be used in clinical trials to help the development of disease-­specific therapy and, ultimately, as a clinical tool in identifying patients who would benefit most from new and existing treatments (Ostridge et al. 2016). International clinical guidelines reinforce the opinion that COPD is a manageable and treatable disease for patients with all grades of disease severity. Precision medicine depends on a deeper understanding of the individual patient’s genetic background and the gene-environment interactions that lead to specific underlying biological processes – the disease “endotype” (i.e., inflammation). By understanding how these distinct underlying biological processes manifest in the individual patient – the disease “phenotype” (i.e., emphysema) – it is thought that precision medicine will enable the “right” treatment to be provided to the right patient at the right time. A better understanding of the underlying disease endotypes will not only allow for better patient management but will also open the doors to discover new drugs/interventions and to start clinical trials that target the underlying diseases embodied in chronic lung disease. Furthermore, linking the endotype to the phenotype will enable the identification and stratification of patients with certain disease features for clinical trials focused on targeted treatments, as well as on the evaluation of treatment efficacy. In light of this, there is renewed interest in the development and evaluation of the COPD phenotype measurements, which currently include clinical and functional biomarkers as well as the recently discovered pulmonary imaging biomarkers (Kirby et al. 2016).

To validate CT imaging as a biomarker of disease, it is necessary to compare it to underlying disease mechanisms and outcome measures and track longitudinal changes.

3

Exacerbation and Comorbidities

Acute exacerbations of COPD are increasingly being recognized as a major and increasing burden to patients. Exacerbations are associated with impaired quality of life, a more rapid decline in lung function, and higher mortality. Being able to predict which patients are at the greatest risk for acute COPD exacerbations will enable healthcare providers to better target these individuals for preventive therapy. Also in this case, spirometry is inadequate as the sole procedure for risk assessment of COPD exacerbations. There are subsets of patients who have a severely reduced FEV1 but do not experience frequent exacerbations. Recently, Han et al. (2011) highlighted that the quantitative measures of lung structural changes identified with volumetric CT are associated with COPD exacerbation frequency, a clinical outcome of public health importance. By summarizing, independent of the severity of airflow obstruction, a 5% increase in total lung emphysema in those with 35% or greater emphysema is associated with a 1.18-fold increase in COPD exacerbation frequency; a 1-mm increase in segmental airway wall thickness is associated with a 1.84-fold increase in COPD exacerbation frequency (Han et al. 2011). As regards comorbidities, it is well demonstrated that COPD and lung cancer share a common risk factor, cigarette smoking, and that COPD is associated with a two- to fourfold increased risk for lung cancer regardless of smoking habits, even among never-smokers (Turner et al. 2007). In a recent study by Schwartz (2016), focused on the risk of lung cancer associated with COPD phenotypes evaluated on quantitative image analysis, lung cancer risk was significantly and consistently associated with each of the COPD measures. The odds of lung cancer were increased approximately 1.4- to 3.1-fold among those with COPD com-

COPD

pared to those without. Both quantitative CT air trapping and FEV1/FVC 180° Present or absent

Abutment: ≤180° Encasement: >180°

Tethering or tear drop

Present or absent

Note: MPD main pancreatic duct, SMV superior mesenteric vein, SMA superior mesenteric artery, RHA right hepatic artery, CHA common hepatic artery, MPV main portal vein

J.M. Lee and H.-J. Kang

498

3.1

Morphologic Evaluation of Main Tumor and Secondary Signs

On CT, PDACs most often present as an ill-­defined, solid hypoattenuating masses compared to normal pancreatic parenchyma (Francis 2007) (Figs. 2 and 4). Pancreatic cancers can occasionally appear to be cystic or necrotic, and in rare cases they can contain calcium (Callery et al. 2009) Cystic-necrotic degeneration, an uncommon feature of adenocarcinoma, is present in 8% of cases (Kosmahl et al. 2005). With recent MDCT technology and increasing use of CT for evaluation of abdominal abnormalities, it is not rare to encounter non-contour deforming tumors, which underscores the importance of pancreatic phase images in which the lesion to background pancreatic parenchymal contrast is greatest (Fletcher et al. 2003) (Fig. 4).

a

Approximately 5.4–10% of pancreatic adenocarcinomas are isoattenuating relative to the background pancreatic parenchyma (Prokesch et al. 2002a; Kim et al. 2010), especially in small tumors 2 cm or less (Yoon et al. 2011), and thus making diagnosis more difficult. In these situations, indirect (secondary) signs such as upstream pancreatic duct dilation or the double-duct sign caused by pancreatic and common bile duct obstruction are helpful for the diagnosis (Yoon et al. 2011; Francis 2007; Blouhos et al. 2015). In addition, other secondary signs of pancreatic cancer include focal pancreatic enlargement, extension of tumor beyond the pancreas, and upstream pancreatic atrophy secondary to ductal obstruction (Tamm et al. 2013). As the tumor reaches its advanced stage, it typically infiltrates the peripancreatic structures and involves adjacent vasculature, and in some cases adjacent organs.

b

c

Fig. 4  MDCT of a 72-year-old woman with clearly resectable pancreas cancer. (a) Axial CT scan that obtained during pancreatic phase shows a 2.5 cm hypovascular mass is in the pancreas head, which is confined to pancreatic parenchyma (arrow). (b) This mass show delayed enhance-

ment in portal phase and major vessels including the superior mesenteric artery (arrowhead) and superior mesenteric vein (empty arrow) show the clear fat plane with the mass. (c) Even though the mass invade duodenal second portion (arrows), it does not affect tumor resectability

Pancreatic Tumors

3.2

Vascular Evaluation

In the absence of distant metastasis, the presence of degree of contact between the tumor and the peripancreatic vessels is of paramount importance in determining surgical resectability. In addition, it is important to recognize variants of vascular anatomy such as celiac and mesenteric arterial variants and variants of SMV-PV in the preoperative planning of extended pancreatic resection (Zakharova et al. 2012). According to the National Comprehensive Cancer Network (NCCN) guideline, less than or equal to 180° tumor contact of the vessel circumferential is described as “abutment” (Fig. 5) and more than 180° tumor contact of the vessel circumference is referred to as “encasement” (Fig. 6). The utility of these terms includes the ability to differentiate clearly resectable tumor, from “borderline resectable tumor, from clearly unresectable tumor

a

499

(Tamm et al. 2012; Varadhachary et al. 2006). According to the previous study by Lu et al. (1997), more than 180° of tumor-vessel contact is highly specific (a sensitivity of 84% and specificity of 98%) for vascular invasion by the tumor and for tumor unresectability if the involved vessels are either celiac artery or superior mesenteric artery (Fig. 6). In addition, another sign of vascular invasion by pancreatic cancers is irregularity of the vessel contour (including “tear drop” deformity) or changes in caliber, and when irregularity of the vessel contour is seen, regardless of the degree of contact between tumor and vessel, vascular invasion should be considered (Wong and Lu 2008; Brugel et al. 2004). The irregularity of vessel contour by vascular invasion occurred more often than that of the artery, because the wall of the vein is much thinner and weaker than the wall of the artery (Li et al. 2005). On the contrary, as the wall of the artery is thicker and more

b

c

Fig. 5  A 78-year-old woman with borderline resectable pancreas cancer. Approximately 3 cm hypovascular mass (arrow) is seen in the pancreas head, and it abuts SMA (arrowhead) in 80° in late arterial phase image (a) and

encases SMV (empty arrow) with subsequent luminal narrowing in portal phase (b). Length of SMV involvement (empty arrows) is 1.5 cm on the coronal image (c)

J.M. Lee and H.-J. Kang

500

a

b

c

d

Fig. 6  A 54-year-old woman with unresectable pancreas cancer. (a) Approximately 5 cm hypovascular soft tissue mass (arrows) is seen in the pancreas body, and it abuts to the superior mesenteric artery (arrowheads). Axial (b) and oblique coronal MPR images (c) display diffuse soft

tissue infiltration encasing the proper hepatic artery (arrows). On coronal MIP image (d), the main portal vein and proximal superior mesenteric vein show severe luminal narrowing (arrow) over 2 cm due to tumor involvement and the splenic vein is not opacified

flexible than the vein wall, invaded arteries may show regular wall, and may appear stretched on MDCT because of the presence of focal tissue fibrosis (Zakharova et al. 2012). In addition, it would be helpful to evaluate the presence and pattern of collateral venous channels surrounding the pancreatic head such as short gastric varices, gastrohepatic ligament varices, and gastroepiploic to gastrocolic trunk, as CT may fail to reveal any direct contact with the vein (Vedantham et al. 1998; Yamada et al. 2000). With technological developments of MDCT, perilymphatic and perineural tumor infiltration of pancreatic cancer can be depicted as “reticular,” “tubular,” or “soft tissue mass” appearances in the peripancreatic fat tissues on high-quality thin-section imaging (Sai et al. 2010; Makino

et al. 2008; Deshmukh et al. 2010). When tumor spread along these pathways is suspected, the surgeon should be alerted, and neoadjuvant therapy can be considered before surgery.

3.3

Extrapancreatic Evaluation

The presence of extrapancreatic tumor extension either local or distant, need to be cautiously evaluated as it can affect the surgical decision-­ making. If focal hepatic lesions are present that demonstrate suspicious features concerning for metastasis (poorly defined margins, rim enhancement) or are indeterminate if the lesion is too small to characterize by means of CT, then further imaging such as MRI or tissue sampling to

Pancreatic Tumors

arrive at a final diagnosis may be warranted (Al-Hawary et al. 2014). With regard to lymph node staging, the presence and location of suspicious lymph nodes (defined as short axis >1 cm, abnormal round morphology, heterogeneity, or central necrosis) should be noted (Al-Hawary et al. 2014). This is especially true for enlarged lymph nodes which are outside the immediate local drainage pathways based on tumor location (i.e., aortocaval or paraaortic lymph nodes), as these can alter s­ taging from local node involvement to metastatic disease. However, unfortunately, both CT and MRI are not accurate at lymph node staging in patients with pancreatic cancer (Kauhanen et al. 2009).

501

3.4

Recently published pancreatic adenocarcinoma oncology guidelines by the NCCN describe grouping patients based on radiographic criteria into those with clearly resectable disease, borderline resectable disease, or clearly unresectable disease (Tempero et al. 2014) (Figs. 5, 6, and 7). Clearly resectable disease corresponds to AJCC stages I and II (Fig. 5), while clearly unresectable disease represents AJCC stage III and IV (Edge et al. 2010) (Fig. 7). According to the NCCN guideline, borderline resectable patients have no distant

a

b

c

d

Fig. 7  A 53-year-old man who underwent neoadjuvant concurrent chemoradiation therapy (CCRTx) for histologically confirmed pancreatic adenocarcinoma. (a) Axial CT scan that obtained during pancreatic phase demonstrates approximately 3 cm hypovascular mass (arrow) in the pancreas body which widely abuts the celiac trunk between 90° and 180° (arrowhead). (b) The mass show an increased maximum standardized uptake value (4.8) on

 erformance of CT for Staging P and Resectability of Pancreatic Cancer

PET-CT fusion image. (c, d) After CCRTx, the mass still abuts the celiac trunk (arrowhead) without tumor size change. However, the maximum standardized uptake value of the mass (arrow) is decreased on PET-CT fusion image. After surgical resection, there is no celiac trunk invasion on operative findings and pathologic examination of the resected specimen confirmed no tumor involvement of the celiac trunk

J.M. Lee and H.-J. Kang

502

metastases, short segmental venous involvement with suitable vessel above and below the point of involvement allowing for safe and complete resection and vein reconstruction, SMA and CA abutment (≤180° of the circumference of involvement) and CHA involvement without extension to CA or HA bifurcation (Tempero et al. 2014) (Fig. 6). As of now, there are various definitions of “borderline resectable” pancreatic cancers which have been proposed by different organizations, and the consensus is not yet reached. MDCT is very effective for detecting and staging adenocarcinoma, with a sensitivity of up to 90% for the detection and an accuracy of 80–90% for the staging (Schima et al. 2007). In addition, MDCT has shown excellent performance for evaluating vascular involvement thanks to its high spatial resolution and good delineation of perivascular fat plane in many studies (Lepanto et al. 2002; Zhao et al. 2009; Li et al. 2006; Vargas et al. 2004; Lu et al. 1997; Manak et al. 2009; Gusmini et al. 2009; Catalano et al. 2003). However, it is of important note that the positive predictive value of CT for determining nonresectability based on vascular involvement is very high (89–100%), but it is lower for predicting resectability (45–79%) (Al-Hawary et al. 2014; Lu et al. 1997; Wong and Lu 2008; Brugel et al. 2004; Chen et al. 2016; Tamm et al. 2012; Park et al. 2009). This is because the diagnostic criteria for vascular invasion have been developed for being more specific than sensitive to minimize the number of patients inappropriately denied surgery and potential cure (Li et al. 2006). But with improving MDCT technology, CT has shown a higher diagnostic accuracy, and the pooled sensitivity and specificity increased to 85% and 82%, respectively (Zhao et al. 2009). However, a serious diagnostic dilemma occurs following neoadjuvant chemotherapy and radiation therapy, as the vascular contact by pancreatic cancer may be replaced by perivascular haziness or fat stranding (Fig. 7). In fact, as perivascular haziness developed after neoadjuvant treatments can be caused by either posttreatment fibrosis or viable tumor, neoadjuvant therapy significantly decreases the accuracy of CT scan in determining resectability R0 of pancreatic carcinoma, and results in an overestimation of vascular invasion (Ly and Miller 2002; Cassinotto et al. 2013).

Therefore, given that overestimation of vascular invasion may significantly reduce CT scan specificity for resectability after preoperative treatment (Kim et al. 2009; Morgan et al. 2010; Cassinotto et al. 2013), increased hazy attenuation or stranding contact with the major peripancreatic vessels in patients with prior radiation therapy or combined chemoradiation therapy need to be considered in conjunction with the treatment response of the main tumor and changes of tumor markers such as CA-19-9 (Fig.  7). In addition, baseline studies are useful for identifying the extent of the tumor before radiation therapy (Tamm et al. 2006), and partial regression of tumor-vessel contact may indicate suitability for surgical exploration (Cassinotto et al. 2014). As of now, however, there are no clear diagnostic criteria to differentiate perivascular invasion from tumor progression from posttreatment fibrosis after neoadjuvant treatments.

4

Pancreatic Neuroendocrine Tumors

Pancreatic neuroendocrine tumors (PNETs) are a group of tumors that arise from the endocrine pancreas and account for 1–10% of all pancreatic tumors. Although PNETs may produce distinct clinical syndromes, most of these neoplasms are asymptomatic and present as an incidental finding (Amin and Kim 2016). Functional tumors present with a defined clinical syndrome secondary to hormone hypersecretion, and included insulinomas, gastrinomas, glucagonomas, VIPomas, and somatostatinomas. Nonfunctional tumors, compromise 60–90% of all PNETs, and are often diagnosed later in the course of the disease when they present with symptoms of compression or metastatic disease (Amin and Kim 2016). Functional tumors represent the minority of PNETs, and insulinomas are the most common followed by gastrinomas. Glucagonomas, somatostatinomas, and VIPomas are rare and represent fewer than 10% of all functional PNETs. PNETs can present at any age; however, the incidence peaks in the sixth and seventh decades (Halfdanarson et al. 2008). PNETs are found with equal frequency throughout the pancreas. Although 90% of PNETs occur sporadically,

Pancreatic Tumors

these tumors are also well-recognized features of four familial syndromes: multiple endocrine neoplasia type I (MEN1), von Hippel–Lindau syndrome (VHL), neurofibromatosis type 1 (NF1), and tuberous sclerosis complex (TSC) (Amin and Kim 2016). Tumors tend to be multiple, especially when associated with syndromes such as multiple endocrine neoplasia type 1 and von Hippel–Lindau syndrome. Single lesions are seen in 90% of insulinomas, whereas multiple lesions are present in 20–40% of gastrinomas (Low et al. 2011). Regarding classification of PNET, the 2010 WHO classification of PNETs is based on the proliferative index of the tumor as measured by the Ki-67 index: endocrine tumors with a Ki-67 index ≤2% (G1, well-differentiated endocrine tumor), those with a Ki-67 index from 2% to 20% (G2, well-differentiated endocrine carcinoma), and endocrine carcinomas with a Ki-67 index a

503

>20% (G3, poorly differentiated neuroendocrine carcinoma) (Ricci et al. 2016). Several imaging modalities exist to aid the clinician in the localization and staging of PNETs. MDCT, MRI, endoscopic ultrasound (EUS), somatostatin-receptor scintigraphy (SRS), and functional PET (fPET) are all commonly used in clinical practice (Amin and Kim 2016). Among them, MDCT is widely available, and is commonly used as the first line imaging modality for detection of PNET. The most common imaging finding of PNET is a hyperenhancing solid tumor beset seen in the pancreatic phase of contrast-­ enhanced MDCT, and 5–10% of PNETs can show cystic tumor appearance with thing enhancing peripheral rim (Fig. 8). In addition, they can show calcifications (20–50%) and may show unusual appearance including central necrosis or cystic degeneration, venous tumor thrombus, and b

c

Fig. 8  A 59-year-old man with grade I PNET. (a, b) Axial CT scan during the pancreatic phase (a) demonstrates a 0.8 cm hypervascular mass (arrow) in pancreas body, which becomes isoenhanced during the portal venous phase (b) without dilatation of the main pancreatic

duct. (c) This mass present hypermetabolism on Ga-68 DOTA-TOC PET (arrow). Pathologic evaluation of the resected specimen confirmed this lesion as grade I PNET with 1.02% of Ki-67 index

J.M. Lee and H.-J. Kang

504

a

b

c

d

Fig. 9  A 57-year-old woman with multiple endocrine neoplasia (MEN) type I and multiple pancreatic neuroendocrine tumors (PNETs). (a) Axial postcontrast CT scan shows a 2 cm peripheral enhancing mass with cystic degeneration in the pancreas neck (arrow), and another 1.5 cm hyperenhancing solid mass in pancreas body-tail

junction (arrowhead). (b) CT scan that obtained inferior to (a) shows another 1 cm hypervascular mass in the pancreas uncinate process (empty arrow). (c, d) All lesions present hypermetabolism on Ga-68 DOTA-TOC PET. Pathologic examinations of the surgical specimen revealed that all three lesions were grade I neuroendocrine tumors

intraductal tumor within the main pancreatic duct (Balachandran et al. 2014) (Fig. 9). The staging system for PNET is the same as that for pancreatic ductal adenocarcinoma.

sensitivity in nonfunctioning PNETs because of their larger sizes at presentation. PNETs have a rich vascular supply and therefore enhance avidly during the arterial phase, enhancing more rapidly and intensely than the normal pancreas (Low et al. 2011). Nonfunctioning PNETs typically demonstrate heterogeneous enhancement. Small PNETs (90

25%

surgical treatment (Stassen et al. 2012). Nonoperative management is preferred in the hemodynamically stable patient. However, there is a higher failure rate with increasing grade of injury. The preferred treatment for hemodynamically stable patients with AAST

grade IV or V injury is still controversial. Nonoperative management contains splenic artery embolization (SAE) and surveillance. SAE is indicated if active extravasation or a pseudoaneurysm/AV fistula is evident on CT (Olthof et al. 2013). SAE is performed either as coil embolization of the proximal splenic artery or as superselective embolization of the bleeding arterial branch. Complications of SAE are persistent hemorrhage, coil migration, and splenic infarction or abscess. The rate of severe complications is less than 4% (Skattum et al. 2013; Frandon et al. 2014). Surveillance is a treatment option in hemodynamically stable patients without active contrast extravasation or blush in CT and AAST grade I-III. A study by Saksobhavivat et al. showed that the CT-based classification scale proposed by Marmery et al. was the best individual predictor for decision making between surveillance and splenic intervention in hemodynamically stable patients (Saksobhavivat et al. 2015). A close monitoring is required due to the risk of secondary splenic rupture. The risk of secondary rupture declines over time, and 92% of secondary ruptures take place in the first 6 days after injury (Peitzman

Spleen

563

a

b

c

Fig. 10  AAST V and Baltimore 4b: shattered spleen (a) with active intraperitoneal bleeding (b and arrows in c)

et al. 2000). Follow-up CT imaging should only be performed selectively based on the patient’s clinical status. The duration of observation should be based on the clinical presentation and severity of the initial injury.

3

Benign Lesions

3.1

General Principles

Benign splenic lesions expose certain specific imaging characteristics. They are usually well circumscribed, and the majority is hypodense to the splenic parenchyma in unenhanced and early contrast medium phases (Cave: hamartoma).

3.2

Cysts

Splenic cysts are the most common focal lesions of the spleen. They can be subdivided into primary/congenital cysts and secondary cysts. Furthermore, from an etiological standpoint, they can be divided into parasitic and nonparasitic cysts. About 20% of splenic cystic lesions are primary cysts, which are typically incidental findings at imaging. They are true cysts lined with epithelial cells. Secondary splenic cysts or pseudocysts account for about 80% of splenic cystic lesions and are typically posttraumatic after hematoma or splenic infarction. Secondary cysts present more often calcifications compared to primary cysts. In CT

A. Euler and S.T. Schindera

564

a

b

Fig. 11  Axial image of a simple splenic cyst in unenhanced (a) and portal-venous phase (b) as a round, well-­demarcated, hypodense lesion without enhancement after application of contrast medium

imaging, both types present as well-defined, rounded masses with w ­ ater-­equivalent attenuation (Fig. 11). Cysts do not exhibit an enhancement in contrast-enhanced phases. Differential diagnoses of nonparasitic cysts are splenic abscess and splenic hydatid cyst. Here, the patient’s history can help to differentiate among these entities.

3.3

Epidermoid Cyst

Splenic epidermoid cyst belongs to the family of primary, true splenic cysts. They are congenital and show a variation of clinical ­ symptoms from asymptomatic to nausea, abdominal pain, and splenomegaly, especially at young ages (Rana et al. 2014). In CT, they present as hypodense, well-demarcated lesions with a thin wall and without uptake of contrast medium. Calcifications of the wall may occur. Partial splenectomy is the treatment of choice in patients with symptomatic epidermoid cysts.

3.4

Hemangioma

Hemangiomas are the second most frequent focal splenic lesions (Luna et al. 2006). They are usually asymptomatic and found incidentally. Hemangiomas are slow-flow venous malformations, and the majority show a cavernous type. Typically, a hypo- to isodense mass is seen on unenhanced CT (Fig. 12a). Hemangiomas are usually smaller than 2 cm and may show calcifications and a central, stellate scar (Fotiadis et al. 2009). After administration of intravenous contrast medium, they show peripheral enhancement, followed by centripetal filling and persistent contrast enhancement in late phase images (Fig. 12b, c). Secondary hemangiomas can be present in the course of a systemic angiomatosis like Klippel-Trénaunay-Weber syndrome or Beckwith-Wiedemann syndrome. The presence of multiple splenic hemangiomas which replace the entire parenchyma of the spleen is known as splenic hemangiomatosis. Splenic hemangiomatosis is also associated with the Klippel-Trénaunay-Weber syndrome and the

Spleen

565

a

b

c

Fig. 12  Axial image of the spleen in unenhanced phase exhibits an iso- to slightly hypodense mass (a). The classic peripheral enhancement is seen in arterial phase (b),

followed by centripetal filling and persistent contrast enhancement in late phase (c)

Kasabach-Meritt syndrome. Complications of hemangiomas include hemorrhage, infarction, thrombosis, and very rarely rupture.

spleen. Histologically they are divided into three subtypes: capillary, cavernous, or cystic (Rodriguez-Montes et al. 2016). Clinical presentation ranges from asymptomatic to left upper abdominal pain due to a mass effect and compression of adjacent structures. Multiple ­ lymphangiomas of the spleen can occur in systemic lymphangiomatosis. In CT, they present as thin-­walled, hypodense masses without enhancement. They are typically located subcapsular. While septations can be seen in ultrasound and MRI, they are usually too thin to be demarcated in CT. Cystic lymphangiomas may express peripheral calcifications. Complications of very large lymphangiomas are bleeding, splenomegaly, and secondary portal hypertension. Very

3.5

Lymphangioma

Lymphangioma is a rare, benign neoplasm, which is primarily seen in children. They are congenital malformations of the lymphatic system, which show cystic dilatation of the lymphatic vessels (Kim et al. 2015). The cystic appearance develops slowly, due to an abnormal or absence communication between the lymphatic vessels. Lymphangiomas typically occur in the neck or axillary region, but they are rarely found in the

A. Euler and S.T. Schindera

566

rarely a transformation into a malignant lymphangiosarcoma has been described (Feigenberg et al. 1983).

3.6

other malignant solid masses of the spleen like lymphomas or metastases. However, splenectomy is often needed to exclude malignancy (Carlomagno et al. 2015).

Hamartoma 3.7

Hamartomas are benign, solid lesions that can have a size of up to 19 cm (Lam et al. 1999). They consist of an anomalous mixture of tumor tissue and normal splenic tissue with red and white pulp. They are usually incidental findings, but larger lesions may lead to splenomegaly and rupture. Clinical symptoms are therefore linked to the mass effect of larger lesions. Hamartomas of the spleen are associated with solid (thymoma) and hematological malignancies, with tuberous sclerosis and WiskottAldrich-like syndrome (Lee and Maeda 2009). In CT, hamartomas are often isodense to the spleen in unenhanced imaging as well as after administration of contrast medium, which makes them difficult to detect. A change or distortion of the splenic contour may be the only finding in these cases. However, some hamartomas show a heterogenous enhancement after i.v. contrast ingestion (Fig. 13). Imaging is important to differentiate hamartomas from

a

Fig. 13  Splenic hamartomas. Axial CT images of the spleen in arterial (a) and venous phase (b). In the arterial phase, two round masses with heterogeneous enhance-

Sclerosing Angiomatoid Nodular Transformation

Sclerosing angiomatoid nodular transformation (SANT) is a solitary, non-neoplastic vascular tumor, which develops secondary to a vascular injury or an inflammation. Histopathologically, it is composed out of multiple angiomatoid nodules surrounded by fibrous tissue (Martel et al. 2004). SANT is usually asymptomatic. However, like most benign lesions of the spleen, large lesions can lead to abdominal pain and splenomegaly due to mass effect. After administration of contrast medium, the lesion is hypodense compared to the splenic parenchyma in arterial and portal-­ venous phase and becomes isodense in late phase due to centripetal filling. The first differential diagnosis is hemangioma. Diagnosis is usually established by splenectomy, rather than percutaneous biopsy because of the increased rate of complications in a vascular lesion (Imamura et al. 2016).

b

ment are seen (arrows in a). In the venous phase, these masses are slightly hyperdense to almost isodense to the splenic parenchyma (arrows in b)

Spleen

3.8

567

Angiomyolipoma

Angiomyolipoma of the spleen is extremely rare and usually associated with tuberous sclerosis. In tuberous sclerosis, additional angiomyolipomas can usually be found in the kidneys. In CT, angiomyolipomas show areas with negative, fat-­ equivalent attenuation, and they may show areas of contrast medium uptake due to increased vascularity (Thipphavong et al. 2014).

4

Semi-malignant Lesions

4.1

Littoral Cell Angioma

Littoral cell angioma is a rare vascular tumor, which is characterized by multiple spongelike vascular spaces. It can occur at any age and is usually benign. However, malignant transformation has been reported (Ben-Izhak et al. 2001). Littoral cell angioma is typically symptomatic leading to anemia, thrombocytopenia, and splenomegaly. Characteristic in CT are multiple, hypodense nodules in unenhanced, arterial, and portal-venous phases (Fig. 14a). However, littoral cell angioma is isodense to the splenic parenchyma in late phase (Fig. 14b). The

a

differential diagnosis is broad and includes benign and malignant lesions. Therefore, splenectomy is usually performed to establish the diagnosis.

4.2

Hemangioendothelioma

Hemangioendothelioma is a very rare, borderline-­ malignant, vascular tumor. It occurs in pediatric patients or more commonly in young adults. Clinical presentation is nonspecific. In CT, hemangioendothelioma is hypodense in unenhanced images and a radiative peripheral enhancement in arterial phase followed by centripetal filling in late phase has been described (Wang et al. 2015). However, imaging alone will usually not allow to establish the diagnosis. Partial or total splenectomy has been described as treatment options.

5

Malignant Lesions

5.1

Lymphoma

Lymphoma is the most common malignancy in the spleen. It can occur primary or secondary to a systemic lymphatic disease. The spleen is

b

Fig. 14  Littoral cell angioma. Axial image of the upper abdomen in portal-venous (a) and late phase (b). Portal-venous phase shows multiple hypodense nodules which become isodense to the splenic parenchyma in late phases

A. Euler and S.T. Schindera

568

a

b

Fig. 15  Axial CT image of a patient with a diffuse large B-cell non-Hodgkin lymphoma demonstrates multiple, round, hypodense lesions in the spleen and liver in portal-­

venous phase (a). PET scan exhibits increased uptake of the abovementioned lesions in the liver and the spleen and in additional lesions in the right axillary region (b)

involved in about one third of Hodgkin and non-­Hodgkin lymphomas (Saboo et al. 2012), while primary lymphoma is extremely rare (3 and ≤10 mm and/or intense enhancement of the esophageal wall, without stenosis. The outer borders of the tumor are smooth (Fig. 2). T2: Focal, polypoid, or diffuse circumferential thickening of the esophageal wall >10 and ≤15  mm, with or without the presence of mild stenosis. The outer borders of the tumor are either smooth or show stranding for less than one-third of the tumor circumference or length (Fig. 3). T3: Tumor appears symmetric or asymmetric with markedly diffuse or circumferential wall thickening of ≥15  mm, with mild-to-severe stenosis, and marked stranding over one-third of the tumor extension, or extensive blurring of the outer border (Fig. 4). T4: Tumor shows invasion into one of the adjacent structures, such as the pericardium, the diaphragm, the pleura (T4a), the tracheobronchial tree, or the aorta and spine (T4b), using the

A. Ba-Ssalamah et al.

584

a

b

c

Fig. 1  Hydro-MDCT of the esophagus. Axial (a), coronal (b), and sagittal (c) reformations along the course of the esophagus, demonstrating the normal wall thickness of the esophagus (≤3  mm) and homogeneous enhancement (arrows). Note: on the coronal reformation, the

a

b

c

Fig. 2  The value of fused FDG-PET/Hydro-MDCT of the esophagus in visualizing small tumors. The axial (a) hydro-MDCT image of the esophagus shows only nonspecific wall thickening. However, the axial and coronal (b, c), fused FDG-PET/Hydro-MDCT reformations show

a

physiologic angulation caused by the aortic arch with pseudothickening (partial volume) of the esophageal wall (arrowhead). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

b

Fig. 3  Hydro-MDCT of the esophagus. Axial (a), coronal (b), and sagittal (c) reformations show circumferential wall thickening with a maximum 10  mm depth in the upper third of the esophagus (arrows), with little stranding

a small focus of tracer uptake in the distal third of the esophagus, consistent with biopsy-proven T1 tumor (arrows). The outer borders of the esophagus are smooth. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

c

of the paraesophageal fat but less than one-third the tumor extension consistent with a T2 tumor (short arrow). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques

a

b

c

d

585

Fig. 4  Hydro-MDCT of the esophagus. Axial (a) and coronal (b) reformations show a large mass with inhomogeneous enhancement in the mediastinum arising from the esophageal wall, with marked infiltration of the paraesophageal fat, but no infiltration of the adjacent organs

(arrowheads) consistent with a T3 tumor, according to CT staging. The corresponding fused FDG-PET-CT image shows strong tracer uptake, but neither lymph nodes nor distant metastases. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

criteria described in the literature (Thompson and Halvorsen 1994) (Fig. 5). These structures may be invaded by contiguous tumor spread. However, it is often difficult to reliably distinguish infiltration into these adjacent organs (T4) from broad contact without infiltration. Evaluating direct invasion by esophageal cancer into adjacent vital structures by MDCT is based upon two criteria: mass effect and loss of fat planes. When the trachea or bronchial wall is indented or displaced away from the spine by the

tumor mass, then mass effect is present and invasion is presumed (Lagergren et al. 2000). Coronal or sagittal reformatted images often can best visualize tumor invasion. The cardia is directly involved by carcinoma of the distal esophagus and adenocarcinomas of the esophagogastric junction (AEG I–III) in about 60% of patients according to the Siewert classification (Siewert 2007). The precise depiction of the anatomic location of the tumor and assessment of the degree of cardia involvement are crucial for

A. Ba-Ssalamah et al.

586

a

b

Fig. 5  MDCT of the esophagus. Axial (a) and coronal (b) reformations show a huge mass with inhomogeneous enhancement in the mediastinum arising from the esophageal wall. Encasement of the aorta (a, arrowheads) and infiltration of the pericardium (a, thin arrows) are consis-

tent with a T4 tumor. Note enlarged right perihilar and perigastric lymph nodes consistent with stage N2 (more than three LN) disease (short arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

surgical planning. Since the stomach is the organ usually used as the first choice for reconstruction after esophagectomy, the radiologist should comment on the status of the cardia, i.e., AEG I-III, in the radiological report (Fig. 6). When fused FDG-PET/hydro-MDCT is used, tracer uptake is markedly helpful in the detection of even a small primary tumor, and may better estimate local staging (Fig. 2). The sensitivity of FDG-PET for the detection of primary esophageal tumors has been reported to be 91–95% in prospective studies (Lowe et  al. 2005; Meyers et al. 2007). Since PET scanners have a limited spatial resolution of about 5–8  mm, lesions smaller than 1  cm might not be detected; however, the combination of PET/CT and the hydro technique may improve accuracy.

of lymphatic metastases is related to the tumor’s local T stage, including its size and depth of penetration (Stein et al. 2005). The extensive mediastinal lymphatic network of the esophagus, which communicates with the abdominal and cervical collateral vessels, is responsible for the findings of mediastinal, supraclavicular, and celiac lymph node metastases in at least 75% of patients (Thompson and Halvorsen 1994; Thompson et al. 1983). According to the American Joint Committee on Cancer, the 8th edition (Rice et al. 2017), N staging depends on the presence of positive locoregional or periesophageal lymph nodes (affected lymph nodes) as follows:

3.2.1 N Staging Lymphatic spread is found in 74–88% of patients at the time of diagnosis with esophageal carcinoma because of the abundant lymphatic channels in the esophagus (Siewert 2007). The frequency

• • • •

N0: no regional lymph node metastasis N1: 1–2 positive regional lymph nodes N2: 3–6 positive regional lymph nodes N3: ≥7 positive regional lymph nodes

Accurate lymph node assessment for metastatic spread is challenging, even with PET/ MDCT.  However, improved evaluation appears

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques

a

Fig. 6  Hydro-MDCT of the esophagus, coronal reformations, show a large adenocarcinoma in the lower-third of the esophagus at the gastroesophageal junction (AEG I) (a), with fused PET-CT image (b). The differentiation between distal esophageal cancer and gastric cancer

possible if morphology, including size and shape, contrast enhancement pattern, and tracer uptake of lymph nodes, is considered (Blom et al. 2012; Okada et al. 2009). On MDCT, periesophageal lymph nodes are considered positive if they are ≥6 mm in diameter, are round rather than bean-­shaped, and show marked or inhomogeneous contrast enhancement (Ba-Ssalamah et al. 2003) (Fig. 7). For FDG-PET, there are no uniformly accepted SUV cutoffs for lymph node metastases, although a few institutions have defined their own cutoff values (Yu et al. 2011; Kato et al. 2009). Lymph nodes, in general, however, are considered positive if they show FDG uptake that is higher than the background. A meta-analysis of 12 studies (n  =  490) examined the diagnostic accuracy of FDG-PET in the preoperative staging of esophageal cancer and reported a sensitivity and specificity for detecting locoregional lymph node involvement of 51% and 84%, respectively (van Westreenen et al. 2004).

587

b

located in the cardia can be difficult. Note the perigastric lymph node involvement (arrowhead). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

Fig. 7  Hydro-MDCT of the stomach and distal esophagus. Coronal reformatted image shows polypoid tumorous thickening of the gastric wall at the gastroesophageal junction (short arrows). Whereas the multiple perigastric lymph nodes are regional, i.e., N2 (more than three, thin arrows), the retroperitoneal lymph nodes are outside of the expected drainage and therefore represent metastatic spread of cancer, i.e., M1 (arrowheads). The lymph nodes are round, ≥6 mm in diameter, and enhance inhomogenously, greater than the surrounding muscles, consistent with malignant lymph nodes. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

A. Ba-Ssalamah et al.

588

3.2.2 M Staging Hematogenous metastases from esophageal carcinoma most commonly involve the liver because the esophagus is drained by the portal vein (Fig. 8). Less common sites of hematogenous spread include the lungs, adrenal glands, kidneys, bones, and brain. Lymph node involvement outside a periesophageal location, for example, within the supraclavicular region, is considered M1 disease (Nomura et  al. 2012). a

c

Fig. 8  Hydro-PET-MDCT of the esophagus. Axial and coronal reformations show a mild diffuse wall thickening with inhomogeneous enhancement at the gastroesophageal junction, and paraesophageal fat-stranding (a, short arrow). A stent has been inserted due to the significant stenosis. Axial (c) and coronal (b, d) PET-CT images

Advanced distal esophageal cancers can develop peritoneal metastases, similar to gastric cancers. FDG-PET is most helpful in distinguishing potentially resectable, locally advanced disease (T3–4, N0, M0) from distant disease (M1). In prospective studies, M1 disease was detected by FDG-PET and missed by CT (with or without EUS) in 5–7% of cases (Meyers et  al. 2007; Heeren et al. 2004). M1 disease was detected by b

d

show mild nodular FDG uptake at the left shoulder, left elbow, and left scapula, consistent with bone metastases making this stage M1 disease (thin arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques

a

b

589

c

Fig. 9  Hydro-MDCT of the esophagus. Axial (a), coronal (b), and curved sagittal reformations (c) in a patient who is status post esophagectomy and gastric pull-up for esophageal cancer. The band-like solid inhomogeneously

enhancing mass at the anastomosis is biopsy-proven recurrent tumor (right side cervical, arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

FDG-PET and undiagnosed on CT in 6–15% of patients (Flamen et al. 2000; Meyers et al. 2007; Heeren et al. 2004) (Fig. 8).

and superior to conventional CT for detecting distant metastatic disease. It can also accurately delineate the neoesophagus and its surroundings (Fig.  9). Distinguishing between fibrosis and tumor tissue on CT is based on indirect signs, such as retraction versus positive mass effect, and is often difficult, even in the best of circumstances. FDG/PET/MDCT can overcome this limitation (Sun et al. 2009; Carlisle et al. 1993; Tunaci 2002) to a large extent. Early postoperative cases with possible inflammatory reactions or early postradiation changes may be FDGavid, and therefore must be interpreted with caution.

3.3

Follow-Up After Esophagectomy

Unfortunately, even after curative surgery, there is a high rate of esophageal cancer recurrence, categorized as either locoregional recurrence and/or distant metastatic disease. Because the anatomy of the posterior mediastinum is markedly changed post-esophagectomy and gastric excision, the detection of local tumor recurrence can be quite difficult. Therefore, the choice of 3.3.2 CT and PET-CT for Predicting Survival imaging modality is crucial to early detection of recurrent disease. Again, FDG-PET/hydro-­ Dysphagia and weight loss are common in MDCT is the modality of choice. Mural thicken- patients with esophageal cancer and these are ing, adjacent mass, and/or suspicious lymph associated with sarcopenia, i.e., loss of skeletal nodes are highly predictive of recurrent disease muscle mass. Sarcopenia has been defined by (Guo et al. 2007). international consensus as muscle loss that results in a skeletal muscle index (SMI) of ≤39 cm2/m2 3.3.1 CT Findings for women and ≤55  cm2/m2 for men (Fearon Locoregional recurrent esophageal tumor is well et al. 2011). demonstrated by MDCT. A smooth or spiculated The SMI, defined as the total cross-sectional area of extrinsic mass effect on the mediastinal skeletal muscle area at the level of lumbar verteborder can be visualized by MDCT. Furthermore, bra L3, has been shown to correlate well with MDCT with multiplanar reformations is accu- whole-body muscle mass. It can be easily calcurate in detecting masses after esophageal surgery lated on any CT scanner.

A. Ba-Ssalamah et al.

590

a

b

c

d

e

f

Fig. 10  Hydro-MDCT of the esophagus, pre-­neoadjuvant chemotherapy shows a mild circumferential tumorous wall thickening with inhomogeneous enhancement in the middle third of the esophagus, with little paraesophageal fat-stranding (a, arrowhead). Axial and coronal fused PET-CT images (b, c) show vigorous FDG uptake. 3 months after neoadjuvant chemotherapy, the axial CT

image (d) showed no significant change. There is persistent, mild inhomogeneous circumferential wall thickening. The fused PET-CT images (e, f) show strong tracer uptake at the corresponding site (arrows), indicating that the patient was a nonresponder. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

More recently, our group found that sarcopenia measured using CT is associated with reduced overall survival after esophageal cancer surgery and that sarcopenia provides a better estimation of cachexia than bone mineral index (BMI) (Tamandl et  al. 2016a). Furthermore, we found that other body composition parameters (total muscle area [TMA], fat-free mass index [FFMi], fat mass index [FMi], subcutaneous, visceral, and retrorenal fat [RRF], and muscle attenuation) are also associated with shorter survival, independent of traditional clinical parameters (Tamandl et al. 2016a). In reevaluating PET-CT’s role in predicting the overall survival of patients with esophageal cancer scheduled for neoadjuvant chemotherapy and radiation therapy, recent studies have found that patients with little or no response to neoadjuvant treatment do poorly and might not be the best candidates to proceed with surgical resection

(Figs.  10 and 11) (Lordick et  al. 2007; Kelsen et al. 1998). Our group found that a change in PET parameters shows quite a good correlation to survival in esophageal cancer, with a close association with overall survival, which is independent of changes in SUVmax and CT volume (Tamandl et al. 2016b). Furthermore, we showed that metabolic parameters after neoadjuvant chemotherapy correlate with pathologic response and nodal status. Therefore, metabolic parameters may be better suited than maximum standardized uptake value (SUVmax) for response assessment, which is still the most commonly employed parameter with which to assess treatment response. In addition, because CT is problematic when attempting to differentiate posttreatment changes from residual or recurrent tumor tissue, CT volumetry of the esophagus is often a useful adjunct tool

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques

a

591

b

d

c

e

f

Fig. 11 Hydro-MDCT of the esophagus, preneoadjuvant chemotherapy. This axial reformation shows a marked circumferential tumorous wall thickening with inhomogeneous enhancement along the gastroesophageal junction, and marked paraesophageal fat-stranding (a, short arrow). Axial and coronal fused PET-CT images (b, c) showed vigorous FDG uptake. 3

months after neoadjuvant chemotherapy, the axial CT image (d) showed a significant decrease in wall thickening and fused PET-CT images (e, f) showed no radiotracer uptake at the site of the previous cancer (arrows), indicating that the patient was a responder. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

with which to assess response; however, the combination of CT volumetry and metabolic parameters has been shown to be more accurate (Li et al. 2013). The volumetric parameters that take into account both tumor volume (metabolic tumor volume, MTV) and metabolic activity (total lesion glycolysis, TLG; derived from MTV and mean SUV of the entire mass) can be used to overcome the abovementioned drawbacks. (Tamandl et al. 2016b).

lowing criteria: (a) predominantly esophageal involvement with only regional lymph node involvement; (b) no definite enlarged mediastinal lymph nodes; (c) lack of hepatic and splenic involvement; and (d) absence of superficial lymphadenopathy (Kaplan 2004).

4

Other Esophageal Malignancies

4.1

Esophageal Lymphoma

Although esophageal lymphoma is rare, its histologic appearance is quite diverse since any histologic subtype may affect the esophagus (Mendelson and Fermoyle 2005). The diagnosis of primary esophageal lymphoma rests on the fol-

4.1.1 CT and PET/CT Findings CT may demonstrate a homogeneously enhancing mass with irregular borders or sharply delineated, pronounced, polypoid wall thickening in any part of the esophagus, with or without associated lymphadenopathy. There is no specific CT finding for esophageal lymphoma. Lymphoma may infiltrate the entire esophagus diffusely. While splenic involvement is suggestive of lymphoma, hepatic metastases are characteristic of esophageal cancer. PET/CT scans can also be used in staging patients with primary esophageal lymphoma, as well as for monitoring tumor response to therapy (Suga et al. 2009).

A. Ba-Ssalamah et al.

592

4.2

Leiomyoma and GIST

Leiomyomas, the most common benign esophageal tumor, account for 60–70% of all benign esophageal neoplasms. They are distinctly rare in the rest of the gastrointestinal tract, however (Hatch et al. 2000; Seremetis et al. 1976; Simmang et al. 1989). The tumor has a male predilection (2:1) and a median age of occurrence at 30–35 years. Almost always solitary, leiomyomas typically range between 2 and 8  cm in diameter. More often asymptomatic than not, typical complaints include dysphagia or substernal chest pain due to obstruction of esophageal bolus transit. Gastrointestinal stromal tumors (GISTs) are the most common nonepithelial tumors of the gastrointestinal tract. In contrast to leiomyomas, GISTs are extremely rare in the esophagus (Monges et al. 2010).

4.2.1 CT Findings Contrast-enhanced CT scans reveal a smooth or lobulated tumor margin, with either iso- high-, or homogeneously low attenuation. Leiomyoma and GIST may appear as a well-circumscribed, intensely enhancing mass or may be a sessile, pedunculated, polypoidal, exophytic intraluminal solid mass, sometimes with secondary ulceration. Leiomyomas are the only tumors that may contain calcification (Fig. 12).

a

The absence of infiltration of the esophageal wall or the absence of the typical circumferential growth pattern enables differentiation from esophageal cancer. GISTs may remain constant in size or even enlarge during therapy, but show a decrease in CT attenuation values (Hounsfield units, HU) (Choi et  al. 2004). PET/CT is able to show the early effects in patients who are undergoing treatment. Functional imaging proved significantly more accurate than CT alone when assessing the GIST response to therapy. Combined PET/CT imaging is, therefore, a valuable diagnostic tool for the primary diagnosis and posttreatment follow-up of GISTs (Suga et  al. 2009; Antoch et al. 2004).

4.3

Fibrovascular Polyps

Fibrovascular polyps are rare masses, comprising about 1% of all benign esophageal tumors. Even so, they are the most common benign intraluminal esophageal mass (Sargent and Hood 2006). Giant fibrovascular polyps, defined as polyps larger than 5 cm in diameter, may be lethal should they bleed or, rarely, cause asphyxiation if regurgitated. Patients commonly present with dysphagia or hematemesis.

b

Fig. 12  Hydro-MDCT of the esophagus. Axial (a, b) images show an almost 3  cm polypoid, soft-tissue esophageal mass, with smooth margins (arrows). Its homogeneously low-attenuation background and few

scattered calcifications within the middle-third of the esophagus are consistent with a leiomyoma. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques

4.3.1 CT Findings The polyps may not be well visualized on endoscopy, and imaging plays a vital role in aiding diagnosis, as well as in providing important information for preoperative planning, such as the location of the pedicle, the vascularity of the polyp, and the tissue elements of the mass. These polyps contain predominantly fibrovascular and fatty tissue, which gives them their typical CT appearance of a pedunculated intraluminal mass of fat density, which expands the esophagus (Ascenti et al. 1999).

4.4

Esophageal Fistula

593

radiochemotherapy. Although cough, aspiration, and fever are usually present, such vague symptoms, even when culminating in pneumonia, infrequently lead to the diagnosis. A high index of suspicion is key to making the diagnosis of a fistula. More than half such fistulas involve the trachea; alternatively, a connection with the left or right main or lower lobe bronchus may be formed. Patients with esophageal-airway fistulas are treated with covered stents to seal off the leak. CT may be necessary to localize the fistula and to aid in treatment planning. CT can also be used to detect pleuro-plumonary or mediastinal inflammatory reactions to esophageal fistulae (Liu et al. 2006; Peyrin-Biroulet et al. 2006).

Esophageal fistulas are classified according to 4.4.1 CT Findings their anatomic relationship with the surrounding CT can demonstrate a fistulous connection between structures, i.e., esophageal-tracheal, esophageal-­ the esophagus and the tracheobronchial system, bronchial, esophago-pleural, aorto-esophageal, pleura, pericardium, or mediastinal fat if the fistuand esophago-pericardial. When congenital, lous tract is of sufficient size and contains air or esophageal-airway fistulas are typically called oral contrast medium. Oral administration of dilute tracheo-esophageal fistulas. Acquired iodinated contrast material (contrast material: esophageal-­ airway fistulas may be a life-­ water, 1:100) can help to delineate the fistula. CT threatening complication of esophageal cancer or can also detect perifocal reactions in the form of may occur after esophageal trauma, infection, or empyema, pneumonia, or mediastinitis (Fig. 13). a

b

Fig. 13  Hydro-MDCT of the esophagus. Sagittal (a), axial (b), and coronal (c) reformations in a patient with known esophageal cancer. A fistula is suspected, due to continuous coughing and recurrent pneumonias. These

c

images clearly show a tract between the tumor and the left main bronchus consistent with a fistula (arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

A. Ba-Ssalamah et al.

594

a

b

c

Fig. 14  Hydro-MDCT of the esophagus in a patient with primary achalasia shows markedly diffuse dilatation of the entire fluid-filled esophagus. There is no significant wall thickening to suggest an obstructing mass (arrows).

Barium esophagogram, in the same patient, also showed diffuse esophageal dilatation (arrowheads). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

4.5

(Fig. 14). In contrast to a stricture, the esophageal wall is not thinned at the site of the narrowing, and the wall is not thickened as it is with the esophageal tumor or esophagitis. Most pseudoachalasia patients have CT findings of esophageal dilation, which is usually more marked than in true achalasia. Moreover, asymmetric wall thickening or a mass is often found in pseudoachalasia. In this group, asymmetric or marked thickening (>10 mm) indicates pseudoachalasia.

Achalasia

Achalasia, a primary motor disorder of the esophagus, is rare. Although patients usually present in adulthood, the disorder may be seen in children. Idiopathic achalasia is characterized by incomplete relaxation of the lower esophageal sphincter (LES) on swallowing and a peristalsis of the esophageal body (Gelfand and Botoman 1987). It should be distinguished from pseudoachalasia, described in 1947 by Ogilvie, as submucosal infiltration of the lower esophagus and cardia by carcinoma (Carter et  al. 1997). CT usually differentiates achalasia from pseudoachalasia, whereas endoscopy and biopsy are used to confirm tumor extension in pseudoachalasia. However, CT may be used in suspect cases, when submucosal tumor growth escapes endoscopic detection. In addition, CT may delineate ancillary features of malignancy, such as lymph nodes, solid organ metastases, or peritoneal carcinomatosis (Carter et al. 1997).

4.5.1 CT Findings CT shows uniform dilatation that affects a long segment of the esophagus, with no wall thickening and with normal-appearing boundary surfaces and mediastinal fat. The esophagus narrows abruptly at the esophagogastric junction, with no evidence of an intramural or extrinsic, obstructive lesion

4.6

Dysphagia Lusoria

Rarely, dysphagia may be caused by anatomic variants, such as deviations of the aortic arch or supraaortic branches, which may displace or compress the proximal esophagus. The most frequent cause is an anomalous right subclavian artery, arising from the descending aorta as a fourth supra-aortic branch that passes behind the esophagus. When the aortic origin of the anomalous right subclavian artery is congenitally wide, it is called a Kommerell’s diverticulum. Although rare, the aortic diverticulum is the more likely of the two to cause dysphagia (dysphagia lusoria). Other causes include a duplicated aortic arch or an aortic aneurysm (Keum et al. 2006). CT angiography is the procedure of choice in the evaluation of vascular thoracic anomalies.

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques

a

b

595

c

Fig. 15  Hydro-MDCT of the esophagus. Paraaxial (a), paracoronal (b), and parasagittal (c) reformations show an aberrant right subclavian artery (arrowheads) that arises more posteriorly from the descending aorta and

compresses the esophageal lumen consistent with the clinical diagnosis of dysphagia lusoria. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

4.6.1 CT Findings The diagnosis of aortic arch variants is easily made on contrast-enhanced MDCT. An aberrant right subclavian artery, or dysphagia lusoria, demonstrates a typical pattern on contrast-­ enhanced MDCT and is more easily diagnosed using multiplanar reconstructions. It arises more posteriorly than normal, and runs behind the esophagus (Fig.  15). An aortic diverticulum appears as a circumscribed, asymmetric aneurysm, such as protrusion from a wide, funnel-­ shaped origin of the subclavian artery in the distal aortic arch.

sia, and tumors, predispose the esophagus to perforation. Esophageal perforation is associated with high mortality, and postoperative leaks occur frequently after primary surgical repair (Chao et al. 2005). Prompt and accurate diagnosis of esophageal and gastric perforation is critical, because the consequences of missed esophageal and gastric injury are devastating, with potential progression to fulminant mediastinitis, septic shock, and/or peritonitis. Delay in treatment beyond 24 h after onset may adversely affect prognosis. Fluoroscopy, with a water-soluble contrast agent, such as Gastrograffin, is the method of choice to demonstrate esophageal rupture. However, CT has been increasingly used for the diagnosis of gastric and esophageal perforations (LeBlang and Nunez 1999).

5

Gastroesophageal Diseases

5.1

Gastro- and Esophageal Perforation

Esophageal and gastric injuries include penetrating injuries, blunt traumatic perforation, iatrogenic perforation, as well as spontaneous perforation due to a sudden rise in intraluminal pressure in the esophagus during vomiting. Such a tear may be either transmural, as in Boerhaave syndrome, or limited to the mucosa and submucosa, as in Mallory-Weiss syndrome. Perforations, sometimes life-threatening, may also occur as a consequence of a gastric ulcer. Most often, esophageal and gastric perforations occur during endoscopic investigation of malignant disease. Esophageal diseases, such as strictures, achala-

5.1.1 CT Findings Radiographic detection of esophageal injuries relies on the presence of indirect radiological signs, including subcutaneous or muscular, thoracic, or cervical emphysema, a widened mediastinum, pneumomediastinum, pneumopericardium, left-sided pneumothorax, pleural effusion, and left lower lobe atelectasis. Furthermore, an attempt to pass a nasogastric tube will show that its trajectory is abnormal. CT can also display subtle signs, such as localized esophageal wall thickening, mucosal hyperemia, mucosal dissection, and intramural hematoma or edema (De Lutio di Castelguidone et al. 2005). CT may show small pockets of mediastinal air in the presence of a small tear (Fig. 16).

A. Ba-Ssalamah et al.

596

a

b

c

Fig. 16  Hydro-MDCT of the esophagus. Axial (a), coronal (b), and sagittal (c) reformations in a patient with esophageal perforation after dilation of distal tumor stenosis. CT scan shows free air (a, b, thin arrows), as well as a soft tissue defect within the esophageal wall (arrowheads)

with extensive contrast media extravasation in the left lateral spread of the esophagus (a, b, thick arrows) is shown. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

A high index of suspicion is critical in making the diagnosis, as findings are often very subtle. In patients with gastric perforation, MDCT is a valuable modality for the diagnosis and localization of the site of a gastric perforation (Young et al. 2008). In addition to the presence of a pneumoperitoneum, the other significant CT findings that help to localize the site of perforation include a collection of extraluminal gas in close proximity to the perforated viscus, focal gastric wall thickening, and focal discontinuity of the gastric wall. Extravasation of oral contrast on a CT scan is considered diagnostic of gastric perforation (Fig. 16). However, the reported sensitivity is low and ranges between 19% and 42%. Hence, the use of oral contrast may not provide much additional information and may delay surgical exploration.

diverticulum (GD) is rare, with a prevalence of approximately 0.04% in upper gastrointestinal studies (Rashid et  al. 2012). GDs are classified into congenital and acquired types. GD is characterized by the presence of a pouch protruding from the gastric wall; however, the majority of GD cases are asymptomatic and are diagnosed incidentally during routine examinations.

5.2.1 CT Findings Diverticula appear as an air-, water-, or contrast-­ filled outpouching. Mid-esophageal and epiphrenic diverticula are better visualized on coronal or sagittal (Fig.  17) reformations on hydro-MDCT. The most frequent location is posteroinferior to the cricoid cartilage, the so-called Zenker’s diverticulum, which actually is a pharyngeal diverticulum. Hydro-MDCT may detect the presence of GDs as thin-walled cystic masses in the left adre5.2 Diverticula nal area filled with water. Conventional CT scans, Diverticula are incidental findings at CT with the patient in a prone position, may further (Pearlberg et  al. 1983). Esophageal diverticula aid in diagnosis, by forcing gastric air into the are divided into the pulsion and traction types. diverticulum cavity, leading to the formation of Whereas traction diverticula are considered true an air-fluid level in the mass (Fig.  18). Positive diverticula because they contain all three layers oral contrast agents are helpful as well. GDs may be misdiagnosed as adrenal tumors, of the esophageal wall, pulsion diverticula, containing only the musocal and submucosal layers, in addition to a number of normal structures, anaare referred to as false diverticula. The two pre- tomic variations, and lesions of adjacent organs dominant locations of esophageal diverticula are including: hepatic tumors; fluid-filled colon; the mid-esophagus (at the level of the tracheal splenic lobulation; tortuous or dilated splenic bifurcation) and the distal esophagus (the so-­ arteries or veins; exophytic upper pole renal mass; called epiphrenic diverticula). The gastric suprarenal fat; thickening of the diaphragmatic

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques

a

b

Fig. 17  Hydro-MDCT of the esophagus. Axial, coronal, and sagittal (a–c) reformations in a patient with a Zenker’s diverticulum, lateroposterior to the cervical esophagus. It

a

597

c

is filled with air and barium due to formerly performed esophagogram (arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

b

Fig. 18  Hydro-MDCT of the stomach. Axial (a) and coronal (b) reformatted images in a patient with a gastric diverticulum. A well-defined saccular fluid-filled

outpouching arises from the gastric fundus consistent with a diverticulum. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

crura; and pancreatic tail masses. Accordingly, it is important to be cautious when diagnosing left para-adrenal masses on CT images (Jebasingh et al. 2014; Schramm et al. 2014).

cysts (GDCs) are rare and represent 4% of all alimentary tract duplications. Approximately 67% manifest within the first year of life. GDCs in adults are generally encountered as incidental findings at endoscopy or laparotomy, or other imaging. Greater than 80% of gastric duplications are cystic and do not communicate with the lumen of the stomach. The remainder are tubular, maintaining some communication with the gastric lumen (Tjendra et  al. 2016). Since most cases occur along the greater curvature of the stomach, the cysts can potentially compress the adjacent organs, such as the pancreas, kidney, spleen, and

5.3

Duplication Cyst

Duplication cysts of the esophagus are rare congenital anomalies that may be discovered incidentally on conventional chest radiographs as an indeterminate mediastinal mass, and require further investigation by CT (Kuhlman et  al. 1985; Kolomainen et  al. 2017). Gastric duplication

A. Ba-Ssalamah et al.

598

a

b

c

Fig. 19  Hydro-MDCT of the stomach. Axial (a), coronal (b), and sagittal (c) reformatted images in a patient with a duplication cyst. A well-defined, thick-walled fluid-­density

structure (arrows) is noted along the gastroesophageal junction line. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

adrenal gland. Accordingly, the differential diagnosis includes lesions arising from these organs (Lee et  al. 2010). Like cysts elsewhere, GDCs may develop complications, including infection, bleeding, perforation, ulceration, fistula formation, obstruction, compression, or secondary carcinoma (Lee et  al. 2010). Up to 10% of GDCs may contain ectopic pancreatic tissue, putting them at risk for pancreatitis and mimicking a pancreatic pseudocyst (Oeda et al. 2010).

troesophageal junction remains in its normal position. Hemorrhage, incarceration, obstruction, and strangulation of the stomach and intestine are the most common complications.

5.4.1 CT Findings Axial herniation of the stomach results in a large retrocardiac mass as the gastric cardia is displaced into the thoracic cavity. With CT, the demonstration of gastric folds within a retrocardiac mass is frequent and pathognomonic. Good distension of 5.3.1 CT Findings the stomach and esophagus is very helpful in the Duplication cysts are smoothly marginated, differential diagnosis. The much more common homogeneous masses of water density that most paraesophageal hernia is associated with fixation commonly occur in the lower esophagus (60%). of the gastric cardia. Portions of the herniated They are intimately related to the esophagus, but stomach sit alongside the esophagus (Fig. 20). The rarely communicate with its lumen. The cyst may “upside-down stomach” is an organo-axial volvube paraesophageal or intramural (Fig. 19). Gastric lus where the stomach rotates around its long axis. duplication cysts appear on CT scans as fluid-­ All of the stomach sits in the thoracic cavity, with attenuated cystic masses in close contact with the none of it detected below the diaphragm (Fig. 21). stomach. They may have a higher attenuation This may occur with a paraesophageal hernia. value, however, if they have bled or are infected. Whether it occurs in isolation, or with a paraesophageal hernia, it is a surgical emergency.

5.4

Hiatal Hernia

Esophageal hiatal hernias comprise two types: sliding axial hernia and paraesophageal hernia (Eren and Ciris 2005). Sliding hiatal hernia is a displacement of the gastric fundus and cardioesophageal junction upward into the posterior mediastinum. In the paraesophageal hernia, all or part of the stomach slides up into the thorax, while the gas-

5.5

Esophagitis, Gastritis, and Ulcer

Inflammation of the esophagus and stomach is not an indication for CT. Esophagitis and gastritis may be noted incidentally during the course of a CT staging examination or follow-up (Berkovich et al. 2000). High-grade esophagitis manifests in

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques

a

b

Fig. 20  Hydro-MDCT of the stomach and esophagus. Axial (a) and coronal (b) reformations show a small axial herniation of the stomach that appears as a small

a

599

b

retrocardiac mass. The cardia is displaced into the thoracic cavity (arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

c

Fig. 21  Hydro-MDCT of the stomach and esophagus. Axial (a), sagittal (b), and coronal (c) reformations show a large paraesophageal hernia, with so-called “upside down stomach,” which is an extreme condition in which

all of the stomach has herniated into the thoracic cavity and no portions of the stomach can be detected below the diaphragm (arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

33–41% of patients with malignancies, who are treated with concurrent chemoradiotherapy. Painful esophagitis decreases the nutritional status of patients and can lead to treatment interruptions, which, in turn, adversely affect survival. Gastritis is a very common disease. Predisposing factors include alcohol consumption, ingestion of aspirin or nonsteroidal anti-­ inflammatory drugs, as well as stress, viral or fungal infections, and H. pylori infection. Helicobacter gastritis is identified in nearly 80% of patients with gastric ulcers and in nearly 100%

of patients with chronic gastritis. In many cases, gastric carcinoma and lymphoma are attributed to chronic Helicobacter gastritis (Levine et  al. 1996). A gastric ulcer is a mucosal defect that extends down to the muscularis mucosae and beyond. Ulcers are usually solitary. Gastric ulcer disease with or without perforation is one of the most important causes of acute abdomen (Jacobs et al. 1991). Most gastric ulcers (90%) develop along the lesser curvature or the posterior wall of the antrum or body of the stomach.

A. Ba-Ssalamah et al.

600

a

b

Fig. 22  Hydro-MDCT of the stomach. Axial (a) and coronal (b) reformatted images in a patient with gastroenteritis. There are markedly thickened folds throughout the entire stomach and small bowel. Note the enhancing

mucosal lining and irregular folds (arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

5.5.1 CT Findings The inflamed mucosa shows uniform, circumferential wall thickening that usually involves a relatively long esophageal segment. Inflammatory and neoplastic wall changes cannot be reliably distinguished based upon CT morphology. Short segments of ulcerative wall thickening, especially with an abrupt zone of transition, are more suggestive of a malignant lesion, while longer segments are more consistent with an inflammatory process. The most common CT findings are a thickened esophageal wall and a target sign. Although endoscopy is a more sensitive modality for the detection of this condition, the CT finding of a relatively long segment of circumferential esophageal wall thickening, with or without a target sign, should suggest the diagnosis of esophagitis in the proper clinical setting (Fig. 22). Risk factors for Candida and cytomegalovirus (CMV) esophagitis include HIV infection and organ transplantation. Thickened gastric wall folds are the best CT sign of conventional or H. pylori-related gastritis. Polypoid and lobulated folds are difficult to distinguish from gastric cancer and lymphoma. Biopsy is required in ambiguous cases. A gastric ulcer is a mucosal defect that reaches the muscularis mucosae and beyond (Fig. 23).

5.6

Gastroesophageal Varices

Varices are commonly associated with splanchnic obstruction or portal hypertension (Chen et al. 2013). The presence of gastric varices without esophageal varices is considered a sign of isolated splenic vein occlusion, most commonly secondary to pancreatitis or pancreatic carcinoma. Gastric varices communicate with the esophageal and periesophageal veins which drain, via the azygos/hemiazygos venous system, to the superior vena cava (Balthazar et al. 1987).

5.6.1 CT Findings Gastroesophageal varices are best visualized in the portal venous or delayed phases after the administration of intravenous contrast material. Esophageal varices present as intraluminal (intramural, submucosal) tubular, often dot-like structures that show marked pooling of intravenous contrast medium (Fig. 24). Paraesophageal varices are often larger and more serpiginous. CT criteria for esophageal varices are defined more specifically as nodular or tubular enhancing structures within the esophageal wall that abut the intraluminal ­ surface, thus distinguishing esophageal from

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques Fig. 23  Gastric ulcer with perforation. Axial and coronal reformatted images show a mucosal defect in the gastric fundus, which extends down to the serosa (arrows). There is diffuse wall thickening due to the associated inflammation and fluid collection. This was a surgically proven gastric ulcer. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

601

a

b

a

Fig. 24  Hydro-MDCT of the esophagus. Coronal (a) and axial (b) reformatted images in a patient with gastric and paraesophageal varices which appear as brightly enhanc-

b

ing serpiginous veins adjacent to the gastric and esophageal wall (arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

A. Ba-Ssalamah et al.

602

paraesophageal varices. While esophageal varices can be appreciated easily by endoscopy, paraesophageal varices are seen only on CT or endoscopic sonography. In some patients, MDCT may indicate the etiology of the varices by demonstrating such conditions as cirrhosis, pancreatitis, or pancreatic carcinoma.

6

Gastric Adenocarcinoma

Gastric cancer remains a deadly disease, with overall 5-year survival rates of less than 20%. However, early gastric cancers are curable lesions, with 5-year survival rates exceeding 90%. The peak incidence is between 50 and 70 years of age (Moore 1986). Gastric carcinoma has striking geographic variations, with the highest prevalence reported in Japan and Korea. Conditions that predispose to the development of gastric carcinoma include atrophic gastritis, pernicious anemia, gastric polyps, partial gastrectomy, and Menetrier’s disease (Oiso 1975). About 30% of cancers are located in the antrum, 30% in the body, and 30% in the fundus or cardia. The remaining 10% are diffusely infiltrating lesions that involve the entire stomach. Most gastric cancers are adenocarcinomas of mucous cell origin (Parker et  al. 1996). Signet-­ ring-­cell carcinomas account for 5–15% of all gastric cancers and typically cause scirrhous infiltration of the gastric wall (Balthazar et  al. 1980). Scirrhous carcinomas frequently involve the distal half of the stomach, arise near the pylorus, and gradually extend upward from the antrum into the body and fundus. In advanced cases, the entire stomach is infiltrated by tumor.

6.1

Tumor Detection and Classification

Gastric carcinomas similar to esophageal carcinoma, may manifest as a focal area of mural thickening with or without ulceration, as a polypoid lesion, or as generalized mural thickening. In early gastric cancers, malignant invasion is limited to the mucosa or submucosa, regardless of

the presence of lymph node metastases (Maruyama and Baba 1994). By definition, advanced gastric cancer invades the muscularis propria. Advanced cancer (D'Elia et al. 2000) may manifest as large, segmental or diffuse wall thickening with irregular lobulation, and, often, ulceration. Signet-ring-cell cancer usually manifests as a scirrhous tumor of the stomach that leads to obliteration of gastric folds, diffuse wall thickening, and rigidity of the gastric lumen (i.e., linitis plastica). Gastric carcinoma can manifest as large, polypoid, fungating lesions as well. Carcinoma of the gastric cardia may be difficult to appreciate at CT because of the normal soft-tissue thickening that occurs at the gastroesophageal junction due to the reflections of the phreno-esophageal ligament and the attachments of the gastrohepatic ligament on the adjacent lesser curvature. Proper distention of the stomach, absence of perigastric linear infiltration, and pathologic lymph nodes help distinguish a focal or diffuse tumor from the normal gastroesophageal junction.

6.2

T Staging

In early gastric cancer, T1, a small focal thickening of the gastric wall can be seen, with no transmural enhancement (Fig. 25). In early-advanced gastric cancers (T2), malignant invasion is limited to the muscularis propria or serosa and the outer border may be smooth or show few soft-­tissue linear strands that extend into the surrounding fat, as seen with a desmoplastic or inflammatory reaction (Fig.  26). The probability of transmural extension of the tumor (T3) is directly correlated with mural thickness. In transmural extension, the serosal contour becomes blurred and strand-like areas of increased attenuation may be seen that extend into the perigastric fat (Fig.  27). Tumor spread frequently occurs via ligamentous and peritoneal reflections to adjacent organs (T4) (Fig. 28). The liver may be invaded via the gastrohepatic ligament, the pancreas via the lesser sac, and the transverse colon via the gastrocolic ligament. The distal esophagus is directly involved in carcinoma of the cardia in about 60% of patients,

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques Fig. 25 Hydro-MDCT of the stomach. Axial (a) and coronal (b) reformatted images show an early gastric cancer (pT1). This small polypoid lesion (arrows) protrudes more than 5 mm into the lumen with marked focal enhancement of the inner layer of the gastric wall. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

603

a

b

a

Fig. 26  Hydro-MDCT of the stomach. Axial (a) and coronal (b) reformatted images show an early advanced gastric cancer (pT2). There is a large, contrast-enhanced soft-tissue mass along the lesser curvature. Note that the

b

fat plane around the tumor is maintained, without evidence of perigastric fatty infiltration (arrowhead). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

A. Ba-Ssalamah et al.

604

a

b

Fig. 27  Hydro-MDCT of the stomach. Axial (a) and coronal (b) reformatted images show advanced gastric cancer (pT3). A large, polypoid carcinoma extends from the cardia down to the fundus of the stomach and up into

the distal esophagus with mild infiltration of the perigastric fatty tissue (arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

a

b

c

d

Fig. 28  Hydro-MDCT of the stomach. Axial (a) and coronal (b–d) reformatted images show diffuse, extensive, polypoid tumorous thickening of the gastric wall along the posterior aspect of the fundus and body. This is

a T4 tumor due to direct infiltration of the splenic hilum (arrows) and pancreatic tail (arrowhead). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques

whereas the duodenum is involved in carcinoma of the antrum in 13–18% of patients. It is often difficult to distinguish infiltration into the transverse mesocolon from infiltration of the mesenteric fat. Coronal or sagittal reformatted images are best suited for this purpose.

6.3

N Staging

Lymphatic spread is found in 74–88% of patients with gastric carcinoma because of the abundant lymphatic channels draining the stomach (Fukuya et al. 1995). The frequency of lymphatic metastases is related to the tumor size and depth of penetration. According to the American Joint Committee on Cancer (O'Sullivan et al. 2017), N staging depends on the number of positive perigastric lymph nodes (N1 1–2, N2 3–6, N3a ≥ 7–15, N3b ≥ 16 affected lymph nodes) Lymph node assessment for metastatic spread remains a challenge, even with MDCT. However, improved evaluation appears possible if both morphologic and enhancement criteria are used (Fig. 7).

6.4

M Staging

Hematogenous metastases from gastric carcinoma most commonly involve the liver since the stomach is drained by the portal vein (Fig. 29). a

Fig. 29  Contrast-enhanced FDG-PET-CT of the stomach. Axial reformatted images (a) and fused PET-CT images (b) show a solitary gastric small metastasis in the left liver lobe. Although it appears as a slightly nonspecific hypodense lesion on CT, it showed vigorous FDG avidity

605

Other less common sites of hematogenous spread include the lungs, adrenal glands, kidneys, bones, and brain. Lymph node involvement outside the perigastric location is considered M1 disease (Fig. 7). Advanced cancers may develop peritoneal metastases. Some patients with signet-ring-­ cell gastric carcinomas may have intraperitoneal metastastatic implants to the ovaries, known as Krukenberg tumors (Fig. 30).

6.5

Follow-Up After Partial Gastrectomy

A gastric “stump” cancer is defined as a primary carcinoma of the gastric remnant that occurs after latent periods of 15–25 years from the time of partial gastrectomy for gastric ulcers or other benign disease. Affected individuals usually have undergone a Billroth II procedure rather than a Billroth I procedure. These tumors tend to be located in the distal portion of the gastric remnant near the gastrojejunal anastomosis. It has been postulated that recurrent bile reflux proximal to the anastomosis causes chronic gastritis, intestinal metaplasia, and, eventually, gastric carcinoma (Fig.  31). Not all thickening of the anastomotic region is attributable to tumor involvement. Gastric surgery, such as a Nissen fundoplication procedure for gastric reflux, can sometimes simulate marked fold thickening or postoperative complications, such as a seroma. b

on the fused image, consistent with a metastasis (arrows). Note the tracer uptake in the lymph node metastasis pre-aortic. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

A. Ba-Ssalamah et al.

606

a

b

c

Fig. 30  Hydro-MDCT of the stomach. Coronal (a, c) and axial (b) reformatted images show tumor spread to the pelvis. The coronal reformatted image show the cancer in the gastric body as diffuse wall thickening (arrowheads), as well

a

Fig. 31  Hydro-MDCT of the stomach. Axial (a) and coronal (b) reformatted images show recurrence of gastric cancer after partial gastrectomy. The images display the recurrent cancer as a moderate wall thickening of the gastric remnant (arrowheads) which is accompanied by carcinomatosis along

as huge pelvic masses (arrows). At surgery the tumors, localized to the ovaries, were consistent with Krukenberg tumors. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

b

the hepatoduodenal ligament causing dilatation of the intrahepatic bile ducts (thin arrows), as well as ascites (thick black arrows), small-bowel dilatation and mesenteric lymphadenopathy (thick white arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques

7

Other Gastric Malignancies

7.1

Gastric Lymphoma

Lymphoma involves the stomach more frequently than any other portion of the gastrointestinal tract (Lewin et al. 1978). Primary gastric lymphomas are confined to the stomach and regional lymph nodes (about 35% of gastrointestinal lymphomas) and are predominantly non-Hodgkin’s lymphomas of B-cell origin (Lewin et  al. 1978). Lymphoma of mucosa-associated lymphoid tissue (MALT) is a distinct type of extranodal lymphoma that is characterized by a relatively indolent clinical course and has a much better prognosis than gastric carcinoma, with overall 5-year survival rates of 50–60% (Isaacson et  al. 1986). There is evidence linking Helicobacter pylori gastritis with the development of MALT lymphoma (Wotherspoon et  al. 1991). It is thus proposed that H. pylori infection may trigger the acquisition of MALT and that the subsequent inflammatory response may be a prerequisite for the development of MALT lymphoma. It has further been proposed that MALT lymphoma may be a precursor of high-grade B-cell non-­Hodgkin’s lymphoma in gastric tissue and that most highgrade lymphomas follow this evolutionary pathway (Chan et  al. 1990). However, high-grade, B-cell gastric lymphoma may also arise de novo (Parsonnet et  al. 1994). Although the clinical symptoms in high-grade and MALT lymphoma may be similar, they differ in several aspects. High-grade, B-cell lymphoma has a relatively aggressive course as opposed to the more indolent and favorable outcome of MALT lymphoma (Jung 2016). In addition, in high-grade gastric lymphomas, the extent of disease is usually greater at presentation, with involvement of adjacent organs and perigastric lymph nodes (Jung 2016). Lymphomas may involve any portion of the stomach. Transpyloric spread of tumor into the duodenum occurs in about 30% of patients (Cho et  al. 1996). Despite extensive lymphomatous infiltration, the stomach usually remains pliable and distensible without significant luminal narrowing. Whereas early gastric lymphoma is confined to the mucosa and submucosa, with an

607

average size of only 3.5  cm at diagnosis (Sato et al. 1986) (Fig. 32), advanced lesions can have a mean diameter of 10 cm. Most cases involve the antrum and body, although the entire stomach can be involved. There are four radiologic patterns of gastric lymphoma (Fischbach et al. 1992): (1) infiltrative lymphomas manifest as focal or diffuse thickened gastric folds due to the submucosal spread of tumor; (2) one or more ulcerated lesions are characteristic of ulcerative gastric lymphoma; (3) polypoid gastric lymphomas appear as intraluminal masses that may simulate polypoid adenocarcinomas; and, finally, (4) multiple submucosal nodules, ranging from several millimeters to several centimeters in size, are the hallmark of nodular gastric lymphoma. 18 F-FDG PET/CT has the advantages of detecting gastric lymphoma that is limited to the submucosal stage, which may be missed by gastroscopy, and in finding unanticipated lesions outside the stomach (Li et al. 2016).

7.1.1 CT Findings CT is the primary imaging modality for the pretreatment evaluation of abdominal lymphoma. In patients with suspected gastric lymphoma, MDCT can depict the gastric lesion and aid in staging of generalized lymphoma in the abdomen and chest. Furthermore, MDCT may aid in the early diagnosis of disease progression in patients undergoing therapy and follow-up for low-grade MALT lymphoma, which may progress to high-­grade, B-cell lymphoma. The CT appearances of lymphoma and gastric carcinoma may be very similar. Regarding the 18F-FDG PET/CT pattern, gastric lymphoma patients predominantly present with diffuse/segmental tracer uptake, whereas gastric carcinoma patients show mainly local tracer uptake (Li et al. 2016). Furthermore, the presence of splenomegaly, or involvement of lymph nodes in the retroperitoneal space below the renal hilus, may provide additional clues for the diagnosis of gastric lymphoma (Li et al. 2016). 18 F-FDG PET-CT appears to be highly accurate for the detection of relapse in patients with primary gastric lymphoma posttreatment (sensi-

A. Ba-Ssalamah et al.

608

a

b

c

d

Fig. 32  Hydro-MDCT of the stomach. Axial (a) and coronal (b) reformatted images in a patient with biopsy-­ proven gastric lymphoma show diffuse thickening of the gastric wall (arrows). Corresponding fused PET-CT

images (c, coronal and d, axial) show extensive FDG uptake. Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

tivity 96%, specificity 91%, and accuracy 93%). This high accuracy is observed in patients in whom relapse is suspected clinically or radiologically, as well as in patients who are undergoing routine follow-up (Sharma et al. 2013). In patients with a previous history of surgery and distorted anatomy, CT cannot reliably differentiate relapse from postoperative changes. Moreover, after chemotherapy/radiotherapy, the post-therapy fibrosis may persist on CT, which cannot be distinguished from residual disease (Fig. 32).

tumors were called carcinoids (Rindi et al. 1993). Carcinoid tumors of the stomach are rare (prevalence, approximately 0.3%) (Rindi et  al. 1993). Clinicopathological characterization of gastric carcinoid tumors has revealed three subtypes, which have unique endoscopic appearances, predisposing conditions, and clinical outcomes. Type 1 gastric carcinoid tumors are associated with enterochromaffin-like cell hyperplasia, hypergastrinemia, and chronic atrophic gastritis, with or without pernicious anemia. Type 1 tumors generally represent benign disease. Nodal and hepatic metastases are very rare (Rindi et al. 1996). Type 2 tumors are the least common type, representing 5–10% of gastric carcinoid tumors. They are seen in the hypergastrinemia referred to as Zollinger-Ellison syndrome in association with multiple endocrine neoplasia (MEN) type 1. Approximately 30% of patients with MEN 1 have gastric carcinoid tumors (Modlin and Tang 1996). Type 2 tumors also arise from

7.2

Gastro-Entero-Pancreatic Neuroendocrine Tumors (GEPNETs)

GEPNETs are increasing in incidence, and accurate staging is important for the selection of the appropriate treatment. Prior to the 2010 World Health Organization re-classification, these

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques

enterochromaffin-­ like cells in the setting of hyperplasia. These tumors are multicentric and variable in size but are prone to spread to local lymph nodes. Tumor-related death is rare, as is carcinoid syndrome. The appearance of these tumors on CT and contrast upper gastrointestinal (UGI) series can be striking because there are multiple masses in the setting of diffusely thickened gastric wall folds (Binstock et al. 2001). Type 3 gastric carcinoid tumors are sporadic tumors and are not associated with a hypergastrinemic state. They represent about 13% of gastric carcinoid tumors (Binstock et  al. 2001). Unlike type 1 and 2 tumors, type 3 tumors are large, solitary tumors that may show ulceration and are more likely to be invasive with distant metastases. The likelihood of metastases is dependent upon tumor size (Modlin and Tang 1996). Carcinoid syndrome, which manifests with the triad of flushing, blushing, and diarrhea, occurs when tumorproduced serotonin and kallirein enter the systemic circulation. The presence of carcinoid syndrome, then, suggests that the liver is unable to breakdown serotonin, either due to underlying liver disease or metastatic disease to the liver. The prognosis is poor, with a 5-year survival rate of 20% (Rindi et al. 1996; Modlin and Tang 1996). Like neuroendocrine tumors (NET), gastric carcinoids are graded, histologically, depending upon either their mitotic rate or the percentage of the tumor cells that label positively for Ki-67 antigen. Those with the lowest mitotic rate and/or the smallest percentage of Ki-67 antigen labelling are referred to as grade 1 (G1). The higher the grade, the less well differentiated and those classified as G3 are known as neuroendocrine carcinomas.

7.2.1 CT Findings The improved understanding of gastric carcinoid tumors has important implications for the radiologic evaluation of patients. When these tumors are suspected, contrast material and water-­ enhanced MDCT should be used to detect small mucosal masses. The discovery of polyps in a patient known to have chronic atrophic gastritis should alert the radiologist to the possibility of type 1 gastric carcinoid tumors. MDCT is necessary to properly assess type 2 (MEN 1-­associated) and type 3 (sporadic) gastric carcinoid tumors,

609

given the increased predisposition for spread to local lymph nodes and hepatic metastases (Binstock et al. 2001). 68 Ga-DOTATATE PET/CT imaging is more sensitive for staging and detecting unknown primary GEPNETs than CT alone Fig.  33). Therefore, we believe it should be implemented in the initial management and follow-up of patients with GEPNETs, as it can significantly optimize patient care decisions (Sadowski et al. 2016). The degree of NET differentiation is important not only because of its impact in determining prognosis but also because it plays a key role in NET detection. When imaging with PET-CT, the ideal method of detecting and staging NETs, G1 and G2 NETs take up 68Ga-DOTATATE radiotracer, due to the presence of somatostatin receptors. However, the less well-differentiated NETs become, the fewer somatostatin receptors they have, until G3 tumors lack sufficient receptors to accumulate 68Ga-DOTATATE. This ­false-­negative for NETs can be avoided by performing a conventional FDG PET scan that is based upon tumor glucose activity. Naturally, the FDG scan should be done in settings where there is sufficient clinical evidence, for example, chromogranin A serum levels or serotonin breakdown products in urine, or radiological evidence, such as a hypervascular gastric mass on the arterial phase of dynamic CT or hypervascular hepatic and/or osteosclerotic bony metastases, to suggest that a NET is likely. Other causes for falsenegative PET imaging when a GEPNET is, in fact, present, include small tumor size, poor quality study, and/or high background gastric uptake of radiotracer. Tumors less than 1 cm cannot reliably be detected. The exam quality can be judged by looking at the intensity of splenic activity relative to liver, and by checking for physiologic uptake in the adrenal and pituitary glands (Deppen et al. 2016).

7.3

Gastrointestinal Stromal Tumors (GISTs)

GISTs have recently been recognized as the most common mesenchymal neoplasm of the gastrointestinal tract. Many, but not all, mesenchymal

A. Ba-Ssalamah et al.

610

a

b

c

d

Fig. 33  Hydro-MDCT of the stomach. Axial (a), coronal (b), and fused PET-CT (c, d) reformatted images. The biopsy-proven carcinoid of the stomach (gastrinoma) appears as a small slightly hypervascular lesion on the arterial phase (a, b, arrows). It shows vigorous uptake of

radionuclide tracer because its somatostatin receptors are strongly Dotatoc-avid (c, d, arrowheads). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

tumors previously diagnosed as leiomyomas, leiomyoblastomas, and leiomyosarcomas are now considered GISTs on the basis of specific immunohistochemical criteria, specifically, c-kit positivity (Miettinen and Lasota 2001). The malignant variety of GISTs represents only about 3% of all malignant gastrointestinal tumors (Davis et al. 2000). Approximately 60–70% are found in the stomach (Miettinen et al. 1999). It is known that 10–30% of GISTs are malignant and the risk of malignancy increases with extragastric location, a diameter greater than 5 cm, and extension to adjacent organs (Miettinen et  al. 1999;

DeMatteo et  al. 2000). Before and during surgery, in the absence of metastases, it is difficult to distinguish benign from malignant lesions. As with leiomyomas and leiomyosarcomas, intramural, intraluminal, and exophytic GISTs can be distinguished (Fig. 34).

7.4

Neural Tumors

Neural tumors constitute about 5–10% of benign gastric tumors (Hoare and Elkington 1976). The majority are nerve sheath tumors (neurinomas,

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques

a

611

b

c

Fig. 34  Hydro-MDCT of the stomach. Axial (a), coronal (b), and sagittal (c) reformatted images in a patient with GIST of the stomach shows a large, inhomogeneous, lobulated and partially calcified mass located between the left

liver lobe and the fundus of the stomach (arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

schwannomas, or neuromas). Most nerve sheath tumors are benign, but sarcomatous degeneration in these lesions has occasionally been reported. Neural tumors in the stomach usually appear on CT scans as submucosal masses (with or without ulceration) that are indistinguishable from other mesenchymal tumors (Fig. 35).

leading cause (30–35% of cases) (Fig.  36), but other infiltrating primary malignant tumors or metastatic lesions can also produce gastric outlet obstruction. Infrequently, the condition is caused by mural infiltration or spasm resulting from inflammatory disorders, such as severe pancreatitis or cholecystitis. Fibrous scarring after ingestion of corrosive substances may cause antral narrowing. CT demonstrates a markedly distended stomach. Differentiation between a benign and a malignant cause of gastric outlet obstruction is based on evidence of a mass in the region of the gastric outlet or by demonstration of inflammatory disease in the vicinity of these structures (Goodman and Halpert 1991).

7.5

Gastric Outlet Obstruction

In adults, peptic ulcer disease is, by far, the most common cause of gastric outlet obstruction (60–65% of cases). An annular constricting carcinoma of the distal antrum or pylorus is the second

A. Ba-Ssalamah et al.

612

a

c

Fig. 35  Hydro-MDCT of the stomach. Axial (a), coronal (b), and fused PET-CT images (c, axial and d, coronal) show a 3  cm biopsy-proven schwannoma. The lesion is

b

d

lobulated, hyperdense, and exhibits strong FDG uptake (arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques Fig. 36 Hydro-MDCT of the stomach. Coronal (a), sagittal (b), and axial (c) reformatted images in a patient who presented with gastric outlet obstruction. The grossly distended stomach contains a large volume of debris (arrowheads). The thickening and enhancement of the gastric antrum/pylorus was confirmed as gastric cancer at surgery (arrows). Published with kind permission of Medical University of Vienna 2019. All Rights Reserved

a

613

b

c

References Antoch G, Kanja J, Bauer S, Kuehl H, Renzing-Koehler K, Schuette J  et  al (2004) Comparison of PET, CT, and dual-modality PET/CT imaging for monitoring of imatinib (STI571) therapy in patients with gastrointestinal stromal tumors. J Nucl Med 45(3):357–365 Ascenti G, Racchiusa S, Mazziotti S, Bottari M, Scribano E (1999) Giant fibrovascular polyp of the esophagus: CT and MR findings. Abdom Imaging 24(2):109–110

Balthazar EJ, Rosenberg H, Davidian MM (1980) Scirrhous carcinoma of the pyloric channel and distal antrum. AJR Am J Roentgenol 134(4):669–673 Balthazar EJ, Naidich DP, Megibow AJ, Lefleur RS (1987) CT evaluation of esophageal varices. AJR Am J Roentgenol 148(1):131–135 Bar-Shalom R, Guralnik L, Tsalic M, Leiderman M, Frenkel A, Gaitini D et al (2005) The additional value of PET/CT over PET in FDG imaging of oesophageal cancer. Eur J  Nucl Med Mol Imaging 32(8): 918–924

614 Ba-Ssalamah A, Prokop M, Uffmann M, Pokieser P, Teleky B, Lechner G (2003) Dedicated multidetector CT of the stomach: spectrum of diseases. Radiographics 23(3):625–644 Ba-Ssalamah A, Zacherl J, Noebauer-Huhmann IM, Uffmann M, Matzek WK, Pinker K et  al (2009) Dedicated multi-detector CT of the esophagus: spectrum of diseases. Abdom Imaging 34(1):3–18 Ba-Ssalamah A, Matzek W, Baroud S, Bastati N, Zacherl J, Schoppmann SF et  al (2011) Accuracy of hydro-­ multidetector row CT in the local T staging of oesophageal cancer compared to postoperative histopathological results. Eur Radiol 21(11):2326–2335 Berkovich GY, Levine MS, Miller WT Jr (2000) CT findings in patients with esophagitis. AJR Am J Roentgenol 175(5):1431–1434 Binstock AJ, Johnson CD, Stephens DH, Lloyd RV, Fletcher JG (2001) Carcinoid tumors of the stomach: a clinical and radiographic study. AJR Am J Roentgenol 176(4):947–951 Blom RL, Vliegen RF, Schreurs WM, Belgers HJ, Stohr I, Oostenbrug LE et al (2012) External ultrasonography of the neck does not add diagnostic value to integrated positron emission tomography-computed tomography (PET-CT) scanning in the diagnosis of cervical lymph node metastases in patients with esophageal carcinoma. Dis Esophagus 25(6):555–559 Boellaard R, van Lingen A, Lammertsma AA. Experimental and clinical evaluation of iterative reconstruction (OSEM) in dynamic PET: quantitative characteristics and effects on kinetic modeling. J Nucl Med 2001;42(5):808–817 Bradley J, Bae K, Choi N, Forster K, Siegel BA, Brunetti J et al (2012) A phase II comparative study of gross tumor volume definition with or without PET/CT fusion in dosimetric planning for non-small-cell lung cancer (NSCLC): primary analysis of radiation therapy oncology group (RTOG) 0515. Int J Radiat Oncol Biol Phys 82(1):435–441.e1 Carlisle JG, Quint LE, Francis IR, Orringer MB, Smick JF, Gross BH (1993) Recurrent esophageal carcinoma: CT evaluation after esophagectomy. Radiology 189(1):271–275 Carter M, Deckmann RC, Smith RC, Burrell MI, Traube M (1997) Differentiation of achalasia from pseudoachalasia by computed tomography. Am J  Gastroenterol 92(4):624–628 Cerfolio RJ, Bryant AS (2006) Maximum standardized uptake values on positron emission tomography of esophageal cancer predicts stage, tumor biology, and survival. Ann Thorac Surg 82(2):391–394; discussion 94–5 Chan JK, Ng CS, Isaacson PG (1990) Relationship between high-grade lymphoma and low-grade B-cell mucosa-­ associated lymphoid tissue lymphoma (MALToma) of the stomach. Am J Pathol 136(5):1153–1164 Chao YK, Liu YH, Ko PJ, Wu YC, Hsieh MJ, Liu HP et al (2005) Treatment of esophageal perforation in a referral center in Taiwan. Surg Today 35(10):828–832

A. Ba-Ssalamah et al. Chen BC, Wang HH, Lin YC, Shih YL, Chang WK, Hsieh TY (2013) Isolated gastric variceal bleeding caused by splenic lymphoma-associated splenic vein occlusion. World J Gastroenterol 19(40):6939–6942 Cheng JC, Laforest R, O'Sullivan JA (2011) A scatter and randoms weighted (SRW) iterative PET reconstruction. Med Phys 38(6):3186–3192 Cho KC, Baker SR, Alterman DD, Fusco JM, Cho S (1996) Transpyloric spread of gastric tumors: comparison of adenocarcinoma and lymphoma. AJR Am J Roentgenol 167(2):467–469 Choi H, Charnsangavej C, de Castro Faria S, Tamm EP, Benjamin RS, Johnson MM et  al (2004) CT evaluation of the response of gastrointestinal stromal tumors after imatinib mesylate treatment: a quantitative analysis correlated with FDG PET findings. AJR Am J Roentgenol 183(6):1619–1628 Choi J, Kim SG, Kim JS, Jung HC, Song IS (2010) Comparison of endoscopic ultrasonography (EUS), positron emission tomography (PET), and computed tomography (CT) in the preoperative locoregional staging of resectable esophageal cancer. Surg Endosc 24(6):1380–1386 Davis GB, Blanchard DK, Hatch GF 3rd, Wertheimer-­ Hatch L, Hatch KF, Foster RS Jr et al (2000) Tumors of the stomach. World J Surg 24(4):412–420 De Lutio di Castelguidone E, Pinto A, Merola S, Stavolo C, Romano L (2005) Role of Spiral and Multislice Computed Tomography in the evaluation of traumatic and spontaneous oesophageal perforation. Our Exp Radiol Med 109(3):252–259 D'Elia F, Zingarelli A, Palli D, Grani M (2000) Hydro-­ dynamic CT preoperative staging of gastric cancer: correlation with pathological findings. A prospective study of 107 cases. Eur Radiol 10(12):1877–1885 DeMatteo RP, Lewis JJ, Leung D, Mudan SS, Woodruff JM, Brennan MF (2000) Two hundred gastrointestinal stromal tumors: recurrence patterns and prognostic factors for survival. Ann Surg 231(1):51–58 Deppen SA, Liu E, Blume JD, Clanton J, Shi C, Jones-­ Jackson LB et  al (2016) Safety and efficacy of 68Ga-DOTATATE PET/CT for diagnosis, staging, and treatment Management of Neuroendocrine Tumors. J Nucl Med 57(5):708–714 Engeroff B, Kopka L, Harz C, Grabbe E (2001) Impact of different iodine concentrations on abdominal enhancement in biphasic multislice helical CT (MS-CT). RöFo 173(10):938–941 Eren S, Ciris F (2005) Diaphragmatic hernia: diagnostic approaches with review of the literature. Eur J Radiol 54(3):448–459 Fearon K, Strasser F, Anker SD, Bosaeus I, Bruera E, Fainsinger RL et  al (2011) Definition and classification of cancer cachexia: an international consensus. Lancet Oncol 12(5):489–495 Fischbach W, Kestel W, Kirchner T, Mossner J, Wilms K (1992) Malignant lymphomas of the upper gastrointestinal tract. Results of a prospective study in 103 patients. Cancer 70(5):1075–1080

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques Flamen P, Lerut A, Van Cutsem E, De Wever W, Peeters M, Stroobants S et al (2000) Utility of positron emission tomography for the staging of patients with potentially operable esophageal carcinoma. J  Clin Oncol 18(18):3202–3210 Fukuya T, Honda H, Hayashi T, Kaneko K, Tateshi Y, Ro T et  al (1995) Lymph-node metastases: efficacy for detection with helical CT in patients with gastric cancer. Radiology 197(3):705–711 Gelfand MD, Botoman VA (1987) Esophageal motility disorders: a clinical overview. Am J  Gastroenterol 82(3):181–187 Goodman P, Halpert RD (1991) Radiological evaluation of gastric stapling procedures for morbid obesity. Crit Rev Diagn Imaging 32(1):37–67 Guo H, Zhu H, Xi Y, Zhang B, Li L, Huang Y et al (2007) Diagnostic and prognostic value of 18F-FDG PET/ CT for patients with suspected recurrence from squamous cell carcinoma of the esophagus. J  Nucl Med 48(8):1251–1258 Haley M, Konski A, Li T, Cheng JD, Maurer A, Haluszka O et al (2009) Influence of diabetes on the interpretation of PET scans in patients with esophageal cancer. Gastrointest Cancer Res 3(4):149–152 Halpern BS, Dahlbom M, Quon A, Schiepers C, Waldherr C, Silverman DH et al (2004) Impact of patient weight and emission scan duration on PET/CT image quality and lesion detectability. J Nucl Med 45(5):797–801 Halvorsen RA, Thompson WM (1984) Computed tomographic evaluation of esophageal carcinoma. Semin Oncol 11(2):113–126 Hatch GF 3rd, Wertheimer-Hatch L, Hatch KF, Davis GB, Blanchard DK, Foster RS Jr et al (2000) Tumors of the esophagus. World J Surg 24(4):401–411 Heeren PA, Jager PL, Bongaerts F, van Dullemen H, Sluiter W, Plukker JT. Detection of distant metastases in esophageal cancer with (18)F-FDG PET.  J Nucl Med 2004;45(6):980–987 Hoare AM, Elkington SG (1976) Gastric lesions in generalized neurofibromatosis. Br J Surg 63(6):449–451 Hough DM, Yu L, Shiung MM, Carter RE, Geske JR, Leng S et al (2013) Individualization of abdominopelvic CT protocols with lower tube voltage to reduce i.v. contrast dose or radiation dose. AJR Am J Roentgenol 201(1):147–153 Hsu WH, Hsu PK, Wang SJ, Lin KH, Huang CS, Hsieh CC et  al (2009) Positron emission tomography-­ computed tomography in predicting locoregional invasion in esophageal squamous cell carcinoma. Ann Thorac Surg 87(5):1564–1568 Isaacson PG, Spencer J, Finn T (1986) Primary B-cell gastric lymphoma. Hum Pathol 17(1):72–82 Jacobs JM, Hill MC, Steinberg WM (1991) Peptic ulcer disease: CT evaluation. Radiology 178(3):745–748 Jebasingh FK, Naik D, Chandramohan A, Paul MJ (2014) Posterior gastric diverticulum mimicking adrenal adenoma on imaging. BMJ Case Rep 2014 Jeganathan R, McGuigan J, Campbell F, Lynch T (2011) Does pre-operative estimation of oesophageal tumour metabolic length using 18F-fluorodeoxyglucose PET/

615

CT images compare with surgical pathology length? Eur J Nucl Med Mol Imaging 38(4):656–662 Jung JT (2016) Gastric mucosa-associated lymphoid tissue lymphoma based on outcome of domestic treatment. Korean J Gastroenterol 68(4):175–178 Kaplan KJ (2004) Primary esophageal lymphoma: a diagnostic challenge. South Med J 97(4):331–332 Kato H, Nakajima M, Sohda M, Tanaka N, Inose T, Miyazaki T et  al (2009) The clinical application of (18)F-fluorodeoxyglucose positron emission tomography to predict survival in patients with operable esophageal cancer. Cancer 115(14):3196–3203 Kelsen DP, Ginsberg R, Pajak TF, Sheahan DG, Gunderson L, Mortimer J et al (1998) Chemotherapy followed by surgery compared with surgery alone for localized esophageal cancer. N Engl J Med 339(27):1979–1984 Keum B, Kim YS, Jeen YT, Chun HJ, Um SH, Kim CD et al (2006) Dysphagia lusoria assessed by 3-­dimensional CT. Gastrointest Endosc 64(2):268–269 Kim HW, Won KS, Song BI, Kang YN (2015) Correlation of primary tumor FDG uptake with histopathologic features of advanced gastric cancer. Nucl Med Mol Imaging 49(2):135–142 Kobori O, Kirihara Y, Kosaka N, Hara T (1999) Positron emission tomography of esophageal carcinoma using (11)C-choline and (18)F-fluorodeoxyglucose: a novel method of preoperative lymph node staging. Cancer 86(9):1638–1648 Kolomainen D, Hurley PR, Ebbs SR (2017) Esophageal duplication cyst: case report and review of the literature. Dis Esophagus 11(1):62–65 Kontaxakis G, Strauss LG, Thireou T, Ledesma-Carbayo MJ, Santos A, Pavlopoulos SA et  al (2002) Iterative image reconstruction for clinical PET using ordered subsets, median root prior, and a web-based interface. Mol Imaging Biol 4(3):219–231 Krause BJ, Herrmann K, Wieder H, zum Buschenfelde CM (2009) 18F-FDG PET and 18F-FDG PET/CT for assessing response to therapy in esophageal cancer. J Nucl Med 50(Suppl 1):89S–96S Kuhlman JE, Fishman EK, Wang KP, Siegelman SS (1985) Esophageal duplication cyst: CT and transesophageal needle aspiration. AJR Am J  Roentgenol 145(3):531–532 Lagergren J, Ye W, Lindgren A, Nyren O (2000) Heredity and risk of cancer of the esophagus and gastric cardia. Cancer Epidemiol Biomark Prev 9(7):757–760 Lea JW, Prager RL, Bender HW Jr (1984) The questionable role of computed tomography in preoperative staging of esophageal cancer. Ann Thorac Surg 38(5):479–481 LeBlang SD, Nunez DB Jr (1999) Helical CT of cervical spine and soft tissue injuries of the neck. Radiol Clin N Am 37(3):515–532. v-vi Lee NK, Kim S, Jeon TY, Kim HS, Kim DH, Seo HI et  al (2010) Complications of congenital and developmental abnormalities of the gastrointestinal tract in adolescents and adults: evaluation with multimodality imaging. Radiographics 30(6):1489–1507

616 Levine MS, Elmas N, Furth EE, Rubesin SE, Goldwein MI (1996) Helicobacter pylori and gastric MALT lymphoma. AJR Am J Roentgenol 166(1):85–86 Lewin KJ, Ranchod M, Dorfman RF (1978) Lymphomas of the gastrointestinal tract: a study of 117 cases presenting with gastrointestinal disease. Cancer 42(2):693–707 Li R, Chen TW, Hu J, Guo DD, Zhang XM, Deng D et al (2013) Tumor volume of resectable adenocarcinoma of the esophagogastric junction at multidetector CT: association with regional lymph node metastasis and N stage. Radiology 269(1):130–138 Li XF, Fu Q, Dong YW, Liu JJ, Song XY, Dai D et  al (2016) (18)F-fluorodeoxyglucose positron emission tomography/computed tomography comparison of gastric lymphoma and gastric carcinoma. World J Gastroenterol 22(34):7787–7796 Liu PS, Levine MS, Torigian DA (2006) Esophagopleural fistula secondary to esophageal wall ballooning and thinning after pneumonectomy: findings on chest CT and esophagography. AJR Am J  Roentgenol 186(6):1627–1629 Lordick F, Ott K, Krause BJ, Weber WA, Becker K, Stein HJ et al (2007) PET to assess early metabolic response and to guide treatment of adenocarcinoma of the oesophagogastric junction: the MUNICON phase II trial. Lancet Oncol 8(9):797–805 Lowe VJ, Booya F, Fletcher JG, Nathan M, Jensen E, Mullan B et al (2005) Comparison of positron emission tomography, computed tomography, and endoscopic ultrasound in the initial staging of patients with esophageal cancer. Mol Imaging Biol 7(6):422–430 Ludeman L, Shepherd NA (2005) Serosal involvement in gastrointestinal cancer: its assessment and significance. Histopathology 47(2):123–131 Luketich JD, Schauer PR, Meltzer CC, Landreneau RJ, Urso GK, Townsend DW et al (1997) Role of positron emission tomography in staging esophageal cancer. Ann Thorac Surg 64(3):765–769 Mani NB, Suri S, Gupta S, Wig JD (2001) Two-phase dynamic contrast-enhanced computed tomography with water-filling method for staging of gastric carcinoma. Clin Imaging 25(1):38–43 Maruyama M, Baba Y (1994) Diagnosis of the invasive depth of gastric cancer. Abdom Imaging 19(6):532–536 Mendelson RM, Fermoyle S (2005) Primary gastrointestinal lymphomas: a radiological-pathological review. Part 1: Stomach, oesophagus and colon. Australas Radiol 49(5):353–364 Meyers BF, Downey RJ, Decker PA, Keenan RJ, Siegel BA, Cerfolio RJ et  al (2007) The utility of positron emission tomography in staging of potentially operable carcinoma of the thoracic esophagus: results of the American College of Surgeons Oncology Group Z0060 trial. J Thorac Cardiovasc Surg 133(3):738–745

A. Ba-Ssalamah et al. Miettinen M, Lasota J  (2001) Gastrointestinal stromal tumors--definition, clinical, histological, ­immunohistochemical, and molecular genetic features and differential diagnosis. Virchows Arch 438(1):1–12 Miettinen M, Sarlomo-Rikala M, Lasota J  (1999) Gastrointestinal stromal tumors: recent advances in understanding of their biology. Hum Pathol 30(10):1213–1220 Modlin IM, Tang LH (1996) The gastric enterochromaffin-­ like cell: an enigmatic cellular link. Gastroenterology 111(3):783–810 Monges G, Bisot-Locard S, Blay JY, Bouvier AM, Urbieta M, Coindre JM et al (2010) The estimated incidence of gastrointestinal stromal tumors in France. Results of PROGIST study conducted among pathologists. Bull Cancer 97(3):E16–E22 Moore JR (1986) Gastric carcinoma: 30-year review. Can J Surg 29(1):25–28 Moss AA, Schnyder P, Thoeni RF, Margulis AR (1981) Esophageal carcinoma: pretherapy staging by computed tomography. AJR Am J  Roentgenol 136(6):1051–1056 Nagaki A, Onoguchi M, Matsutomo N (2011) Patient weight-based acquisition protocols to optimize (18) F-FDG PET/CT image quality. J  Nucl Med Technol 39(2):72–76 Nomura M, Shitara K, Kodaira T, Hatooka S, Mizota A, Kondoh C et  al (2012) Prognostic impact of the 6th and 7th American joint committee on cancer TNM staging systems on esophageal cancer patients treated with chemoradiotherapy. Int J Radiat Oncol Biol Phys 82(2):946–952 Oeda S, Otsuka T, Akiyama T, Ario K, Masuda M, Taguchi S et al (2010) Recurrent acute pancreatitis caused by a gastric duplication cyst communicating with an aberrant pancreatic duct. Intern Med 49(14):1371–1375 Oiso T (1975) Incidence of stomach cancer and its relation to dietary habits and nutrition in Japan between 1900 and 1975. Cancer Res 35(11 Pt. 2):3254–3258 Okada M, Murakami T, Kumano S, Kuwabara M, Shimono T, Hosono M et  al (2009) Integrated FDG-PET/CT compared with intravenous contrast-enhanced CT for evaluation of metastatic regional lymph nodes in patients with resectable early stage esophageal cancer. Ann Nucl Med 23(1):73–80 O'Sullivan B, Brierley J, Byrd D, Bosman F, Kehoe S, Kossary C et  al (2017) The TNM classification of malignant tumours-towards common understanding and reasonable expectations. Lancet Oncol 18(7):849–851 Panebianco V, Grazhdani H, Iafrate F, Petroni M, Anzidei M, Laghi A et al (2006) 3D CT protocol in the assessment of the esophageal neoplastic lesions: can it improve TNM staging? Eur Radiol 16(2):414–421 Parker SL, Tong T, Bolden S, Wingo PA (1996) Cancer statistics, 1996. Cancer J Clin 46(1):5–27

Imaging of the Stomach and Esophagus Using CT and PET/CT Techniques Parsonnet J, Hansen S, Rodriguez L, Gelb AB, Warnke RA, Jellum E et al (1994) Helicobacter pylori infection and gastric lymphoma. N Engl J Med 330(18):1267–1271 Pearlberg JL, Sandler MA, Madrazo BL (1983) Computed tomographic features of esophageal intramural pseudodiverticulosis. Radiology 147(1):189–190 Peyrin-Biroulet L, Bronowicki JP, Bigard MA, Regent D, Walter S, Platini C (2006) Contribution of computed tomography with oral media contrast to the diagnosis of esophago-pericardial fistula. Clin Imaging 30(5):347–349 Picus D, Balfe DM, Koehler RE, Roper CL, Owen JW (1983) Computed tomography in the staging of esophageal carcinoma. Radiology 146(2):433–438 Prokop M (2005) New challenges in MDCT. Eur Radiol 15(Suppl 5):E35–E45 Quint LE, Glazer GM, Orringer MB, Gross BH (1985) Esophageal carcinoma: CT findings. Radiology 155(1):171–175 Rampin L, Nanni C, Fanti S, Rubello D (2005) Value of PET-CT fusion imaging in avoiding potential pitfalls in the interpretation of 18F-FDG accumulation in the distal oesophagus. Eur J Nucl Med Mol Imaging 32(8):990–992 Rashid F, Aber A, Iftikhar SY (2012) A review on gastric diverticulum. World J Emerg Surg 7(1):1 Rice TW, Patil DT, Blackstone EH (2017) 8th edition AJCC/UICC staging of cancers of the esophagus and esophagogastric junction: application to clinical practice. Ann Cardiothorac Surg 6(2):119–130 Rindi G, Luinetti O, Cornaggia M, Capella C, Solcia E (1993) Three subtypes of gastric argyrophil carcinoid and the gastric neuroendocrine carcinoma: a clinicopathologic study. Gastroenterology 104(4):994–1006 Rindi G, Bordi C, Rappel S, La Rosa S, Stolte M, Solcia E (1996) Gastric carcinoids and neuroendocrine carcinomas: pathogenesis, pathology, and behavior. World J Surg 20(2):168–172 Sadowski SM, Neychev V, Millo C, Shih J, Nilubol N, Herscovitch P et  al (2016) Prospective study of 68Ga-DOTATATE positron emission tomography/ computed tomography for detecting gastro-Entero-­ pancreatic neuroendocrine tumors and unknown primary sites. J Clin Oncol 34(6):588–596 Sargent RL, Hood IC (2006) Asphyxiation caused by giant fibrovascular polyp of the esophagus. Arch Pathol Lab Med 130(5):725–727 Sato T, Sakai Y, Ishiguro S, Furukawa H (1986) Radiologic manifestations of early gastric lymphoma. AJR Am J Roentgenol 146(3):513–517 Schramm D, Bach AG, Zipprich A, Surov A (2014) Imaging findings of gastric diverticula. ScientificWorldJournal 2014:923098 Seremetis MG, Lyons WS, deGuzman VC, Peabody JW Jr (1976) Leiomyomata of the esophagus. An analysis of 838 cases. Cancer 38(5):2166–2177

617

Sharma NK, Silverman JS, Li T, Cheng J, Yu JQ, Haluszka O et al (2011) Decreased posttreatment SUV on PET scan is associated with improved local control in medically inoperable esophageal cancer. Gastrointest Cancer Res 4(3):84–89 Sharma P, Suman SK, Singh H, Sharma A, Bal C, Malhotra A et  al (2013) Primary gastric lymphoma: utility of 18F-fluorodeoxyglucose positron emission tomography-­computed tomography for d­ etecting relapse after treatment. Leuk Lymphoma 54(5):951–958 Siewert JR (2007) [Esophageal carcinoma]. Der Chirurg. Zeitschrift fur alle Gebiete der Operativen Medizen 78(5):475–484; quiz 85 Simmang CL, Reed K, Rosenthal D (1989) Leiomyomas of the gastrointestinal tract. Mil Med 154(1):45–47 Skehan SJ, Brown AL, Thompson M, Young JE, Coates G, Nahmias C (2000) Imaging features of primary and recurrent esophageal cancer at FDG PET. Radiographics 20(3):713–723 Smithers BM, Fahey PP, Corish T, Gotley DC, Falk GL, Smith GS et al (2010) Symptoms, investigations and management of patients with cancer of the oesophagus and gastro-oesophageal junction in Australia. Med J Aust 193(10):572–577 Stein HJ, Feith M, Bruecher BL, Naehrig J, Sarbia M, Siewert JR (2005) Early esophageal cancer: pattern of lymphatic spread and prognostic factors for long-term survival after surgical resection. Ann Surg 242(4):566–573; discussion 73–5 Suga K, Yasuhiko K, Hiyama A, Takeda K, Matsunaga N (2009) F-18 FDG PET/CT findings in a patient with bilateral orbital and gastric mucosa-­associated lymphoid tissue lymphomas. Clin Nucl Med 34(9):589–593 Sun L, Su XH, Guan YS, Pan WM, Luo ZM, Wei JH et  al (2009) Clinical usefulness of 18F-FDG PET/ CT in the restaging of esophageal cancer after surgical resection and radiotherapy. World J Gastroenterol 15(15):1836–1842 Talanow R, Shrikanthan S (2010) Imaging protocols for 18F-FDG PET/CT in overweight patients: limitations. J Nucl Med 51(4):662; author reply 62 Tamandl D, Paireder M, Asari R, Baltzer PA, Schoppmann SF, Ba-Ssalamah A (2016a) Markers of sarcopenia quantified by computed tomography predict adverse long-term outcome in patients with resected oesophageal or gastro-oesophageal junction cancer. Eur Radiol 26(5):1359–1367 Tamandl D, Gore RM, Fueger B, Kinsperger P, Hejna M, Paireder M et  al (2016b) Change in volume parameters induced by neoadjuvant chemotherapy provide accurate prediction of overall survival after resection in patients with oesophageal cancer. Eur Radiol 26(2):311–321 Thompson WM, Halvorsen RA Jr (1994) Staging esophageal carcinoma II: CT and MRI. Semin Oncol 21(4):447–452

618 Thompson WM, Halvorsen RA, Foster WL Jr, Williford ME, Postlethwait RW, Korobkin M (1983) Computed tomography for staging esophageal and gastroesophageal cancer: reevaluation. AJR Am J  Roentgenol 141(5):951–958 Tjendra Y, Lyapichev K, Henderson J, Rojas CP (2016) Foregut duplication cyst of the stomach: a case report and review of the literature. Case Rep Pathol 2016:7318256 Tunaci A (2002) Postoperative imaging of gastrointestinal tract cancers. Eur J Radiol 42(3):224–230 Ulla M, Cavadas D, Munoz I, Beskow A, Seehaus A, Garcia-Monaco R (2010) Esophageal cancer: pneumo-­ 64-­MDCT. Abdom Imaging 35(4):383–389 Umeoka S, Koyama T, Watanabe G, Saga T, Kataoka M, Togashi K et al (2010) Preoperative local staging of esophageal carcinoma using dual-phase contrast-­ enhanced imaging with multi-detector row computed tomography: value of the arterial phase images. J Comput Assist Tomogr 34(3):406–412 Weber WA, Ott K, Becker K, Dittler HJ, Helmberger H, Avril NE et  al (2001) Prediction of response to preoperative chemotherapy in adenocarcinomas of the

A. Ba-Ssalamah et al. esophagogastric junction by metabolic imaging. J Clin Oncol 19(12):3058–3065 van Westreenen HL, Westerterp M, Bossuyt PM, Pruim J, Sloof GW, van Lanschot JJ et al (2004) Systematic review of the staging performance of 18F-fluorodeoxyglucose positron emission tomography in esophageal cancer. J Clin Oncol 22(18):3805–3812 Wotherspoon AC, Ortiz-Hidalgo C, Falzon MR, Isaacson PG (1991) Helicobacter pylori-associated gastritis and primary B-cell gastric lymphoma. Lancet 338(8776):1175–1176 Wu CX, Zhu ZH (2014) Diagnosis and evaluation of gastric cancer by positron emission tomography. World J Gastroenterol 20(16):4574–4585 Young CA, Menias CO, Bhalla S, Prasad SR (2008) CT features of esophageal emergencies. Radiographics 28(6):1541–1553 Yu W, Fu XL, Zhang YJ, Xiang JQ, Shen L, Chang JY (2011) A prospective evaluation of staging and target volume definition of lymph nodes by 18FDG PET/CT in patients with squamous cell carcinoma of thoracic esophagus. Int J  Radiat Oncol Biol Phys 81(5):e759–e765

Small Bowel MDCT Marco Rengo, Simona Picchia, and Andrea Laghi

Contents

Abstract

1    Introduction

 619

2    General Recommendations for Imaging 2.1  Patient Preparation 2.2  Acquisition Protocol 2.3  Enteral Contrast Medium

 620  620  620  621

3    Small Bowel Diseases 3.1  Tumours 3.2  Crohn’s Disease 3.3  Intestinal Ischemia and Infarction 3.4  Obstructions and Perforations 3.5  Other Entities

 621  621  627  631  633  635

References

 636

Imaging has always been critical in the evaluation of the small bowel, because of poor accessibility with conventional endoscopic techniques. And also new endoscopic imaging methods (i.e. capsule endoscopy and double-balloon and push enteroscopy) have not completely solved the diagnostic problem in many clinical situations involving the small bowel, and they are still considered complimentary, second- or third-level tests, to be used in restricted clinical scenarios (Flamant et al. 2009).

1

The original version of this chapter was revised. Water marks and line numbers have been removed. M. Rengo (*) • S. Picchia • A. Laghi Department of Radiological Sciences, Oncology and Pathology, Faculty of Medicine and Dentistry, University of Rome Sapienza I.C.O.T., Via F. Faggiana 34, 04100 Latina, Italy e-mail: [email protected]

Introduction

Imaging has always been critical in the evaluation of the small bowel, because of poor accessibility with conventional endoscopic techniques. And also new endoscopic imaging methods (i.e. capsule endoscopy and double-balloon and push enteroscopy) have not completely solved the diagnostic problem in many clinical situations involving the small bowel, and they are still considered complimentary, second- or third-level tests, to be used in restricted clinical scenarios (Flamant et al. 2009). In the past, optimal diagnostic results were obtained with traditional barium studies, like small bowel follow-through and small bowel enterography/enteroclysis. But it was with the advent of cross-sectional imaging techniques that

Med Radiol Diagn Imaging (2017) DOI 10.1007/174_2017_22, © Springer International Publishing AG Published Online: 15 February 2017

619

M. Rengo et al.

620

imaging made a clear step forward. In fact, imaging moved from the evaluation of the lumen and mucosal surface to direct analysis of the small bowel wall, with a consequent clear improvement in diagnostic accuracy, in both inflammatory and neoplastic diseases. The evaluation of the small bowel with MDCT was introduced by Raptopoulos et al. (1997) for the assessment of inflammatory bowel diseases (Raptopoulos et al. 1997; Paulsen et al. 2006). At that time, the necessity of the combination of bowel distention, with a low-density contrast to maximize bowel wall enhancement, was recognized as well as the need for optimization of enhancement with a specific “enteric phase” to improve conspicuity of small bowel wall pathology. Since then, advances in technology of CT scanners, from single- to multi-slice, opened a new era in small bowel imaging. Robustness, reproducibility, high spatial resolution and scanning speed are the main features of MDCT making it the method of choice in many clinical settings. Because of concerns for dose exposure, MRI of the small bowel gained advantages particularly in the evaluation of young patients affected by inflammatory bowel diseases, but recent improvements in low-dose exams, based on iterative reconstruction algorithms, are now making MDCT feasible again also in those situations. In this chapter, the authors will illustrate main technical features concerning the study of the small bowel with CT as well as the findings of the most frequent pathologies encountered in clinical practice.

2

General Recommendations for Imaging

2.1

Patient Preparation

A 4–6 h fasting period is necessary as well as the evaluation of renal function by means of estimated glomerular filtration rate (GFR) calculation since the IV injection of contrast medium is recommended in all settings. There is no evidence that routine medications should be stopped.

The use of spasmolytic agents is preferred since it improves the bowel distension magnitude, especially of the proximal segments. Among the recommended agents, the first choice is hyoscine butylbromide (0.5 mg/kg IV), and the second choice is glucagon (1 mg IV). However, since CT acquisition is relatively fast and motion artefacts due to peristaltic contractions are rare, there is no evidence of a real benefit in using spasmolytic agents (Taylor et al. 2016).

2.2

Acquisition Protocol

The acquisition of unenhanced phase is recommended for the evaluation of ischemic or oncologic diseases, while it is not necessary for inflammatory disease. Enhanced phases should be optimized according to the clinical setting. For all enhanced phases, the use of a CM bolus detection technique is recommended. The monitoring of the CM bolus arrival should be placed at the level of the origin of the superior mesenteric artery. A circular region of interest (ROI) should be placed in the aorta, and a threshold of 100 HU represents the reference value for synchronization with scan beginning. After the threshold image acquisition should start with a variable delay according to the clinical needs. For the evaluation of suspected arterial ischemia or arterial bleeding, an early arterial phase should be acquired. For early arterial phase, a delay of 6 s should be set after the trigger (Huprich et al. 2008; Huprich et al. 2013). A late arterial phase is recommended for the evaluation of neoplastic disease. In this setting, a delay of 10 s from the trigger should be used. For all clinical indications, especially for IBD, an enteric phase is mandatory setting a delay of 14 s from the trigger (Schindera et al. 2007). A portal venous phase is recommended in case of bleedings and ischemia. A portal venous phase should be acquired 45 s after the trigger. CM injection protocol should be optimized according to patient’s size and acquisition protocol. CM should be injected at an iodine delivery rate (IDR) ranging between 1.2 and 1.6 g of iodine per second depending on the tube voltage

Small Bowel MDCT

used (Rengo et al. 2011). The CM volume should be calculated on the basis of patient’s size. Both total body weight (TBW) and lean body weight (LBW) can be used for the quantification; in the first case, 0.625 mg of iodine per Kg of TBW should be injected and in the second, 0.750 mg of iodine per Kg of LBW (Rengo et al. 2011). The acquisition of thin slice images is recommended as well as the minimization of radiation exposure by optimization of tube voltage (Guimaraes et al. 2010), tube current and iterative reconstruction techniques (Kambadakone et al. 2011; Gandhi et al. 2016).

2.3

Enteral Contrast Medium

Bowel distension is mandatory in most clinical settings, with the exception of emergency. The small bowel can de distended both by oral administration of CM (CT enterography) or by injection of CM through a nasojejunal catheter (CT enteroclysis) (Minordi et al. 2011) (Fig. 1). There is no single recommended enteral CM. Usually, water with hyperosmolar agents is recommended like mannitol (with or without locust a

Fig. 1 (a) CT enterography. Distention of the small bowel obtained with oral neutral contrast media (*). (b) CT enteroclysis. Distention (*) of the small bowel obtained

621

bean gum), polietilenglicole (PEG), sorbitol and lactulose (Borthne et al. 2006) (Fig. 2). The volume of CM to achieve an optimal bowel distension should range between 1 l and 2.5 l, and it should be ingested in 45–60 min (Kuehle et al. 2006).

3

Small Bowel Diseases

3.1

Tumours

Primary tumours of the small bowel are rare (less than 6% of tumours of the gastrointestinal tract), even if the incidence is increasing over time; this change is primarily driven by the rising incidence of carcinoid tumours (Pan and Morrison 2011; Rondonotti et al. 2017). The most frequent malignant tumours are adenocarcinoma, carcinoid tumour, lymphoma and sarcoma (Pan and Morrison 2011), and the most frequent benign tumours are leiomyoma, adenoma and lipoma (Pourmand and Itzkowitz 2016). The prevalence of small bowel cancer tends to increase with age, and men have higher rates than women (Haselkorn et al. 2005). Metastases are more common in the small bowel than primary tumours and may occur by b

with the injection of neutral contrast media through nasojejunal catheter (arrow)

M. Rengo et al.

622

a

b

Fig. 2 (a) Distention of the small bowel obtained with neutral contrast media. (b) Distention of the small bowel obtained with positive contrast media

i­ntraperitoneal seeding (most common), haematogenous spread, lymphatic spread or local extension. Diagnosis is often delayed because symptoms are often absent or nonspecific (abdominal pain, anaemia, gastrointestinal bleeding, weight loss). Sometimes the tumour presents in the emergency setting with small bowel obstruction. Both MRI and CT enterography/enteroclysis have good sensitivity and specificity for the diagnosis of small bowel tumours, respectively, 0.91 and 0.95 and 0.92 and 0.99 (Van Weyenberg et al. 2010; Soyer et al. 2013a). MRI is preferred for the absence of radiation, but it has less spatial resolution than that achieved with CT; therefore, it is less useful for the evaluation of small lesions (Amzallag-Bellenger et al. 2012). Capsule endoscopy has been introduced recently in clinical practice to facilitate early diagnosis of small bowel tumours in patients without symptoms of obstruction (Cobrin et al. 2006; Rondonotti et al. 2017). PET-CT is not a first-line test, but it can be useful when CT or MR is uncertain and for the evaluation of disease response to treatment and small bowel metastasis (Cronin et al. 2012).

CT enterography and enteroclysis are the methods of choice to investigate the small bowel wall. Studies about the adequacy of distention of small bowel obtained with the two methods are controversial: some authors found that distention of the small bowel with CT enterography may not be as optimal as with enteroclysis, especially for the jejunum and proximal ileum (van der Merwe et al. 2013); others found the distention not significantly different (Wold et al. 2003). However, when tolerated, CT enterography is often preferred due to reduced invasiveness (Dave-Verma et al. 2008). Intravenous CM is also administered to evaluate lesion enhancement.

3.1.1 CT Findings of Benign Tumours Leiomyoma is the most common small bowel benign tumour (incidence 20–45%) (Sokhandon et al. 2016), often detected in the jejunum. It is often asymptomatic (usually an incidental finding). Among symptomatic patients, bleeding is the most common presentation because of high vascularity and high probability of ulceration of the lesion. Obstruction in case of intraluminal growth is the

Small Bowel MDCT

second most common clinical presentation (Sokhandon et al. 2016). Leiomyoma typically presents as a solitary lesion, rounded, with sharp margins and homogeneous aspect. The origin may be submucosal or polypoid, intramural and extraluminal (less common) (Gourtsoyiannis and Mako 1997). At CT leiomyoma, it presents as intramural solitary masses with moderate contrast enhancement and no signs of invasion of adjacent organs. Calcifications inside the lesion are pathognomonic. Adenoma is the second most common benign tumour (incidence 15–20%) (Sokhandon et al. 2016). Familial adenomatous polyposis is associated with an increased incidence of this lesion (Kopacova et al. 2013). The size is usually 1–3 cm (only villous adenomas can be sized over 3 cm), and it involves predominantly the duodenum and ileocaecal valve. Histological classification consists of three types: villous, the most common, which can be sessile or polypoid, tubular and tubulovillous (Sokhandon et al. 2016). Villous adenomas may undergo malignant transformation. They are often solitary. When multiple, they usually affect a single segment, are of different sizes and may be pedunculated; this appearance is in contrast with familial polyposis, where they are ubiquitously localized in the small bowel and colon, sessile and usually of about equal size (Gourtsoyiannis and Mako 1997). Adenomas are usually asymptomatic; only large polyps may cause intussusception. At CT their shape is round, oval or lobulated; margins are usually regular. Most of them show homogenous and moderate enhancement after intravenous CM. Lipoma is the third most common benign tumour (incidence ranging between 8% and 20%) (Sokhandon et al. 2016), most often located in the ileum (region of the ileocaecal valve). Usually lipoma is solitary, with submucosal origin (from the adipose tissue of the submucosa) and intraluminal projection; it is well circumscribed, encapsulated and with ovoid or lobulated shape. The size is about 1–6 cm in diameter (Miao et al. 2010). Lipoma is usually asymptomatic and incidentally found; large-size lipoma can cause intussusception, and this is frequently the primary symptom (Thompson 2005). CT attenuation is similar to

623

adipose tissue, which is the pathognomonic CT feature (Fig. 3), with no enhancement following intravenous injection of CM (Fang et al. 2010); density can be inhomogeneous when necrosis, cystic degeneration or calcification is present (more common when lesions are larger than 2 cm).

3.1.2 C  T Findings of Malignant Tumours Adenocarcinoma is mostly located near the papilla in the duodenum (Fig. 4), less frequently in the jejunum and rarely in the ileum (Dabaja et al. 2004). Crohn’s disease is associated with an

Fig. 3  Axial CT image acquired during venous phase. The arrow shows small bowel lipoma with its typical attenuation similar to the adipose tissue

Fig. 4  Axial CT image acquired during arterial phase. (*) shows duodenal adenocarcinoma with moderate and inhomogeneous enhancement

624

M. Rengo et al.

Fig. 6  Coronal CT image acquired during venous phase. Figure shows several common CT findings associated to adenocarcinoma: typical eccentric wall thickening of the lesion (thick arrow), multiple peritoneal metastasis (thin arrows), lymphadenopathies (arrow head) and ascites (*)

Fig. 5  Coronal CT image acquired during arterial phase. The arrow shows adenocarcinoma associated with Crohn’s disease

increased risk of developing small intestinal adenocarcinoma (Fig. 5) (Gourtsoyiannis and Mako 1997), and in this case, the lesion is more frequently located in the ileum (Sokhandon et al. 2016). It is often symptomatic, but symptoms are not specific (abdominal pain, obstruction with proximal dilatation especially if located in the duodenum and chronic blood loss). For this reason, the diagnosis is often made with the detection of metastasis (usually the liver and peritoneum). Adjacent lymphadenopathies are found in less than 50% (Dudiak et al. 1989). CT findings are eccentric wall thickening, with solitary polypoid or ulcerated (ulceration is frequent) mass and with soft tissue attenuation and moderate and inhomogeneous contrast enhancement (Fig. 6). Carcinoid tumours are a group of neuroendocrine tumours arising from submucosa. The incidence of carcinoid tumours increased over the last decades. Ninety percent are located in the gastrointestinal tract, especially in the distal ileum and appendix (Pelage et al. 1999). Clinical presentation is not specific (abdominal pain is the most common

Fig. 7  Axial CT image obtained during arterial phase. The arrow shows ileal carcinoid tumour with avid arterial enhancement

symptom). Carcinoid syndrome occurs only in a minority of patients. Lesion can be solitary (more common) or multiple and typically less than 2 cm (Gourtsoyiannis and Mako 1997). The main CT features are the presence of internal calcifications and avid arterial ­ enhancement after intravenous CM (Fig. 7). There is also typical invasion of the serosa with desmoplastic reaction that causes wall thickening, retraction of the adjacent small bowel loops, lymphadenopathies and mesenteric masses (that can show internal calcifications like the primary tumour) (Fig. 8) (Soyer et al. 2013b). Non-Hodgkin lymphoma is the third most common neoplasm, mostly located in the distal

Small Bowel MDCT

a

625

b

c

Fig. 8 (a) Axial CT image obtained during arterial phase. The arrow shows mesenteric reaction with internal calcification associated with carcinoid tumour. (b) Coronal CT

a

image obtained during arterial phase. The arrow shows mesenteric reaction associated with small bowel carcinoid tumour. (c) Surgical resection of carcinoid tumour

b

Fig. 9 (a) Axial CT image obtained during venous phase. The arrow shows ileal lymphoma with homogenous and moderate enhancement. (b) Surgical resection of lymphoma

ileum (Fig. 9). Hodgkin lymphoma is rare (Morgan et al. 2004). Celiac disease and immunocompromised state are risk factors. Symptoms are nonspecific (abdominal pain, weight loss, chronic blood loss). Typical CT feature is circumferential and asymmetrical wall thickening with “intestinal aneurysm” (luminal dilatation)

without obstruction (Fig. 10); it is caused by destruction of the autonomic nerve plexus with consequent peristalsis block. This tumour can have central necrosis and ulcerations (especially large-size lesions) (Fig. 11); for these reason, there can be internal cavitation (Gore et al. 2006). Lymphoma usually does not have desmoplastic

626

M. Rengo et al.

Fig. 10  Axial CT image obtained during venous phase. The arrow shows ileal lymphoma with circumferential and asymmetrical wall thickening and typical “intestinal aneurysm” Fig. 12  Coronal CT image obtained during venous phase. The figure shows ileal lymphoma (thick arrow) with associated multiple bulky mesenteric lymphadenopathies (thin arrows)

Fig. 11  Coronal CT image obtained during venous phase. The arrow shows ileal lymphoma with internal necrosis

reaction. Bulky mesenteric lymphadenopathy (Fig. 12) and splenomegaly support the diagnosis. Enhancement is homogenous and moderate (Shinya et al. 2016). Less common appearances are multifocal sites of localizations along the

small bowel (in the same or in separate segments) or a polypoid mass protruding in the lumen that can cause obstruction or intussusceptions (Anzidei et al. 2011). The most frequent small bowel sarcomas are gastrointestinal stromal tumours (GIST) and leiomyosarcoma (Shenoy 2015). GIST most frequently affects the stomach and small bowel. They are not always malignant lesions: malignant GISTs are more frequent when larger than 5 cm. Usually symptoms are nonspecific (abdominal pain and nausea); obstruction is rare. For these reasons, the diagnosis is often made in case of metastases or large-size lesions. Malignant GIST usually grows as large extraluminal mass; it displaces rather than invades adjacent structures, and it can have internal haemorrhage, necrosis and cavitation. Enhancement is generally high and heterogeneous (for the internal areas of haemorrhage or necrosis) (Sokhandon et al. 2016); homogenous enhancement is usually observed in small lesions (Fig. 13). Metastases

Small Bowel MDCT

a

Fig. 13 (a) Coronal CT image obtained during arterial phase. The arrow shows small ileal GIST with early homogeneous enhancement. (b) Axial CT image obtained

are more frequent in the liver and peritoneum (Hong et al. 2006). Lymphadenopathies are uncommon. The differential diagnosis between leiomyosarcoma and leiomyoma can be difficult; both occur more frequently in the ileum and jejunum. As leiomyoma, leiomyosarcoma presents as a single large lesion (>5 cm), with extraluminal growth and low incidence of obstruction (as the GIST); however, ulcerations, causing bleeding, and abdominal pain are more frequent. Moreover, irregular margins, enlarged lymph nodes and intraperitoneal seeding are specific for leiomyosarcoma. In rare cases, leiomyosarcoma can cause a palpable abdominal mass (Gourtsoyiannis and Mako 1997). Metastases are usually incidental findings in patients with known primary tumour. They most commonly arise from intraperitoneal seeding, in particular from the ovary, appendix and colon. Haematogenous metastases are more common in patients with melanoma, lung, breast, colon and renal cancers. It could be difficult to differentiate metastases from primary tumours on CT because they often present similar characteristics (Sokhandon et al. 2016). They do not have uniform growth pattern: may be single or multifocal, with or without mesenteric involvement. Larger lesions may cause luminal stenosis, surface

627

b

during arterial phase. The arrow shows ileal GIST with early inhomogeneous enhancement due to central necrosis

Fig. 14  Coronal CT image obtained during venous phase. The figure shows ileal metastasis with internal necrosis (thick arrow) and ulceration (thin arrow)

ulcerations and internal necrosis (Fig. 14). Metastases are often associated with nodal disease (Fig. 15).

3.2

Crohn’s Disease

Crohn’s disease (CD) is an idiopathic inflammatory bowel disease (IBD) which can involve any portion of the gastrointestinal tract, although it is more common in the terminal ileum and colon. The diagnosis is made usually in young patients (the peak is 15–25 years) with chronic diarrhoea and recurrent

628

Fig. 15  Coronal CT image obtained during venous phase. The arrow shows ileal metastasis (arrow head), jejunal metastasis (*) and associated lymphadenopathy (arrow)

abdominal pain. Patients with older diagnosis (about 15% percent) are less likely to have complications (Quezada et al. 2013). Although many hypotheses were made, CD remains idiopathic. It could be due to both a genetic and secondary cause (i.e. infective agents, geographical location, etc.) (Ventham et al. 2013; Fogel et al. 2017). There is not yet a cure for CD. Immunosuppressive drugs can be favouring the remission of active disease and prevent reactivation. Despite advances in medical treatment, a large number of patients require surgery (Maxwell et al. 2016). Initial diagnosis of CD is not simple and straightforward since a single and reliable diagnostic test, considered to be a gold standard, is missing. Thus, diagnosis is based on a combination of endoscopic, histological, radiological and/ or biochemical investigations (Panes et al. 2013). Ileocolonoscopy with biopsies of the terminal ileum and of different colonic segments is mandatory to assess the diagnosis, and cross-sectional imaging is a natural complement, able to detect, stage and characterize the disease.

M. Rengo et al.

Initially only the mucosa is involved with shallow aphthoid ulceration. Over time, the entire bowel wall is involved with transmural lesions, which can lead to complications like stenosis, fistulas and abscesses. Intestinal fibrosis is the most common complication of CD, resulting in stricture formation in the small intestine and colon. It is important to distinguish inflammatory strictures from fibrotic strictures because clinical management is different: the first are usually responsive to treatment, while fibrotic strictures require surgery (Spinelli et al. 2010). Small bowel is affected in 80% of the cases (Furukawa et al. 2004). CT and MR have a similar diagnostic accuracy (Lee et al. 2009), and they are better than other techniques for the assessment of transmural lesions and active inflammation and evaluation of extra-intestinal complications. Capsule endoscopy is better for assessing early mucosal disease (Flamant et al. 2009; Lu et al. 2010). Both CT/MR enterography and CT/MR enteroclysis diagnose CD with similar diagnostic accuracy (Negaard et al. 2007; Lalitha et al. 2011; Lang et al. 2015). The lack of ionizing radiation from MR would make it a better option, because patients with CD are often young, and CD is a chronic disease, which requires repeated imaging examinations to assess the status (Lee et al. 2009; Wallihan et al. 2015). However, recent iterative reconstruction algorithms allow to obtain high-quality images with significant reduction of radiation dose around 40% (Wallihan et al. 2015). Distension is necessary to evaluate small bowel wall in these patients. Despite the controversy about the accuracy of distension (Wold et al. 2003; van der Merwe et al. 2013), the diagnostic accuracy for Crohn’s disease of CT enterography and CT enteroclysis is similar (Mazzeo et al. 2001). Thus, CT enterography is preferred over CT enteroclysis because of its minor invasivity, if a patient is able to drink a reasonable amount of water (Ilangovan et al. 2012). Intravenous CM is essential in diagnosing active inflammatory small bowel Crohn’s disease and also complications (Baker et al. 2015). Sensitivity of CT is reported to be 98% in the diagnosis of transmural and extramural Crohn’s disease but is

Small Bowel MDCT

only 70% for an early-stage disease (Furukawa et al. 2004).

3.2.1 CT Findings Bowel wall thickening is the most frequent feature of Crohn’s disease, both in the active inflammatory and fibrostenosing disease. A wall thickness > 4 mm is considered pathological (Furukawa et al. 2004), and it is due to the presence, within the submucosal fat, of pus, blood, oedema or fibrosis. Bowel thickening is more consistent in the active inflammation, and it usually ranges from 1 to 2 cm (Furukawa et al. 2004; Park and Lim 2013) (Fig. 16). A thick wall (present in up to 83% of patients) is observed most

Fig. 16  Coronal CT image obtained during enteric phase. The arrow shows bowel wall thickening in recurrent Crohn’s disease at ileocolonic anastomosis

629

frequently in the terminal ileum, though other portions of the small bowel may be similarly affected (Maxwell et al. 2016). It is usually circumferential and discontinuous with typical skip lesions; it can be asymmetric because of preferential involvement along the mesenteric border of the bowel wall in Crohn’s disease (Ilangovan et al. 2012; Fernandes et al. 2014). Three predominant phenotypes of Crohn’s disease can be observed: predominantly inflammatory, predominantly fibrostenosing and penetrating. The enhancement of involved segments can suggest the activity of the disease and can be used to discriminate between inflammatory and fibrostenosing disease. The predominantly inflammatory disease may have a marked homogeneous (the intensity of the enhancement is correlated with the grade of activity) or a stratified enhancement (alternating layers of higher or lower attenuation) (Fig. 17). The stratified enhancement may have “target” or “double-halo” appearance (Al-Hawary et al. 2013). The “target” sign is characterized by three concentric rings; the inner one is represented by the enhancement of the

Fig. 17  Coronal CT image obtained during venous phase. The figure shows different possible enhancement of active inflammatory Crohn’s disease: stratified (thick arrow) and homogenous (thin arrow)

630

mucosa and the outer one by the enhancement of the serosa; between these two layers, the unenhanced submucosa (infiltrated by oedema or fat) can be observed (Fig. 18). The “double-halo” sign is characterized by two concentric rings; the inner one is represented by the enhancement of the mucosa and the outer by the unenhancing submucosa (Fig. 19). Layered mural enhancement is usually lost in fibrostenosing disease, where homogenous, less intense and delayed

Fig. 18  Axial CT obtained during venous phase. The arrow shows stratified enhancement with typical “target sign” in Crohn’s disease

Fig. 19  Coronal CT obtained during arterial phase. The arrows show typical skip lesions and stratified enhancement with “double-halo sign” in Crohn’s disease

M. Rengo et al.

enhancement is more frequent, because it is suggestive of transmural fibrosis. A typical finding of active inflammation is the presence of penetrating ulcers which appears as the presence of focal defect of the mucosa and the submucosa is limited to the serosa. The involvement of mesentery is typical of active inflammation. Two findings can be observed in this setting: the fibrofatty proliferation and the “comb” sign. Fibrofatty proliferation is represented by the inhomogeneous enhancement of the fat surrounding the involved segment associated or not with the presence of lymph nodes. The “comb” sign consists of the dilatation and tortuosity of the vasa recta on the mesenteric side of the involved segment (Hill and DiSantis 2015). Fibrostenosing disease is characterized by the presence of strictures and upstream bowel dilatation, usually >3 cm. Stenosis may occur also in patients with active inflammatory disease; however, they are more frequent in patients with fibrosis (Park and Lim 2013); evaluating the enhancement behaviour of strictures and thus distinguishing active from chronic disease are important in treatment management, as mentioned before (Fig. 20). Penetrating disease is characterized by the presence of fistulas (Fig. 21), abscess and phlegmons.

Fig. 20  Axial CT image during venous phase. The figure shows Crohn’s disease complicated with stricture (small arrow) and consequent pre-stenotic dilatation (big arrow)

Small Bowel MDCT

a

631

b

Fig. 21 (a) Axial CT image during arterial phase. The figure shows Crohn’s disease complicated with Ileosigmoid fistula. (b) Axial CT image during venous phase. The arrow shows Crohn’s disease complicated with cutaneous fistula

Most frequently, fistulas occur between the bowel loops (entero-enteric), with the abdominal wall (entero-cutaneous), in the perineum or with any adjacent organ (the bladder, uterus, colon). In most cases, the sinus tract or the fistula can be detected due to the high spatial resolution of MDCT (Paquet et al. 2016).

3.3

Intestinal Ischemia and Infarction

Small bowel ischemia arises from any cause of vascular compromise of the small bowel wall. It can be acute or chronic (less common). The main causes of acute ischemia are occlusion of the superior mesenteric artery (proximal or distal), occlusion of the superior mesenteric vein (proximal or distal), mechanical small bowel ­obstruction,

low flow (shock bowel) and acute enteritis induced by chemotherapy or radiotherapy. The main causes of chronic ischemia are atherosclerotic stenosis of the superior mesenteric artery and chronic enteritis induced by radiotherapy. Occlusion of the superior mesenteric artery due to embolization or thrombosis on atherosclerosis is responsible for 60–70% of cases of acute ischemia, followed by non-occlusive conditions and occlusion of the superior mesenteric vein due to thrombosis. Acute occlusion of the superior mesenteric artery mostly results from an embolus that originates from the left atrium consequent to atrial fibrillation (Wiesner et al. 2003). Acute mesenteric ischemia (AMI) can be partially mural (i.e. involves only the mucosa and submucosa) or transmural (i.e. involves all bowel wall layers and leads to infarction with necrosis of the bowel segment/s). Acute bowel infarction

632

has still a high mortality rate, approximately 50–70% (Tendler 2003; Tilsed et al. 2016). Early diagnosis can recognize AMI before intestinal necrosis and determine optimal treatment, but often diagnosis is difficult, particularly in the early phase. The main symptom of AMI is severe acute abdominal pain; other symptoms can be nausea, vomiting and diarrhoea. The development of transmural infarction may also present with fever, bloody diarrhoea and shock (Tilsed et al. 2016). Contrast-enhanced computed tomography (CT) has replaced angiography as the first diagnostic step, largely because not only vessels but also bowel diseases and other related abdominal findings can be diagnosed (Liu and Platt 2014). It also has the advantage to find alternative diagnosis of acute abdominal pain. CT of the abdomen and pelvis is generally performed with intravenous CM (unenhanced, arterial and portal phases) for the evaluation of vessels and bowel wall enhancement. A critical decision in patients with possible acute ischemia is to administer enteral CM for bowel distention. CT with intravenous CM alone avoids the waiting time that requires the oral contrast to arrive in the small intestine; however, without oral contrast, the bowel is poorly distended, and the evaluation of the bowel wall could be more difficult. It therefore depends essentially on the evaluation of a patient’s clinical condition. The question about which luminal CM is better (either positive or neutral) in possible acute bowel ischemia is also open. On the one hand, positive CM is probably better to delineate fluid collections, hematomas and bowel leakage (Dhatt et al. 2015); on the other hand, neutral CM (e.g. water) is probably better to evaluate bowel wall enhancement, especially without wall thickening (Wiesner et al. 2003).

3.3.1 CT Findings Acute small bowel ischemia can be due to several causes. CT findings depend on the cause (e.g. embolus in the superior mesenteric artery and thrombosis of the superior mesenteric vein), and others can be found in all acute cases of ischemia/infarction such as bowel wall thickening or

M. Rengo et al.

thinning, luminal dilatation, bowel wall’s altered attenuation in the unenhanced phase and altered enhancement, intraperitoneal fluid, pneumatosis intestinalis and pneumatosis portalis and signs of perforation (e.g. pneumoperitoneum). Normal small bowel wall measures 2.5–3 cm between the two outer walls) with or without air-­ fluid levels in the context and normal calibre or collapse of distal loops. Sudden transition from dilated to normal or collapsed loops (called “transition point” or “transition zone”) facilitates the localization of the obstruction. Dilated loops (with or without air-fluid levels) were also seen in adynamic ileus, without mechanical obstruction and transition point. The grade of bowel dilatation alone is not a criterion for distinguishing mechanical small bowel obstruction from adynamic ileus (Furukawa et al. 2001). Another CT finding in mechanical small bowel obstruction is

Small Bowel MDCT

“small bowel faeces sign”, found in 5–7% of patients (Paulson and Thompson 2015) and more common in the high-grade obstruction, which arises from the stasis of gas bubbles in the lumen of the small intestine that gives it a colon’s appearance. A “closed-loop obstruction” with its characteristic U-shaped or C-shaped configuration is diagnosed when bowel segment of variable length is occluded at two adjacent points along its course, often consequent to adhesions or hernias; the dilatation of the involved segment can be consistent and lead to mesenteric ischemia (in this case, it is called “strangulation”). Main CT findings of bowel perforation are free extraluminal air (entity and localization) and focal discontinuity of the bowel wall on the perforation site. Extraluminal leakage in case of oral contrast media administration is other highly sensitive finding. Perforation site is not always found; free air bubbles are usually concentrated near the perforation site, especially when the amount of air is small, and this finding can help the localization (Kim et al. 2009). In contrast to gastroduodenal perforation, the amount of extraluminal free air in jejunum and ileum perforation can be very small, usually localized among mesenteric folds (Furukawa et al. 2005). Extraluminal air originates not only from gastrointestinal perforation; other possible causes are mechanical ventilation, pneumothorax, peritoa

Fig. 25 (a) Axial CT image obtained in unenhanced phase. The arrow shows fish spine that causes intestinal perforation with associated abscess. (b). Oblique CT

635

neal lavage and chest injury (Kim et al. 2009). These causes had to be searched, especially when the perforation site is not found. Other CT findings of bowel perforation are segmental bowel wall thickening (more common), bowel wall enhancement, perivisceral fat stranding, adjacent abscess (Fig. 25) and intraperitoneal fluid, in addition to signs related to the specific adjacent cause.

3.5

Other Entities

3.5.1 Graft-Versus-Host Disease (GVHD) Graft-versus-host disease is still a frequent complication of allogeneic post-haematopoietic stem cell transplantation. It can be acute or chronic on the basis of post-procedural time of onset: the chronic form traditionally arises beyond the first 3 months (Brodoefel et al. 2010). The liver, skin and gastrointestinal tract (especially small bowel) are the principal affected organs. Symptoms are aspecific (abdominal pain, diarrhoea, nausea, vomiting and fever). It requires an early diagnosis and treatment with immunosuppressive drugs. Abdominal CT with intravenous contrast media is generally performed. Intestinal CT features are in order of frequency: bowel wall thickening (the most b

image obtained in arterial phase. The arrow shows abscess associated with intestinal perforation caused by fish spine

636

M. Rengo et al.

phy for the assessment of small-bowel diseases beyond f­ requent), bowel enhancement (mucosal enhanceCrohn disease. Radiographics 32(5):1423–1444 ment is more common than “target sign” derived Anzidei M, Napoli A, Zini C, Kirchin MA, Catalano C, from mucosal and serosal enhancement) and Passariello R (2011) Malignant tumours of the small bowel dilatation of multiple segments. Extra-­ intestine: a review of histopathology, multidetector CT and MRI aspects. Br J Radiol 84(1004):677–690 intestinal CT findings are in order of frequency: Baker ME, Hara AK, Platt JF, Maglinte DD, Fletcher JG engorgement of adjacent vasa recta (the most fre(2015) CT enterography for Crohn’s disease: optimal quent), mesenteric fat stranding, ascites and spletechnique and imaging issues. Abdom Imaging nomegaly (Kalantari et al. 2003). Biliary 40(5):938–952 abnormalities and periportal oedema can also be Barlow JM, Johnson CD, Stephens DH (1996) Celiac disease: how common is jejunoileal fold pattern reversal seen (Shimoni et al. 2012).

3.5.2 Coeliac Disease Coeliac disease is a chronic autoimmune disorder caused by intolerance of gluten. Often diagnosis is late because symptoms are aspecific and have slow progression. Principal symptoms are abdominal pain, irregular alvus and weight loss. CT or MR enterography/enteroclysis with intravenous contrast media is usually performed. However, CT and MR cannot always see the eventual ulcerations (condition called “ulcerative jejunitis”, more frequent in refractory to a gluten-­diet disease), which can be seen with endoscopy. The disease tends to start in the duodenum and extends into the ilium. The small bowel mucosa is primarily affected, because of progressive loss of villi and hypertrophy of crypts with consequent intraluminal fluid excess. Main CT features are dilated fluid-filled loops, bowel wall thickening, vascular engorgement and intramural fat. A characteristic feature is partial or complete “jejunoileal fold pattern reversal” when there are more ileal than jejunal folds, and it is often not associated with other findings of malabsorption (Barlow et al. 1996). Mesenteric lymphadenopathies, especially upper mesenteric lymph nodes, are commonly found. If the small bowel malabsorption is severe, large fluid volumes passed also into the colon (Scholz et al. 2011).

References Al-Hawary MM, Kaza RK, Platt JF (2013) CT enterography: concepts and advances in Crohn’s disease imaging. Radiol Clin N Am 51(1):1–16 Amzallag-Bellenger E, Oudjit A, Ruiz A, Cadiot G, Soyer PA, Hoeffel CC (2012) Effectiveness of MR enterogra-

found at small-bowel follow-through? AJR Am J Roentgenol 166(3):575–577 Borthne AS, Abdelnoor M, Storaas T, Pierre-Jerome C, Klow NE (2006) Osmolarity: a decisive parameter of bowel agents in intestinal magnetic resonance imaging. Eur Radiol 16(6):1331–1336 Brodoefel H, Bethge W, Vogel M, Fenchel M, Faul C, Wehrmann M, Claussen C, Horger M (2010) Early and late-onset acute GvHD following hematopoietic cell transplantation: CT features of gastrointestinal involvement with clinical and pathological correlation. Eur J Radiol 73(3):594–600 Cobrin GM, Pittman RH, Lewis BS (2006) Increased diagnostic yield of small bowel tumors with capsule endoscopy. Cancer 107(1):22–27 Cognet F, Ben Salem D, Dranssart M, Cercueil JP, Weiller M, Tatou E, Boyer L, Krause D (2002) Chronic mesenteric ischemia: imaging and percutaneous treatment. Radiographics 22(4):863–879. discussion 879-880 Cronin CG, Scott J, Kambadakone A, Catalano OA, Sahani D, Blake MA, McDermott S (2012) Utility of positron emission tomography/CT in the evaluation of small bowel pathology. Br J Radiol 85(1017):1211–1221 Dabaja BS, Suki D, Pro B, Bonnen M, Ajani J (2004) Adenocarcinoma of the small bowel: presentation, prognostic factors, and outcome of 217 patients. Cancer 101(3):518–526 Dave-Verma H, Moore S, Singh A, Martins N, Zawacki J (2008) Computed tomographic enterography and enteroclysis: pearls and pitfalls. Curr Probl Diagn Radiol 37(6):279–287 Dhatt HS, Behr SC, Miracle A, Wang ZJ, Yeh BM (2015) Radiological evaluation of bowel ischemia. Radiol Clin N Am 53(6):1241–1254 Di Nardo G, Di Lorenzo C, Lauro A, Stanghellini V, Thapar N, Karunaratne TB, Volta U, De Giorgio R (2017) Chronic intestinal pseudo-obstruction in children and adults: diagnosis and therapeutic options. Neurogastroenterol Motil 29(1) Dudiak KM, Johnson CD, Stephens DH (1989) Primary tumors of the small intestine: CT evaluation. AJR Am J Roentgenol 152(5):995–998 Fang SH, Dong DJ, Chen FH, Jin M, Zhong BS (2010) Small intestinal lipomas: diagnostic value of multi-­ slice CT enterography. World J Gastroenterol 16(21):2677–2681 Fernandes T, Oliveira MI, Castro R, Araujo B, Viamonte B, Cunha R (2014) Bowel wall thickening at CT: simplifying the diagnosis. Insights Imaging 5(2):195–208

Small Bowel MDCT Flamant M, Trang C, Bourreille A (2009) Wireless capsule in inflammatory bowel disease. Gastroenterol Clin Biol 33(Suppl 3):S183–S189 Fogel O, Richard-Miceli C, Tost J (2017) Epigenetic changes in chronic inflammatory diseases. Adv Protein Chem Struct Biol 106:139–189 Furukawa A, Yamasaki M, Furuichi K, Yokoyama K, Nagata T, Takahashi M, Murata K, Sakamoto T (2001) Helical CT in the diagnosis of small bowel obstruction. Radiographics 21(2):341–355 Furukawa A, Saotome T, Yamasaki M, Maeda K, Nitta N, Takahashi M, Tsujikawa T, Fujiyama Y, Murata K, Sakamoto T (2004) Cross-sectional imaging in Crohn disease. Radiographics 24(3):689–702 Furukawa A, Sakoda M, Yamasaki M, Kono N, Tanaka T, Nitta N, Kanasaki S, Imoto K, Takahashi M, Murata K, Sakamoto T, Tani T (2005) Gastrointestinal tract perforation: CT diagnosis of presence, site, and cause. Abdom Imaging 30(5):524–534 Gandhi NS, Baker ME, Goenka AH, Bullen JA, Obuchowski NA, Remer EM, Coppa CP, Einstein D, Feldman MK, Kanmaniraja D, Purysko AS, Vahdat N, Primak AN, Karim W, Herts BR (2016) Diagnostic accuracy of CT enterography for active inflammatory terminal ileal Crohn disease: comparison of full-dose and half-dose images reconstructed with FBP and half-dose images with SAFIRE. Radiology 280(2):436–445 Geffroy Y, Boulay-Coletta I, Julles MC, Nakache S, Taourel P, Zins M (2014) Increased unenhanced bowel-wall attenuation at multidetector CT is highly specific of ischemia complicating small-bowel obstruction. Radiology 270(1):159–167 Gore RM, Mehta UK, Berlin JW, Rao V, Newmark GM (2006) Diagnosis and staging of small bowel tumours. Cancer Imaging 6:209–212 Gourtsoyiannis N, Mako E (1997) Imaging of primary small intestinal tumours by enteroclysis and CT with pathological correlation. Eur Radiol 7(5):625–642 Guimaraes LS, Fletcher JG, Yu L, Huprich JE, Fidler JL, Manduca A, Ramirez-Giraldo JC, Holmes DR Jr, McCollough CH (2010) Feasibility of dose reduction using novel denoising techniques for low kV (80 kV) CT enterography: optimization and validation. Acad Radiol 17(10):1203–1210 Hainaux B, Agneessens E, Bertinotti R, De Maertelaer V, Rubesova E, Capelluto E, Moschopoulos C (2006) Accuracy of MDCT in predicting site of gastrointestinal tract perforation. AJR Am J Roentgenol 187(5):1179–1183 Haselkorn T, Whittemore AS, Lilienfeld DE (2005) Incidence of small bowel cancer in the United States and worldwide: geographic, temporal, and racial differences. Cancer Causes Control 16(7):781–787 Hill NS, DiSantis DJ (2015) The comb sign. Abdom Imaging 40(5):1010 Hong X, Choi H, Loyer EM, Benjamin RS, Trent JC, Charnsangavej C (2006) Gastrointestinal stromal tumor: role of CT in diagnosis and in response evaluation and surveillance after treatment with imatinib. Radiographics 26(2):481–495

637 Huprich JE, Fletcher JG, Alexander JA, Fidler JL, Burton SS, McCullough CH (2008) Obscure gastrointestinal bleeding: evaluation with 64-section multiphase CT enterography – initial experience. Radiology 246(2):562–571 Huprich JE, Barlow JM, Hansel SL, Alexander JA, Fidler JL (2013) Multiphase CT enterography evaluation of small-bowel vascular lesions. AJR Am J Roentgenol 201(1):65–72 Ihedioha U, Alani A, Modak P, Chong P, O'Dwyer PJ (2006) Hernias are the most common cause of strangulation in patients presenting with small bowel obstruction. Hernia 10(4):338–340 Ilangovan R, Burling D, George A, Gupta A, Marshall M, Taylor SA (2012) CT enterography: review of technique and practical tips. Br J Radiol 85(1015):876–886 Kalantari BN, Mortele KJ, Cantisani V, Ondategui S, Glickman JN, Gogate A, Ros PR, Silverman SG (2003) CT features with pathologic correlation of acute gastrointestinal graft-versus-host disease after bone marrow transplantation in adults. AJR Am J Roentgenol 181(6):1621–1625 Kambadakone AR, Chaudhary NA, Desai GS, Nguyen DD, Kulkarni NM, Sahani DV (2011) Low-dose MDCT and CT enterography of patients with Crohn disease: feasibility of adaptive statistical iterative reconstruction. AJR Am J Roentgenol 196(6):W743–W752 Kim SH, Shin SS, Jeong YY, Heo SH, Kim JW, Kang HK (2009) Gastrointestinal tract perforation: MDCT findings according to the perforation sites. Korean J Radiol 10(1):63–70 Kopacova M, Rejchrt S, Bures J, Tacheci I (2013) Small intestinal tumours. Gastroenterol Res Pract 2013:702536 Kuehle CA, Ajaj W, Ladd SC, Massing S, Barkhausen J, Lauenstein TC (2006) Hydro-MRI of the small bowel: effect of contrast volume, timing of contrast administration, and data acquisition on bowel distention. AJR Am J Roentgenol 187(4):W375–W385 Lalitha P, Reddy M, Reddy KJ, Kumari MV (2011) Computed tomography enteroclysis: a review. Jpn J Radiol 29(10):673–681 Lang G, Schmiegel W, Nicolas V, Brechmann T (2015) Impact of small bowel MRI in routine clinical practice on staging of Crohn’s disease. J Crohns Colitis 9(9):784–794 Lee SS, Kim AY, Yang SK, Chung JW, Kim SY, Park SH, Ha HK (2009) Crohn disease of the small bowel: comparison of CT enterography, MR enterography, and small-bowel follow-through as diagnostic techniques. Radiology 251(3):751–761 Liu PS, Platt JF (2014) CT angiography in the abdomen: a pictorial review and update. Abdom Imaging 39(1):196–214 Lu XH, Qin MW, Wen XH, Liu W, Shi JH, Qian JM, Li JN (2010) The diagnosis of Crohn’s disease of the small bowel: comparing CT enterography, capsule endoscopy, small-bowel follow through and ileocolonoscopy. Zhonghua Nei Ke Za Zhi 49(9):746–749 Macari M, Balthazar EJ (2001) CT of bowel wall thickening: significance and pitfalls of interpretation. AJR Am J Roentgenol 176(5):1105–1116

638 Maxwell EC, Dawany N, Baldassano RN, Mamula P, Mattei P, Albenberg L, Kelsen JR (2016) Diverting ileostomy for the treatment of severe, refractory, pediatric inflammatory Bowel disease. J Pediatr Gastroenterol Nutr doi: 10.1097/MPG.0000000000001498. [Epub ahead of print] Mazzeo S, Caramella D, Battolla L, Melai L, Masolino P, Bertoni M, Giusti P, Cappelli C, Bartolozzi C (2001) Crohn disease of the small bowel: spiral CT evaluation after oral hyperhydration with isotonic solution. J Comput Assist Tomogr 25(4):612–616 Miao F, Wang ML, Tang YH (2010) New progress in CT and MRI examination and diagnosis of small intestinal tumors. World J Gastrointest Oncol 2(5):222–228 Miller G, Boman J, Shrier I, Gordon PH (2000) Etiology of small bowel obstruction. Am J Surg 180(1):33–36 Minordi LM, Vecchioli A, Mirk P, Bonomo L (2011) CT enterography with polyethylene glycol solution vs CT enteroclysis in small bowel disease. Br J Radiol 84(998):112–119 Morgan PB, Kessel IL, Xiao SY, Colman M (2004) Uncommon presentations of Hodgkin’s disease. Case 1. Hodgkin’s disease of the jejunum. J Clin Oncol 22(1):193–195 Negaard A, Paulsen V, Sandvik L, Berstad AE, Borthne A, Try K, Lygren I, Storaas T, Klow NE (2007) A prospective randomized comparison between two MRI studies of the small bowel in Crohn’s disease, the oral contrast method and MR enteroclysis. Eur Radiol 17(9):2294–2301 Pan SY, Morrison H (2011) Epidemiology of cancer of the small intestine. World J Gastrointest Oncol 3(3):33–42 Panes J, Bouhnik Y, Reinisch W, Stoker J, Taylor SA, Baumgart DC, Danese S, Halligan S, Marincek B, Matos C, Peyrin-Biroulet L, Rimola J, Rogler G, van Assche G, Ardizzone S, Ba-Ssalamah A, Bali MA, Bellini D, Biancone L, Castiglione F, Ehehalt R, Grassi R, Kucharzik T, Maccioni F, Maconi G, Magro F, Martin-Comin J, Morana G, Pendse D, Sebastian S, Signore A, Tolan D, Tielbeek JA, Weishaupt D, Wiarda B, Laghi A (2013) Imaging techniques for assessment of inflammatory bowel disease: joint ECCO and ESGAR evidence-based consensus guidelines. J Crohns Colitis 7(7):556–585 Paquet N, Glickman JN, Erturk SM, Ros PR, Heverhagen JT, Patak MA (2016) Crohn’s disease activity: abdominal computed tomography histopathology correlation. Eur J Radiol Open 3:74–78 Park MJ, Lim JS (2013) Computed tomography enterography for evaluation of inflammatory bowel disease. Clin Endosc 46(4):327–366 Paulsen SR, Huprich JE, Fletcher JG, Booya F, Young BM, Fidler JL, Johnson CD, Barlow JM, Earnest F 4th (2006) CT enterography as a diagnostic tool in evaluating small bowel disorders: review of clinical experience with over 700 cases. Radiographics 26(3):641–657. discussion 657–662 Paulson EK, Thompson WM (2015) Review of small-­ bowel obstruction: the diagnosis and when to worry. Radiology 275(2):332–342

M. Rengo et al. Pelage JP, Soyer P, Boudiaf M, Brocheriou-Spelle I, Dufresne AC, Coumbaras J, Rymer R (1999) Carcinoid tumors of the abdomen: CT features. Abdom Imaging 24(3):240–245 Pourmand K, Itzkowitz SH (2016) Small bowel neoplasms and polyps. Curr Gastroenterol Rep 18(5):23 Quezada SM, Steinberger EK, Cross RK (2013) Association of age at diagnosis and Crohn’s disease phenotype. Age Ageing 42(1):102–106 Raptopoulos V, Schwartz RK, McNicholas MM, Movson J, Pearlman J, Joffe N (1997) Multiplanar helical CT enterography in patients with Crohn’s disease. AJR Am J Roentgenol 169(6):1545–1550 Reginelli A, Iacobellis F, Berritto D, Gagliardi G, Di Grezia G, Rossi M, Fonio P, Grassi R (2013) Mesenteric ischemia: the importance of differential diagnosis for the surgeon. BMC Surg 13(Suppl 2):S51 Rengo M, Bellini D, De Cecco CN, Osimani M, Vecchietti F, Caruso D, Maceroni MM, Lucchesi P, Iafrate F, Paolantonio P, Ferrari R, Laghi A (2011) The optimal contrast media policy in CT of the liver. Part I: technical notes. Acta Radiol 52(5):467–472 Rha SE, Ha HK, Lee SH, Kim JH, Kim JK, Kim JH, Kim PN, Lee MG, Auh YH (2000) CT and MR imaging findings of bowel ischemia from various primary causes. Radiographics 20(1):29–42 Rondonotti E, Koulaouzidis A, Yung DE, Reddy SN, Georgiou J, Pennazio M (2017) Neoplastic diseases of the small bowel. Gastrointest Endosc Clin N Am 27(1):93–112 Schindera ST, Nelson RC, DeLong DM, Jaffe TA, Merkle EM, Paulson EK, Thomas J (2007) Multi-detector row CT of the small bowel: peak enhancement temporal window – initial experience. Radiology 243(2):438–444 Scholz FJ, Afnan J, Behr SC (2011) CT findings in adult celiac disease. Radiographics 31(4):977–992 Shenoy S (2015) Small bowel sarcoma: tumor biology and advances in therapeutics. Surg Oncol 24(3):136–144 Shimoni A, Rimon U, Hertz M, Yerushalmi R, Amitai M, Portnoy O, Guranda L, Nagler A, Apter S (2012) CT in the clinical and prognostic evaluation of acute graft-­ vs-­host disease of the gastrointestinal tract. Br J Radiol 85(1016):e416–e423 Shinya T, Inai R, Tanaka T, Akagi N, Sato S, Yoshino T, Kanazawa S (2016) Small bowel neoplasms: enhancement patterns and differentiation using post-contrast multiphasic multidetector CT. Abdom Radiol (NY) [Epub ahead of print] Sokhandon F, Al-Katib S, Bahoura L, Copelan A, George D, Scola D (2016) Multidetector CT enterography of focal small bowel lesions: a radiological-pathological correlation. Abdom Radiol (NY) doi: 10.1007/s00261016-1015-1. [Epub ahead of print] Soyer P, Aout M, Hoeffel C, Vicaut E, Place V, Boudiaf M (2013a) Helical CT-enteroclysis in the detection of smallbowel tumours: a meta-analysis. Eur Radiol 23(2):388–399 Soyer P, Dohan A, Eveno C, Dray X, Hamzi L, Hoeffel C, Kaci R, Boudiaf M (2013b) Carcinoid tumors of the small-bowel: evaluation with 64-section CT-enteroclysis. Eur J Radiol 82(6):943–950

Small Bowel MDCT Spinelli A, Correale C, Szabo H, Montorsi M (2010) Intestinal fibrosis in Crohn’s disease: medical treatment or surgery? Curr Drug Targets 11(2):242–248 Taylor SA, Avni F, Cronin CG, Hoeffel C, Kim SH, Laghi A, Napolitano M, Petit P, Rimola J, Tolan DJ, Torkzad MR, Zappa M, Bhatnagar G, Puylaert CA and Stoker J (2016). The first joint ESGAR/ ESPR consensus statement on the technical performance of cross-­ sectional small bowel and colonic imaging. Eur Radiol [Epub ahead of print] Tendler DA (2003) Acute intestinal ischemia and infarction. Semin Gastrointest Dis 14(2):66–76 Thompson WM (2005) Imaging and findings of lipomas of the gastrointestinal tract. AJR Am J Roentgenol 184(4):1163–1171 Tilsed JV, Casamassima A, Kurihara H, Mariani D, Martinez I, Pereira J, Ponchietti L, Shamiyeh A, Al-Ayoubi F, Barco LA, Ceolin M, D’Almeida AJ, Hilario S, Olavarria AL, Ozmen MM, Pinheiro LF, Poeze M, Triantos G, Fuentes FT, Sierra SU, Soreide K, Yanar H (2016) ESTES guidelines: acute mesenteric ischaemia. Eur J Trauma Emerg Surg 42(2):253–270 van der Merwe BS, Ackermann C, Els H (2013) CT enteroclysis in the developing world: how we do it, and the pathology we see. Eur J Radiol 82(8):e317–e325

639 Van Weyenberg SJ, Meijerink MR, Jacobs MA, Van der Peet DL, Van Kuijk C, Mulder CJ, Van Waesberghe JH (2010) MR enteroclysis in the diagnosis of small-­ bowel neoplasms. Radiology 254(3):765–773 Ventham NT, Kennedy NA, Nimmo ER, Satsangi J (2013) Beyond gene discovery in inflammatory bowel disease: the emerging role of epigenetics. Gastroenterology 145(2):293–308 Wallihan DB, Podberesky DJ, Sullivan J, Denson LA, Zhang B, Salisbury SR, Towbin AJ (2015) Diagnostic performance and dose comparison of filtered back projection and adaptive iterative ­ dose reduction three-­ dimensional CT enterography in children and young adults. Radiology 276(1):233–242 Wiesner W, Khurana B, Ji H, Ros PR (2003) CT of acute bowel ischemia. Radiology 226(3):635–650 Wold PB, Fletcher JG, Johnson CD, Sandborn WJ (2003) Assessment of small bowel Crohn disease: noninvasive peroral CT enterography compared with other imaging methods and endoscopy – feasibility study. Radiology 229(1):275–281 Yeung KW, Chang MS, Hsiao CP, Huang JF (2004) CT evaluation of gastrointestinal tract perforation. Clin Imaging 28(5):329–333

Imaging of Large Bowel with Multidetector Row CT Jay D. Patel, Heather I. Gale, and Kevin J. Chang

Contents 1    Colorectal Cancer 1.1  Colorectal Cancer Pathophysiology 1.2  Colorectal Staging 1.3  Colorectal Screening 1.4  CT Colonography 1.5  CTC Technique 1.6  CTC: Polyps and Cancer 1.7  CTC Reporting 1.8  CTC Screening 1.9  CT Colonography Indications/ Contraindications

 642  642  642  643  646  646  648  649  649

2    Colonic Lymphoma

 650

3    Colitis 3.1  Inflammatory Bowel Disease: Ulcerative Colitis and Crohn’s Disease 3.2  Infectious Colitis 3.3  Pseudomembranous Colitis

 651

 650

 651  653  653

J.D. Patel, MD (*) Department of Diagnostic Imaging, The Warren Alpert Medical School of Brown University/Rhode Island Hospital, 593 Eddy St., Providence, RI 02903, USA e-mail: [email protected] H.I. Gale, MD Assistant Professor, Department of Diagnostic Imaging, The Warren Alpert Medical School of Brown University/Rhode Island Hospital, 593 Eddy St., Providence, RI 02903, USA e-mail: [email protected] K.J. Chang, MD, FSAR Division of Body Imaging, Department of Diagnostic Imaging, The Warren Alpert Medical School of Brown University/Rhode Island Hospital, 593 Eddy St., Providence, RI 02903, USA e-mail: [email protected]

3.4  I schemic Colitis 3.5  T  yphlitis 3.6  S  tercoral Colitis

 653  654  654

4    Acute Diverticulitis

 654

5    Appendix 5.1  Appendicitis 5.2  Primary Neoplasms of the Appendix

 655  656  656

6    Epiploic Appendagitis

 658

7    Colonic Volvulus 7.1  Cecal Volvulus 7.2  Sigmoid Volvulus

 659  659  661

8    Lower Gastrointestinal Bleeding: Role of CTA

 662

Conclusion

 663

References

 663

Abstract

The use of high-resolution multidetector CT (MDCT) has revolutionized evaluation of the large bowel in both the acute emergency room setting and in chronic conditions. The physical exam is often limited and CT can help differentiate between conditions that may mimic each other clinically. Patients often present with vague abdominal symptoms, and CT can help elucidate the etiology and help guide management and treatment. The pathology is vast, and some of the more common acute conditions include appendicitis, diverticulitis, inflammatory bowel disease, and bowel obstruction. More recently, CT has also come

Med Radiol Diagn Imaging (2017) DOI 10.1007/174_2017_7, © Springer International Publishing AG Published Online: 15 February 2017

641

J.D. Patel et al.

642

to play a significant role in the evaluation of acute lower gastrointestinal bleeding. Primary evaluation with CTA has become accepted as an alternative initial screening exam and has been incorporated into the algorithm and work-up of lower gastrointestinal bleeding in many large medical centers. CTA allows for a quick and efficient survey of the abdomen and can triage patients appropriately, ensuring accurate, timely, and safe management. Considerable improvements have also been made in colorectal cancer screening with CT colonography (CTC, also known as virtual colonoscopy). The American Cancer Society (ACS) and US Preventive Services Task Force (USPSTF) now recognize CTC as an acceptable primary screening option for colorectal cancer, which should pave the way for more widespread usage.

1

Colorectal Cancer

Multidetector CT plays an essential role in the diagnosis, staging, and follow-up treatment of colon cancer. Colorectal cancer is the third most commonly diagnosed cancer and second leading cause of cancer death in the USA with an estimated 4.5% lifetime risk of developing the disease. For the year 2016, the American Cancer Society (ACS) estimates there will be 134,490 new cases of colorectal cancer resulting in approximately 49,190 deaths in the USA. This accounts for 8.0% of all new cancer cases and 8.3% of all cancer-related deaths (SEER Cancer Statistics Review. Available from: https://seer.cancer.gov/data/).

1.1

Colorectal Cancer Pathophysiology

Virtually all colon cancers arise from polyps. Even though there are individuals who are prone to developing polyps such as individuals with a personal or family history of colorectal cancer, those with a history of inflammatory bowel disease or hereditary forms of colorectal cancer, 75–95% of all colon cancers develop in individuals with little or no genetic predisposition for malignancy.

There are two key models or pathways proposed for colorectal cancer development. The vast majority arise from mucosal epithelial cells which undergo a series of mutations according to a well-established adenoma-carcinoma sequence. In this pathway, colorectal cancers arise from precursor lesions known as adenomatous polyps, which undergo a series of stepwise mutational activation of oncogenes and inactivation of tumor suppressor genes ultimately leading to abnormal cell proliferation, apoptosis, and subsequently carcinoma (Bond 2000). More recently, a serrated neoplastic pathway for colorectal carcinogenesis has also been identified accounting for up to one third of all colorectal cancers (Rex et al. 2012). Serrated lesions are a group of polyps that can be classified pathologically according to the World Health Organization as hyperplastic polyps, sessile serrated adenoma/ polyps, or traditional serrated adenomas. While most hyperplastic polyps are typically benign, small subsets, particularly large hyperplastic polyps in the right colon, have been shown to be precursors to sessile serrated adenomas that can ultimately progress to cancer themselves.

1.2

Colorectal Staging

The advent of multidetector CT has played a crucial role in the diagnosis, staging, and treatment of colon cancer. Colon cancer spreads through a variety of patterns including direct infiltration and extension through the serosa, lymphatic drainage to regional lymph nodes, hematogenous spread through the portal venous system to the liver, as well as intraperitoneal seeding. CT has become routine for preoperative staging and surgical planning (Nerad et al. 2016). Currently, the TNM staging system established by the American Joint Committee on Cancer (AJCC) is the most widely used staging system for colorectal cancer (Fig. 1). This system essentially evaluates three key components in determining staging of the cancer. T – indicates how invasive the primary tumor is and degree of extension into the wall of the intestine and surrounding structures. N – indicates the extent of spread to regional lymph nodes.

Imaging of Large Bowel with Multidetector Row CT

643

M – indicates whether the cancer has metastasized to other organ systems. TNM Colon and Rectum Cancer Staging: Seventh Edition (AJCC) Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor Tis Carcinoma in situ: intraepithelial or invasion of lamina propria T1 Tumor invades submucosa T2 Tumor invades muscularis propria T3 Tumor invades through the muscularis propria into pericolorectal tissues T4a Tumor penetrates to the surface of the visceral peritoneum T4b Tumor directly invades or is adherent to other organs or structures Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis in 1–3 regional lymph nodes N1a Metastasis in one regional lymph node N1b Metastasis in 2–3 regional lymph nodes N1c Tumor deposit(s) in the subserosa, mesentery, or nonperitonealized pericolic or perirectal tissues without regional nodal metastasis

N2 Metastasis in 4 or more regional lymph nodes N2a Metastasis in 4–6 regional lymph nodes N2b Metastasis in 7 or more regional lymph nodes Distant Metastasis (M) M0 No distant metastasis M1 Distant metastasis M1a Metastasis confined to one organ or site (e.g., liver, lung, ovary, nonregional node) M1b Metastases in more than one organ/site or the peritoneum The staging of cancer at presentation greatly impacts treatment and survival. Based on the National Cancer Institute’s SEER database from 2004 to 2010, the 5-year relative survival rate for individuals with stage I colon cancer was about 92%, 87% for stage IIA, 63% for IIB, 89% for stage IIIA, 69% for IIIB, 53% for stage IIIC, and a dismal 11% for those with stage IV distant metastatic disease.

1.3

Colorectal Screening

The earlier colorectal cancer diagnosis can be made, the better the prognosis. The ACS c­ urrently recommends screening for colon cancer ­beginning at the age of 50 years in asymptomatic men and

a T1

T2

T3

T4a

T4b

Mucosa

Submucosa Muscularis externa

Serosa (visceral peritoneum)

Fig. 1  TNM colon-rectum cancer staging. (a) T-staging. (b) N-staging. (c) M-staging

644

b

Fig. 1 (continued)

J.D. Patel et al.

Imaging of Large Bowel with Multidetector Row CT

645

c

Fig. 1 (continued)

women at average risk. High-risk patients, such as those with either a personal or family history of prior colonic adenomatous polyps, prior colon cancer, Peutz-Jeghers syndrome, hereditary nonpolyposis colorectal cancer (HNPCC), familial adenomatous polyposis (FAP), juvenile polyposis syndrome (JPS), or chronic inflammatory bowel disease, should obtain screening at an earlier stage. There are a number of colorectal screening tests available and the American Cancer Society endorses a variety of screening regimens based on the examination used. Tests can be divided into cancer prevention and cancer detection. Cancer prevention tests have the potential to image both cancer and polyps, whereas cancer detection tests have lower sensitivity for polyps and typically lower sensitivity for cancer ­detection (Levin et al. 2008.

Tests that detect polyps and cancer include: • Colonoscopy – recommended every 10 years • CT colonography (virtual colonoscopy) – recommended every 5 years • Flexible sigmoidoscopy – recommended every 5 years • Double-contrast barium enema – ­recommended every 5 years Tests that detect cancer include: • Guaiac-based fecal occult blood test (gFOBT) – recommended every year • Fecal immunochemical test (FIT) – ­recommended every year • Stool DNA test (sDNA) – recommended every 3 years

J.D. Patel et al.

646

1.4

CT Colonography

CT colonography (CTC), also known as virtual colonoscopy, was first described and proposed as an imaging modality for the evaluation of colonic mucosa and colon cancer detection by Vining and Gelfand in 1994. Since then, with the advent of thin-section MDCT, automated insufflation, and oral tagging, CTC screening protocols and technique have been significantly refined. While optical colonoscopy has traditionally been used as the gold standard for colorectal cancer screening as polypectomy can be performed concurrently, it has several disadvantages. It is invasive and resource intensive and often involves the use of sedation. It is also potentially inconvenient to both the patient and his or her driver requiring significant time spent away from the daily routine. In addition, although small, there is a risk for perforation and bleeding, with an overall complication rate of approximately 0.4% (Nelson et al. 2002). Some feel the complication rate may be significantly underreported with hospital visitation rates as high as 2% within the first week after colonoscopy (Ranasinghe et al. 2016). As a result, CT colonography has emerged as a safe, effective, and efficient alternative means for screening asymptomatic adults.

1.5

CTC Technique

There are four essential components to performing CT colonography. CT colonography routinely consists of (1) patient preparation, (2) colonic distension, (3) multidetector CT scanning, and (4) interpretation using dedicated CTC 3D rendering software. The procedure begins with careful bowel preparation. Dietary restrictions generally include maintenance of a clear liquid diet 1–2 days prior to examination. While there are several variations and protocols devised for patient preparation and examination performance, the most crucial aspect of performing high-quality CT colonography involves a thorough colonic cathartic preparation (bowel prep) for at least 1 day. The bowel prep utilized may differ from that

used in optical colonoscopy. Instead of a high-­ volume “wet” prep involving agents such as polyethylene glycol, a “drier” prep can be used which leaves less residual fluid in the colon allowing for better visualization of the colonic wall air-mucosal interface. In general, patients are better able to tolerate these “dry” lower-­ volume bowel preps than high-volume iso-­ osmolar preps (typically 2–4 L of fluid). In addition to cleansing with laxatives, fecal and fluid tagging is also important in patient preparation. Many centers use iodinated water-soluble contrast medium for fluid tagging with or without dilute barium (2%) for fecal tagging. This improves polyp detection by raising the inherent CT densities of residual fluid and stool, helping to discriminate these residua from the underlying soft-tissue density of submerged polyps and cancers (Pickhardt and Choi 2003). Although a fully cleansed colon is recommended for patients who can tolerate the preparation, noncathartic or reduced-cathartic CTC preparations are also available, which may reduce patient discomfort at the cost of study sensitivity. The most crucial technical component of the study lies in adequate gaseous distension of the entire colon. This is usually performed via insertion of a small flexible balloon-tipped rectal catheter through which CO2 is insufflated (Shinners et al. 2006). Less preferred, although acceptable, is manual insufflation with room air. In general, a thin-collimation low-radiation dose technique is then employed on a multidetector CT scanner (≥16 slice) in both supine and prone or lateral decubitus positions following scout topogram confirmation of adequate colonic insufflation. A section thickness of 1–1.25 mm with a reconstruction interval of ≤1 mm is optimal. Image acquisition is obtained during a single breath hold in end-expiration to limit pressure-related effects of inflated lungs on the transverse colon. The scan is then repeated with the patient in the prone or decubitus position. The data is reformatted into two-dimensional images in axial and other planes as well as reconstructed into three-dimensional endoluminal images that can simulate the view obtained during conventional colonoscopy via commercially available

Imaging of Large Bowel with Multidetector Row CT

a

647

b

c d

Fig. 2  A 78-year-old man for colorectal screening. (a) Axial supine CT shows a lobular pedunculated polyp in the ascending colon (arrow). (b) 3D endoluminal view confirms the polyp morphology and its location on a haustral fold. (c) 3D surface-rendered “colon map” simulates the look of a double-contrast barium enema and guides

the colonoscopist or surgeon to the polyp’s exact location. Also note the malignant stricture in the sigmoid colon at the bottom of the image. (d) Photograph of a portion of the colectomy specimen confirming the lobular appearance of this villous polyp with high-grade dysplasia

software programs (Fig. 2). Many alternative 3D rendering techniques have also been developed including virtual filet, unfolded cubes, and panoramic views (Chang and Soto 2010). Image interpretation and evaluation of colonic polyps is performed with either a primary 2D or primary 3D approach. The ACRIN trial by Johnson et al. demonstrated no statistical difference in sensitivity between a primary 2D or 3D interpretation (Johnson et al. 2008). However,

others such as Pickhardt et al. have suggested there are multiple limitations to a 2D-only approach and advocate the use of a 3D approach for a primary survey of the colon as it may increase sensitivity for polyp detection, followed by a 2D evaluation for confirmation of suspected lesions as it is more specific (Pickhardt 2007). Regardless, both 2D and 3D evaluation should be utilized together for polyp detection. In addition, software for automatic polyp detection

J.D. Patel et al.

648

a

b

c

d

Fig. 3  A 42-year-old male presented for CT colonography due to incomplete colonoscopy to the level of an obstructing sigmoid mass. (a) Supine axial CT shows circumferential bowel wall thickening in the sigmoid colon (arrows). (b) Coronal CT demonstrates a large mass/ necrotic lymphadenopathy adjacent to the colon (arrow).

(c) 3D surface-rendered “colon map” shows a typical “apple core” appearance of an annular constricting mass (arrows). (d) 3D endoluminal view confirms an annular constricting mass in the sigmoid colon highly suspicious for malignancy

(computer-­aided detection [CAD]) is also available and helps to reduce interobserver variability and perception errors, especially for readers with limited CTC experience.

on the other hand appear as larger intraluminal masses with an irregular and/or nodular contour. They may also appear as “apple core” or “saddle” lesions which are annular or s­emiannular constricting masses with irregular wall thickening and luminal narrowing (Fig. 3). Calcifications within the tumor or metastasis is a finding that can be associated with a mucinous histologic subtype. Inflammation and stranding of the adjacent fat planes often is a sign of tumor extension through the bowel. Regional lymph nodes >1 cm in size are suspicious for metastatic disease; however,

1.6

CTC: Polyps and Cancer

Most colon cancers are thought to develop from adenomatous polyps and on CT are generally seen as well-defined oval or round soft-tissue masses which project into the lumen. Frank carcinomas

Imaging of Large Bowel with Multidetector Row CT

even small nodes may harbor disease. Distant metastasis is most commonly seen in the liver as the colon has a vast portal venous drainage system. Additional sites of metastasis include the lung, adrenal gland, and peritoneum. CTC has repeatedly demonstrated sensitivities equivalent to that of optical colonoscopy in the detection of clinically relevant polyps. Polyp detection exceeds 90% for clinically relevant lesions 6 mm or larger, while polyps 1 cm Abiding by this reporting system prevents confusion among reports and provides standardized guidelines for the management of various imaging findings.

In regard to reporting of findings, the CT Colonography Reporting and Data System (C-RADS) has been developed to ensure consistency and clear communication of results between readers of CT colonography. It is a well-­ established standard for reporting CTC findings and divides findings into colonic and extraco- 1.8 CTC Screening lonic categories as described below (Zalis et al. 2005; Yee et al. 2016): In 2008, the American Cancer Society guideline for colorectal cancer screening was revised Colonic Findings jointly with the US Multi-Society Task Force C0: Inadequate study/awaiting prior compari- on Colorectal Cancer and the American College sons. Inadequate colonic preparation or of Radiology (ACR) to include CTC every insufflation. 5 years as an option for screening average-risk C1: Normal colon or benign lesion. No polyp individuals. greater than 6 mm. Continue routine screening There are several potential advantages (every 5 years per American Cancer Society and benefits of CT colonography over optical screening guidelines). colonoscopy. Compared to traditional optical

J.D. Patel et al.

650

c­ olonoscopy, there is no need for sedation, and patients are able to avoid the cardiopulmonary risks associated with anesthesia. In addition, virtual colonoscopy requires less technical staff as it can be performed without the presence of anesthetists and nurses. It is a quick examination, requiring approximately 10–15 min table time. Most patients tolerate the entire examination without the need for sedation and can thus return to work immediately without the need of a ­separate driver. CT colonography is also extremely safe with a reported overall perforation rate of 0.009%, significantly lower than that of optical colonoscopy (Pickhardt 2006). In addition to colonic neoplasia, CTC also allows for the detection of potentially life-­ threatening extracolonic findings. While not all extracolonic findings are clinically significant, they are important in approximately 10% of patients who require further follow-up (Pickhardt et al. 2008). This includes incidental findings such as extracolonic cancers (most commonly renal, lung, and lymphoma) as well as abdominal aortic aneurysms and adrenal lesions.

the same day as the colonoscopy, unless the reason for failure is inadequate bowel preparation. CT colonography may also be particularly useful in patients who are at increased risk for complications during optical colonoscopy such as patients of advanced age, on anticoagulant therapy, or who have a high sedation risk. CTC is generally contraindicated in acute abdominal conditions such as acute diverticulitis and acute inflammatory bowel disease due to an increased risk of perforation (Bellini et al. 2014). In addition, CTC is not recommended for routine surveillance imaging of inflammatory bowel disease, evaluation of anal canal disease, or in the pregnant or potentially pregnant patient. The relative contraindications include symptomatic acute colitis, acute diarrhea, recent diverticulitis, recent colorectal surgery, symptomatic colon-containing abdominal wall hernia, recent deep endoscopic biopsy or polypectomy, colonic perforation, and high-grade small-bowel obstructions.

2 1.9

 T Colonography Indications/ C Contraindications

According to the ACR Practice Parameters and Technical Standards (the American College of Radiology 2014), the indications for CTC include, but are not limited to, the following: screening individuals who are at average or moderate risk for developing colorectal carcinoma, surveillance examination in patients with prior history of colonic neoplasm, or diagnostic examination in symptomatic patients (Kim et al. 2010; Yee et al. 2010). CT colonography is also indicated in patients following incomplete screening, surveillance, or diagnostic colonoscopy and for characterization of colorectal lesions indeterminate on optical colonoscopy. Incomplete or failed colonoscopy may be secondary to a variety of factors including colonic tortuosity, nonvisualization of the colon proximal to an obstructive lesion, or colonic spasm. In general, these can be performed

Colonic Lymphoma

Colonic lymphoma is rarer than gastrointestinal lymphoma and much more rare than colonic adenocarcinoma. The most common subtype of colonic lymphoma is diffuse large B-cell lymphoma. The incidence of disease is much more common in patients with acquired immunodeficiency syndrome and inflammatory bowel disease and those who are immunocompromised such as individual posttransplantation. Compared to colon adenocarcinoma, colonic lymphoma presents as marked bowel wall thickening with aneurysmal luminal dilatation rather than stenosis. As a result, bowel obstruction is exceedingly rare. In addition, it tends to affect longer colonic segments than adenocarcinoma. Unlike other tumors, necrosis is also uncommon. The vast majority of large bowel lymphoma occurs in the right colon, whereas colonic adenocarcinoma is most common in the rectosigmoid colon (Quayle and Lowney 2006). Regional and diffuse lymphadenopathy is often an accompanied finding.

Imaging of Large Bowel with Multidetector Row CT

3

Colitis

Patients with colitis present with nonspecific abdominal pain, and CT is almost universally the initial study of choice performed for suspected colonic disease, especially in the emergency room setting. Its widespread availability and ease of performance make it an excellent modality for screening patients with nonspecific symptoms. While the subtype of colitis is based on the combination of clinical, laboratory, and pathologic data, CT can help narrow the differential diagnosis by evaluating the extent and distribution of inflammation. The primary hallmark of colitis on CT is bowel wall thickening, mural edema, pericolonic inflammatory stranding, and mucosal enhancement. The evaluation of the colonic wall can be challenging when the colon is not appropriately distended, as a decompressed colon may mimic wall thickening. Notably, the added benefit of performing CT is its ability to accurately evaluate for complications of colitis such as abscess formation and perforation.

3.1

651

a

b

I nflammatory Bowel Disease: Ulcerative Colitis and Crohn’s Disease

Ulcerative colitis (UC) is typically a disease of young adults aged 15–40 years; however, a second peak between 50 and 60 years old is also common with slight male predominance (Ekbom et al. 1991; Loftus 2004). It is a disease of colonic inflammation and mucosal ulceration that typically afflicts the rectum and progresses in a retrograde continuous fashion without “skip” lesions (Fig. 4). The hallmark of this condition is colonic wall thickening and luminal narrowing. With the administration of IV contrast, the “halo or target sign” can often be observed in which there is a stratification of the layers of the wall. In acute disease, although nonspecific, there is hyperattenuation and hyperenhancement of the inner mucosa, low water attenuation of submucosal edema, and outer hyperattenuation of the muscularis propria. In chronic disease, there can be a similar halo or target appearance of the wall;

Fig. 4  A 23-year-old with history of ulcerative colitis presents with worsening abdominal pain and diarrhea. (a) Axial CT following intravenous and oral contrast administration demonstrates marked contiguous rectosigmoid wall thickening, and (b) coronal reformatted images demonstrate continuity to the level of the descending colon with engorgement of mesenteric vessels (arrows)

however, instead of submucosal water attenuation, there is fatty infiltration (Jones et al. 1986). In addition, with increasing chronicity there is increased perirectal fat and widening of the presacral space. While there is primary involvement of the left colon, the terminal ileum can also be involved via backwash ileitis resulting in a patulous, dilated ileocecal valve. Similar to ulcerative colitis, Crohn’s disease typically affects young adults in their twenties;

J.D. Patel et al.

652

however, late onset in adulthood has also been observed. While there is considerable overlap in the CT imaging features of Crohn’s disease and ulcerative colitis such as mural stratification, mural enhancement, and bowel wall thickening, there are important differences. Crohn’s colitis is notorious for transmural inflammation and most commonly affects the terminal ileum and ­proximal colon. In such cases, there is thickwalled small bowel and proximal colon with narrowed lumen resulting in a so-called string sign (Fig. 5). In addition, in contrast to UC, inflammation often leads to stenosis of the ileocecal valve and proximal dilatation of the terminal ileum. Crohn’s disease can also involve “skip lesions” and affect any region in the alimentary tract. CT enterography (CTE) may be particularly helpful in the evaluation of small-bowel pathology in individuals suspected of having Crohn’s disease. CT enterography involves the use of intravenous contrast in combination with large-­ volume neutral oral contrast agent for luminal distension. Adequate small-bowel distension is necessary as collapsed bowel can pose diagnostic challenges and even mimic small-bowel pathology. Using neutral oral contrast agents such as VoLumen (Bracco Diagnostics, Inc., Monroe Twp, NJ) permits optimal small-bowel distension and improves assessment of mucosal enhancement, mural thickening, and evaluation of strictures. In addition, CT enterography is also useful for evaluating disease activity, and detecting active inflammation as measuring small-bowel mural attenuation has been correlated with disease activity (Bodily et al. 2006). Although nonspecific, the presence of engorged mesenteric vessels also suggests active disease and results in a so-called comb sign from hyperemic, congested vessels that are widely spaced (Lee et al. 2002). While CTE is useful in the initial presentation or diagnosis of inflammatory bowel disease, MR enterography (MRE) may be preferred for follow-­up surveillance of disease activity, particularly in younger individuals where cumulative lifetime radiation dose from multiple CTs should be minimized. In the acute setting, CT is helpful not only in diagnosis but also in the assessment of disease

a

b

Fig. 5 (a) A 45-year-old female with history of Crohn’s disease presents with worsening right lower quadrant abdominal pain. Axial CT demonstrates marked mural wall thickening of the terminal ileum and luminal narrowing consistent with terminal ileitis (arrows). (b) Different patient, 38-year-old with history of Crohn’s disease, underwent fluoroscopic small-bowel follow-through series demonstrating marked luminal narrowing and fibrosis of an 8–9 cm segment of terminal ileum consistent with a “string sign”

complication. Although rare, one of the most feared complications of both types of inflammatory bowel disease is toxic megacolon. CT demonstrates thin-walled, marked colonic dilatation with loss of normal haustral folds and irregular shaggy mucosa. This can ultimately lead to bowel ischemia and perforation. Other complications include phlegmon, which is an ill-defined inflammatory mass without discrete walled-off collection or abscess in which there is a well-defined

Imaging of Large Bowel with Multidetector Row CT

rim-enhancing wall with central low attenuation. These collections may also contain air-fluid levels, scattered foci of gas, or internal septations. Depending on the size and location of these collections in conjunction with patient’s clinical history, findings on CT can help guide treatment whether it is conservative management with antibiotics or, if large and accessible, percutaneous image-guided drainage. Other complications of Crohn’s disease include sinus tracts and fistulas, which can be readily visualized on multidetector CT by outlining communicating fluid tracts. Finally, there is also increased risk of colon cancer particularly with long-standing ulcerative colitis, but also Crohn’s disease.

3.2

Infectious Colitis

Infectious colitis may be caused by a host of different bacteria (Shigella, Salmonella, Campylobacter, Yersinia, tuberculosis), fungi, viruses (herpes, Cytomegalovirus), and parasites. The imaging findings are generally nonspecific with considerable overlap in the CT findings of different infectious agents. Imaging findings include bowel wall thickening, pericolonic inflammatory stranding, and ascites. While many infectious agents produce diffuse pancolitis such as CMV and E. coli, others have a predilection for the right colon including or excluding the ileum such as Salmonella, Yersinia, tuberculosis, and amebiasis. Others may have predominately left colonic involvement including schistosomiasis, shigellosis, and herpes (Thoeni and Cello 2006).

3.3

Pseudomembranous Colitis

Pseudomembranous colitis is a type of infectious colitis that results from bacterial overgrowth of Clostridium difficile within the colon. The bacteria release cytotoxic enterotoxins that produce an exudative inflammatory process within the colonic mucosa. It usually results as a complication of antibiotic use, which disrupts normal gut flora and allows C. difficile to colonize the colon. CT findings include bowel wall thickening, a

653

Fig. 6  A 22-year-old male with recent antibiotic use presents with abdominal pain and watery diarrhea and was found to have C. difficile colitis. Axial CT image following intravenous contrast administration demonstrates marked intramural edema and transverse colonic wall thickening producing an “accordion”-like appearance (arrows)

shaggy mucosal outline due to sloughed mucosal cells, and marked submucosal edema resulting in characteristic “accordion sign” (Fig. 6) (Macari et al. 1999). Typically, there is pancolonic involvement; however, it may also manifest as isolated segmental disease. Untreated, C. difficile colitis may progress to toxic megacolon and ultimately bowel perforation. Treatment is supportive therapy and antibiotics consisting of metronidazole and oral vancomycin.

3.4

Ischemic Colitis

Ischemic injury of the colon most commonly occurs in the elderly population older than 70 years old. In this age group, ischemic colitis most commonly occurs in the setting of low flow states and decreased cardiac output on a background of extensive atherosclerotic disease. In the younger population, the disease may occur secondary to a vasculitis or hypercoagulable state. Regardless, diminished blood flow leads to colonic ischemia and secondary inflammation. The CT appearance of ischemic colitis will vary depending on the severity; however, the distribution of colonic involvement is crucial in making the diagnosis. Afflicted segments typically follow a vascular distribution and most commonly affect watershed areas including the splenic flexure and rectosigmoid colon. CT will demonstrate wall

J.D. Patel et al.

654

thickening and submucosal edema producing a “target” or “halo” sign. In addition, pericolonic inflammation may be present. If necrosis and ulceration develops, this can lead to perforation.

3.5

Typhlitis

Typhlitis is also known as neutropenic enterocolitis and is an infectious colitis usually confined to the cecum and ascending colon in patients that are neutropenic and severely immunocompromised. This includes patients with acquired immunodeficiency syndrome and posttransplantation or on chemotherapy for malignancy. Classically, the disease is associated with patients with leukemia undergoing chemotherapy treatment. CT is the modality of choice for diagnosis, and imaging demonstrates marked circumferential thickening of the cecum and ascending colon with pericolonic inflammatory stranding and edema (Fig. 7). Intramural areas of low attenuation may represent edema or hemorrhage. The disease may also occasionally extend into the terminal ileum. It is important to make the diagnosis in a timely fashion as it can quickly progress to ischemia and necrosis with pneumatosis intestinalis and ultimately bowel perforation.

3.6

a

b

Stercoral Colitis

Stercoral colitis is a rare cause of inflammatory colitis that results from ischemic pressure necrosis secondary to chronic fecal impaction. Long-­ standing fecal impaction results in increased intraluminal pressure, ulceration of the mucosa, and ultimately bowel perforation if not managed appropriately. CT findings of chronic fecal impaction include a distended colon with wall thinning; however, in cases of stercoral ulceration, there may be focal thickening of the colonic wall. In addition, there is pericolonic inflammatory stranding and edema adjacent to the site of fecal impaction, and extraluminal gas may be present suggesting microperforation (Fig. 8) (Heffernan et al. 2005).

Fig. 7  A 70-year-old female with CLL status post bone marrow transplantation and neutropenia presents with right upper and lower quadrant abdominal pain. (a) Axial CT shows marked thickening of the cecal pole with circumferential pneumatosis intestinalis (arrow). (b) Coronal CT demonstrates cecal and ascending colonic wall thickening and circumferential pneumatosis intestinalis (arrow) consistent with typhlitis in this neutropenic patient

4

Acute Diverticulitis

Diverticulitis is a recognized complication of diverticulosis. Diverticula are small sac-like outpouchings of the mucosa and submucosa through areas of weakness and defect in the muscularis,

Imaging of Large Bowel with Multidetector Row CT

a

Fig. 8  A 78-year-old female with chronic constipation presents with abdominal pain and was found to have stercoral colitis. (a) Axial CT following intravenous contrast

which are still covered by serosa. Since they do not contain all layers of the colonic wall, they are known as “false” or “pulsion diverticula.” Most diverticula occur along the mesenteric surface of the colon, typically where the vasa recta penetrate the muscular layer (Meyers et al. 1973). Due to their close proximity to vessels, this also explains their propensity for colonic bleeding. Although the exact prevalence is difficult to determine as many individuals are asymptomatic, there is increased predilection with advanced age, particularly after the age of 50. Colonic diverticula result from increased intraluminal pressure and shortening and thickening of the colon known as myochosis coli. The most common location for diverticula to form is in the sigmoid colon. Diverticulitis develops when there is inflammation within a diverticulum, usually caused by obstruction of the neck by stool or food particles and is followed by subsequent microperforation. On CT, diverticulitis appears as bowel wall thickening and pericolonic inflammatory stranding centered around a diverticulum (Fig. 9). Complications include diverticular perforation; abscess formation; fistulization to nearby structures including the bladder, bowel, vagina, and skin; as well as bowel obstruction from adhesions. Treatment for mild acute uncomplicated diverticulitis is conservative management with antibiotics and supportive care; however, in the case of complicated or repeated bouts of diverticulitis, surgery may be required.

655

b

administration demonstrates a large stool ball within the rectum with marked rectal distension and wall thickening. (b) In addition, there is perirectal edema (arrow)

Fig. 9  A 47-year-old male with leukocytosis and acute-­ onset left lower quadrant abdominal pain. Axial CT following intravenous contrast administration shows sigmoid diverticulosis with wall thickening and adjacent mesenteric inflammatory stranding (arrow) consistent with acute uncomplicated diverticulitis

5

Appendix

The normal appendix is a thin-walled blind-­ ending tubular structure consistently arising between the ileocecal valve and apex of the cecal pole. The length of the appendix and location of the tip are much more variable with roughly one third of cases coursing inferomedial to the cecum and two thirds coursing retrocecal. The normal appendix typically measures 6 mm or less, is surrounded by homogenous mesenteric fat, and maintains a well-defined outer contour.

J.D. Patel et al.

656

5.1

Appendicitis

phlegmon or abscess. Appendicitis can also be confined to the distal tip with wall thickening Acute appendicitis is one of the most common and peri-inflammation isolated to the tip, while causes of acute abdominal pain in children and the proximal portion appears collapsed or normal young adults. The classic presentation consists of in caliber. Complications of perforation include periumbilical pain followed by focal tenderness phlegmon, which appears as a soft-­tissue inflamat McBurney’s point with associated fever, nau- matory mass without walled-off collection or sea/vomiting, and leukocytosis. abscess, which appears as a discrete peripheral Appendicitis typically results from obstruc- rim-enhancing collection with central low attenution of the tip of the appendix followed by fluid ation. While appendectomy has traditionally been distension, venous engorgement, and ultimately the definitive curative treatment for appendicitis, ischemia and perforation. Obstruction is most there has been an increasing use of a trial of anticommonly caused by lymphoid hyperplasia or biotics without surgery for the management of an appendicolith. CT findings include a fluid-­ less severe cases. In general, phlegmon or small distended appendix >6 mm in diameter, thickened abscess may be treated with antibiotics and interand enhancing wall, and periappendiceal inflam- val appendectomy, while larger abscesses may matory fat stranding (Fig. 10). Appendicoliths require percutaneous or surgical drainage prior to on CT appear as calcification within the lumen appendectomy to control the spread of infection. of the appendix; however, if the appendix is ruptured, it may also present adjacent to or within

5.2

Primary Neoplasms of the Appendix

Primary neoplasms of the appendix are uncommon, found in 0.5–1.0% of all appendectomy specimens (Deans and Spence 1995; Connor et al. 1998; Hananel et al. 1998). Approximately 30–50% of all appendiceal neoplasms manifest clinically with signs and symptoms of acute appendicitis; however, it is important to accurately differentiate the two entities as the surgical approach, and management is quite different, often involving hemicolectomy in the case of appendiceal neoplasm (Carr et al. 1995; Connor et al. 1998; Pickhardt et al. 2002).

Fig. 10  A 19-year-old febrile with acute right lower quadrant abdominal pain. Coronal contrast-enhanced CT demonstrates a blind-ending tubular structure in the right lower quadrant with fluid distension, mucosal hyperemia, periappendiceal inflammatory stranding (arrow), and calcified appendicolith (arrow) consistent with acute, uncomplicated appendicitis

5.2.1 M  ucinous Epithelial Neoplasm: Mucocele of the Appendix The majority of epithelial tumors of the appendix are mucin rich with propensity to form mucoceles (Carr et al. 1995) and account for the majority of appendiceal tumors detected at imaging (Pickhardt et al. 2003). The viscous mucous results in chronic obstruction at the neck of the appendix with subsequent dilatation of the lumen. There are both benign and malignant causes of mucoceles, the most common of which are mucinous neoplasms. Causes include

Imaging of Large Bowel with Multidetector Row CT

mucosal hyperplasia, mucinous neoplasms (mucinous cystadenoma and mucinous cystadenocarcinoma), appendiceal carcinoid, and adjacent cecal tumor. On CT, mucoceles appear as ­well-­circumscribed blind-ending, thin-walled a

657

tubular or spherical cystic masses with central low attenuation arising from the base of the cecum (Fig. 11). Curvilinear mural calcification within the wall is suggestive of the diagnosis, but is seen in less than 50% of patients (Dachman c

b

Fig. 11 (a) A 42-year-old male with nonspecific abdominal pain. Non-contrast coronal CT depicts a fluid-­ distended appendix in the right lower quadrant with internal calcifications (arrow) and soft-tissue density at the base (arrow). Patient subsequently underwent right hemicolectomy with pathology consistent with mucinous adenocarcinoma of the appendix. (b) 56-year-old male presented with increasing abdominal distension. Axial CT

depicts a markedly dilated fluid-filled appendix with mural calcifications (arrow) as well as diffuse ascites. (c) Coronal CT shows the blind-ending appendix arising from the colon (long arrow) as well as scalloping of the liver surface by gelatinous pseudomyxoma peritonei (short arrow), findings consistent with ruptured mucinous neoplasm of the appendix

J.D. Patel et al.

658

et al. 1985; Madwed et al. 1992). Mucoceles smaller than 2 cm in diameter are usually caused by nonneoplastic occlusion and result in simple retention cysts, while those larger than 2 cm are usually caused by a mucinous neoplasm (Carr et al. 1995; Carr and Sobin 1996). Reliably differentiating benign and malignant causes of mucoceles is difficult on imaging alone; however, irregularity of the wall and soft-tissue thickening are features suggestive of malignancy (Wang et al. 2013). Due to the ambiguity in imaging diagnosis, the treatment is surgical. Complications include ileocolic intussusception, gastrointestinal bleeding, ureteral obstruction, and superimposed infection, and if neoplastic in origin, appendiceal rupture may lead to diffuse seeding of the peritoneum with accumulation of gelatinous ascites known as pseudomyxoma peritonei (Fig. 11b, c).

5.2.2 Nonmucinous Epithelial Neoplasm The nonmucinous adenomas and adenocarcinomas are characteristically similar to colorectal neoplasia elsewhere; however, they are exceedingly rare. On CT, these appear as a focal soft-­ tissue mass involving the appendix without mucocele formation. There may be direct invasion of adjacent organs. 5.2.3 Carcinoid Tumor Carcinoid tumors of the appendix arise from neuroendocrine cells and, although rare, are the most common of all appendiceal neoplasms, comprising nearly 80% of primary appendiceal neoplasms (Deans and Spence 1995; Sandor and Modlin 1998). Compared to other neoplasms of the appendix, carcinoid tumors tend to occur more often in young adults (Modlin et al. 2003). Carcinoids have a varied appearance and are often barely discernable, incidentally found at appendectomy. On imaging, most are small in size (usually 10 cm, abnormal positioning of the cecum with a gas-filled loop of colon, and cecal apex directed toward the left upper quadrant referred to as a “coffee bean” sign (Fig. 13). Additional findings include distal colonic decompression, proximal small-bowel ­distension, “whirl sign” with twisting of the mesentery, and “X marks the spot” sign when there are crossing transition points (Rosenblat et al. 2010).

J.D. Patel et al.

660

a

b

c

Fig. 13  A 68-year-old presents to the ER with abdominal pain, nausea, and vomiting. (a) Scout image from CT demonstrates gaseous distension of a loop of bowel directed toward the left upper quadrant. (b) Coronal CT shows the cecum is fluid filled and massively dilated with

cecal apex directed toward the left upper quadrant similar to the scout image. Distal colon is also decompressed. (c) “Whirl sign” (arrows) is noted within the right lower quadrant compatible with twisting of the mesentery and CT findings of cecal volvulus

Imaging of Large Bowel with Multidetector Row CT

a

661

b

c

Fig. 14  A 78-year-old female presented with obstipation, nausea, and vomiting. (a) Scout image from CT shows marked gaseous distension of sigmoid colon. (b) Coronal CT confirms the findings on scout demonstrating a mark-

edly dilated sigmoid colon with apex pointed toward the right upper quadrant, referred to as a “coffee bean” sign. (c) Coronal CT demonstrates a “bird’s beak” appearance at the point of volvulus (arrow)

7.2

loop of bowel extends cranially from the pelvis beyond the level of the transverse colon, it is referred to as the “northern exposure” sign. Sigmoid volvulus tends to occur more commonly in the elderly population compared to cecal volvulus, and risk factors include chronic constipation, redundant sigmoid colon, and high-fiber diet. Complications include closedloop obstruction and ultimately bowel ischemia and perforation.

Sigmoid Volvulus

Sigmoid volvulus is the most common type of colonic volvulus and accounts for 60–75% of all cases of colonic volvulus (Peterson et al. 2009). The sigmoid colon twists on its mesocolon and results in large bowel obstruction with a large distended loop of bowel directed at the right upper quadrant, also referred to as a “coffee bean” sign (Fig. 14). When this

J.D. Patel et al.

662

8

Lower Gastrointestinal Bleeding: Role of CTA

While colonoscopy has traditionally been considered the first-line modality for the diagnosis and evaluation of lower gastrointestinal bleeding, it has its limitations. It may not be feasible in the hemodynamically unstable patient and even in patients that undergo colonoscopy; the ­ underlying etiology may be obscured by overwhelming bleeding or inadequate bowel ­ preparation. MDCT with CT angiography offers an alternative modality for the evaluation of patients with lower gastrointestinal bleeding and has gained greater acceptance over the years as an effective first-line alternative. It is particularly useful in the emergency room setting where it is a quick and readily available test. Patients can be triaged in a timely manner and if actively bleeding, management and treatment can be directed appropriately. In the urgent care setting, evaluation of lower gastrointestinal bleeding with colonoscopy can be very challenging, and some studies suggest the source of bleeding may only be identified in 13% of cases, although a wide range has been reported in the literature (Lee et al. 2011). In fact, a specific cause may not be identified on endoscopy or subsequent work-up in as many as 20% of patients (Whelan et al. 2010). This is at least partially related to the fact that 75–80% of all gastrointestinal bleeding stops spontaneously without intervention (Lee et al. 2011). High-resolution CT allows for short acquisition times and the ability to image at different time intervals following contrast administration. While exact technical parameters vary between institutions, a three-phase examination is generally performed including non-contrast, arterial, and portal venous phase imaging. Imaging is performed without oral contrast as intraluminal positive contrast can obscure or mask active bleeding. Intravenous contrast is administered via a power

injector at a rate of 4 mL/s. Arterial phase imaging can be performed via automated Hounsfield unit triggering when the abdominal aorta reaches 100–150 HU. Portal venous imaging is then performed approximately 70–90 s after initial injection. The non-contrast study is initially used to evaluate for any pre-existing hyperdense material or substance within the colon that may mimic contrast extravasation or blood products on subsequent post-contrast imaging. Occasionally, clotted blood related to recent hemorrhage may appear hyperdense. Active bleeding within the colon is identified on CT angiography by the presence of intraluminal contrast extravasation or “blush,” which may take on a variety of appearances including jetlike stream, pooling of contrast between folds, or more amorphous high-density material within the lumen (Fig. 15) (Artigas et al. 2013). The diagnosis is made when the extravasation is present on arterial phase imaging and changes in shape, size, or location on delayed imaging. CT angiography can detect bleeding at a threshold rate of 0.3–0.5 mL/min. In comparison, conventional catheter-directed angiography can detect bleeding at a threshold rate of 0.5 mL/min and nuclear medicine scintigraphy with tagged 99 Tc-labeled red blood cells at a rate of 0.1 mL/ min (Artigas et al. 2013). Although the nuclear medicine tagged RBC scan is more sensitive, it is more time-consuming and may not always be readily available in the emergency setting. Similarly, catheter-directed angiography is usually reserved for the hemodynamically unstable patient with severe bleeding and is best utilized as a targeted therapeutic procedure. As a result, CTA may be ideally situated to screen and triage patients with lower gastrointestinal bleeding. Many times CTA can also accurately diagnose the underlying etiology, the most common causes of which include diverticulosis, angiodysplasia, ulcers, and malignancy.

Imaging of Large Bowel with Multidetector Row CT

a

663

b

c

Fig. 15  A 78-year-old presents to the ER with lower gastrointestinal bleeding. (a) Axial non-contrast CT shows no acute abnormality. (b) Arterial phase shows contrast pooling within the distal descending colon (arrow). (c)

Conclusion

The pathology of the large bowel is vast; however, MDCT offers an accurate, efficient, and versatile modality for diagnosis. It can be used in virtually any setting whether it is chronic multisystemic diseases or in the acute emergency room setting. When tailored appropriately, MDCT can aid in the diagnosis, management, and treatment of patients. As discussed, CT enterography can be used to evaluate individuals with inflammatory bowel disease, triple-phase CTA for acute lower gastrointestinal bleeding, MDCT for staging and follow-up surveillance of colorectal ­cancer, as well as an emerging role for widespread cancer screening with CT colonography.

Delayed venous phase shows increased pooling and slight change in morphology consistent with active gastrointestinal bleeding (arrow)

References Almeida AT, Melao L, Viamonte B, Cunha R, Pereira JM (2009) Epiploic appendagitis: an entity frequently unknown to clinicians–diagnostic imaging, pitfalls, and look-alikes. AJR Am J Roentgenol 193(5):1243–1251 Artigas JM, Marti M, Soto JA, Esteban H, Pinilla I, Guillen E (2013) Multidetector CT angiography for acute gastrointestinal bleeding: technique and findings. Radiographics 33(5):1453–1470 Bellini D, Rengo M, De Cecco CN, Iafrate F, Hassan C, Laghi A (2014) Perforation rate in CT colonography: a systematic review of the literature and meta-analysis. Eur Radiol 24(7):1487–1496 Bodily KD, Fletcher JG, Solem CA, Johnson CD, Fidler JL, Barlow JM, Bruesewitz MR, McCollough CH, Sandborn WJ, Loftus EV Jr, Harmsen WS, Crownhart BS (2006) Crohn disease: mural attenuation and thickness at contrast-enhanced CT Enterography–­

664 correlation with endoscopic and histologic findings of inflammation. Radiology 238(2):505–516 Bond JH (2000) Clinical evidence for the adenoma-­ carcinoma sequence, and the management of patients with colorectal adenomas. Semin Gastrointest Dis 11(4):176–184 Carr NJ, Sobin LH (1996) Unusual tumors of the appendix and pseudomyxoma peritonei. Semin Diagn Pathol 13(4):314–325 Carr NJ, McCarthy WF, Sobin LH (1995) Epithelial noncarcinoid tumors and tumor-like lesions of the appendix. A clinicopathologic study of 184 patients with a multivariate analysis of prognostic factors. Cancer 75(3):757–768 Chang KJ, Soto JA (2010) Computed tomographic colonography: image display methods. In: A.H. Dachman and A. Laghi (eds) Atlas of virtual colonoscopy. Springer, New York/Dordrecht/Heidelberg/London, pp 111–132 Collins DC (1955) A study of 50,000 specimens of the human vermiform appendix. Surg Gynecol Obstet 101(4):437–445 Connor SJ, Hanna GB, Frizelle FA (1998) Appendiceal tumors: retrospective clinicopathologic analysis of appendiceal tumors from 7,970 appendectomies. Dis Colon Rectum 41(1):75–80 Dachman AH, Lichtenstein JE, Friedman AC (1985) Mucocele of the appendix and pseudomyxoma peritonei. AJR Am J Roentgenol 144(5):923–929 Deans GT, Spence RA (1995) Neoplastic lesions of the appendix. Br J Surg 82(3):299–306 Ekbom A, Helmick C, Zack M, Adami HO (1991) The epidemiology of inflammatory bowel disease: a large, population-based study in Sweden. Gastroenterology 100(2):350–358 Hananel N, Powsner E, Wolloch Y (1998) Adenocarcinoma of the appendix: an unusual disease. Eur J Surg 164(11):859–862 Hatch KF, Blanchard DK, Hatch GF 3rd, Wertheimer-­ Hatch L, Davis GB, Foster RS Jr, Skandalakis JE (2000) Tumors of the appendix and colon. World J Surg 24(4):430–436 Heffernan C, Pachter HL, Megibow AJ, Macari M (2005) Stercoral colitis leading to fatal peritonitis: CT findings. AJR Am J Roentgenol 184(4):1189–1193 Johnson CD, Chen MH, Toledano AY, Heiken JP, Dachman A, Kuo MD, Menias CO, Siewert B, Cheema JI, Obregon RG, Fidler JL, Zimmerman P, Horton KM, Coakley K, Iyer RB, Hara AK, Halvorsen RA Jr, Casola G, Yee J, Herman BA, Burgart LJ, Limburg PJ (2008) Accuracy of CT colonography for detection of large adenomas and cancers. N Engl J Med 359(12):1207–1217 Jones B, Fishman EK, Hamilton SR, Rubesin SE, Bayless TM, Cameron JC, Siegelman SS (1986) Submucosal accumulation of fat in inflammatory bowel disease: CT/pathologic correlation. J Comput Assist Tomogr 10(5):759–763 Kim HJ, Park SH, Pickhardt PJ, Yoon SN, Lee SS, Yee J, Kim DH, Kim AY, Kim JC, Yu CS, Ha HK (2010) CT

J.D. Patel et al. colonography for combined colonic and extracolonic surveillance after curative resection of colorectal cancer. Radiology 257(3):697–704 Lee SS, Ha HK, Yang SK, Kim AY, Kim TK, Kim PN, Lee MG, Myung SJ, Jung HY, Kim JH, Min YI (2002) CT of prominent pericolic or perienteric vasculature in patients with Crohn's disease: correlation with clinical disease activity and findings on barium studies. AJR Am J Roentgenol 179(4):1029–1036 Lee SS, Oh TS, Kim HJ, Chung JW, Park SH, Kim AY, Ha HK (2011) Obscure gastrointestinal bleeding: diagnostic performance of multidetector CT enterography. Radiology 259(3):739–748 Levin B, Lieberman DA, McFarland B, Smith RA, Brooks D, Andrews KS, Dash C, Giardiello FM, Glick S, Levin TR, Pickhardt P, Rex DK, Thorson A, Winawer SJ, G. American Cancer Society Colorectal Cancer Advisory, U. S. M.-S. T. Force, C. American College of Radiology Colon Cancer (2008) Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. CA Cancer J Clin 58(3):130–160 Loftus EV Jr (2004) Clinical epidemiology of inflammatory bowel disease: incidence, prevalence, and environmental influences. Gastroenterology 126(6):1504–1517 Macari M, Balthazar EJ, Megibow AJ (1999) The accordion sign at CT: a nonspecific finding in patients with colonic edema. Radiology 211(3):743–746 Madwed D, Mindelzun R, Jeffrey RB Jr (1992) Mucocele of the appendix: imaging findings. AJR Am J Roentgenol 159(1):69–72 Meyers MA, Volberg F, Katzen B, Alonso D, Abbott G (1973) The angioarchitecture of colonic diverticula. Significance in bleeding diverticulosis. Radiology 108(2):249–261 Miettinen M, Sobin LH (2001) Gastrointestinal stromal tumors in the appendix: a clinicopathologic and immunohistochemical study of four cases. Am J Surg Pathol 25(11):1433–1437 Modlin IM, Lye KD, Kidd M (2003) A 5-decade analysis of 13,715 carcinoid tumors. Cancer 97(4):934–959 Nelson DB, McQuaid KR, Bond JH, Lieberman DA, Weiss DG, Johnston TK (2002) Procedural success and complications of large-scale screening colonoscopy. Gastrointest Endosc 55(3):307–314 Nerad E, Lahaye MJ, Maas M, Nelemans P, Bakers FC, Beets GL, Beets-Tan RG (2016) Diagnostic accuracy of CT for local staging of colon cancer: a systematic review and meta-analysis. AJR Am J Roentgenol 207(5):984–995 Peterson CM, Anderson JS, Hara AK, Carenza JW, Menias CO (2009) Volvulus of the g­astrointestinal tract: appearances at multimodality imaging. Radiographics 29(5):1281–1293 Pickhardt PJ (2006) Incidence of colonic perforation at CT colonography: review of existing data and implications for screening of asymptomatic adults. Radiology 239(2):313–316

Imaging of Large Bowel with Multidetector Row CT Pickhardt PJ (2007) Screening CT colonography: how I do it. AJR Am J Roentgenol 189(2):290–298 Pickhardt PJ, Choi JH (2003) Electronic cleansing and stool tagging in CT colonography: advantages and pitfalls with primary three-dimensional evaluation. AJR Am J Roentgenol 181(3):799–805 Pickhardt PJ, Levy AD, Rohrmann CA Jr, Kende AI (2002) Primary neoplasms of the appendix manifesting as acute appendicitis: CT findings with pathologic comparison. Radiology 224(3):775–781 Pickhardt PJ, Levy AD, Rohrmann CA Jr, Abbondanzo SL, Kende AI (2002) Non-Hodgkin's lymphoma of the appendix: clinical and CT findings with pathologic correlation. AJR Am J Roentgenol 178(5):1123–1127 Pickhardt PJ, Choi JR, Hwang I, Butler JA, Puckett ML, Hildebrandt HA, Wong RK, Nugent PA, Mysliwiec PA, Schindler WR (2003) Computed tomographic virtual colonoscopy to screen for colorectal neoplasia in asymptomatic adults. N Engl J Med 349(23):2191–2200 Pickhardt PJ, Levy AD, Rohrmann CA Jr, Kende AI (2003) Primary neoplasms of the appendix: ­radiologic spectrum of disease with pathologic correlation. Radiographics 23(3):645–662 Pickhardt PJ, Hanson ME, Vanness DJ, Lo JY, Kim DH, Taylor AJ, Winter TC, Hinshaw JL (2008) Unsuspected extracolonic findings at screening CT colonography: clinical and economic impact. Radiology 249(1):151–159 Pickhardt PJ, Hassan C, Halligan S, Marmo R (2011) Colorectal cancer: CT colonography and colonoscopy for detection–systematic review and meta-analysis. Radiology 259(2):393–405 Quayle FJ, Lowney JK (2006) Colorectal lymphoma. Clin Colon Rectal Surg 19(2):49–53 Ranasinghe I, Parzynski CS, Searfoss R, Montague J, Lin Z, Allen J, Vender R, Bhat K, Ross JS, Bernheim S, Krumholz HM, Drye EE (2016) Differences in colonoscopy quality among facilities: development of a post-colonoscopy risk-standardized rate of unplanned hospital visits. Gastroenterology 150(1):103–113 Rao PM, Wittenberg J, Lawrason JN (1997) Primary epiploic appendagitis: evolutionary changes in CT appearance. Radiology 204(3):713–717 Rex DK, Ahnen DJ, Baron JA, Batts KP, Burke CA, Burt RW, Goldblum JR, Guillem JG, Kahi CJ, Kalady MF, O'Brien MJ, Odze RD, Ogino S, Parry S, Snover DC, Torlakovic EE, Wise PE, Young J, Church J (2012) Serrated lesions of the colorectum: review and recommendations from an expert panel. Am J Gastroenterol 107(9):1315–1329 ; quiz 1314, 1330

665 Rosenblat JM, Rozenblit AM, Wolf EL, DuBrow RA, Den EI, Levsky JM (2010) Findings of cecal volvulus at CT. Radiology 256(1):169–175 Sandor A, Modlin IM (1998) A retrospective analysis of 1570 appendiceal carcinoids. Am J Gastroenterol 93(3):422–428 Shinners TJ, Pickhardt PJ, Taylor AJ, Jones DA, Olsen CH (2006) Patient-controlled room air insufflation versus automated carbon dioxide delivery for CT colonography. AJR Am J Roentgenol 186(6):1491–1496 Singh AK, Gervais DA, Hahn PF, Rhea J, Mueller PR (2004) CT appearance of acute appendagitis. AJR Am J Roentgenol 183(5):1303–1307 The American College of Radiology (2014) ACR–SAR– SCBT-MR practice parameter for the performance of Computed Tomography (CT) colonography in adults. ACR Practice Parameters and Technical Standards, Reston Thoeni RF, Cello JP (2006) CT imaging of colitis. Radiology 240(3):623–638 van Eeden S, Offerhaus GJ, Peterse HL, Dingemans KP, Blaauwgeers HL (2000) Gangliocytic paraganglioma of the appendix. Histopathology 36(1):47–49 Wang H, Chen YQ, Wei R, Wang QB, Song B, Wang CY, Zhang B (2013) Appendiceal mucocele: a diagnostic dilemma in differentiating malignant from benign lesions with CT. AJR Am J Roentgenol 201(4):W590–W595 Whelan CT, Chen C, Kaboli P, Siddique J, Prochaska M, Meltzer DO (2010) Upper versus lower gastrointestinal bleeding: a direct comparison of clinical presentation, outcomes, and resource utilization. J Hosp Med 5(3):141–147 Yee J, Rosen MP, Blake MA, Baker ME, Cash BD, Fidler JL, Grant TH, Greene FL, Jones B, Katz DS, Lalani T, Miller FH, Small WC, Sudakoff GS, Warshauer DM (2010) ACR Appropriateness Criteria on colorectal cancer screening. J Am Coll Radiol 7(9):670–678 Yee J, Chang KJ, Dachman AH, Kim DH, McFarland EG, Pickhardt PJ, Cash BD, Bruining DH, Zalis ME (2016) The added value of the CT colonography reporting and data system. J Am Coll Radiol 13(8):931–935 Zalis ME, Barish MA, Choi JR, Dachman AH, Fenlon HM, Ferrucci JT, Glick SN, Laghi A, Macari M, McFarland EG, Morrin MM, Pickhardt PJ, Soto J, Yee J, C. Working Group on Virtual (2005) CT colonography reporting and data system: a consensus proposal. Radiology 236(1):3–9

Peritoneal Surface Malignancy Davide Bellini, Paolo Sammartino, and Andrea Laghi

Contents

Abstract

1    Introduction

 667

2    Definition and Clinical Features of PC

 668

3    CT Imaging 3.1  Technical Consideration 3.2  CT Appearance with Pathological Correlation 3.3  Quantification

The term peritoneal surface malignancies comprises any cancer originated from the peritoneum itself (primary peritoneal malignancy) or metastasized to the peritoneum from a different primary site (secondary peritoneal malignancy). A major problem in treating peritoneal metastases (PM) originating from the various intra-abdominal tumors (gastric, colorectal, ovarian) is how to identify these malignant implants early so as to stage patients accurately (Cotte et  al. 2010). Multidetector computed tomography (MDCT) is considered the imaging modality of choice for the evaluation of these patients. An accurate depiction of peritoneal implants and staging is essential to guide patients’ management.

 669  669  670  671

4    Role of Laparoscopy, US, MR, and PET

 672

References

 673

1 D. Bellini (*) Department of Radiological Sciences, Oncological and Pathological Sciences, University of Rome “Sapienza”, Latina, Italy e-mail: [email protected] P. Sammartino Department of Surgery ‘P. Valdoni’, Sapienza University of Rome, Rome, Italy A. Laghi Department of Radiological Sciences, Oncology and Pathology, University of Rome “Sapienza”, Sant’Andrea University Hospital Via di Grottarossa, Rome, Italy

Introduction

The term peritoneal surface malignancies comprises any cancer originated from the peritoneum itself (primary peritoneal malignancy) or metastasized to the peritoneum from a different primary site (secondary peritoneal malignancy). A major problem in treating peritoneal metastases (PM) originating from the various intra-­ abdominal tumors (gastric, colorectal, ovarian) is how to identify these malignant implants early so as to stage patients accurately (Cotte et al. 2010). Multidetector computed tomography (MDCT) is considered the imaging modality of choice for

Med Radiol Diagn Imaging (2018) https://doi.org/10.1007/174_2018_185, © Springer International Publishing AG Published Online: 17 May 2018

667

668

the evaluation of these patients. An accurate depiction of peritoneal implants and staging is essential to guide patients’ management.

D. Bellini et al.

metastases from gastric and colon cancer) characterized by unusual biological aggressivity and (compared with other indications) by more modest therapeutic results (Goere et al. 2015; Canbay et al. 2014). For this reason, accurately diagnos2 Definition and Clinical ing the extent of existing peritoneal spread as a criterion for selecting candidates to schedule for Features of PC CRS has become a key factor in the preoperative Over the past 30 years, the clinical approach to a management of these patients. This requirement patient with malignant peritoneal spread has holds particularly true if, in the era of valueundergone radical changes. Previously consid- based medicine, we interpret the outcome results ered without hope of survival, regardless of other obtained in relation to the costs incurred. In this clinical factors, in some circumstances, and given connection, the experience the Gustave Roussy rigid selection criteria, these patients can now Institute has acquired in treating peritoneal undergo therapeutic options with curative intent metastases from colorectal cancer offers some (Sadeghi et  al. 2000; Sugarbaker 1999a). Paul useful information. Analyzing data for a large Sugarbaker, who pioneered these treatments, was series of patients who consecutively underwent the first to standardize, for peritoneal malignant CRS combined with HIPEC and investigating spread, a multimodal therapeutic option that fore- oncological outcome and expenses born (Goere sees aggressive cytoreductive surgery (CRS) et  al. 2013; Bonastre et  al. 2008), they undercombined with hyperthermic endoperitoneal che- lined the high overall procedural costs, and the motherapy (HIPEC). The term “Carcinomatosis” notable overall costs compared with the few has itself been progressively abandoned for two (16%) long-term survivors all having an reasons. First, because it is undoubtedly linked in extremely low PCI.  For this reason, whenever the collective imaginary to an incurable disease, the preoperative workup provides a reliable PCI, and also because in day-to-day practice perito- one could already select those patients most neal metastases and primary peritoneal tumors likely to achieve therapeutic benefits, thus notahave similar clinical manifestations and are more bly saving the available resources. Being able to rationally represented by the term peritoneal sur- determine, but also quantify, the extent of peritoface malignancies (PSM) (Di Giorgio and Pinto neal involvement in a patient candidated to CRS 2018; Sugarbaker 2017). combined with HIPEC is therefore the greatest Presently, an international committee of diagnostic challenge in current therapeutic stratexperts belonging to the Peritoneal Surface egies to combat PSM. The task includes the need Oncology Group International (PSOGI) orga- to compare available imaging techniques and nizes biannual meetings that arouse ever-­ consider laparoscopy. A recent meta-analysis we increasing interest (20 participants in the first in conducted (Laghi et al. 2017) on the diagnostic 1998, more than 700  in the last in 2016) and performance of the main radiological imaging have helped to lay down criteria and indications techniques (CT/MRI/PET-CT) tends to show for treating the various neoplasias that in their that the diagnostic technique most often used is clinical history have caused malignant peritoneal CT.  CT performs well despite an undervalued spread (Sugarbaker 2017). Identifying reliable peritoneal tumor load that ranges from 12 to prognostic factors able to influence these 33%, as others have already noted (Esquivel patients’ outcome is, as in other oncological et al. 2010). Exceedingly controversial remains fields, a preeminent need. Most important, in the role of laparoscopy in preoperative PSM PSM the extent of peritoneal spread, assessed staging. If on the one side some investigators with the peritoneal cancer index (PCI) (Jacquet support this technique based on their personal and Sugarbaker 1996), is a determinant prognos- experience (high diagnostic accuracy for stagtic factor especially in these forms (peritoneal ing) (Valle et al. 2012; Fagotti et al. 2013), oth-

Peritoneal Surface Malignancy

ers remain more skeptical, underlining the risk of complications and parietal metastases (portsite metastases) able to interfere with outcome results(Ataseven et al. 2016; Nunez et al. 2014). Although laparoscopy can yield good results, its performance seems too closely linked to the personal experience of specific operators in special circumstances. The decision to use it for staging before CRS combined with HIPEC must also be made on a case-by-case basis. Lastly, apart from the various staging procedures, quantifying the exact amount of peritoneal disease present remains an unusually challenging undertaking involving the disease histotype, the patients’ complex clinical history, the numerous procedures done, and last, the objective limitations of the classification criteria.

3

CT Imaging

3.1

Technical Consideration

CT imaging is routinely performed using a multidetector CT; 16 row-detector CT scanner should be considered the minimum requirement to ensure the acquisition of the entire abdomen and pelvis in a single breath hold, which is extremely important to decrease motion artifacts. Patient preparation includes fasting for at least 6 h and oral administration of 500–1000  mL of water 15–20  min prior to the study. While the patient is lying on the table inside the scanner room a spasmolytic agent is administered intravenously to minimize motion artifacts. The two agents mainly used in clinical routine are glucagon and hyoscine N-butylbromide. Their effects, following an intravenous injection, start approximately 30  s after administration. One milligram of glucagon is enough to reach the desired effect while around 20 mg of hyoscine N-butylbromide are needed to achieve the same results. CT scanning is performed in the supine position from the diaphragm to the ischial tuberosity before and after i.v. administration of iodinated contrast media. Since cancer patients undergo several follow-up examinations, it is advisable

669

to optimize the number of CT phases acquired. Pre-­contrast scan is useful to detect dystrophic calcification, quite common for mucinous tumors. Arterial phase (18–23 s after a threshold of 100 HU is reached using a bolus-tracking monitoring technique) should be acquired to better evaluate vascular infiltration in the case of implants located adjacent to vascular structures. Portal venous phase (60–70 s from the start of the CM injection) is considered the most important phase in cancer patients and cannot be skipped. A delayed phase, acquired from 5 to 10 min after contrast injection, can be indicated in case of some small implants that usually increase conspicuity and contrast enhancement over the time. Axial and other multiplanar reformatted images are useful to detect peritoneal disease and check the common peritoneal site of pathological involvement. CT acquisition parameters and contrast media injection protocol should be carefully selected according to patients’ characteristics in order to reduce the risks for the patients. Vascular and liver parenchymal enhancements are generally affected by different kinetics. Vascular enhancement is determined by the iodine dose delivered per unit of time (Iodine Delivery Rate (IDR)), whereas parenchymal enhancement is influenced by the total iodine dose, which is strictly related to patient body size (Awai et al. 2002; Heiken et al. 1995). Recent studies recommend tailoring the amount of CM according to patients’ lean body weight (LBW) (Ho et al. 2007; Rengo et al. 2011). In particular, Kondo and coworkers recommend injecting 750 mg I per kg of patients’ LBW to maximize the lesion detection rate (Kondo et al. 2011). The rationale for this novel approach is related to the vascular characteristics of adipose tissue, which is poorly perfused compared to parenchymal organs (Ho et al. 2007; Awai et al. 2016). In the case of hypervascular liver lesions or peritoneal implants, enhancement is directly proportional to IDR (gI/s). Radiologists can easily control the IDR by modifying the injection flow rate (FR) according to the iodine concentration of a given contrast medium. As a general rule, if the

D. Bellini et al.

670

acquisition of an arterial phase for body imaging is required, the IDR should not be less than 1.2 g I/s (Yanaga et al. 2008; Matoba et al. 2009). Another very important aspect to consider is the directly proportional relationship between low kV acquisition protocol and tissue contrast enhancement. Lowering the tube voltage during the CT acquisition protocol increases tissue contrast enhancement. Iodine better absorbs X-ray photons at low energy because the X-rays are closer to the iodine k-edge. This principle has been widely used, mainly in CT angiography, and it has potential advantages in clinical practice to reduce both radiation exposure and the amount of CM while maintaining optimal tissue enhancement, especially adjusting image quality using iterative reconstruction technique. As shown in the study by Bae and colleagues (2010), an increase in concentration by 1 mg of iodine per milliliter yields a contrast enhancement of 41.12 HU at 80 kVp, 31.74  HU at 100  kVp, and 26.18  HU at 120  kVp. Moreover, based on the results of Botsikas and colleagues (Botsikas et al. 2016), the total amount of CM in g I could be reduced by 20% if low kVs are used.

3.2

CT Appearance with Pathological Correlation

MDCT plays a vital role in the evaluation of patients with suspected or proven malignant peritoneal disease and it is established as the primary imaging modality of choice in this setting. According to the primary tumor origin, the histology, and the anatomical site, a wide spectrum of imaging appearances of peritoneal surface malignancies exists. From a structural standpoint, peritoneal implants can be classified in three broad categories: solid nodules, cystic nodules, and mixed nodules with either a solid or a cystic component. All types of implant categories may be partially calcified showing hyperdense spots at CT (Pannu et al. 2003; Coakley et al. 2002) and may present with different patterns that depict typical aspects of peritoneal carcinomatosis. From a morphological standpoint and distribution in the peritoneal cavity, six different patterns can be identified (Fig.  1): ascites, micronodular pattern, nodular pattern, omental cake, plaque-like pattern, and mass-like pattern. In patients with peritoneal carcinomatosis, intra-abdominal fluid is a common finding, con-

MORPHOLOGICAL PATTERN MICRONODULAR

Small (1-5mm) peritoneal implants involving serosa and Subserosal adipose tissue.

NODULAR

>5mm, they might acquire oval shape with irregular edge or star shape with irregular edge.

Fig. 1  Morphological pattern

PLAQUE-LIKE

Irregular edge tissue located along visceral wall and peritoneum.

OMENTAL CAKE

Nodular omental involvement with fibrotic streaks; smudged appearance of the omental fat.

MASSES

Confluence of several nodular implants. If > 10 cm is defined «bulky tumor».

ASCITES

Free fluid into abdominal cavity.

Peritoneal Surface Malignancy

sequent to an increased capillary permeability with fluid production and /or obstructed lymphatic vessels with decreased absorption (Forstner et al. 1995). For that reasons, the presence of ascites is usually one of the first “sign and way” of carcinomatosis. Micronodular pattern is characterized by the presence of tiny 1–5 mm spots on peritoneal surface, involving the tunica serosa and subserosal fat. Usually, this feature affects the greater omentum, lesser omentum, and mesentery. Nodular pattern is characterized by the presence of nodular implants larger than 5  mm, involving the tunica serosa and subserosal fat. Nodules may resemble an oval shape with smooth contours or have a star-shaped appearance with spiculated margins. Omental cake consists of a diffuse nodular involvement of the greater omentum in association with fibrotic tissue reaction leading to a consolidation of the omental fat that seems to be stratified (Pannu et al. 2003; Coakley et al. 2002; Forstner et al. 1995). Plaque-like pattern is typically found in the subdiaphragmatic spaces and is due to the confluence of multiple nodular implants. Plaques are irregular soft-tissue thickenings of inconstant extension that coat abdominal viscera and peritoneal walls, usually appearing as soft tissue scalloping the liver and splenic surfaces and presenting a lower attenuation than the parenchyma on contrast-enhanced scans (Walkey et al. 1988). Mass-like pattern is typically found in the pelvis and arises from the same mechanism of “plaque-like” appearance. In this case, the confluence of multiple nodular implants can lead to the formation of tissue mass which can reach sizes of several centimeters. When a single mass is larger than 10 cm, it is called “bulky tumor.”

3.3

Quantification

Several studies demonstrated that it is possible to estimate the probability of complete cytoreduction with preoperative computed tomography (Yan et al. 2005).

671

The initial extent of peritoneal carcinomatosis represents one of the most important prognostic factors. Localization of peritoneal implants is crucial to obtain a precise map of peritoneal implants, necessary for the presurgical and pretreatment evaluation of the disease. To provide a staging as precise as possible, radiologists must carefully check the peritoneal surface, spaces, and ligaments, specifying in the report each site involved. Several methods of classification have been used to investigate the extent of carcinomatosis. Among them, the best seems to be the Peritoneal Cancer Index (PCI). The PCI of Sugarbaker (Fig. 2) was chosen by an expert panel as a useful quantitative prognostic tool (Portilla et al. 2008), taking into account size, location, and number of lesions. According to this method the abdomen is subdivided into 12 abdomino-pelvic areas (9 areas  +  4 relating to small bowel). A score from 0 to 3 is assigned to each specific area and it is calculated according with lesions’ size reaching a maximum score of 39 (Rengo et al. 2011; Sugarbaker 1999b). This score allows an estimation of the probability of a complete cytoreduction and guides patients’ management. PCI threshold for surgery varies according to the origin of primary tumor: PCI  1000

hsst4 >1000 260 ± 74 300 ± 140

hsst5 7321 7.2 ± 1.6 377 ± 18

For imaging NET, however, irrespective of its higher affinity to sstr2, 68Ga-DOTA-TATE was not superior over 68Ga-DOTA-TOC PET/CT in a comparative study (Poeppel et al. 2011), both showing comparable diagnostic accuracy for the detection of NET lesions. The approximately tenfold higher affinity for the sstr2 of 68Ga-DOTA-­ TATE did not prove to be clinically relevant. Quite unexpectedly, SUVmax of 68Ga-DOTA-­ TOC scans even tended to be higher than their 68 Ga-DOTA-TATE counterparts. Results for 68Ga DOTA-NOC which is known to bind also to other sstr subtypes (mainly sstr 3 and 5) were also inconsistent (Wild et al. 2013; Kabasakal et al. 2012); therefore, all three tracers can most probably be regarded as quite equally efficient in NET imaging. False negative SSTR imaging results mainly occur in very small lesions or lesions with low SSTR expression (e.g., poorly differentiated NET or insulinoma). False positive results may occur in inflammation. Furthermore, physiological tracer uptake in the pituitary gland, the thyroid gland, the uncinate process of the pancreas, often also in the caudate, in adrenal glands, and accessory spleen should be considered when interpreting PET results.

3.1.1 Detection of Primary Besides tumor staging, where SSTR imaging is routinely recommended in most well-­ differentiated NET (Pavel et al. 2016), detection of primary site in patients presenting with suspicion for NET or with proven metastases of unknown primary is one of the major fields of SSTR PET/CT. In a series of 38 patients SSTR PET/CT demonstrated a significantly higher sensitivity (94%

Multislice PET/CT in Neuroendocrine Tumors

vs. 63%) and accuracy (87% vs. 68%) than contrast-­enhanced CT alone (Kazmierczak et al. 2016). These findings are in line with other studies investigating the applicability and clinical value of SSTR-targeted hybrid imaging for primary detection in CUP-NET (Alonso et al. 2014; Naswa et al. 2012; Prasad et al. 2010; Ambrosini et al. 2010; Lee et al. 2015) and add to the literature as they directly compare 68 Ga-DOTA-TATE-­PET and contrast-enhanced CT in the setting of a blinded ­diagnostic accuracy study. Naswa et al. demonstrated the usefulness of 68Ga-DOTA-­NOC-PET/CT for primary detection in CUP-­NET patients but did not quantify the additional diagnostic accuracy of SSTR-targeted PET compared to CT (Naswa et al. 2012). Additionally, the CT part of the examination was performed without using an intravenous contrast agent, potentially accounting for the lower primary detection rate (60%) compared to other studies with contrastenhanced CT. In another series of 29 patients with proven NET metastases SSTR PET/CT detected so far unknown primary lesions in 59% of the patients (Prasad et al. 2010). Also in patients with suspected NET due to clinical symptoms, elevated levels of tumor markers, or indeterminate tumors suggestive of NET, SSTR PET/CT is highly accurate for tumor detection presenting with a sensitivity of 81% and specificity of 90% in 104 consecutive patients (Haug et al. 2012). Therefore, SSTR PET/CT can be recommended in patients presenting with neuroendocrine tumor of unknown origin or strong suspicion for NET.

3.1.2 Suspicion of Tumor Recurrence As expected based on the results for primary staging SSTR PET/CT has also been demonstrated to be an excellent tool for detection of tumor recurrence in NET patients. Haug et al. demonstrated a sensitivity of 90% and a specificity of 82% for SSTR PET/CT in a series of 63 patients with 29 documented relapses

681

where PET/CT was performed for the reasons of ­ regular follow-up, of increasing tumor marker or clinical suspicion for recurrence (Haug et al. 2014).

3.1.3 Theranostics Another strong indication for SSTR PET/CT is in vivo quantification of SSTR expression with regard to the possibility of SSTR-based treatment (theranostic concept). The term theranostics has been introduced to the field of personalized medicine to summarize the integration of diagnostics and therapeutics in patient management. For well-differentiated NET several dedicated treatment options exist, including SSTR-targeted therapy (biotherapy with cold SS analogues or peptide receptor radionuclide therapy (PRRT)) (Singh et al. 2016). PRRT is a novel treatment strategy for advanced NET with a tumor response rate of about 20% and tumor stabilization in 60% for intervals up to 3 years (Kjaer and Knigge 2015). Sufficient SSTR expression determined with 68 Ga-DOTA-peptide PET/CT has to be documented before PRRT can be recommended (Bodei et al. 2013; Caplin et al. 2014). SSTR PET/CT offers straightforward quantification of all tumor lesions since some groups could show a high correlation between quantitative uptake and SSTR expression in immunohistochemistry (Kaemmerer et al. 2011). Therefore SSTR PET/ CT even allows for a more robust patient selection in these often heterogeneous tumors compared to immunohistochemistry, mostly performed only for one or two lesions. Usually the Krenning score (Kwekkeboom and Krenning 2002) is used to grade the uptake intensity of NET on SSTR imaging, using liver and spleen as reference organs. Figure 3 shows four patient examples representing the four grades of SSTR expression, ranging from Grade 1 (uptake < normal liver) over Grade 2 (uptake = normal liver) and Grade 3 (uptake > normal liver) to Grade 4 (uptake > spleen). Typically, peptide receptor radionuclide therapy (PRRT) is considered when the Krenning score is Grade 3 or 4.

682

G. Pöpperl and C. Cyran

Fig. 3  68Ga-DOTA-TOC PET/CT images (MIP, CT, and PET/CT fused images) of four different NET patients representing the four grades of SSTR expression according to the Krenning score (Kwekkeboom and Krenning 2002),

ranging from Grade 1 (uptake < normal liver) over Grade 2 (uptake = normal liver) and Grade 3 (uptake > normal liver) to Grade 4 (uptake > spleen)

3.1.4 Response Prediction to PRRT Several groups have shown that uptake at baseline SSTR PET/CT can predict the delivered absorbed dose or the response after PRRT. Ezziddin and colleagues evaluated 61 lesions in 21 patients treated with 177Lu-DOTA-­TATE and found a significant correlation between 68Ga-DOTA-TOC SUVmax or SUV mean and the tumor-absorbed dose during the first treatment cycle (Ezziddin et al. 2012). Kratochwil et al. even defined a cutoff value for optimal baseline SUVmax by demonstrating that an SUVmax above 16.4 on 68Ga-DOTA-TOC PET/CT before PRRT (using 90Y–DOTA-TOC or 177 Lu-DOTA-TATE) was a sensitive predictor of lesion stabilization or shrinkage of liver metastases in 30 patients (Kratochwil et al. 2015). The excellent results of the NETTER-1 trial (Strosberg

et al. 2017) provide strong support for wider application of 177Lu-DOTA-TATE therapy in patients with metastatic or inoperable progressive intestinal G1 and G2 NET. Further prospective studies are warranted to define semiquantitative thresholds below which a benefit from PRRT is unlikely and thus should be refrained.

3.1.5 Therapy Monitoring The value of SSTR PET/CT for assessing response to PRRT compared to RECIST criteria in conventional imaging, however, is still a matter of debate. Therapy monitoring and the early prediction of therapy response to treatment is essential for ­therapy guidance, to minimize unnecessary side effects of potentially ineffective therapies and

Multislice PET/CT in Neuroendocrine Tumors

683

improve the cost effectiveness of treatment regimens. PRRT using somatostatin analogues has shown promising treatment results reducing signs and symptoms of hormone hypersecretion, improving quality of life, and slowing tumor growth in patients with inoperable or metastasized NET (Bodei et al. 2009; Teunissen et al. 2011). Via internalization through SSTR (mainly subtype 2) targeted radiopharmaceuticals such as DOTATATE or DOTA-TOC labelled with high-­energy β-emitters (90Yttrium, 177Lutetium) irradiate primary tumors and metastases achieving significant volume reductions with partial and complete objective responses in up to 30% of the patients (Muros et al. 2009; Grozinsky-Glasberg et al. 2011). As morphology-based assessment of therapy response using CT criteria alone does not correlate well with progression-free survival, clinical outcome, and quality of life in NET, other molecular imaging biomarkers of therapy response are particularly sought after. The currently available data is inconclusive leaving it unclear whether assessing the response to PRRT with 68Ga-DOTApeptides, either after one treatment cycle or at the end of treatment, offers an advantage over conventional anatomic imaging alone, since studies on this topic demonstrated heterogeneous results (Haug et al. 2010; Gabriel et al. 2009; Wulfert et al. 2014). Haug and colleagues evaluated

Ga-DOTA-TATE PET/CT in the prediction of progression-free survival and clinical outcome in NET patients after PRRT. The authors report that patients with a decline in SUV tumor/spleen ratio after PRRT demonstrated a significantly longer time-to-progression compared to patients without SUVT/S decline, concluding the parameter may be a relevant predictor of patient outcome in welldifferentiated NET. Additionally, pretherapeutic SUVmax was an accurate predictor of time-toprogression, potentially because high receptor density causes high accumulation of activity and good response during PRRT (Haug et al. 2010). Similarly Öksüz et al. reported that therapeutic response to PRRT is associated with pre-therapeutic tumor uptake of 68Ga-DOTA-TOC with defined SUVmax cutoff values for 68Ga-DOTA-TOC and assumed 90Y–DOTA-TOC uptake to predict response to PRRT (Öksüz et al. 2014). However, these authors critically discuss that 68Ga-DOTATOC SUVmax is not directly related to 90Y– DOTA-­TOC tumor uptake and is only a rough estimate of the subsequently absorbed dose delivered by PRRT (Öksüz et al. 2014). One example for treatment monitoring with PET using 68Ga-labelled somatostatin analogues is shown in Fig. 4.

Fig. 4 Follow-up 68Ga-DOTA-TOC PET images and chromogranin A levels of a patient with metastasized pNET during and after PRRT. Baseline investigation (11/14) shows extensive abdominal lymph node metastases, small disseminated liver metastases, and some bone

metastases. Follow-up investigations during and after PRRT (with 177Lu-DOTA-TOC) show a marked decrease of tumor load in PET as well as a decrease in tumor marker level

68

G. Pöpperl and C. Cyran

684

3.2

Somatostatin Receptor Antagonists

Most somatostatin-based radiotracers used for diagnostic or therapeutic purposes were agonists so far. Their internalization was thought to be the essential mechanism for high accumulation and long-lasting tracer uptake. However, Ginj et al. were the first demonstrating in animal studies that radiolabelled SSTR antagonists are superior to agonists for tumor targeting due to the fact that antagonists bind to a larger number of binding sites compared to agonists (Ginj et al. 2006). In vitro receptor autoradiography showed four times higher accumulation in SSTR expressing human tumor samples of a SSTR antagonist compared to a SSTR agonist (Cescato et al. 2011). These results were also confirmed clinically in a first human imaging study comparing 111In-DOTABASS (antagonist) and 111In-pentreotide (agonist) showing higher tumor and lower renal uptake for the antagonist (Wild et al. 2011). The preliminary results for antagonists seem to have major impact especially on therapeutic applications of antagonists. In a first therapeutic approach significantly higher tumor doses were observed for 177Lu-DOTA-JR11 (antagonist) compared to 177Lu-DOTA-TATE (agonist) during PRRT. Therefore, antagonists seem to have vast potential especially for theranostics (imaging and PRRT) of NET.

4

PET Using Fluorodeoxyglucose (FDG)

F–fluorodeoxyglucose (FDG) is the most commonly used radiopharmaceutical for PET imaging in oncology. It is taken up via a glucose transporter, phosphorylated, and trapped in the cell according to its metabolic rate. FDG uptake therefore reflects increased glucose metabolism and is linked to the cellular proliferative activity (Bombardieri et al. 2003; Boellaard et al. 2010). High uptake is usually associated with aggressive tumor growth and a less favorable prognosis. The diagnostic value of FDG PET for imaging well-differentiated NET, which compose the 18

majority of NET, therefore, is limited since it frequently fails to visualize well-differentiated and slow-growing NET with low proliferation rate (Eriksson et al. 2000). High FDG uptake usually is found in poorly differentiated NET resulting in a high detection rate of G3 tumors (Kayani et al. 2008). Furthermore, FDG PET may be useful in less well-differentiated tumors offering prognostic information depending on FDG uptake (Pasquali et al. 1998) that can influence therapeutic management. In some studies, a variable tracer uptake intensity of the different tumor manifestations in metastatic NET was described as a so-called flipflop phenomenon with high uptake of somatostatin-based tracers in somatostatin receptor positive tumor parts and high FDG uptake of somatostatin receptor negative NET manifestations. Thereby FDG and somatostatin analogues provide noninvasive information reflecting the different degrees of differentiation of the often heterogeneous NET. Two representative patient examples are shown in Fig. 5, one patient with well-­differentiated SSTR positive NET lesions without FDG uptake and one patient with only minor SSTR expression and coexistent FDG uptake of NET lesions. More recent prospective data on FDG PET/ CT suggest that positive FDG uptake provides excellent prognostic information predicting early progression and a more unfavorable course of disease also in patients with low grade NET (G1/2) (Garin et al. 2009; Bahri et al. 2014). FDG PET seems to be a strong and independent prognostic predictor even exceeding traditional markers such as Ki-67, CgA or the presence of liver metastases (Binderup et al. 2010; Garin et al. 2009) and FDG PET/CT was shown to be of greater prognostic utility than positive sstr expression in somatostatin receptor imaging (Bahri et al. 2014). Authors concluded that positive SSTR imaging does not eliminate the need for performing FDG PET. In the same context, recent data demonstrated that negative FDG uptake is a positive predictor for response to PRRT (Severi et al. 2013), while FDG positive NET more often show progression after PRRT. The role of FDG PET as predictor of tumor response and prognosis still has to be investigated in larger series of patients.

Multislice PET/CT in Neuroendocrine Tumors

685

a

b

Fig. 5  18F–FDG PET (left; MIP) and 68Ga-DOTA-TOC (right; MIP) in two NET patients presenting with (a) a well-differentiated (KI-67: 5%) metastasized midgut NET showing positive SSTR expression of lymph node metas-

tases but no positive FDG uptake. (b) Less well-­ differentiated pNET (KI 67: 15%) with only minor SSTR expression of liver and bone metastases and co-existing positive FDG uptake

5

proven to be superior over CT or conventional SRS for detection of metastases (Koopmans et al. 2006). ­However, it is doubtable, whether there is really an advantage over SSTR PET/CT, since only the latter offers a theranostic concept. In an intraindividual comparison of both methods in a mixed group of NET patients (Haug et al. 2009) 68 Ga-DOTA-TATE PET proved clearly superior over FDOPA PET for detection and staging of NET. FDOPA uptake tended to be increased only in those patients with elevated plasma serotonin. Therefore, FDOPA PET seems to have a promising role only in patients with gastroenteropancreatic NET with negative or inconclusive findings at conventional morphological or conventional SRS (Ambrosini et al. 2007), where it could help to detect the primary tumor or other unsuspected lesions in the majority of patients. Carbidopa pretreatment is recommended for FDOPA PET since it was shown to enhance the sensitivity of FDOPA PET by reducing the conversion of FDOPA and the excretion of F-­dopamine in the urine. Furthermore, carbidopa lowers physiological FDOPA uptake in the

 ET Using P Fluorodihydroxyphenylalanine (FDOPA)

Based on the concept of amine precursor uptake and decarboxylation (APUD), the amine precursor l-dihydroxyphenylalanine (l-DOPA) has also been utilized for PET imaging of NET. Only few studies on FDOPA PET/CT in NET have been published so far demonstrating the latter to be particularly suited to visualize carcinoid tumors with elevated serotonin levels (Koopmans et al. 2006). The increased activity of aromatic L-amino acid decarboxylase, involved in the synthesis of serotonin in these tumors, explains the high sensitivity in serotonin positive carcinoid tumors. However, sensitivity of FDOPA PET/CT in mostly serotonin negative high grade NET or in foregut or hindgut NET is comparably low (Montravers et al. 2006). Therefore, FDOPA PET/CT seems to be a sensitive imaging tool only for the detection of functional NET, especially when located in midgut, where it has been

G. Pöpperl and C. Cyran

686

p­ ancreas, both leading to a higher availability of FDOPA for tumor uptake increasing the tumor-­ to-­background ratio of tracer uptake (Hoffman et al. 1992). Controversial reports have been published for localizing pancreatic insulin-secreting tumors where somatostatin receptor imaging in general is known to be less useful, since insulinomas express SSTR only in about 50%. On the one hand promising results have been published for patients with insulinoma and negative CT, MRI, and ultrasound results. In a small series FDOPA PET imaging was clearly superior over morphological imaging for the detection of beta-cell hyperplasia in adults and was recommended for the detection of insulinoma or betacell hyperplasia in patients with confirmed hyperinsulinemic hypoglycemias when other diagnostic workup was negative (Kauhanen et al. 2007). In a more recent report of Tessonnier et al., however, FDOPA PET failed to detect the tumor in most patients and was only of very limited value in localizing pancreatic insulinsecreting tumors in adult hyperinsulinemic hypoglycemia (Tessonnier et al. 2010). FDOPA PET/CT showed to be a very useful tool also in the detection of pheochromocytoma and paraganglioma showing even higher sensitivity and specificity in patients with extraadrenal, predominantly noradrenaline secreting, and hereditary types of pheochromocytoma and paraganglioma compared to the standard MIBG scintigraphy (Fottner et al. 2010).

6

Glucagon-like Peptide-1 Receptor (GLP-1R) Imaging

Strong overexpression of glucagon-like peptide-1 receptor (GLP-1R) is the basis for this kind of molecular imaging (Körner et al. 2012) of clinically often life-threatening benign insulinomas which mostly are difficult to localize by conventional imaging methods due to their small lesion size. Up to now 111In-labelled GLP-1R ligands for SPECT/CT imaging have been prospectively evaluated in patients with benign insulinomas showing a relatively high detection rate despite

the inconvenient imaging properties of 111Indium (Christ et al. 2013). Recently, 68Ga-labelled compounds have been developed, as expected showing higher spatial resolution and sensitivity. Currently, two prospective clinical trials evaluate the impact of two 68Ga-labelled ligands for GLP-1R in patients with endogenous hyperinsulinemic hypoglycemia. Results of these studies have to be awaited to define the clinical role of this imaging modality in the future. Conclusions

Conventional imaging modalities such as ultrasound, CT, MRI, and EUS are the workhorses for abdominal imaging and therefore also relevant for imaging NET. However, especially in these tumors, molecular imaging techniques have gained more and more impact on patient management. With the introduction of hybrid PET/CT scanners the combined morphological and functional imaging approach has been one of the fastest growing techniques. The lack of anatomic information, especially when using highly ­specific PET tracers as in NET, requires exact anatomic correlation, which is offered by hybrid scanners with high quality CT devices and scanning protocols. The better spatial resolution of PET clearly argues for a complete replacement of planar scintigraphy and SPECT since it has led to a better localization of occult tumors in the small intestine and pancreas as well as to an improved staging and restaging especially with regard to bone metastases, small lymph node metastases, or peritoneal carcinosis. Concerning the various tracers, the broad variety of NET demands for a diagnostic workup which is tailored to the specific patient’s pathology. Especially SSTR imaging mainly enhanced the diagnostic workup of patients with well-­differentiated NET (G1/2) and therefore tracers targeting SSTR are the first choice in this tumor entity. The development of somatostatin antagonists instead of agnoists may further improve this approach especially also with regard to the theranostic concept and evaluation of patients eligible for PRRT, but prospective data in larger series of patients are still lacking.

Multislice PET/CT in Neuroendocrine Tumors

FDG PET primarily is useful in patients with poorly differentiated, high proliferative NET associated with rapidly increasing tumor markers. Quite expected, however, it performs inferior to SSTR imaging in well-differentiated NET. In the latter, however, it seems to be useful for detection of less differentiated parts of the tumor and, therefore, FDG PET may act as prognostic marker even in well-differentiated NET. Promising data also exist for FDG as a predictive marker regarding response to PRRT. PET using amine precursors such as FDOPA performs also inferior to SSTR imaging and will only play an inferior role in imaging NET. FDOPA could potentially be applied as first line molecular imaging probe in pheochromocytomas/paragangliomas and may play a role in SSTR negative midgut tumors with positive immunohistochemical detection of serotonin as well as in insulinomas. The clinical role of SSTR antagonist and of GLP-1R ligands for NET imaging needs to be further evaluated.

References Ahmed A, Turner G, King B et al (2009) Midgut neuroendocrine tumours with liver metastases: results of the UKINETS study. Endocr Relat Cancer 16:885–894 Alonso O, Rodríguez-Taroco M, Savio E et al (2014) 68 Ga-DOTATATE PET/CT in the evaluation of patients with neuroendocrine metastatic carcinoma of unknown origin. Ann Nucl Med 28:638–645 Ambrosini V, Nanni C, Zompatori M et al (2010) 68 Ga-DOTA-NOC PET/CT in comparison with CT for the detection of bone metastasis in patients with neuroendocrine tumours. Eur J Nucl Med Mol Imaging 37:722–727 Ambrosini V, Tomassetti P, Rubello D et al (2007) Role of 18 F-dopa PET/CT imaging in the management of patients with 111In-pentetreotide negative GEP tumours. Nucl Med Commun 28:473–477 Antunes P, Ginj M, Zhang H et al (2007) Are radiogallium-­ labelled DOTA-conjugated somatostatin analogues superior to those labelled with other radiometals? Eur J Nucl Med Mol Imaging 34:982–993 Auernhammer CJ, Göke B (2011) Therapeutic strategies for advanced neuroendocrine carcinomas of jejunum/ ileum and pancreatic origin. Gut 60:1009–1021 Bahri H, Laurence L, Edeline J et al (2014) High prognostic value of 18F-FDG PET for metastatic gastroentero-

687 pancreatic neuroendocrine tumors: a long-term evaluation. J Nucl Med 55:1786–1790 Baumann T, Rottenburger C, Nicolas G et al (2016) Gastroenteropancreatic neuroendocrine tumours (GEP-NET) - imaging and staging. Best Pract Res Clin Endocrinol Metab 30:45–57 Binderup T, Knigge U, Loft A et al (2010) 18 F-fluorodeoxyglucose positron emission tomography predicts survival of patients with neuroendocrine tumors. Clin Cancer Res 16:978–985 Bodei L, Ferone D, Grana CM et al (2009) Peptide receptor therapies in neuroendocrine tumors. J Endocrinol Investig 32:360–369 Bodei L, Mueller-Brand J, Baum RP et al (2013) The joint IAEA, EANM, and SNMMI practical guidance on peptide receptor radionuclide therapy (PRRNT) in neuroendocrine tumours. Eur J Nucl Med Mol Imaging 40:800–816 Boellaard R, O'Doherty MJ, Weber WA et al (2010) FDG PET and PET/CT: EANM procedure guidelines for tumour PET imaging: version 1.0. Eur J Nucl Med Mol Imaging 37:181–200 Bombardieri E, Aktolun C, Baum RP et al (2003) FDG-­ PET: procedure guidelines for tumour imaging. Eur J Nucl Med Mol Imaging 30:115–124 Breeman WA, de Jong M, de Blois E et al (2005) Radiolabelling DOTA-peptides with 68Ga. Eur J Nucl Med Mol Imaging 32:478–485 Buchmann I, Henze M, Engelbrecht S et al (2007) Comparison of 68Ga-DOTATOC PET and 111In-DTPAOC (Octreoscan) SPECT in patients with neuroendocrine tumours. Eur J Nucl Med Mol Imaging 34:1617–1626 Caplin ME, Pavel M, Ruszniewski P (2014) Lanreotide in metastatic enteropancreatic neuroendocrine tumors. N Engl J Med 371:1556–1557 Cescato R, Waser B, Fani M et al (2011) Evaluation of 177 Lu-DOTA-sst2 antagonist versus 177Lu-DOTA-sst2 agonist binding in human cancers in vitro. J Nucl Med 52:1886–1890 Christ E, Wild D, Ederer S et al (2013) Glucagon-like peptide-1 receptor imaging for the localisation of insulinomas: a prospective multicentre imaging study. Lancet Diabetes Endocrinol 1(2):115–122 Cimitan M, Buonadonna A, Cannizzaro R et al (2003) Somatostatin receptor scintigraphy versus chromogranin A assay in the management of patients with neuroendocrine tumors of different types: clinical role. Ann Oncol 14:1135–1141 Decristoforo C, Mather SJ, Cholewinski W et al (2000) 99m Tc-EDDA/HYNIC-TOC: a new 99mTc-labelled radiopharmaceutical for imaging somatostatin receptor-­ positive tumours; first clinical results and intra-patient comparison with 111In-labelled octreotide derivatives. Eur J Nucl Med 27:1318–1325 Dromain C, de Baere T, Lumbroso J et al (2005) Detection of liver metastases from endocrine tumors: a prospective comparison of somatostatin receptor scintigraphy, computed tomography, and magnetic resonance imaging. J Clin Oncol 23:​ 70–78

688 Eriksson B, Bergström M, Orlefors H et al (2000) Use of PET in neuroendocrine tumors. In vivo applications and in vitro studies. Q J Nucl Med 44:68–76 Ezziddin S, Lohmar J, Yong-Hing CJ et al (2012) Does the pretherapeutic tumor SUV in 68Ga DOTATOC PET predict the absorbed dose of 177Lu octreotate? Clin Nucl Med 37:e141–e147 Fottner C, Helisch A, Anlauf M et al (2010) 6-18F-fluoro-­ L-dihydroxyphenylalanine positron emission tomography is superior to 123I-metaiodobenzyl-guanidine scintigraphy in the detection of extraadrenal and hereditary pheochromocytomas and paragangliomas: correlation with vesicular monoamine transporter expression. J Clin Endocrinol Metab 95:2800–2810 Frilling A, Sotiropoulos GC, Radtke A et al (2010) The impact of 68Ga-DOTATOC positron emission tomography/computed tomography on the multimodal management of patients with neuroendocrine tumors. Ann Surg 252:850–856 Gabriel M, Decristoforo C, Donnemiller E et al (2003) An intrapatient comparison of 99mTc-EDDA/HYNIC-­ TOC with 111In-DTPA-octreotide for diagnosis of somatostatin receptor-expressing tumors. J Nucl Med 44:708–716 Gabriel M, Decristoforo C, Kendler D et al (2007) 68 Ga-DOTA-Tyr3-octreotide PET in neuroendocrine tumors: comparison with somatostatin receptor scintigraphy and CT. J Nucl Med 48:508–518 Gabriel M, Muehllechner P, Decristoforo C et al (2005) 99m Tc-EDDA/HYNIC-Tyr(3)-octreotide for staging and follow-up of patients with neuroendocrine gastro-­ entero-­pancreatic tumors. Q J Nucl Med Mol Imaging 49:237–244 Gabriel M, Oberauer A, Dobrozemsky G et al (2009) 68 Ga-DOTA-Tyr3-octreotide PET for assessing response to somatostatin-receptor-mediated radionuclide therapy. J Nucl Med 50:1427–1434 Garin E, Le Jeune F, Devillers A et al (2009) Predictive value of 18F-FDG PET and somatostatin receptor scintigraphy in patients with metastatic endocrine tumors. J Nucl Med 50:858–864 Ginj M, Zhang H, Waser B et al (2006) Radiolabelled somatostatin receptor antagonists are preferable to agonists for in vivo peptide receptor targeting of tumors. Proc Natl Acad Sci 103:16436–16441 Grozinsky-Glasberg S, Barak D, Fraenkel M et al (2011) Peptide receptor radioligand therapy is an effective treatment for the long-term stabilization of malignant gastrinomas. Cancer 117:1377–1385 Haug A, Auernhammer CJ, Wängler B et al (2009) Intraindividual comparison of 68Ga-DOTA-TATE and 18 F-DOPA PET in patients with well-differentiated metastatic neuroendocrine tumours. Eur J Nucl Med Mol Imaging 36:765–770 Haug AR, Auernhammer CJ, Wangler B et al (2010) 68 Ga-DOTATATE PET/CT for the early prediction of response to somatostatin receptor-mediated radionuclide therapy in patients with well-differentiated neuroendocrine tumors. J Nucl Med 51:1349–1356

G. Pöpperl and C. Cyran Haug AR, Cindea-Drimus R, Auernhammer CJ et al (2012) The role of 68Ga-DOTATATE PET/CT in suspected neuroendocrine tumors. J Nucl Med 53:1686–1692 Haug AR, Cindea-Drimus R, Auernhammer CJ et al (2014) Neuroendocrine tumor recurrence: diagnosis with 68 Ga-DOTATATE PET/CT. Radiology 270:517–525 Hellman P, Lundstrom T, Ohrvall U et al (2002) Effect of surgery on the outcome of midgut carcinoid disease with lymph node and liver metastases. World J Surg 26:991–997 Hoffman JM, Melega WP, Hawk TC et al (1992) The effects of carbidopa administration on 6-[18F]fluoro-L-­ dopa kinetics in positron emission tomography. J Nucl Med 33:1472–1477 Hope TA, Pampaloni MH, Nakakura E et al (2015) Simultaneous 68Ga-DOTA-TOC PET/MRI with gadoxetate disodium in patients with neuroendocrine tumor. Abdom Imaging 40:1432–1440 Ichikawa T, Peterson MS, Federle MP et al (2000) Islet cell tumor of the pancreas: biphasic CT versus MR imaging in tumor detection. Radiology 216:163–171 Kabasakal L, Demirci E, Ocak M et al (2012) Comparison of 68Ga-DOTATATE and 68Ga-DOTANOC PET/CT imaging in the same patient group with ­neuroendocrine tumours. Eur J Nucl Med Mol Imaging 39:1271–1277 Kaemmerer D, Peter L, Lupp A et al (2011) Molecular imaging with 68Ga-SSTR PET/CT and correlation to immunohistochemistry of somatostatin receptors in neuroendocrine tumours. Eur J Nucl Med Mol Imaging 38:1659–1668 Kauhanen S, Seppanen M, Minn H et al (2007) Fluorine-­ 18-­L-dihydroxyphenylalanine (18F-DOPA) positron emission tomography as a tool to localize an insulinoma or beta-cell hyperplasia in adult patients. J Clin Endocrinol Metab 92:1237–1244 Kayani I, Bomanji JB, Groves A et al (2008) Functional imaging of neuroendocrine tumors with combined PET/CT using 68Ga-DOTATATE (DOTA-DPhe1,Tyr3-­ octreotate) and 18F-FDG. Cancer 112:2447–2455 Kazmierczak PM, Rominger A, Wenter V et al (2016) The added value of 68Ga-DOTA-TATE-PET to contrast-­ enhanced CT for primary site detection in CUP of neuroendocrine origin. Eur Radiol. doi:10.1007/ ­ s00330-016-4475-3 Kjaer A, Knigge U (2015) Use of radioactive substances in diagnosis and treatment of neuroendocrine tumors. Scand J Gastroenterol 50:740–747 Koopmans KP, de Vries EG, Kema IP et al (2006) Staging of carcinoid tumours with 18F-DOPA PET: a prospective, diagnostic accuracy study. Lancet Oncol 7:728–734 Körner M, Christ E, Wild D et al (2012) Glucagon-like peptide-1 receptor overexpression in cancer and its impact on clinical applications. Front Endocrinol 3:158 Kratochwil C, Stefanova M, Mavriopoulou E et al (2015) SUV of [68Ga]DOTATOC-PET/CT predicts response probability of PRRT in neuroendocrine tumors. Mol Imaging Biol 17:313–318 Kwee TC, van Ufford HM, Beek FJ et al (2009) Whole-­ body MRI, including diffusion-weighted imaging, for

Multislice PET/CT in Neuroendocrine Tumors the initial staging of malignant lymphoma: comparison to computed tomography. Investig Radiol 44:683–690 Kwekkeboom DJ, Krenning EP (2002) Somatostatin receptor imaging. Semin Nucl Med 32:84–91 Lee JR, Kim JS, Roh JL et al (2015) Detection of occult primary tumors in patients with cervical metastases of unknown primary tumors: comparison of 18F FDG PET/CT with contrast-enhanced CT or CT/MR imaging-­prospective study. Radiology 274:764–771 Look Hong NJ, Petrella T, Chan K (2015) Cost-­ effectiveness analysis of staging strategies in patients with regionally metastatic melanoma. J Surg Oncol 111:423–430 Modlin IM, Oberg K, Chung DC et al (2008) Gastroenteropancreatic neuroendocrine tumours. Lancet Oncol 9:61–72 Montravers F, Grahek D, Kerrou K et al (2006) Can fluorodihydroxyphenylalanine PET replace somatostatin receptor scintigraphy in patients with digestive endocrine tumors? J Nucl Med 47:1455–1462 Muros MA, Varsavsky M, Iglesias Rozas P et al (2009) Outcome of treating advanced neuroendocrine tumours with radiolabelled somatostatin analogues. Clin Transl Oncol 11:48–53 Naswa N, Sharma P, Kumar A et al (2012) 68Ga-DOTANOC PET/CT in patients with carcinoma of unknown primary of neuroendocrine origin. Clin Nucl Med 37:245–251 Öksüz MÖ, Winter L, Pfannenberg C et al (2014) Peptide receptor radionuclide therapy of neuroendocrine tumors with 90Y-DOTATOC: is treatment response predictable by pre-therapeutic uptake of 68 Ga-DOTATOC? Diagn Interv Imaging 95:289–300 Pasquali C, Rubello D, Sperti C et al (1998) Neuroendocrine tumor imaging: can 18F-fluorodeoxyglucose positron emission tomography detect tumors with poor prognosis and aggressive behavior? World J Surg 22:588–592 Pavel M, O'Toole D, Costa F et al (2016) ENETS consensus guidelines update for the management of distant metastatic disease of intestinal, pancreatic, bronchial neuroendocrine neoplasms (NEN) and NEN of unknown primary site. Neuroendocrinology 103:172–185 Poeppel TD, Binse I, Petersenn S et al (2011) 68 Ga-DOTATOC versus 68Ga-DOTATATE PET/CT in functional imaging of neuroendocrine tumors. J Nucl Med 52:1864–1870 Prasad V, Ambrosini V, Hommann M et al (2010) Detection of unknown primary neuroendocrine tumours (CUPNET) using 68Ga-DOTA-NOC receptor PET/CT. Eur J Nucl Med Mol Imaging 37:67–77 Rappeport ED, Hansen CP, Kjaer A et al (2006) Multidetector computed tomography and neuroendocrine pancreaticoduodenal tumors. Acta Radiol 47:248–256

689 Reubi JC, Waser B (2003) Concomitant expression of several peptide receptors in neuroendocrine tumours: molecular basis for in vivo multireceptor tumour targeting. Eur J Nucl Med Mol Imaging 30:781–793 Rindi G, Arnold R, Bosman FT et al (2010) Nomenclature and classification of neuroendocrine neoplasms of the digestive system. In: Bosman TF, Carneiro F, Hruban RH, Theise ND (eds) WHO classification of tumours of the digestive system, 4th edn. International Agency for Research on cancer (IARC), Lyon, p 13 Ruf J, Heuck F, Schiefer J et al (2010) Impact of multiphase 68Ga-DOTATOC-PET/CT on therapy management in patients with neuroendocrine tumors. Neuroendocrinology 91:101–109 Sadowski SM, Neychev V, Millo C et al (2016) Prospective study of 68Ga-DOTATATE positron emission tomography/computed tomography for detecting gastro-­ Entero-­ pancreatic neuroendocrine tumors and unknown primary sites. J Clin Oncol 34:588–596 Severi S, Nanni O, Bodei L et al (2013) Role of 18FDG PET/CT in patients treated with 177Lu-DOTATATE for advanced differentiated neuroendocrine tumours. Eur J Nucl Med Mol Imaging 40:881–888 Singh S, Asa SL, Dey C et al (2016) Diagnosis and management of gastrointestinal neuroendocrine tumors: An evidence-based Canadian consensus. Cancer Treat Rev 2016; 47: 32–45 Strosberg J, El-Haddad G, Wolin E et al (2017) Phase 3 trial of 177Lu-Dotatate for Midgut neuroendocrine tumors. N Engl J Med 376:125–135 Tessonnier L, Sebag F, Ghander C et al (2010) Limited value of 18F-F-DOPA PET to localize pancreatic insulin-­secreting tumors in adults with hyperinsulinemic hypoglycemia. J Clin Endocrinol Metab 95:303–307 Teunissen JJ, Kwekkeboom DJ, Valkema R et al (2011) Nuclear medicine techniques for the imaging and treatment of neuroendocrine tumours. Endocr Relat Cancer 18(Suppl 1):S27–S51 Wild D, Bomanji JB, Benkert P et al (2013) Comparison of 68Ga-DOTANOC and 68Ga-DOTATATE PET/CT within patients with gastroenteropancreatic neuroendocrine tumors. J Nucl Med 54:364–372 Wild D, Fani M, Behe M et al (2011) First clinical evidence that imaging with somatostatin receptor antagonists is feasible. J Nucl Med 52:1412–1417 Wulfert S, Kratochwil C, Choyke PL et al (2014) Multimodal imaging for early functional response assessment of 90 Y−/177Lu-DOTATOC peptide receptor targeted radiotherapy with DW-MRI and 68Ga-DOTATOC-PET/ CT. Mol Imaging Biol 16:586–594

Adrenals Christoph Schabel and Daniele Marin

Contents 1    Introduction

 691

2    Morphology

 693

3    CT Densitometry

 693

4    Contrast Media Kinetics

 693

5    Multi-energy CT

 694

References

 696

1

Introduction

Adrenal masses are often incidentally detected in imaging studies acquired for indications unrelated to suspected adrenal disease (Young Jr 2007). With rising numbers of imaging studies, the incidence of incidentalomas of the adrenal

C. Schabel Department of Radiology, Duke University Medical Center, Durham, NC, USA Department of Diagnostic and Interventional Radiology, University Hospital of Tuebingen, Tuebingen, Germany D. Marin (*) Department of Radiology, Duke University Medical Center, Durham, NC, USA e-mail: [email protected]

gland has increased (Kaltsas et al. 2012; Lee and Lee 2014; Elsayes et al. 2017a). Adrenal lesions can be classified as malignant or benign lesions with or without hormonal hypersecretion, which can be assessed clinically or biochemically (Young Jr 2007; Boland et al. 2008). Non-hypersecreting adrenal tumors are more common and comprise both malignant and benign lesions (Young Jr 2007; Elsayes et al. 2017a; Boland et al. 2008). Owed to the high number of incidental findings, accurate lesion characterization is mandatory (Blake et al. 2010; Korobkin 2000; Mayo-Smith et al. 2001; Slattery et al. 2006; Angeli et al. 1997) (Fig. 1). Adrenal gland lesions are seen in 5–8% of patients undergoing CT with a higher prevalence in patients of older age (Ettinghausen and Burt 1991). The majority of findings are benign, while the malignant masses are predominantly metastases of extra-adrenal cancer (Young Jr 2007; Ettinghausen and Burt 1991; Caoili et al. 2002; Hahn et al. 2006). CT has proven a very accurate modality to characterize adrenal lesions. In particular, CT can differentiate benign adenomas from non-adenomatous lesions based on the fat content, which is measurable in up to 70% of adrenal adenomas (Ettinghausen and Burt 1991; Caoili et al. 2002; Hahn et al. 2006; Boland 2010; Korobkin et al. 1996; Boland et al. 2008; Schurch et al. 1992; Lee et al. 1991). In CT imaging, mor-

Med Radiol Diagn Imaging (2017) DOI 10.1007/174_2017_171, © Springer International Publishing AG Published Online: 29 December 2017

691

C. Schabel and D. Marin

692

a

b

c

Fig. 1 A seventy-six-year-old female patient presented with painless hematuria and underwent triple phase genitourinary computed tomography. Imaging revealed an incidental finding of the right adrenal gland. The mass had

attenuations of 5 HU (a), 53 HU (b), and 22 HU (c) on the unenhanced, nephrographic, and delayed phase CT acquisitions. Calculated absolute and relative washouts were 64 and 58%, respectively, suggesting an adrenal adenoma

phology, CT densitometry, and contrast media kinetics are most relevant (Lee et al. 1991; Korobkin et al. 1998; Szolar and Kammerhuber 1998). Interpretation should always be performed in conjunction with patients’ clinical history

(potential prior malignancies) and hormonal tests (exclusion of hormone-producing tumors), together with possible prior exams (stable for at least 6 months is a reliable indicator for a benign lesion) (Boland 2010, 2011a, b) (Fig. 2).

Adrenals

693

Attenuation-7 HU

Attenuation-5 HU

Fig. 2  Dual-energy CT imaging of an adrenal mass. Virtual non-contrast (VNC) images simulate the attenuation properties and overall appearance of true non-contrast (TNC) images accurately

2

Morphology

Lesion properties that are associated with a benign adenoma are the size of less than 3 cm, uniform shape, and smooth margins (Blake et al. 2010; Boland et al. 2008; Boland 2011a, b, c).

been advocated to reduce the number of indeterminate lesions, which is based on histogram post-­ processing with a threshold of 10% (Bae et al. 2003; Remer et al. 2006; Ho et al. 2008).

4 3

CT Densitometry

CT densitometry is based on a region-of-interest (ROI) analysis. The ROI should be placed within the lesion, comprising about 50–70% of the lesion avoiding the margins to reduce confounding effects of the partial volume effect (Lee et al. 1991). Mean CT numbers of equal or less than +10 HU on unenhanced CT are highly indicative of a benign adrenal adenoma diagnosis with a sensitivity and specificity of 71 and 98%, respectively (Blake et al. 2010; Boland et al. 2008; Boland 2011c). However, CT values of greater than +10 HU on unenhanced CT images can be observed in malignant adrenal lesions making further test necessary (Blake et al. 2010; Boland 2010, 2011a, b, c; Boland et al. 2008). Histogram analysis has

Contrast Media Kinetics

For these lesions, which exceed CT numbers of +10 HU, contrast media kinetics have been established as a reliable way to differentiate adenomas from malignant lesions (Boland et al. 2008; Korobkin et al. 1998). The washout behavior is calculated as the difference between the portal venous phase (HU) and the delayed phase (HU) divided by either the venous phase, which is referred to as relative washout, or the difference of the portal venous phase (HU) and the unenhanced CT images (HU), which is referred to as the absolute washout. The percentage of washout can be obtained by multiplication by 100. Values of more than 40% or 60% in the relative and absolute washout , respectively, can be regarded as indicators for adenomas, which demonstrate no signs of capillary leak yielding very high

C. Schabel and D. Marin

694

sensitivities and specificities of 96 and 100%, respectively (Blake et al. 2010; Boland 2010, 2011a, b, c; Boland et al. 2008). The disadvantage of this technique is delayed phases, which should be obtained 10–15 min after contrast Absolute washout =

media application (Szolar and Kammerhuber 1998; Boland et al. 1997; Pena et al. 2000; Blake et al. 2006), and shorter delays may negatively impact accuracy (Sangwaiya et al. 2010). Kinetic formulas:

AttenuationPortal venous − AttenuationDelayed

phase

AttenuationPortal venous − AttenuationUnenhanced

×100

Values of >60% can be regarded as indicators for adenomas.

Relative washout =

AttenuationPortal venous − AttenuationDelayed phase AttenuationPortal venous

×100

Values of >40% can be regarded as indicators for adenomas. Further methods for differentiation include histogram analysis, which calculates the number of pixels in an attenuation range and has been found more sensitive than conventional CT densitometry (Bae et al. 2003). However, this method is limited for clinical by the availability of histogram analysis tools.

5

Multi-energy CT

Unenhanced dual-energy CT supplements CT densitometry. Microscopic fat causes lower attenuation values for the low kVp CT acquisition compared to the high kVp CT acquisition (Coursey et al. 2010; Gupta et al. 2010). However, dual-energy imaging does not improve accuracy for lipid-poor adenomas (Gupta et al. 2010). Dual-energy CT can be utilized to calculate virtual unenhanced images after contrast media application, which show a high concor-

dance with true unenhanced images (Ho et al. 2012). However, virtual unenhanced images show slightly elevated attenuation values compared to true unenhanced images, and cutoff values might have to be adjusted not to miss lipid-rich adenomas (Kim et al. 2013). This potential shortcoming can be overcome by acquiring a delayed phase while saving radiation exposure for omitting the unenhanced CT acquisition resulting in a potential radiation dose exposure of up to 30% (Kim et al. 2013) (Figs. 3 and 4). If adenomas have been ruled out, potential differentials comprise adrenal metastases, collision tumors, lymphoma, myelolipoma, cystic adrenal masses, lymphangioma, pheochromocytoma, adrenal carcinoma, adrenal hemorrhage, hemangioma, neuroblastoma, ganglioneuroma, ganglioneuroblastoma, and granulomatous diseases (Elsayes et al. 2017b; Allen and Francis 2015) (Figs. 5 and 6).

Adrenals

695

Fig. 3  Dual-energy CT imaging of a lipid-poor and lipid-rich adenoma. The lipid-rich adenoma shows a fat-fraction of 15% compared to 0% of the lipid-poor adenoma

Fig. 4  A twenty-five-year-old male patient with a history of a diffuse large B-cell non-Hodgkin lymphoma presented with an adrenal manifestation

Fig. 5  An eighty-year-old male patient, who had a pathology-proven renal cell carcinoma, presented with an adrenal metastasis. Measured attenuations are 35 HU and 158 HU of the unenhanced and arterial phase, respectively

Fig. 6  A fifty-year-old male patient, who fell 6 m onto his right side, suffered a level III trauma with diffuse chest, abdomen, and spine pain. CT imaging revealed an adrenal thickening without contrast enhancement, which was interpreted as an adrenal hemorrhage

696

C. Schabel and D. Marin

Ho LM, Paulson EK, Brady MJ, Wong TZ, Schindera ST (2008) Lipid-poor adenomas on unenhanced CT: does histogram analysis increase sensitivity compared Allen BC, Francis IR (2015) Adrenal imaging and interwith a mean attenuation threshold? Am J Roentgenol vention. Radiol Clin North Am 53(5):1021–1035 191(1):234–238 Angeli A, Osella G, Ali A, Terzolo M (1997) Adrenal Ho LM, Marin D, Neville AM et al (2012) Characterization incidentaloma: an overview of clinical and epidemioof adrenal nodules with dual-energy CT: can virtual logical data from the National Italian Study Group. unenhanced attenuation values replace true unenhanced Hormones 47(4–6):279–283 attenuation values? Am J Roentgenol 198(4):840–845 Bae KT, Fuangtharnthip P, Prasad SR, Joe BN, Heiken JP Kaltsas G, Chrisoulidou A, Piaditis G, Kassi E, Chrousos G (2003) Adrenal masses: CT characterization with his(2012) Current status and controversies in adrenal incitogram analysis method. Radiology 228(3):735–742 dentalomas. Trends Endocrinol Metab 23(12):602–609 Blake MA, Kalra MK, Sweeney AT et al (2006) Kim YK, Park BK, Kim CK, Park SY (2013) Adenoma Distinguishing benign from malignant adrenal masses: characterization: adrenal protocol with dual-energy multi–detector row CT protocol with 10-minute delay. CT. Radiology 267(1):155–163 Radiology 238(2):578–585 Korobkin M (2000) CT characterization of adrenal Blake MA, Cronin CG, Boland GW (2010) Adrenal imagmasses: the time has come. Radiology 217(3):629–632 ing. Am J Roentgenol 194(6):1450–1460 Korobkin M, Giordano TJ, Brodeur FJ et al (1996) Adrenal Boland GW (2010) Adrenal imaging: why, when, what, adenomas: relationship between histologic lipid and and how? Part 1. Why and when to image? Am CT and MR findings. Radiology 200(3):743–747 J Roentgenol 195(6):W377–WW81 Korobkin M, Brodeur FJ, Francis IR, Quint L, Dunnick N, Boland GW (2011a) Adrenal imaging: why, when, what, Londy F (1998) CT time-attenuation washout curves and how? Part 2. What technique? Am J Roentgenol of adrenal adenomas and nonadenomas. AJR Am 196(1):W1–W5 J Roentgenol 170(3):747–752 Boland GW (2011b) Adrenal imaging: why, when, what, Lee HJ, Lee J (2014) Differential diagnosis of adrenal mass and how? Part 3. The algorithmic approach to definiusing imaging modality: special emphasis on f-18 flutive characterization of the adrenal incidentaloma. Am oro-2-deoxy-d-glucose positron emission tomography/ J Roentgenol 196(2):W109–WW11 computed tomography. Endocrinol Metab 29(1):5–11 Boland GW (2011c) Adrenal imaging. Abdom Imaging Lee MJ, Hahn PF, Papanicolaou N et al (1991) Benign 36(4):472–482 and malignant adrenal masses: CT distinction with Boland GW, Hahn PF, Peña C, Mueller PR (1997) Adrenal attenuation coefficients, size, and observer analysis. masses: characterization with delayed contrast-­ Radiology 179(2):415–418 enhanced CT. Radiology 202(3):693–696 Mayo-Smith WW, Boland GW, Noto RB, Lee MJ (2001) Boland GWL, Blake MA, Hahn PF, Mayo-Smith WW State-of-the-art adrenal imaging. Radiographics (2008) Incidental adrenal lesions: principles, tech21(4):995–1012 niques, and algorithms for imaging characterization. Pena CS, Boland GW, Hahn PF, Lee MJ, Mueller PR Radiology 249(3):756–775 (2000) Characterization of indeterminate (lipid-poor) Caoili EM, Korobkin M, Francis IR et al (2002) Adrenal adrenal masses: use of washout characteristics at masses: characterization with combined unenhanced contrast-­enhanced CT. Radiology 217(3):798–802 and delayed enhanced CT. Radiology 222(3):629–633 Remer EM, Motta-Ramirez GA, Shepardson LB, Coursey CA, Nelson RC, Boll DT et al (2010) Dual-­ Hamrahian AH, Herts BR (2006) CT histogram energy multidetector CT: how does it work, what can analysis in pathologically proven adrenal masses. Am it tell us, and when can we use it in abdominopelvic J Roentgenol 187(1):191–196 imaging? Radiographics 30(4):1037–1055 Sangwaiya MJ, Boland GW, Cronin CG, Blake MA, Halpern Elsayes KM, Emad-Eldin S, Morani AC, Jensen CT EF, Hahn PF (2010) Incidental adrenal lesions: accuracy (2017a) Practical approach to adrenal imaging. Radiol of characterization with contrast-­enhanced washout multiClin North Am 55(2):279–301 detector CT—10-minute delayed imaging protocol revisElsayes KM, Emad-Eldin S, Morani AC, Jensen CT ited in a large patient cohort. Radiology 256(2):504–510 (2017b) Practical approach to adrenal imaging. Schurch W, Seemayer T, Gabbiani G, Sternberg S (1992) Radiologic. Clinics 55(2):279–301 Histology for pathologists. Raven, New York, NY, p 109 Ettinghausen SE, Burt M (1991) Prospective evaluation of Slattery JM, Blake MA, Kalra MK et al (2006) unilateral adrenal masses in patients with operable nonAdrenocortical carcinoma: contrast washout characsmall-cell lung cancer. J Clin Oncol 9(8):1462–1466 teristics on CT. Am J Roentgenol 187(1):W21–WW4 Gupta RT, Ho LM, Marin D, Boll DT, Barnhart HX, Szolar DH, Kammerhuber FH (1998) Adrenal adenomas Nelson RC (2010) Dual-energy CT for characterand nonadenomas: assessment of washout at delayed ization of adrenal nodules: initial experience. Am contrast-enhanced CT. Radiology 207(2):369–375 J Roentgenol 194(6):1479–1483 Young WF Jr (2007) The incidentally discovered adrenal Hahn PF, Blake MA, Boland GW (2006) Adrenal lesions: mass. N Engl J Med 2007(356):601–610 attenuation measurement differences between CT scanners. Radiology 240(2):458–463

References

Kidneys, Ureters, and Bladder Christoph Schabel and Daniele Marin

Contents 1      Kidneys 1.1   Anatomy 1.2   Renal Imaging 1.3   Incidental Renal Lesions 1.4   Multi-energy Imaging of Incidental Renal Lesions 1.5   Cystic Renal Masses 1.6   Solid Renal Masses 1.7   Renal Cell Carcinoma 1.8   Pseudotumor 1.9   Renal Infections 1.10  Traumatic Renal Injury

 697  697  697  698

2      Ureters

 704

3      Bladder

 706

References

 706

 698  700  701  701  702  702  703

C. Schabel Department of Radiology, Duke University Medical Center, Durham, NC, USA Department of Diagnostic and Interventional Radiology, University Hospital of Tuebingen, Tuebingen, Germany D. Marin (*) Department of Radiology, Duke University Medical Center, Durham, NC, USA e-mail: [email protected]

1

Kidneys

1.1

Anatomy

The kidneys are situated in the perirenal space, which is enclosed by the anterior (Gerota) and posterior renal fasciae and located on the posterior abdominal wall on each side of the vertebral column. The kidney’s three-dimensional orientation is parallel to the lateral border of the psoas muscle with the long axis and adjacent to the quadratus lumbar muscle. The right kidney usually lies slightly lower than the left kidney because of the right lobe of the liver. Normal kidneys measure 11 cm (10th–90th percentile, 9.8, 12.3) and 5.8 cm (10th–90th percentile, 5.1, 6.4) with decreasing size at an older age (Emamian et al. 1993). The kidney shows a bean-shaped structure with a superior and an inferior pole and a hilum, where the renal artery, renal vein, and ureter are located. The organ is covered by the renal capsule containing cortex and medulla (renal parenchyma) and the renal sinus (vasculature, lymphatic system, and collecting system).

1.2

Renal Imaging

Multidetector row CT (MDCT) has taken a rapid evolution since its introduction in the early 2000s, and its application expanded for the genitourinary system. State-of-the-art genitourinary

Med Radiol Diagn Imaging (2017) DOI 10.1007/174_2017_172, © Springer International Publishing AG Published Online: 29 December 2017

697

C. Schabel and D. Marin

698

MDCT protocols exploit the extended coverage along the z-axis and the rapid image acquisition with multiphasic contrast media protocols. Therefore, this imaging modality became well accepted for characterization of renal lesions (Kang and Chandarana 2012; Israel and Bosniak 2008; Israel and Silverman 2011). Imaging protocols consist of an unenhanced and up to three contrast-enhanced CT acquisitions during corticomedullary (approximately 40s), nephrographic (90 s), and excretory phases (8 min after contrast media administration) with the highest diagnostic value for renal lesion characterization attributed to the nephrographic phase, which should be used together with the unenhanced acquisition to calculate lesion enhancement (Kang and Chandarana 2012; Israel and Bosniak 2005, 2008; Israel and Silverman 2011; Silverman et al. 2008). Corticomedullary phases and excretory phases are mostly utilized for surgical planning and evaluation of the collecting system. For an optimal depiction of the urinary collecting system, good oral or intravenous hydration and the use of intravenous diuretics (i.e., furosemide 10–20 mg) result in an improved excretion and favorable dilution of the contrast media (Sanyal et al. 2007; Silverman et al. 2006a).

attenuation (−10 to +20 HU), imperceptible walls, and a change of attenuation between enhanced and unenhanced CT acquisition of less than 10 HU (Emamian et al. 1993; Kang and Chandarana 2012; Israel and Bosniak 2008). On the other hand, lesions exceeding 20 HU of attenuation change can safely be regarded as enhancing (Israel and Bosniak 2005). Unfortunately, a substantial number of renal lesions remain indeterminate following MDCT examinations with an attenuation change ranging from 10 to 20 HU (Kang and Chandarana 2012; Israel and Bosniak 2005, 2008; Israel and Silverman 2011; Silverman et al. 2008) confounded by a phenomenon called pseudoenhancement, a spurious increase of attenuation in renal lesion due to beam hardening (Israel and Silverman 2011; Israel and Bosniak 2005; Silverman et al. 2008; Sanyal et al. 2007). Under such situations, further diagnostics are typically indicated, either as a follow-up or additional MRI imaging to account for the clinical relevance, because renal cell carcinomas are mostly asymptomatic (61%) and in the majority of cases diagnosed as incidental findings (Suh et al. 2003) (Figs. 1 and 2).

1.4 1.3

Multi-energy Imaging of Incidental Renal Lesions

Incidental Renal Lesions

With increasing numbers of MDCT, incidental detection of vastly prevalent incidental renal lesions in the adult population increases (Israel and Silverman 2011; Silverman et al. 2008). In a first step, renal masses should be confirmed as true renal masses and from conditions mimicking renal masses, such as hypertrophied parenchyma, congenital anomalies, vascular findings, infection, infarction, hemorrhage, or trauma (Israel and Silverman 2011). The majority of true renal lesions can be conclusively diagnosed as benign cysts on contrast-enhanced MDCT relating to distinct imaging characteristics, such as homogeneous appearance, fluid

Technical advancements and the introduction of multi-energy imaging, as yet clinically available as dual-energy CT imaging, try to overcome these limitations with post-processing, such as calculation of energy-specific and material-­specific applications (Wortman et al. 2016; Graser et al. 2008, 2009, 2010; Johnson et al. 2006, 2007; Brown et al. 2009; Fletcher et al. 2009; Kaza et al. 2011; Arndt et al. 2012; Ascenti et al. 2011, 2012a, b, 2013a, b; Karlo et al. 2010; Mileto et al. 2012; Miller et al. 2011; Petersilka et al. 2008; Song et al. 2011). Both applications are based on energy-dependent attenuation of matter, which is obtained from polychromatic CT data acquired at two

Kidneys, Ureters, and Bladder

699

a

b

c

d

Fig. 1  A sixty-year-old female patient presented with painless hematuria. Imaging protocol consisted of an unenhanced, nephrographic (90 s), and excretory phases (8 min after contrast media administration) with measured attenuations of 22 HU (a), 30 HU (b), and 28 HU (c), respectively. The increase of 8 HU between unenhanced and nephrographic phase represents a phenomenon called

pseudoenhancement. Dual-energy imaging revealed an iodine uptake of less than 0.5 mg/mL (d) (Siemens), which indicated a renal cyst as well. The lesion was interpreted as a renal cyst and proven by MRI imaging. (a) Unenhanced phase CT imaging. (b) Nephrographic phase CT imaging. (c) Delayed phase CT imaging. (d) Dual-­ energy CT imaging with iodine quantification

distinct energy levels during a dual-energy CT acquisition (Fornaro et al. 2011; Silva et al. 2011; Yu et al. 2012). Currently utilized in renal imaging are energy-specific applications, which extrapolate virtual monochromatic images mimicking CT images acquired with an ideal monochromatic X-ray source within a range from 40 to 190 keV (Fornaro et al. 2011). Virtual monochromatic images have been proven to be more robust and accurate than polychromatic CT images (Fornaro et al. 2011; Silva et al. 2011; Yu et al. 2012; Michalak et al. 2016). Furthermore, studies proposed to reduce

the confounding effects of pseudoenhancement by increasing the energy level of monochromatic images to 90 keV and above and thereby reducing the need for further workups (Mileto et al. 2014). On the other hand, material-specific applications can estimate material composition in a voxel-by-voxel fashion and generate, for example, iodine-density and water-density images (Fornaro et al. 2011; Silva et al. 2011; Yu et al. 2012; Mileto et al. 2014; Apel et al. 2010; Hartman et al. 2012; Kaza et al. 2012a, b; Qu et al. 2012). These water-density or virtual unenhanced images have been demonstrated

C. Schabel and D. Marin

700

a

b

c

d

Fig. 2  An eighty-year-old female patient presented with painless hematuria. Imaging protocol consisted of an unenhanced, nephrographic (90 s), and excretory phases (8 min after contrast media administration) with measured attenuations of 43 HU (a), 158 HU (b), and 90 HU (c), respectively. Dual-energy imaging revealed an iodine

uptake of less than 5.1 mg/mL (d) (Siemens). The lesion was interpreted as a renal cell carcinoma and proven as a clear-cell renal carcinoma by pathology. (a) Unenhanced phase CT imaging. (b) Nephrographic phase CT imaging. (c) Delayed phase CT imaging. (d) Dual-energy CT imaging with iodine quantification

comparable to true unenhanced images and can be used in place of true unenhanced images (Graser et al. 2008, 2009, 2010; Fletcher et al. 2009; Kaza et al. 2011; Arndt et al. 2012; Ascenti et al. 2011, 2012b, 2013a; Johnson et al. 2006; Mileto et al. 2012; Chandarana et al. 2011). Iodine-density images allow for the visualization and the measurement of the iodine within a lesion, with iodine concentrations above 0.5 mg/mL (iodine quantification with Siemens dual-energy CT acquisition) highly suspicious for lesion iodine uptake (Graser et al. 2008, 2009; Chandarana et al. 2011). Thereby, dual-energy CT was reported to have the potential to reduce the radiation dose by up to 50% (no unenhanced CT acquisi-

tion) and the need for follow-up examinations (Graser et al. 2008, 2009, 2010; Kaza et al. 2011; Ascenti et al. 2012b, 2013a; Chandarana et al. 2011).

1.5

Cystic Renal Masses

Cysts are the most frequent renal masses and have been successfully classified by the Bosniak renal cyst classification system, which is based on CT findings and organized into five groups (Bosniak 1986): categories I, II, IIF, III, and IV. Category I masses represents simple cysts and contains the majority of all cysts (Tada et al. 1983) and can safely be regarded as benign.

Kidneys, Ureters, and Bladder

Imaging features include fluid attenuations between 0 and 20 HU, no septa or calcifications, and a hairline-thin wall (Israel and Silverman 2011). Renal cysts show no evidence for soft tissue or enhancement after contrast media application (Bosniak 1986). Category II cysts are minimally complicated masses, which are considered benign. Imaging features contain hairline-thin septa that can mimic enhancement (without being measurable), fine calcification of the septa and walls, or segments of smooth thickened calcification (Israel and Silverman 2011; Bosniak 1986). Small (95% of cases). However, the absence of hematuria does not

exclude substantial renal injuries sting to renal artery thrombosis and injuries of the ureteropelvic junction (Kawashima et al. 2001). MDCT is the most suitable method to differentiate severe from trivial injuries, while a multiphasic CT imaging protocol is recommended if patient conditions are stable (Sandler et al. 2000). Renal injuries can be classified based on the renal organ injury system established by the American Association for the Surgery of Trauma in five categories: category 1, subcapsular hematoma and/or contusion of the parenchyma without injury to the collecting system; category 2, laceration 1 cm in depth and into cortex with a small contained hematoma within Gerota’s fascia but without injury to the collecting system; category 4, laceration through the parenchyma into the urinary collecting system, vascular segmental vein, or artery injury, laceration into the collecting system with urinary extravasation, or renal pelvis laceration and/or complete urethral pelvic disruption; and category 5, main renal artery or vein laceration/avulsion or renal artery/vein thrombosis (Kawashima et al. 2001; Buckley and McAninch 2011) (Fig. 4).

C. Schabel and D. Marin

704 Fig. 4  A sixty-three-­ year-old patient with a history of protracted flank pains presented with symptoms of an infection and hematuria. CT imaging shows an obstructed kidney with a focal pyelo-abscess

2

Ureters

The ureter starts at the ureteropelvic junction and drains the urine into the bladder at the ureterovesical junction. It is located entirely retroperitoneal with an average length of 25 cm and a diameter of 90%) to evaluate patients with painless hematuria (Sadow et al. 2010). Obstruction of the ureter may warrant for additional delayed phase imaging for improved visualization (Potenta et al. 2015; Raman and Fishman 2017). Positive experiences were reported using split bolus techniques to reduce radiation exposure (Potenta et al. 2015). Normal variants can be an origin from an extrarenal renal pelvis and a partial or complete ureteral duplication, which is a common congenital variant (Potenta et al. 2015). A complete ureteral duplication is associated with

Kidneys, Ureters, and Bladder

ureteroceles or an ectopic ureter insertion on the bladder and can result in incontinence in girls and recurring urinary tract infections in boys and girls due to the missing valve mechanism of the ureterovesical junction (Berrocal et al. 2002). Further congenital pathologies refer to smooth muscle dysfunction with the development of a proximal dilatation of the renal pelvis and calyces or ureteral dilatation of 7 mm and more (megaureter) as a result of a proximal or distal dysfunction, respectively (Potenta et al. 2015). Urolithiasis is the most common ureteral pathology (Curhan 2007). Calculi build up in the distal nephron nearby the renal papilla as the junction of renal medulla and minor calyx (Evan 2010). Etiologically, renal stones can be subdivided by their compositions with calcium, uric acid, or struvite being the most common base material (Kambadakone et al. 2010). Dual-energy CT is the only imaging modality that can distinguish different base materials, as composition, size, and location have implications for therapy and outcome (Kaza et al. 2012b; Masch et al. 2016). Unenhanced CT is the gold standard for diagnosis of renal stones, which may exhibit attenuation values of more than 200 HU and can be accompanied by secondary signs of obstruction when dislocated into the ureter (Masch et al. 2016; Smith et al. 1996; Dhar and Denstedt

a

705

2009). Low-dose CT has been utilized successfully and is recommended for imaging of young patients (Masch et al. 2016). Calculi of more than 5 mm diameter are at risk not to pass locations of regular anatomic ureteral narrowing making urologic interventions mandatory (Potenta et al. 2015; Raman and Fishman 2017). Malignancies of the ureter are difficult to assess by CT urography at an early stage without urinary tract obstruction because the distal portion of the ureter cannot be distended adequately, which is the location for most transitional cell carcinoma (Raman and Fishman 2017). However, paying close attention to unilateral urothelial thickening with short segmental constrictions and subtle levels of enhancement during arterial and nephrographic phase can help to identify malignancies of the ureter and distinguish them from urinary tract infections (Raman and Fishman 2017). For the intrarenal collecting system, the diagnostic value of excretory phase and nephrographic phase are interchanged, and transitional cell carcinomas are best depicted by focal destruction and amputation of calyces (Kawamoto et al. 2008; Caoili et al. 2005; Urban et al. 1997a, b), although the specificity is lower compared to the bladder because of confounding processes like blood clots, fungus, infections like tuberculosis, or papillary necrosis (Raman and Fishman 2017) (Figs. 5 and 6).

b

Fig. 5  A fifty-five-year-old patient presented with painless hematuria. CT imaging revealed a transitional cell carcinoma of the right collection system (a), which reached into the right proximal ureter (b)

C. Schabel and D. Marin

706

Fig. 6  A seventy-year-old patient with known non-small cell lung cancer showed bilateral renal metastases

3

Bladder

The best diagnostic modality for the bladder is the cystoscopy. However, CT urography shows a good diagnostic performance to depict malignancies of the bladder, if performed properly. Therefore, caution is warranted particularly in patients not routinely undergoing cystoscopy (Raman and Fishman 2017). For optimal diagnostic performance, the bladder should be distended, and imaging should be performed during the excretory phase and arterial or corticomedullary phase to detect subtle enhancement of hypervascular tumors (Raman and Fishman 2014). Tumor enhancement might be concealed by high attenuation of urine during the excretory phase, particularly in the dorsal circumference of the bladder with imaging in supine position (Raman and Fishman 2014, 2017). Bladder cancer is the most common malignancy of the urinary tract with transitional cell carcinomas being the most prevalent etiology (Lee et al. 2011). Transitional cell carcinomas tend to recur and being multifocal (Vikram et al. 2009a, b). CT’s sensitivity and specificity to detect malignancies of the bladder range from 79 to 93% and 94 to 99%, respectively, when proper protocols are utilized (Sadow et al. 2008; Turney et al. 2006; Blick et al. 2012; Knox et al. 2008; Kim et al. 2004). However, nonneoplastic disorders should be considered for interpretation with idiopathic, infectious, and inflammatory etiolo-

gies, particularly in the presence of signs of infections. Differential diagnosis comprises inflammatory pseudotumor, endometriosis, nephrogenic adenoma, or malacoplakia (Wong-You– Cheong et al. 2006). Cystitis may present with a diffuse wall thickening of the bladder relating to a variety of nonneoplastic conditions such as bacterial or adenovirus infections, exposure to chemotherapy or irradiation, as well as rare causes like tuberculosis, schistosomiasis, or eosinophilic cystitis (Wong-You–Cheong et al. 2006). In general, cystitis may be differentiated from a pyelonephritis by patients’ clinical presentation with upper urinary tract infections causing stronger symptoms of infection and higher elevated infection ­parameters such as CRP (Johansen 2002). A severe progressive form of a cystitis is emphysematous cystitis with gas-generating microorganisms proliferating within the bladder wall.

References Apel A, Fletcher JG, Fidler JL et al (2010) Pilot multi-­ reader study demonstrating potential for dose reduction in dual energy hepatic CT using non-linear blending of mixed kV image datasets. Eur Radiol 21(3):644–652 Arndt N, Staehler M, Siegert S, Reiser MF, Graser A (2012) Dual energy CT in patients with polycystic kidney disease. Eur Radiol 22(10):2125–2129 Ascenti G, Mazziotti S, Lamberto S et al (2011) Dual-­ energy CT for detection of endoleaks after endovascular abdominal aneurysm repair: usefulness of colored iodine overlay. Am J Roentgenol 196(6):1408–1414 Ascenti G, Krauss B, Mazziotti S et al (2012a) Dual-­ energy computed tomography (DECT) in renal masses. Acad Radiol 19(10):1186–1193 Ascenti G, Mazziotti S, Mileto A et al (2012b) Dual-­ source dual-energy CT evaluation of complex cystic renal masses. Am J Roentgenol 199(5):1026–1034 Ascenti G, Mileto A, Gaeta M, Blandino A, Mazziotti S, Scribano E (2013a) Single-phase dual-energy CT urography in the evaluation of haematuria. Clin Radiol 68(2):e87–e94 Ascenti G, Mileto A, Krauss B et al (2013b) Distinguishing enhancing from nonenhancing renal masses with dual-­ source dual-energy CT: iodine quantification versus standard enhancement measurements. Eur Radiol 23(8):2288–2295 Bechtold RE, Zagoria RJ (1997) Imaging approach to staging of renal cell carcinoma. Urol Clin N Am 24(3):507–522 Berrocal T, López-Pereira P, Arjonilla A, Gutiérrez J (2002) Anomalies of the distal ureter, bladder, and

Kidneys, Ureters, and Bladder urethra in children: embryologic, radiologic, and pathologic features. Radiographics 22(5):1139–1164 Blick CG, Nazir SA, Mallett S et al (2012) Evaluation of diagnostic strategies for bladder cancer using computed tomography (CT) urography, flexible cystoscopy and voided urine cytology: results for 778 patients from a hospital haematuria clinic. BJU Int 110(1):84–94 Bosniak MA (1986) The current radiological approach to renal cysts. Radiology 158(1):1–10 Bracken RB, Chica G, Johnson DE, Luna M (1979) Secondary renal neoplasms: an autopsy study. South Med J 72(7):806–807 Brown C, Hartman RP, Dzyubak O et al (2009) Dual-­ energy CT iodine overlay technique for characterization of renal masses as cyst or solid: a phantom feasibility study. Eur Radiol 19(5):1289 Browne R, Zwirewich C, Torreggiani W (2004) Imaging of urinary tract infection in the adult. Eur Radiol Suppl 14(3):E168–E183 Buckley JC, McAninch JW (2011) Revision of current American Association for the Surgery of Trauma Renal Injury grading system. J Trauma Acute Care Surg 70(1):35–37 Caoili EM, Cohan RH, Inampudi P et al (2005) MDCT urography of upper tract urothelial neoplasms. AJR Am J Roentgenol 184(6):1873–1881 Capitanio U, Montorsi F (2016) Renal cancer. Lancet 387(10021):894–906 Chandarana H, Megibow AJ, Cohen BA et al (2011) Iodine quantification with dual-energy CT: phantom study and preliminary experience with renal masses. Am J Roentgenol 196(6):W693–W700 Charig M (1995) Delayed CT findings in acute renal infection. Clin Radiol 50(12):880 Chow W-H, Dong LM, Devesa SS (2010) Epidemiology and risk factors for kidney cancer. Nat Rev Urol 7(5):245–257 Craig WD, Wagner BJ, Travis MD (2008) Pyelonephritis: radiologic-pathologic review. Radiographics 28(1):255–276 Curhan GC (2007) Epidemiology of stone disease. Urol Clin North Am 34(3):287–293 Curry NS, Cochran ST, Bissada NK (2000) Cystic renal masses: accurate Bosniak classification requires adequate renal CT. Am J Roentgenol 175(2):339–342 Dhar M, Denstedt JD (2009) Imaging in diagnosis, treatment, and follow-up of stone patients. Adv Chronic Kidney Dis 16(1):39–47 Emamian SA, Nielsen MB, Pedersen JF, Ytte L (1993) Kidney dimensions at sonography: correlation with age, sex, and habitus in 665 adult volunteers. AJR Am J Roentgenol 160(1):83–86 Evan AP (2010) Physiopathology and etiology of stone formation in the kidney and the urinary tract. Pediatr Nephrol 25(5):831–841 Fletcher JG, Takahashi N, Hartman R et al (2009) Dual-­ energy and dual-source CT: is there a role in the abdomen and pelvis? Radiol Clin North Am 47(1):41–57 Fornaro J, Leschka S, Hibbeln D et al (2011) Dual- and multi-energy CT: approach to functional imaging. Insights Imaging 2(2):149–159

707 Frank I, Blute ML, Cheville JC, Lohse CM, Weaver AL, Zincke H (2003) Solid renal tumors: an analysis of pathological features related to tumor size. J Urol 170(6):2217–2220 Goldman S, Fishman E (1991) Upper urinary tract infection: the current role of CT, ultrasound, and MRI. Semin Ultrasound CT MR 12(4):335–360 Graser A, Johnson TRC, Chandarana H, Macari M (2008) Dual energy CT: preliminary observations and potential clinical applications in the abdomen. Eur Radiol 19(1):13–23 Graser A, Johnson TR, Hecht EM et al (2009) Dual-energy CT in patients suspected of having renal masses: can virtual nonenhanced images replace true nonenhanced images? Radiology 252(2):433–440 Graser A, Becker CR, Staehler M et al (2010) Single-­ phase dual-energy CT allows for characterization of renal masses as benign or malignant. Invest Radiol 45(7):399–405 Hartman R, Kawashima A, Takahashi N et al (2012) Applications of dual-energy CT in urologic imaging: an update. Radiol Clin North Am 50(2):191–205 Helenon O, Merran S, Paraf F et al (1997) Unusual fat-containing tumors of the kidney: a diagnostic dilemma. Radiographics 17(1):129–144 Israel GM, Bosniak MA (2003a) Follow-up CT of moderately complex cystic lesions of the kidney (Bosniak category IIF). Am J Roentgenol 181(3):627–633 Israel GM, Bosniak MA (2003b) Renal imaging for diagnosis and staging of renal cell carcinoma. Urol Clin 30(3):499–514 Israel GM, Bosniak MA (2005) How I do it: evaluating renal masses. Radiology 236(2):441–450 Israel GM, Bosniak MA (2008) Pitfalls in renal mass evaluation and how to avoid them. Radiographics 28(5):1325–1338 Israel GM, Silverman SG (2011) The incidental renal mass. Radiol Clin North Am 49(2):369–383 Johansen TEB (2002) Diagnosis and imaging in urinary tract infections. Curr Opin Urol 12(1):39–43 Johnson TRC, Krauß B, Sedlmair M et al (2006) Material differentiation by dual energy CT: initial experience. Eur Radiol 17(6):1510–1517 Johnson TR, Krauss B, Sedlmair M et al (2007) Material differentiation by dual energy CT: initial experience. Eur Radiol 17(6):1510–1517 Kambadakone AR, Eisner BH, Catalano OA, Sahani DV (2010) New and evolving concepts in the imaging and management of urolithiasis: urologists’ perspective. Radiographics 30(3):603–623 Kang SK, Chandarana H (2012) Contemporary imaging of the renal mass. Urol Clin North Am 39(2):161–170, vi Karlo C, Lauber A, Götti RP et al (2010) Dual-energy CT with tin filter technology for the discrimination of renal lesion proxies containing blood, protein, and contrast-agent. An experimental phantom study. Eur Radiol 21(2):385–392 Kawamoto S, Horton KM, Fishman EK (2008) Transitional cell neoplasm of the upper urinary tract: evaluation with MDCT. Am J Roentgenol 191(2):416–422

708 Kawashima A, Sandler C, Goldman S (2000) Imaging in acute renal infection. BJU Int 86(s1):70–79 Kawashima A, Sandler CM, Corl FM et al (2001) Imaging of renal trauma: a comprehensive review. Radiographics 21(3):557–574 Kaza RK, Caoili EM, Cohan RH, Platt JF (2011) Distinguishing enhancing from nonenhancing renal lesions with fast Kilovoltage-switching dual-energy CT. Am J Roentgenol 197(6):1375–1381 Kaza RK, Platt JF, Cohan RH, Caoili EM, Al-Hawary MM, Wasnik A (2012a) Dual-energy CT with s­ ingleand dual-source scanners: current applications in evaluating the genitourinary tract. Radiographics 32(2):353–369 Kaza RK, Platt JF, Megibow AJ (2012b) Dual-energy CT of the urinary tract. Abdom Imaging 38(1):167–179 Kenney PJ (1990) Imaging of chronic renal infections. AJR Am J Roentgenol 155(3):485–494 Kim JK, Park SY, Ahn HJ, Kim CS, Cho KS (2004) Bladder cancer: analysis of multi-detector row helical CT enhancement pattern and accuracy in tumor detection and perivesical staging. Radiology 231(3):725–731 Knox M, Cowan N, Rivers-Bowerman M, Turney B (2008) Evaluation of multidetector computed tomography urography and ultrasonography for diagnosing bladder cancer. Clin Radiol 63(12):1317–1325 Kopka L, Fischer U, Zoeller G, Schmidt C, Ringert R, Grabbe E (1997) Dual-phase helical CT of the kidney: value of the corticomedullary and nephrographic phase for evaluation of renal lesions and preoperative staging of renal cell carcinoma. AJR Am J Roentgenol 169(6):1573–1578 Lee J, McClennan BL, Melson GL, Stanley RJ (1980) Acute focal bacterial nephritis: emphasis on gray scale sonography and computed tomography. Am J Roentgenol 135(1):87–92 Lee EK, Dickstein RJ, Kamat AM (2011) Imaging of urothelial cancers: what the urologist needs to know. Am J Roentgenol 196(6):1249–1254 Ljungberg B, Bensalah K, Canfield S et al (2015) EAU guidelines on renal cell carcinoma: 2014 update. Eur Urol 67(5):913–924 Lye CM, Fasano L, Woolf AS (2010) Ureter myogenesis: putting Teashirt into context. J Am Soc Nephrol 21(1):24–30 Masch WR, Cronin KC, Sahani DV, Kambadakone A (2016) Imaging in Urolithiasis. Radiol Clin North Am 55(2):209–224 Michalak G, Grimes J, Fletcher J et al (2016) Improved CT number stability across patient size using dual-­ energy CT virtual monoenergetic imaging. Med Phys 43(1):513–517 Mileto A, Mazziotti S, Gaeta M et al (2012) Pancreatic dual-source dual-energy CT: is it time to discard unenhanced imaging? Clin Radiol 67(4):334–339 Mileto A, Nelson RC, Samei E et al (2014) Impact of dual-energy multi–detector row CT with virtual monochromatic imaging on renal cyst pseudoenhancement: in vitro and in vivo study. Radiology 272(3):767–776

C. Schabel and D. Marin Miller CM, Gupta RT, Paulson EK et al (2011) Effect of organ enhancement and habitus on estimation of unenhanced attenuation at contrast-enhanced dual-­ energy MDCT: concepts for individualized and organ-­ specific spectral iodine subtraction strategies. Am J Roentgenol 196(5):W558–W564 Mitnick JS, Bosniak MA, Rothberg M, Megibow AJ, Raghavendra BN, Subramanyam BR (1984) Metastatic neoplasm to the kidney studied by computed tomography and sonography. J Comput Assist Tomogr 9(1):43–49 Petersilka M, Bruder H, Krauss B, Stierstorfer K, Flohr TG (2008) Technical principles of dual source CT. Eur J Radiol 68(3):362–368 Potenta SE, D’Agostino R, Sternberg KM, Tatsumi K, Perusse K (2015) CT urography for evaluation of the ureter. Radiographics 35(3):709–726 Qu M, Jaramillo-Alvarez G, Ramirez-Giraldo JC et al (2012) Urinary stone differentiation in patients with large body size using dual-energy dual-source computed tomography. Eur Radiol 23(5):1408–1414 Raman SP, Fishman EK (2014) Bladder malignancies on CT: the underrated role of CT in diagnosis. Am J Roentgenol 203(2):347–354 Raman SP, Fishman EK (2017) Upper and lower tract urothelial imaging using computed tomography urography. Radiol Clin North Am 55(2):225–241 Rigsby CM, Rosenfield AT, Glickman M, Hodson J (1986) Hemorrhagic focal bacterial nephritis: findings on gray-scale sonography and CT. Am J Roentgenol 146(6):1173–1177 Sadow CA, Silverman SG, O'Leary MP, Signorovitch JE (2008) Bladder cancer detection with CT urography in an Academic Medical Center. Radiology 249(1):195–202 Sadow CA, Wheeler SC, Kim J, Ohno-Machado L, Silverman SG (2010) Positive predictive value of CT urography in the evaluation of upper tract urothelial cancer. Am J Roentgenol 195(5):W337–W343 Sandler C, Amis E, Bigongiari L et al (2000) Diagnostic approach to renal trauma. American College of Radiology. ACR appropriateness criteria. Radiology 215:727–731 Sanyal R, Deshmukh A, Sheorain VS, Taori K (2007) CT urography: a comparison of strategies for upper urinary tract opacification. Eur Radiol 17(5):1262–1266 Sheth S, Scatarige JC, Horton KM, Corl FM, Fishman EK (2001) Current concepts in the diagnosis and management of renal cell carcinoma: role of multidetector CT and three-dimensional CT. Radiographics 21(Suppl_1):S237–S254 Siegel CL, McFarland EG, Brink JA, Fisher AJ, Humphrey P, Heiken JP (1997) CT of cystic renal masses: analysis of diagnostic performance and interobserver variation. AJR Am J Roentgenol 169(3):813–818 Silva AC, Morse BG, Hara AK, Paden RG, Hongo N, Pavlicek W (2011) Dual-energy (spectral) CT: applications in abdominal imaging. Radiographics 31(4):1031–1046 Silverman SG, Akbar SA, Mortele KJ, Tuncali K, Bhagwat JG, Seifter JL (2006a) Multi–detector row

Kidneys, Ureters, and Bladder CT urography of normal urinary collecting system: furosemide versus saline as adjunct to contrast medium. Radiology 240(3):749–755 Silverman SG, Gan YU, Mortele KJ, Tuncali K, Cibas ES (2006b) Renal masses in the adult patient: the role of percutaneous biopsy. Radiology 240(1):6–22 Silverman SG, Israel GM, Herts BR, Richie JP (2008) Management of the incidental renal mass. Radiology 249(1):16–31 Smith R, Verga M, McCarthy S, Rosenfield A (1996) Diagnosis of acute flank pain: value of unenhanced helical CT. AJR Am J Roentgenol 166(1):97–101 Song KD, Kim CK, Park BK, Kim B (2011) Utility of iodine overlay technique and virtual unenhanced images for the characterization of renal masses by dual-­ energy CT. Am J Roentgenol 197(6):W1076–W1082 Stokland E, Hellstrom M, Jakobsson B, Sixt R (1999) Imaging of renal scarring. Acta Paediatr 88(s431):13–21 Stunell H, Buckley O, Feeney J, Geoghegan T, Browne R, Torreggiani W (2007) Imaging of acute pyelonephritis in the adult. Eur Radiol 17(7):1820–1828 Suh M, Coakley FV, Qayyum A, Yeh BM, Breiman RS, Lu Y (2003) Distinction of renal cell carcinomas from high-attenuation renal cysts at portal venous phase contrast-enhanced CT. Radiology 228(2):330–334 Tada S, Yamagishi J, Kobayashi H, Hata Y, Kobari T (1983) The incidence of simple renal cyst by computed tomography. Clin Radiol 34(4):437–439 Talner LB, Davidson A, Lebowitz R, Dalla Palma L, Goldman S (1994) Acute pyelonephritis: can we agree on terminology? Radiology 192(2):297–305 Thrasher J, Paulson D (1993) Prognostic factors in renal cancer. Urol Clin North Am 20(2):247–262 Tsugaya M, Hirao N, Sakagami H et al (1990) Computerized tomography in acute pyelonephritis: the clinical correlations. J Urol 144(3):611–613

709 Turney BW, Willatt JM, Nixon D, Crew JP, Cowan NC (2006) Computed tomography urography for diagnosing bladder cancer. BJU Int 98(2):345–348 Urban BA, Buckley J, Soyer P, Scherrer A, Fishman EK (1997a) CT appearance of transitional cell carcinoma of the renal pelvis: part 2. Advanced-stage disease. AJR Am J Roentgenol 169(1):163–168 Urban BA, Buckley J, Soyer P, Scherrer A, Fishman EK (1997b) CT appearance of transitional cell carcinoma of the renal pelvis: part 1. Early-stage disease. AJR Am J Roentgenol 169(1):157–161 Vikram R, Sandler CM, Ng CS (2009a) Imaging and staging of transitional cell carcinoma: part 1, lower urinary tract. Am J Roentgenol 192(6):1481–1487 Vikram R, Sandler CM, Ng CS (2009b) Imaging and staging of transitional cell carcinoma: part 2, upper urinary tract. Am J Roentgenol 192(6):1488–1493 Volpe A, Finelli A, Gill IS et al (2012) Rationale for percutaneous biopsy and histologic characterisation of renal tumours. Eur Urol 62(3):491–504 Wang L-J, Wong Y-C, Chen C-J, Lim K-E (1997) Pictorial essay. CT features of genitourinary tuberculosis. J Comput Assist Tomogr 21(2):254–258 Wong-You–Cheong JJ, Woodward PJ, Manning MA, Davis CJ (2006) Inflammatory and nonneoplastic bladder masses: radiologic-pathologic correlation. Radiographics 26(6):1847–1868 Wortman JR, Bunch PM, Fulwadhva UP, Bonci GA, Sodickson AD (2016) Dual-energy CT of incidental findings in the abdomen: can we reduce the need for follow-up imaging? Am J Roentgenol 207(4):W58–W68 Yu L, Leng S, McCollough CH (2012) Dual-energy CT– based monochromatic imaging. Am J Roentgenol 199(5_Suppl):S9–S15

Part V Cardiovascular

Technical Innovations and Concepts in Coronary CT Nils Vogler, Mathias Meyer and Thomas Henzler

Contents

Abstract

1    Introduction

 713

2    Coronary CT Angiography: Current Status 2.1  Technical Principle 2.2  Acquisition Techniques

 714  714  714

3    Tube Voltage 3.1  Technical Background 3.2  Impact on Image Luminal Contrast 3.3  Impact on Radiation Exposure

 717  717  717  718

4    Iterative Reconstruction 4.1  Technical Principles 4.2  Technical Evolution 4.3  Impact on Image Appearance/Quality 4.4  Impact on Radiation Exposure 4.5  Limitations

 718  718  719  719  719  719

5    New Concepts in Cardiac CT 5.1  Assessment of Cardiac Function 5.2  Techniques for the Evaluation of Hemodynamic Significance

 720  720

References

 725

By virtue of increased clinical demand, coronary CT angiography (cCTA) has experienced a rapid evolution to overcome the diagnostic challenges in cCTA. This has led to advancements in CT scanner technology, image acquisition techniques, data reconstruction as well as image post-­processing. Furthermore, the potential gain of additional information out of cCTA data such as cardiac function parameters led to the development of sophisticated solutions to fulfill these needs. However, although image quality has increased significantly over the last couple of years, the positive predictive value for the prediction of hemodynamic significant stenosis in cCTA remains rather low. To overcome this limitation, innovative approaches like cardiac perfusion, transluminal attenuation gradient, and CT fractional flow reserve have been developed. This chapter will provide an overview on the most crucial innovations in cCTA, explain their technical background, demonstrate their current status, and point out their major limitations.

 721

N. Vogler, MD • M. Meyer, MD • T. Henzler, MD (*) Institute of Clinical Radiology and Nuclear Medicine, University Medical Center Mannheim, Medical Faculty Mannheim, Heidelberg University, Theodor-Kutzer-Ufer 1-3, 68167 Mannheim, Germany e-mail: [email protected]

1

Introduction

Coronary CT angiography (cCTA) is one of the technical most challenging fields in computed tomography (CT). Over the last decade, the increasing clinical demand to provide diagnostic

Med Radiol Diagn Imaging (2016) DOI 10.1007/174_2016_95, © Springer International Publishing Switzerland Published Online: 17 December 2016

713

N. Vogler et al.

714

cCTA in general patient population created a pressure on CT manufacturers and radiologists to overcome the diagnostic challenges in cCTA. This has led to advancements in CT scanner technology, data reconstruction as well as image post-­ processing. However, although image quality has increased significantly, the remaining challenge lies within the low positive predictive value for hemodynamic significant stenosis in cCTA. In the last decade, vendors and radiologist have tried to improve the detection of hemodynamic significant stenosis and gaining additional information out of cCTA datasets.

2

 oronary CT Angiography: C Current Status

2.1

Technical Principle

The basic principle of cCTA is acquiring a motion-free dataset of the coronary arteries with sufficient intraluminal contrast. This is achieved by co-registration of the patient’s electrocardiogram (ECG) while obtaining image data. The prerequisites to perform diagnostic cCTA differ depending on the dedicated protocol used in the acquisition process. The most crucial prerequisites are sufficient venous access, controllable heart rate and rhythm, sufficient blood pressure, and breath-hold ability (Abbara et al. 2009). Additional functional information may be acquired during the imaging process depending on the protocol used for acquisition (Meyer and Henzler 2016). The specific CT system at disposal is of crucial importance with respect to the acquisition technique. In general, there are two different types of CT scanners, single-source CT (SCCT) and dual-source CT (DSCT). In SCCT the gantry consists of one tube/detector couple with consecutive need of a 180° rotation for image reconstruction of the respective volume. SCCT systems with extremely wide detectors, which warrant extensive z-axis coverage in one rotation (e.g., 16 cm, 256 rows and above), are so-called volume scanners. In contrast, DSCT with two tube/detector couples only require a 90° gantry rotation resulting in superior temporal resolution of up to 66 ms. This high temporal

resolution is beneficial concerning image quality, especially in patients with elevated heart rates (Meyer and Henzler 2016; Fujimura et al. 2014). Furthermore, DSCT is superior in terms of radiation exposure in prospective ECG-triggered low-­ dose cCTA protocols (Yang et al. 2014). The different possibilities of image acquisition in regard to the different CT system at disposal will be further addressed in the following sections.

2.2

Acquisition Techniques

Retrospective ECG Gating In the early days of retrospective ECG gating, the tube current was continuously applied during the whole cardiac cycle over multiple cardiac cycles leading to a high radiation dose. Since identical anatomic areas are scanned at different phases of the cardiac cycle, the pitch of retrospective ECG-­gated cCTA is traditionally lower when compared to nonECG-gated CTA acquisitions (usually around 0.2). Algorithms then sort data from different phases of the cardiac cycles by progressively shifting the temporal window relative to the r-wave of the simultaneous acquired ECG dataset. This enables reconstruction of cCTA datasets at any time point of the cardiac cycle depending on the reconstruction interval chosen (1%, 5%, and 10%) (Desjardins and Kazerooni 2004). Early protocols which applied 100% tube current during the whole cardiac cycle and without other modern measures of dose reduction such as iterative reconstruction (IR) resulted in radiation exposure up to >30 mSv. Newer protocols apply the full tube current within the 40%–80% R-R interval of the cardiac cycle and reduce tube current by up to 70%–80% within the other phases (Fig. 1) (Abbara et al. 2009). This so-called ECGgated tube current modulation results in dose reduction of up to 50% (Abbara et al. 2009; Meyer and Henzler 2016). However, the ventricular function calculation may be less accurate in such a protocol, and in some patients with arrhythmia, this may lead to nondiagnostic scans within the phases in which the tube ­current is downregulated. Retrospective ECG gating is accompanied by some inhered advantages. Even with the most advanced CT systems, retrospective ECG gating is superior to prospectively triggered cCTA in terms of robustness,

Technical Innovations and Concepts in Coronary CT Retrospective ECG gating without ECG dependent tube current modulation

715

Prospective ECG triggering without ECG dependent tube current pulsing

High-pitch spiral acquisition

b

a Retrospective ECG gating with ECG dependent tube current modulation

c

Prospective ECG triggering with ECG dependent tube current pulsing

e

d Heart rate

Radiation dose

Fig. 1  Technical principles in different coronary CT angiography acquisition protocols. The light red bars represent the time and intensity of the applied tube current during the different phases in the cardiac cycle. Retrospective ECG gating without ECG dependent tube current modulation (a). Prospective ECG triggering without ECG dependent tube current pulsing (b). Retrospective ECG gating with ECG dependent tube current modulation (c). Prospective ECG triggering with ECG dependent tube current pulsing (d). High-pitch spiral acquisition (e)

especially in patients with heart rates >80 bpm and consecutive motion artifacts (Meyer and Henzler 2016). Even though radiation exposure has decreased over the years, retrospective gated cCTA using established dose reduction techniques is associated with a significant radiation burden and a common radiation exposure around 9–12 mSv (Menke et al. 2013; Hausleiter et al. 2009). Prospective ECG Triggering (“Step and Shot”) In prospective ECG triggering, the tube voltage and tube current are applied during a prespecified period of the cardiac cycle (Fig. 1) (Desjardins and Kazerooni 2004). To accomplish cCTA, this may be restricted to phases where few motion is to be expected, usually end-systole and/or end-­diastole. However, in contrast to retrospective ECG gating, this restricts the possibility of reconstruction to the prespecified time frame and thus increases vulnerability to motion, even though diagnostic image quality is usually feasible (Abbara et al. 2009; Meyer and Henzler 2016). Depending on the detector width and z-axis coverage, the heart is scanned in several consecutive “steps” at the same time point in cardiac cycle (incremental acquisition technique). In order to cover the whole heart, the table is moved in between the particular “steps.” Depending on the CT system and the patients’ anatomy, usually

two to five steps are required for sufficient z-axis coverage (Meyer and Henzler 2016). Another drawback of prospective triggering was that initially, due to the restriction of tube current to end-systole and/or end-diastole, no additional functional information could be acquired. Modern CT systems have overcome this limitation by applying reduced tube current (e.g., 20%) during a wider time frame in cardiac cycle. This technique is called “ECG pulsing.” Although being similar to retrospective tube current modulation, pulsing allows less radiation exposure since differently to retrospective ECG gating the tube may be shut down in between heartbeat phases, whereas retrospective ECG gating demands tube current over 100% of the whole cardiac cycle, even if partly performed with reduced tube current (Fig. 1) (Takx et al. 2012). Although being more vulnerable to motion artifacts and most sorts of arrhythmia, prospective ECG triggering has advantages in patients with continuous extrasystoles (e.g., bigeminy). By virtue of ­increasingly sophisticated recognition algorithms, modern CT systems may detect extrasystoles with high accuracy and postpone application of tube current to the next regular heartbeat. This results in avoidance of co-­ registration artifacts, which would be hard or impossible to rule out in the reconstruction process in retrospective ECG gating. However, for the

N. Vogler et al.

716

a

c

b

d

Fig. 2  Prospectively triggered high-pitch spiral cCTA in a 34-year-old male with intermediate cardiac risk profile and atypical chest pain. DLP for cCTA 32.1 mGycm, pitch 3.2, table movement 70 cm/s, 70 kV tube ­voltage.

same reason, the acquisition time is usually longer with consecutive need for a longer contrast bolus and potentially more contrast agent (Kondo et al. 2014). Radiation exposure in p­ rospectively triggered cCTA is lower than in retrospective gated cCTA, ranging from 1 to 4 mSv (Menke et al. 2013; Hausleiter et al. 2012).

Performed with 2 × 192 slice third-generation DSCT. MPR MIP projections of the LAD (a) and RCA (b). True axial image of the LAD (c). 3D volume rendering of the coronary vascular tree (d)

try rotation with stationary table position (“singleheartbeat acquisition”). In DSCT the high temporal resolution of up to 66 ms allows a high-pitch spiral acquisition with continuous table movement (“flash acquisition”) (Fig. 1) (Meyer and Henzler 2016). With both techniques, a prospectively triggered cCTA acquisition with sufficient image quality is possible in patients with sinus rhythm Prospective ECG Triggering: CTA in One and heart rates up to 75 bpm (Meyer et al. 2014). Cardiac Cycle Using volume scanners results in radiation expoWith the development of second- and third-­ sure comparable to conventional prospectively generation DSCT and SSCT volume scanners, triggered cCTA. Generally dose ranges from 2.8 to the acquisition of cCTA in a single cardiac cycle 4.27 mSv (Yang et al. 2014; Di Cesare et al. 2016). became feasible (Meyer and Henzler 2016). In DSCT with “flash” acquisition protocols results in SSCT volume scanners, the large detector width lower radiation exposure with a common exposiallows acquisition of the full heart in a single gan- tion 40 mm apart since it is too large for the snare mechanism to cover (Ismail et al. 2015; Bartus et al. 2013). Additionally, as opposed to the favorable CT anatomy of an inferiorly oriented LAA with the LAA tip lateral to the pulmonary trunk, where the LAA tip is easily accessible to the subxiphoid approach (Fig. 6a), a superiorly oriented LAA with the LAA tip direct behind the pulmonary trunk is an unfavorable feature, since this location of the LAA tip makes it inaccessible to the subxiphoid magnet (Laura et al. 2014; Bartus et al. 2013) (Fig. 6b, c). Patients with a posteriorly rotated position heart are also excluded from receiving the LARIAT device due to inability to deliver the device via the transseptal-­ subxiphoid approach (Bartus et al.

2013). Patients with prior cardiac surgery or pericarditis are unsuitable candidates as well, given the high likelihood of residual pericardial adhesions which can be a barrier to LARIAT placement (Wunderlich et al. 2015). CT presence of LAA thrombus is also a contraindication for device placement due to the obvious risk of stroke if the LAA is manipulated (Bartus et al. 2013) (Fig. 6d).

1.8.2 WATCHMAN CT is used pre-procedurally to identify the suitability of the patient for the WATCHMAN and aid in its sizing of the device (Donal et al. 2016). The WATCHMAN device is a self-expanding percutaneous LAA closure system made of nickel

Computed Tomography in the Management of Electrophysiology Procedures

a

b

c

d

765

Fig. 5  Pre-procedural CT imaging and intra-procedural fluoroscopic images for the LARIAT procedure. (a) Multiplanar reformat (MPR) CT image of the inferiorly oriented LAA (white arrow), which is considered favorable anatomy for the LARIAT suture procedure. (b–d)

Fluoroscopic images (b) of the left atrial appendage pre-­ LARIAT closure, (c) of the magnets and snare catheter, and (d) of the left atrium and occluded LAA after LARIAT closure

titanium with fixation barbs for attachment (Kapur and Mansour 2014). The WATCHMAN implantation involves access of the LAA through transseptal cannulation of the atria. Once the access sheath is placed into the distal end of the LAA, the WATCHMAN is then advanced toward the LAA using fluoroscopic guidance (Fig. 7a). The device is then deployed into the LAA to occlude its ostium after fluoroscopic confirmation (Fig. 7b). Both the PREVAIL and PROTECT-AF trials have demonstrated the WATCHMAN device to be non-inferior to warfarin in stroke prevention (Masoudi et al. 2015). Important characteristics for LAA occlusion include correct sizing of the landing zone diameters, the measurement of the depth and orientation of the dominant lobe, and the number of additional lobes present

(Wunderlich et al. 2015). CT accurately reconstructs the LAA ostial morphology, the perimeter of the LAA orifice, and the angle of initial LAA bend (Wunderlich et al. 2015). When comparing CT with TEE and fluoroscopy ostial measurements for WATCHMAN device closure, CT ­correlates well with TEE, resulting in identical success in appropriate device size selection (Saw et al. 2016). Given that the LAA morphology is complex and highly variable, pre-imaging with cardiac CT of the LAA may serve as a complementary role in addition to TEE and fluoroscopy and can additionally evaluate for presence of LAA thrombus (Donal et al. 2016). CT is also used to perform post-procedural assessments after WATCHMAN implantation (Fig. 7c) to identify device embolization and thrombus adherent to the device and can confirm

J. Negusei et al.

766

a

b

c

d

Fig. 6  Pre-procedural CT imaging for LARIAT procedure: anatomical considerations. (a) Favorable anatomy for LARIAT placement, showing an inferiorly oriented LAA. (b) Unfavorable anatomy for LARIAT placement, showing a superiorly oriented LAA apex behind the pul-

monary trunk. (c, d) Two unfavorable features for LARIAT placement with a superiorly oriented LAA apex behind the pulmonary trunk (c, thin arrow) and LAA thrombus (d, thick block arrow)

presence of peri-device leaks (Ismail et al. 2015). CT has increased sensitivity for peri-device leak (Fig. 7d) as compared to TEE (Ismail et al. 2015), but the clinical significance of incomplete occlusion remains uncertain.

in improvement in quality of life, LV function, and reduction in mortality and ventricular arrhythmias (VAs) (Bristow et al. 2004; Cleland et al. 2005). Currently, CRT implantation is a Class I indication for patients with an LV ejection fraction ≤35%, sinus rhythm, left bundle branch block with a QRS duration ≥150 ms, and New York Heart Association Class III/IV heart failure (Tracy et al. 2012). CRT LV leads are typically implanted in a branch of the coronary sinus with the pacing electrodes located ideally in regions of latest LV activation during native atrioventricular conduction. Placement of LV leads via the coronary venous system can be limited by anatomic considerations such as target vessel size and tortuosity. Pre-procedural imaging of the coronary veins by CT is deemed “appropriate” based

2

Cardiac Resynchronization Therapy

2.1

Cardiac Resynchronization Therapy and Coronary Venous Imaging

Cardiac resynchronization therapy is a mainstay treatment for patients with systolic heart failure and electrical ventricular dyssynchrony, resulting

Computed Tomography in the Management of Electrophysiology Procedures

a

b

c

d

767

Fig. 7  Intra-procedural fluoroscopy images and post-­ occlusion (white arrow, LAA). (c, d) Post-WATCHMAN procedural CT images of the WATCHMAN device. (a, b) CT with complete occlusion and thrombosis of the LAA Fluoroscopy of WATCHMAN peri- (a) and post- (b) (c) and with residual peri-device leak (d)

on the 2010 Appropriate Use Criteria for Cardiac Computed Tomography for noninvasive coronary vein mapping prior to placement of a biventricular pacemaker (Taylor et al. 2010).

2.2

Technical Considerations for Computed Tomography Coronary Venous Imaging

The coronary sinus and venous tree with its tributaries can be well visualized by ECG-gated CT (Singh et al. 2005; Abbara et al. 2005; Tops et al. 2005; Van de Veire et al. 2006; Tada et al. 2005; Muhlenbruch et al. 2005; Auricchio et al. 2007). The CT imaging protocol is nearly similar to that

used for a standard coronary CT angiography with a few exceptions to optimize the coronary venous imaging. The coronary veins are best imaged during systole when they are most ­plethoric and can be scanned using ECG-gated prospectively triggered systolic imaging (i.e., 35% or 45%) or retrospectively gated scan with tube current modulation during diastole and maximum tube current during systole (if functional analysis is requested). Radiation dose is less of concern in these end-stage HF patients with 50% mortality in 5 years. Nitroglycerin is not needed since coronary arterial dilatation may interfere with the evaluation of the cardiac veins, which course in similar directions. Depending on the type of CT scanner,

J. Negusei et al.

768

the use of beta-blockers to achieve optimal heart rate may be considered. Caudo-cranial image acquisition is also preferred given that the coronary sinus is inferiorly positioned in the heart, although there does not appear to be much difference in our experience when craniocaudal acquisition is performed. Inspiratory breath-hold is similar to that from a coronary artery study (Truong 2012). After the scout image, a test bolus protocol is preferred over bolus tracking to calculate the contrast agent transit time. For those with severe LV dysfunction, inform patients that the test bolus scan is the longest of the image acquisitions, since breath-hold may be greater than 30 s due to the poor cardiac output and prolonged transit time from the antecubital vein to the ascending or descending aorta. However, the subsequent CT venography breath-hold should be less than 15 s and is based on the scan range volume coverage of the heart (Truong 2012). Similar to that of a coronary artery scan, scan coverage should be from the carina bifurcation to the diaphragm. The most important distinction between coronary arterial and coronary venous imaging is the timing for venous phase imaging to optimally enhance the coronary venous system. While the typical scan delay for coronary venous imaging is 8–10 s after peak arterial opacification in patients with normal cardiac output, in our experience in these HF patients with severely reduced ejection fraction, a 15 s delay should be added to the contrast agent transit time (with minimum of 40 s delay). Additional delay time used for venous phase imaging should be added to the volume of contrast, although this additional volume could be given at a slower rate (i.e., 2–3 milliliters [mL]/second) to minimize the contrast load (Truong 2012).

2.3

Noninvasive Coronary Venous Mapping Prior to Cardiac Resynchronization Therapy

Thorough knowledge of the coronary venous anatomy is necessary prior to CRT placement, particularly the coronary sinus orientation and its ostial measurements, the anatomical characteristics of

venous side branches, and the presence of a Thebesian valve (Van de Veire et al. 2006; Chiribiri et al. 2008; Jongbloed et al. 2005; Mlynarski et al. 2009). The coronary venous system consists of the great cardiac vein, posterior vein of the left ventricle, anterolateral vein, anterior interventricular vein, and the coronary sinus (Mak and Truong 2012). During CRT implantation, the LV lead is placed into the coronary sinus under fluoroscopic guidance and ideally delivered into one of the primary branches of the coronary veins that supply the posterolateral wall (e.g., posterior vein of the left ventricle) to pace the LV. Given the variability in venous anatomy, complete visualization can help identify vessel tortuosity, size, angle of entrance, and presence of a suitable side branch (Mak and Truong 2012; Alikhani et al. 2013). Size extremes in vein diameter or the existence of anatomic barriers such as the Thebesian valve, for instance, can negatively impact CRT outcome (Genc et al. 2013). Knowledge of the surrounding structures is important as well (e.g., phrenic nerve and diaphragm) to prevent complications such as diaphragmatic pacing. Such information can be invaluable by optimizing LV lead positioning and aiding in CRT patient selection (Chiribiri et al. 2008; Jongbloed et al. 2005; Mlynarski et al. 2009). Retrograde venography is the most commonly used technique to assess the coronary venous system via invasive manual cannulation and balloon occlusion of the coronary sinus (Mak and Truong 2012; Genc et al. 2013). Numerous studies have demonstrated excellent correlation between CT (Fig. 8a) and retrograde venography (Fig. 8b) in evaluation of the coronary venous tree (Mak and Truong 2012; Catanzaro et al. 2014). When registered and integrated with real-time fluoroscopy, pre-procedural datasets of CT and MRI were shown to enable successful subsequent CRT implantation in a small case series involving eight patients with one or more previously failed implants (Duckett et al. 2011). A small prospective study of 17 patients examining CT overlaid on live fluoroscopy in CRT planning found statistically significant decreases in both intraoperative fluoroscopy and total procedure times, at the expense of increased radiation exposure (Alikhani

Computed Tomography in the Management of Electrophysiology Procedures

a

769

b

Fig. 8  Pre-procedural CT venography for cardiac resynchronization therapy. (a, b) Volume-rendered CT image (a) of the coronary venous anatomy and corresponding

invasive coronary venography (b) in a patient undergoing cardiac resynchronization therapy

et al. 2013). Additional studies of this approach are warranted to confirm these preliminary findings and assess clinical outcomes.

3

Ventricular Tachycardia

3.1

Ventricular Tachycardia and Myocardial Scar

2.4

Sudden cardiac death, which accounts for approximately half of total cardiac mortality in the general population, is predominately caused by ventricular arrhythmias (VA) in patients with underlying structural heart disease (Rijnierse et al. 2016). Although an LVEF 50% using QCA as the reference standard (Wang et al. 2011). A recent study showed that combining CCTA and DE-CTMPI increased the specificity from 50 to 67% in a population at high risk for CAD (yielding a 17% higher specificity than that of either tests separately) and outperformed CCTA alone in detecting hemodynamically significant coronary artery stenosis (De Cecco et al. 2014b). A trial comparing DE-CTMPI and SPECT for the detection of obstructive CAD with CMR as a gold standard that included 50 patients reported higher sensitivity (90% vs. 85%) and specificity (71% vs. 58%) with DE-CTMPI versus SPECT (Meyer et al. 2012). The most recent studies by Ko et al. and Kim et al. demonstrated similar accuracy compared to CMR (Kim et al. 2014a; Ko et al. 2014a). Ko et al. additionally found fair agreement between rest and stress DECT iodine maps, and described an incremental value of the stress protocol (accuracy of 83%) compared to rest protocol (accuracy of 62%) for the detection of hemodynamically significant CAD (Ko et al. 2014a). A number of studies comparing DE-CTMPI with SPECT observed that the modalities may lead to divergent classification of perfusion defects, such that almost one-half of perfusion defects that are reversible at SPECT are re-classified as fixed with DECT (Meinel et al. 2014a; Vliegenthart et al. 2012b; Wang et al. 2011). When compared to the single energy technique, DE-CTMPI demonstrated superior diagnostic accuracy using iodine maps for the detection of perfusion defects with SPECT as the reference standard (Arnoldi et al. 2011). A multicenter randomized controlled trial, the DECIDE-­Gold trial, with the objective to evaluate the diagnostic accuracy of the DECT and DE-CTMPI in the assessment of hemodynamic significance of CAD compared to ICA, is currently ongoing (Truong et al. 2015). In a comparison between DE and single energy CTMPI in detecting mixed perfusion defects with SPECT as the reference standard, DE-CTMPI showed better results in terms of sen-

789

sitivity (91% vs. 55%), negative predictive value (97% vs. 86%), and accuracy (93% vs. 85%). However, single energy CTMPI showed a higher specificity (98% vs. 94%) (Arnoldi et al. 2011). A recent meta-analysis by Sørgaard et al. (2016) assessed the diagnostic accuracy of stress static CTMPI (both single energy and dual energy) in patients with suspected or known CAD. In a total of 1188 patients, CTMPI reached a pooled sensitivity of 85% and specificity of 81% in the detection of myocardial ischemia, when compared to SPECT and CMR. Moreover, the addition of CTMPI to CCTA significantly improved CCTA specificity on a per-patient (62% to 84%) and per-vessel (72% to 90%) level.

2.3.3 Limitations The main limitation of static single shot CTMPI is the necessity of accurate acquisition timing. As mentioned above, diagnostic accuracy with static CTMPI is highly dependent on the timing of the acquisition. By definition, single shot static CTMPI requires a single data set acquisition. As a result, the contrast attenuation peak can be missed (Bischoff et al. 2013). After the administration of a stressing agent, the consequent high heart rate can cause motion artifacts which can ultimately mimic cardiac perfusion defects. In order to achieve a more robust assessment, it is worth evaluating several cardiac phases. A motion artifact is usually identifiable in a single phase while true hypoperfused areas persist through all cardiac phases. Performing static CT-MPI with a scanner that does not provide full cardiac z-axis coverage may lead to misalignment artifact. Additionally, the iodine contrast uptake can result in a heterogeneous distribution of iodine, thus limiting comparability of myocardial perfusion among different regions of the heart. The lack of standardized myocardial iodine distribution maps results in the high user dependence of a perfusion assessment. Finally, beam-­ hardening artifacts (caused by dense structures as spine, sternum, and iodine in the left ventricle) are an important cause of decreased image quality. The most common area in which beam-hardening artifacts occur is the postero-basal myocardial wall. As a result, detecting true perfusion defects becomes much more difficult result-

D. De Santis et al.

790

ing in increased false positive rates (Kyriakou et al. 2010). However, novel algorithms for beam-­hardening artifact correction were proven to be effective in avoiding false positive-findings and increase the specificity of SE- and DE-CTMPI (Bucher et al. 2016; Kitagawa et al. 2010; Stenner et al. 2010).

2.4

Dynamic CTMPI

2.4.1 Acquisition Technique and Image Analysis Opposite of static CTMPI that acquires a single snap-shot of the myocardium, dynamic CTMPI is characterized by a series of acquisitions during multiple cardiac cycles in order to track the dynamics of the contrast media within the myocardium (Caruso et al. 2016b). The acquisition typically starts few seconds before the arrival of the contrast media in the ascending aorta in order to obtain images of a stable baseline condition and continues for 25–30 s (Meinel et al. 2014b; Huber et al. 2013). Since this technique effectively monitors the dynamics of contrast media within the myocardium, it is able to generate time attenuation curves (TACs). Whole left ventricle scan coverage is a fundamental prerequisite of an effective examination. Modern CT scanners use two different approaches to achieve the goal: 160or 320-row MDCT, with a detector coverage of

a

Fig. 7  Dynamic CT perfusion in a 75-year-old male presenting with chest pain. Myocardial blood flow (a) and myocardial blood volume (b) color coded images showing decreased perfusion of the apical wall (arrows). (c) Derived time-attenuation curves displays the perfusion of

80 mm and 160 mm, respectively, and are able to perform the scan in a stationary mode, without table movement. Alternatively, dual-source CT scanners are fast enough to acquire data of the left ventricle without motion, but their detector coverage is not wide enough to scan the complete left ventricle in one scan (38.4 mm and 57.6 mm for second and third generation, respectively). Thus, to achieve full coverage of the left ventricle, the scan is performed in shuttle mode, consisting of a prospectively ECG-triggered axial acquisition during rapid back and forth table movement. This technique acquires the heart in two alternating slabs with minimum overlap, scanning the lower and the upper part of the heart, respectively, with a table movement in between. This technique reaches a comprehensive scan coverage of 73 mm for the second generation and 105 mm for the third generation, allowing full coverage of the entire left ventricle. The main advantage of dynamic CTMPI over static CTMPI is the possibility to perform both a semiquantitative and a quantitative evaluation. The semiquantitative assessment is obtained by analyzing only the upslope portion of the TAC by means of the upslope method (George et al. 2007; Bastarrika et al. 2010; Al-Saadi et al. 2000), deriving parameters such as peak enhancement, time-to-peak, and upslope. The quantitative assessment allows to calculate myocardial blood flow (MBF), myocardial blood

b

the whole left ventricle (white line) and of the volume of interest (VOI) manually drawn in the apex (yellow line). (d) Quantitative values can also be calculated demonstrating the apical perfusion defect

Functional Cardiac CT Angiography

791

c

d

Fig. 7 (continued)

volume (MBV), and volume transfer constant (Ktrans) through mathematical models applied to TACs (Fig. 7). With dual-source CT scanners, the quantitative analysis is performed by means of a modified bicompartimental model (Bamberg et al. 2010). The computation of those parameters yields a 3D CT dataset represented as color maps that can be analyzed either visually or quantitatively and can also be overlaid onto the traditional grey scale CT dataset. The quality of data and the derived reliability of a fully quantitative analysis depends on an adequate z-axis coverage (10–12 cm in systole), high temporal resolution (≤100 ms for a 360° gantry rotation), and the ability to acquire data from each heartbeat for circa 30 s.

2.4.2 Clinical Results Diagnostic performance of dynamic CTMPI has been extensively assessed in comparison with CMR, SPECT, and ICA with or without FFR measurements (Table 3). Pursuing the achievement of a standardized perfusion image evaluation, Ebersberger et al. analyzed the performance of a semiautomated software for the segmental evaluation of MBF and MBV, demonstrating similar results in comparison to the traditional manual approach but had a significant reduction of analysis times (Ebersberger

et al. 2014). Wichmann et al. investigated the performances of such software for a global evaluation of the left ventricle, demonstrating high diagnostic accuracy for the detection of two or three vessel territories with perfusion defects (Wichmann et al. 2016). Bamberg et al. (2011) and Greif et al. (2013) established a cutoff value for MBF of 75 mL/100 mL/min for differentiating between hemodynamically significant and nonsignificant coronary stenosis. Substantially similar results (78 mL/100 mL/ min) were found by Rossi and coworkers (2014b). Kono et al. (2014) and Wichmann et al. (2015) demonstrated that the relative MBF provides superior diagnostic accuracy in comparison with absolute MBF for the detection of significant coronary stenosis. A recently published multicenter study ­demonstrated that dynamic CTMPI is also able to identify early perfusion disturbances in patients affected by diabetes and hypertension, improving CAD risk stratification in patients without visual perfusion defects (Vliegenthart et al. 2016).

2.4.3 Limitations The major limitation of the dynamic CTMPI is represented by the higher radiation exposure, which is directly related to the longer scan time in comparison to the static technique. A scanning

792

D. De Santis et al.

Table 3  Dynamic CT myocardial perfusion studies CT perfusion protocol Stress

Reference Level of technique analysis CMR Segment

Se (%) 86

Sp (%) 98

PPV NPV (%) (%) 94 96

SPECT, ICA

Segment

83

78

79

82

Stress

9.1 (Stress) 9.1 (Rest) 10

FFR

Vessel

93

87

75

97

2nd DSCT

Stress

9.5

SPECT, ICA

20

2nd DSCT

Stress

12.8

65

2nd DSCT

Stress

9.7

CMR SPECT FFR

Segment Vessel Segment Segment Vessel

85 90 86 84 95

92 81 98 92 74

55 58 94 88 48

98 96 96 92 98

32

Stress

9.5

FFR

Vessel

75

100

100

90

34

256-­ MDCT 2nd DSCT

Stress-­ rest

ICA

Vessel

83

98

98

80

33

2nd DSCT

Stress-­ rest

5.1 (Stress) 3.7 (Rest) 5.3 (Stress)

CMR

Segment Vessel Patient

81 82 77

94 84 60

59 52 56

98 96 80

FFR

Vessel

88

90

77

95

Segment Vessel Segment

78 100 98

75 75 96

51 92 96

91 100 98

Vessel

97

95

95

98

Segment Segment Segment

80 82 88

86 87 87

62 86 78

94 83 93

Vessel

75

78

78

75

Patient population 10

CT technology 2nd DSCT

35

2nd DSCT

Stress-­ rest

33

2nd DSCT

30

Weininger et al. (2012) Greif et al. (2013) Huber et al. (2013) Kim et al. (2013b)

Author Bastarrika et al. (2010) Ho et al. (2010)

Bamberg et al. (2011) Wang et al. (2012)

Kim et al. (2013c)

Average radiation dose (mSv) 10.3a

Rossi et al. (2014b) Bamberg et al. (2014)

80

2nd DSCT

Stress

4.0 (Rest) 9.4

31

2nd DSCT

Stress

11.1

CMR

Baxa et al. (2015)

54

2nd DSCT

Stress-­ rest

ICA

Tanabe et al. (2016a)

53

256-­ MDCT

Stress

8.9 (Stress) 8.4 (Rest) 10.5

Tanabe et al. (2016b) Coenen et al. (2016)

35

256-­ MDCT 2nd and 3rd DSCT

Stress

10.6

SPECT CMR CMR

Stress

9

FFR

43

PPV positive predictive value, NPV negative predictive value, CMR cardiovascular magnetic resonance, DSCT dual-­ source computed tomography, FFR fractional flow reserve, ICA invasive coronary angiography, MDCT multi-detector computed tomography, SPECT single-photon emission computed tomography a Dose of the CT perfusion derived by applying a conversion factor of 0.014

Functional Cardiac CT Angiography

protocol with both rest and stress acquisition reaches up to 18 mSv (Bastarrika et al. 2010; Ho et al. 2010). On the other hand, this technique still provides a lower radiation dose than SPECT and the vendors are actively implementing further dose reduction strategies. Additionally, the longer scan duration requiring a breath hold of approximately 30 s remains challenging for the majority of subjects undergoing the examination (Bamberg et al. 2011).

793

regadenoson are by far the two most commonly used due to their improved safety profiles. The first experimental study investigating stress myocardial CT perfusion with adenosine was performed in 1987 with an electron-beam CT scanner on an animal model (Rumberger et al. 1987). Adenosine directly dilates the coronary arteries through a selective agonist action on the adenosine receptors. Since adenosine has an extremely short half-life of only few seconds, it must be IV administered through a continuous infusion at a recommended dose of 140 μg/kg/ 2.5 Pharmacological Stress min for at least 2 min in order to obtain an Agents increase of the heart rate of 10–20 beats above the resting heart rate. Healthy coronary arteries have a natural dilatory Regadenoson, a selective A2A receptor agonist, response to an increased request for oxygen. is characterized by a half-life of 2 minutes and However, the dilatory response of diseased coro- thus can be IV administered via a pre-drawn sinnary arteries is limited due to the fact that they gle dose syringe, resulting in a more time-effiare already operating in the reserve dilatory cient CT examination (Kurata et al. 2005). capacity in order to compensate for low perfusion Regadenoson is responsible for less systemic (Dole et al. 1985; Gould and Lipscomb 1974; side effects in patients with asthma and chronic Sambuceti et al. 1997). As a result, during the obstructive pulmonary disease compared to adeadministration of a stress agent, the augmented nosine (Mahmarian et al. 2009; Salgado Garcia blood flow in the myocardium supplied by unaf- and Jimenez Heffernan 2014). On the other hand, fected coronary arteries causes a steal phenome- in cases using a stress/rest protocol an adequate non at the level of the myocardium supplied by time interval should be allowed before the rest the affected coronary vessels, reducing the blood scan, considering that the vasodilative effect of pool and, thus, the amount of contrast media regadenoson persists for 15–20 minutes. delivered to the local myocardium. The attenua- Alternatively, theophylline can be IV administion of myocardium during the pharmacological tered to terminate the effect of regadenoson stress is compared to the myocardium at rest. A (Becker and Becker 2013). The most common hypo-attenuating area only present in the stress side effects of adenosine and regadenoson are phase (or significantly increased in the stress headache, dyspnea, flushing, and chest pain. phase) is defined as reversible myocardial perfu- Both molecules can cause moderate to severe sion defect (MPD) and is suspected for reversible complications such as tachycardia and atrio-venmyocardial ischemia. An area with hypo-­ tricular block (0.14% of patients) (Iskandrian attenuating pattern present both in rest and stress et al. 2007; Luu et al. 2013). Rare cases of acute phases are defined as fixed MPDs and correspond myocardial infarct and death have also been to myocardial infarct (Bucher et al. 2014). Since reported (Hsi et al. 2013; Shah et al. 2013). reversible MPDs are usually present earlier than Dobutamine and dipyridamole are other less fixed MPDs, the sensitivity for the detection of common stressor agents. Dobutamine is a synreversible MPDs significantly increases with thetic catecholamine with a strong β1-receptor pharmacological stress tests (Gould and and mild α1- β2-receptor agonist activity. It is Lipscomb 1974). used in low doses for the treatment of heart failStressor agents are represented by adenosine, ure and for the identification of dysfunctional regadenoson, dobutamine, and dipyridamole but viable myocardium, thanks to its strong ino(Table  4) (Kurata et al. 2005). Adenosine and tropic effects. When used in high doses its

D. De Santis et al.

794 Table 4  Overview of pharmacological agents used in stress CT myocardial perfusion Adenosine Nonselective adenosine receptor agonist

Regadenoson A2A selective adenosine receptor agonist

Effect

Direct coronary artery vasodilator

Direct coronary artery vasodilator

Dipyridamole Increases availability of adenosine by inhibiting adenosine deaminase Indirect coronary artery vasodilator

Dosage

140 μg/kg/min for 3–6 min

Bolus of 400 μg in 10–20 s

140 μg/kg/min for 4 min

Contraindications

  – High-grade AV block

  – High-grade AV block

  – High-grade AV block

  – Asthma or COPD

  – Sinus bradycardia

  – Asthma or COPD

  – Sinus bradycardia

  – Systemic hypotension (BP  40 kg) and tube current just before the exam. It may also be advisable (10 mAs per kg up to 5 kg, then 4–5 mAs per to physically immobilize patients, particularly kg up to 92 mAs, then 2–3 mAs per kg up to during contrast injection, which can cause dis138 mAs) (Paul et al. 2011). A further strategy comfort or anxiety. To do this, a soft weight, suggested in the literature that can be associrestraining aid, or vacuum cushion can be ated to reduce radiation dose is (semi-)autoplaced above or around the child’s limbs matic modulation of tube voltage or tube (Booij et al. 2016). In rare cases, general anescurrent (Han et  al. 2011; Litmanovich et  al. thesia may be required. Beta-blocker adminis2014). tration, either oral or intravenous, is generally • Contrast injection protocols: Iodinated contrast indicated for heart rate control in CCTA; agent in CCTA is usually injected through an nonetheless, pharmacological preparation antecubital vein using a dual-syringe power should be carefully applied to children because injector with a biphasic protocol, consisting in beta-blockers substantially prolong patient contrast administration followed by a saline preparation time with no guaranteed success chaser to reduce the amount of contrast given and also newer scanner technologies are able and perivenous artifacts (Hopper et  al. 1997). to obtain diagnostic images even at higher In patients with conotruncal anomalies, it is heart rates (Goo 2015). better to evaluate the coronaries using a tripha• Acquisition Parameters: The portion of the sic protocol, consisting in contrast injection folcardiac cycle to scan as a percentage interval lowed by a mixed bolus of contrast and saline can be automatically defined by the scanner (50–60% diluted contrast) and subsequently by according to the heart rate and dynamically saline (10 mL maximum for infants); this promodified during acquisition, particularly if tocol, according to Litmanovich et al., reduces R-R interval irregularities occur, thanks to an perivenous artifacts and improves opacification automatic algorithm of correction (Pan et  al. of the right heart and pulmonary arteries. A dif2013). In the prospective ECG-triggered ferent injection protocol should be considered sequential and retrospective ECG-gated spiral in patients with Fontan circulation. To achieve scans, ECG triggering or pulsing is usually set the best opacification in these patients, there is at mid-diastole (Dose Range between 70–75% the chance to use a double access through arm of the R-R interval) for heart rates lower than and leg with simultaneous injection of 50%

991

Congenital Heart Disease in Children

mixed contrast and saline (Ghadimi Mahani et al. 2016; Goo 2011). If it is not possible to use a double access, it is suggested to use the triphasic protocol (Litmanovich et  al. 2008). The volume of contrast agent is adapted to the body weight (1.5–3  mL/kg depending on the concentration, with higher volumes for lower concentrations) and the flow rate of injection depends on the size of the angiocatheter (24–18 gauge depending on the age/size of the patient, with lower flows for smaller angiocatheter sizes). Typically, the flow rate ranges between 0.4 and 5 mL/s (Goo 2013). The time from start injection to acquisition (“scan delay”) is usually 15–20 s for young children (under 6 years), and can best be decided based on the duration of contrast agent and saline infusion (the acquisition should start right at the end of the injection), and on the site of cannulation (consider higher amounts of saline and longer delays when injecting from the lower limbs). This may avoid unnecessary radiation exposure due to the acquisition of additional scans to trace the bolus distribution in the body in younger children. In order to track the arrival of contrast and planning the acquisition with the best opacification, particularly in older children with higher inter-variability, it is possible to use the semiautomatic bolus tracking or test bolus systems (Bogaard et  al. 2015; Goo 2013). In patients admitted to intensive care with difficulties in establishing peripheral venous access, it is possible to use central venous catheters with careful injection rates of 0.4–1.2  mL (Plumb and Murphy 2011). However, this may vary according to catheter type.

2.1

Coronary Anomalies

Coronary anomalies have an incidence expected number of alveoli) (Lee et al. 2011b; Michelson 1977). A resulting check-valve mechanism at the bronchial level causes progressive hyperinflation of the lung by allowing more air to enter the involved area on inspiration than leaves on expiration (Berrocal et al. 2004). The most commonly affected lobe is the left upper lobe, followed by the middle lobe. The distribution of lobar involvement is 42% in the left upper lobe, 35% in the right middle lobe, 21% in the right upper lobe, and 1% in each lower lobe (Stocker et al. 1977). Mostly single lobe on one side is involved; however, multilobar involvement (5%) can be seen (Michelson 1977). Pulmonary vascular markings can be seen up to the periphery of the affected lobe and there is no pleural demarcation as we see in pneumothorax.

2.2

Congenital Lobar Emphysema

Congenital lobar emphysema is characterized by progressive overdistention of a lobe, sometimes two lobes (Chowdhury and Chakraborty 2015). Congenital lobar overinflation is more common in males (M:F  =  3:1) (Baert and Donoghue 2007). The underlying pathology is complex and deemed manifold. The exact cause of CLE is difficult to determine and no apparent cause is found in 50% of cases. In the remaining cases, three categories are pathologically described: (1) ­external bronchial obstruction (due to abnormal vessels, enlarged lymph nodes), (2) internal bronchial obstruction (cartilaginous deficiency, bronchial stenosis, redundant bronchial mucosa), and (3) nonobstructive emphysema (Raynor et  al. 1967; Sadaqat et  al. 2011; Kumar et  al. 2008;

S. Ley and J. Ley-Zaporozhan

1022

Affected patients typically present with respiratory distress within the first 6  months of life. The symptoms primarily depend on the size of the affected lobe as well as the degree of compression on adjacent normal lung tissues and mediastinal structures (Lee et al. 2011b; Baert and Donoghue 2007; Man et  al. 1983). Adult presentation is unusual. They usually come with exertional dyspnea or pulmonary infections. Symptoms can be mild and sometimes no symptoms can be seen. The characteristic CT finding of congenital lobar emphysema is a hyperinflated lobe with attenuated pulmonary vessels (Figs.  15 and 16). Recognizing attenuated lung markings within the overinflated lobe can be helpful for differentiating congenital lobar emphysema from a pneumothorax on plain films. There might be also midline substernal lobar herniation and compression of the remain-

ing lung seen. CT is also useful for diagnosing multilobar involvement and excluding secondary causes of lobar overinflation due to vascular rings or mediastinal masses (Lee et al. 2011b). Imaging differential considerations include bronchial atresia, congenital pulmonary airway malformation (CPAM), pulmonary arterial hypoplasia, pulmonary hypoplasia, and Swyer–James syndrome.

2.3

 ongenital Pulmonary Airway C Malformation (CPAM)

Congenital pulmonary airway malformation (CPAM) is the most common (30–40%) of congenital lung malformations and represents a hamartomatous proliferation of cystic spaces that

a

b

c

d

Fig. 15  Congenital emphysema of the right upper lung in a 1-month-old infant (a). Except for some atelectasis otherwise normal appearance of the middle and lower lobe (b), also shown on coronal view (c) and X-ray (d)

Chest CT Imaging in Children

a

1023

b

c

Fig. 16  Congenital overinflation of the right lung with herniation to the left in a 2-week-old infant (a). On CT images severe stenosis of the right main bronchus (b,

arrow) was found, and a normal size of the bronchi distal to the stenosis (c)

resemble terminal bronchioles, being lined by respiratory epithelium, but lacking cartilage or bronchial glands (Berrocal et  al. 2004; Stocker et al. 1977). The previous nomenclature (congenital cystic adenomatoid malformation; CCAM) is superseded based on recognition that lesions may contain solid as well as cystic components. They communicate with the bronchial tree (distinguishing it from bronchogenic cysts and pulmonary sequestrations) and have blood supply derived from the pulmonary artery (distinguishing it from pulmonary sequestrations) as well as pulmonary venous drainage (Chowdhury and Chakraborty 2015). The current Stocker classification of CPAMs, modified on the original classification, comprises five types based on the size of cysts and its resemblance to the bronchoalveolar tree (Chowdhury and Chakraborty 2015; Stocker 2009). Type 0 involves multilobar acinar dysgenesis/dysplasia of major tracheobronchial airways, and being bilateral. This form is incompatible with life. Type 1 (70%) comprises macrocysts of bronchial or bronchiolar origin, with the largest cyst exceeding 2  cm (Fig.  11). Type 2 (15–20%) is similar to type 1 but with cysts 0.5–2 cm in size (Fig. 12). Type 3 (10%) is a predominantly solid lesion associated with microcysts less than 0.5 cm, of bronchoalveolar duct origin, and is the only type that is adenomatoid. Type 4 (1 2.5–3 150–200 1–2 2–2.5 200–250 2–6 2–1.5 250–300 6–15 1–1.5 300–350 Calculation of delay: Injection time × organ factor  Injection time = CM amount divided by flow  Flow depends on size of iv line, usually 1.5– 3.5 mL/s realistically achievable in children  Organ-specific factor (helical CT, 4–16 rows)   Abdomen general 1.8   Pelvis general 2.2   Liver 0.6 (arterial), 2.2 (portal-venous/ parenchymal)   Kidney 0.8 (art), 4.9 (parenchymal)

Table 4  Suggestion for contrast agent (2%) volume for bowel opacification on pediatric abdominal CTs (Sorantin et al. 2013a) Age >6 months 6–12 months 1–3 years 3–10 years >10 years

CA amount (mL) 100 200 300 700 1000

Remember that change of tube voltage will also lead to different Hounsfield unit values, which needs to be considered not only for reading a study, but also for properly setting cutoff values, e.g., for a bolus tracking semiautomated CT angiography start (Nagy et al. 2016); timing scans are usually avoided and not necessary in the child’s abdomen. Reconstruction filters also affect the dose needs as well as whether the query addresses a high or a low contrast issue (Sorantin and Wießpeiner 2010). And finally all other aspects of the “imaging chain” need to be addressed, ending up at proper setting of the workstation and experience and training for being able to read the respective images (Sorantin 2008). After all—when you have set your protocol and did your examination—reconsider quality and radiation dose critically: If you have the impression that the images are “too nice” and a diagnostically sufficient quality could be achieved with less dose, this can be very simply tested by asking the technician to reconstruct the same data set with half the slice thickness—if you still get a diagnostic imaging quality, you probably can reduce the exposure dose for 30–50%. Some basic rules for a pediatric CT are as follows: • Always check for a justifying indication, and if the requested information can be possibly retrieved without CT. • In most cases a single acquisition is sufficient to answer the clinically and therapeutically relevant question; rarely multiphase scans are required: Also consider the use of split bolus scans.

CT of the Pediatric Abdomen

• Use age- and weight/circumference-adapted exposures (“Stöver and Rogala” rule, Image Gently proposals … see Table 1); consider lowering kV in infants and small children or for CTA. • Use automated tube modulation options (if available), but check and adapt upper and lower limits. • Limit scan range to targeted area—not only for the scan, but also for the scout. • Use proper positioning and protective devices; consider p.a. direction for the scout view. • Apply iterative reconstruction, and accept some noise—aiming at a diagnostic, not the most beautiful, image. • Liberally use immobilization and sedation— the risks of sedation are far less than those of radiation; therefore never “try” a CT, even with the fast new-generation CTs (although these techniques such as “flash imaging” or “volume scanning” may significantly reduce the need for sedation). • Use the most modern CT device available with the newest detectors (best sensitivity) for scanning children—create respective protocols beforehand to have them readily available in emergency. Train staff on how to scan and handle the device and the children.

3

Indications, Imaging Findings, and Examples

As mentioned above many queries can be successfully and diagnostically reliable addressed by other imaging reducing the need for pediatric abdominal CT, not only in the liver, the spleen, and the kidneys as well as pancreas but also large parts of the gastrointestinal tract, the reproductive organs, and the lower urinary tract and most vessels can be beautifully assessed by ultrasound (US). This potential has been recently enhanced by modern methods such as harmonic imaging, new speckle reduction filters, image compounding, sono-elastography, the various modern Doppler options, as well as the use of US contrast agents (Pilhatsch and Riccabona 2011; Riccabona 2014a, b). Many other queries can also nicely be imaged by MRI, provided that safe handling is possible in the MR suite for the

1041

respective condition and the study is available within a reasonable time. Let me briefly address some common queries where a CT study may become indicated and is used in childhood: Urolithiasis and complications in urinary tract infection (UTI): One of the most common CT indications in adults is the unenhanced low-­dose CT of the urinary tract for urolithiasis. In children most stones can be diagnosed by US; furthermore ultralow-dose CT protocols bear the risk of not being diagnosed due to the above-­mentioned factors such as lack of visceral fat and poorly calcified “stones.” Thus “stone-uro-CT” is far less commonly indicated in children and serves only as a problem-solving tool using pediatric protocols (e.g., in infected systems with underlying malformations and suspicion of a potentially adherent ureteric stone) or for preoperative assessment such as planning percutaneous lithopraxy (Fig. 1) (Damasio et al. 2013; Riccabona et al. 2009). The imaging findings and diagnostic criteria—however—do not differ from adults, although paucity of fat may hinder perception of the typical perirenal and peri-ureteric stranding or the continuous definition and assessment of the ureter. The assessment of UTI complications is also mostly achievable by US, complemented by MRI—thus only rare conditions such as tuberculous or xanthogranulomatous pyelonephritis or unavailability of MRI may justify a contrast-­enhanced CT scan, in the setting of therapeutically insufficient information on US, and then applying a dedicated pediatric CT protocol (Riccabona et al. 2010). If a CT is needed, usually one late-phase acquisition (with a split bolus protocol for visualization of the collecting system if necessary) is usually sufficient. Again, the imaging findings and diagnostic criteria as seen in adults apply. Abdominal trauma: Particularly for blunt and moderate abdominal pediatric trauma again US has become the initial imaging of choice (Amerstorfer et al. 2015; Riccabona 2013; Sorantin et al. 2013a). Using all of the modern US potential and providing experienced and dedicated examiners a CT study can often be avoided or (depending on the respec-

M. Riccabona and A. Pilhatsch

1042

a

d

b

e

c

f

Fig. 1  Low-dose pediatric CT-stone disease: this axial CT section of the pelvic ureter section (a) shows a large impacted ureteric concrement in a slightly obese child with an atypical, obviously chronic dilatation of the upper collecting system initially considered to be a pelvi-­ureteric junction obstruction on US (b), not confirmed on MR

urography (c). (d–f) Stone CT in an adolescent with medullary nephrocalcinosis due to distal tubular acidosis: this low-dose stone CT was indicated to assess for right-sided urolithiasis, as US could not sufficiently penetrate the calcified medulla to differentiate between calcinosis and urolithiasis; the staghorn calculus on the left side was known

tive findings) can be tailored to the still unanswered relevant questions, helping to reduce radiation burden (Fig. 2). Note that CT is rarely necessary for follow-up in a stable child. Again, a single parenchymal phase acquisition is usually sufficient—with the exception of a complicated urinary tract injury where assessment of the collecting system may render a second delayed scan necessary; however, if not necessary in the acute phase, this can be elegantly also be imaged by a dynamic MRI urography without any radiation burden (Fig. 2). Nevertheless, as in adults, CT is and remains the main imaging tool for multiple and severe trauma. Here an arterial phase of the upper abdomen (usually included in the arterial phase neck and chest scan) and a

delayed parenchymal phase of the entire abdomen are useful and justifiable. Additionally a CT may become indicated for assessing unclear US findings as well as discrepancy between the US and the clini­ cal development—particularly concerning the intestines such as a suspected perforation, ­ bony injuries, and retroperitoneum. Besides contrast leakage in the arterial phase indicating an acute arterial bleeding, periportal stranding and hypoperfused bowel should be particularly observed as they constitute severe conditions and a high probability of a sudden deterioration of the often astonishingly stable child’s circulatory state. But note that particularly in early posttraumatic stages no free gas must be seen even in a perforating bowel

CT of the Pediatric Abdomen

1043

a

d

b

e

c

f

g

Fig. 2  Modern pediatric trauma imaging. (a, b) Single-­ phase trauma CT—renal injury/late parenchymal phase in axial (a) and coronal (b) presentation. (c, d) Ultrasound (c) with power Doppler (d) for early follow-up. (e–g) dynamic MRU for assessing the suspected traumatic urinoma after collecting system injury. (a, b) The initial single-­phase trauma CT (a axial section, b coronal reconstruction) shows the rupture of the right kidney, with contrast extravasation within the hematoma, suspicions of a urinoma formation. (c, d) US for follow-up confirms the unusually hypoechoic liquid-like regional aspect of the perirenal collection (c),

additional to the typical injured and non-perfused residual parenchyma on power Doppler (d) and the inhomogeneously echogenic hematoma in the lower section of the right kidney. (e–g) Dynamic contrast-­enhanced MR urography elegantly demonstrates and confirms the contrast leakage from the collecting system in the late excretory phase (f) of the examination that accumulates in the urinoma in the delayed imaging (g); note that the standard early excretory phase only shows the right-sided renal injury, but still does not demonstrate the contrast leakage (e)

injury (i.e., no “rule-out bowel perforation” is achievable). And that—considering the therapeutic options—often a (contrast-enhanced) US-based imaging approach supplemented by MRI can be sufficient in most cases with moderate blunt abdominal trauma in children (Amerstorfer et al. 2015). Oncology conditions: In oncology, abdominal CT has been (mostly) replaced by MRI for the initial diagnostic workup, confirmation of

a clinically or sonographically suspected or diagnosed tumor including the respective differential diagnosis, characterization of the tumor, as well as follow-up and imaging of complications—provided that there is access and availability. In some situations where MRI is contraindicated or not available, CT will still be performed and sometimes the superior spatial and temporal resolution of CT (e.g., for particularly small vessels or bony i­ nvolvement)

M. Riccabona and A. Pilhatsch

1044

a

b

Fig. 3  Oncology CT—hepatoblastoma in axial (a) and coronal presentation (b); single-phase CT focused on good vessel opacification using a microbolus timing technique (i.e., the scan is started manually when optimal enhancement of both arterial and venous vessels occurs

using visual bolus tracking) for outlining the vascular involvement/anatomy. Note the tumor rupture with hemorrhagic ascites which has indicated this emergency CT, and that the chest was included for assessing potential lung metastasis

may indicate a CT study in spite of or addition to the normally performed MRI (Granata and Magnano 2013) (Fig. 3). If a CT is performed (for example as emergency imaging for acute bleeding in tumor rupture), the respective requirements of the various oncology protocols have to be respected and sometimes multiphase scans may be necessary particularly for characterization of liver lesions; however, usually no unenhanced scan is necessary (da Costa e Silva and da Silva 2007). And often a chest CT is included for completing the staging. Imaging findings do not differ significantly from adults—the major rules for defining the organ of origin, the vascular supply and involvement, or the dynamic enhancement pattern apply and are not discussed. Calcifications may be seen (e.g., typically and often in neuroblastoma, rarely also in Wilms tumors or liver tumors and teratoma), as well as necrotic or hemorrhagic components. Metastases to the regional nodes and to mostly the liver should be searched for and are essential for staging, as well as signs of rupture or infiltration. A rim-like appearance of residual organ tissue (usually in the kidney with a Wilms tumor or a mesoblastic nephroma) should be appreciated as well as usually only moderately enhancing residual embryogenic rests that may represent precancerous lesions (e.g., nephroblastomatosis) or fatty tissue usually seen within a teratoma; however, a

c­omplete and detailed discussion of the CT appearance of all pediatric abdominal tumors is beyond the scope of this chapter. Acute abdomen and miscellaneous queries: The last group of patients who are often referred for CT are those with gastrointestinal queries presenting as an acute abdomen or inflammatory bowel disease. Also with these queries, CT is rarely indicated in children, as most conditions such as appendicitis and abscesses as well as inflammatory bowel disease (IBD)/Crohn’s disease can convincingly be depicted by US (Garcia Peña et al. 2004; Nievelstein et al. 2010; Sorantin et al. 2013a). Again CT has become the problem-­ solving tool in specific acute situations, where US is not able to sufficiently answer the therapeutically relevant questions (e.g., due to obesity, unusual location of the appendix), and MRI is not available or at least is not quickly enough available (Fig. 4). In the latter scenario, CT particularly of suspected appendicitis or (covered) perforation with abscess formation still remains the recommended second step investigation, and a single (potentially contrast-enhanced) scan is sufficient (Fig. 4); the need for bowel opacification by filling with water or diluted contrast is discussed controversially. IBD imaging has completely moved to MRI and US, and there is no routine indication for CT for assessing or following up Crohn’s disease and similar entities, and also not for ovarian torsion or other genital queries.

CT of the Pediatric Abdomen

1045

a

f

b

c

d

e

g

Fig. 4  CT in pediatric intestinal problems: (a) Coronal section of a CT in appendicitis in an obese patient (US was not conclusive)—only moderate contrast enhancement, but obvious periappendiceal changes such as fatty stranding of the mesenteric fat. (b) Axial CT—mid-­ abdomen, demonstrating a small focal free intraperitoneal gas bubble in front of the transverse colon in perforation. (c) Axial CT image: Severe pseudomembranous (clostridium) colitis with an impressively thickened bowel wall and nearly collapsed lumen in a child with acute abdomi-

nal pain and confusing US findings due to the monstrous colon wall thickening. Furthermore, a potential fistula tract even with perforation was suspected clinically. (d–g) An unusual huge abscess formation in the right upper quadrant of a school-age boy seen on ultrasonography (d, e) indicated a preoperative single-phase contrast-enhanced CT (f, g)—somewhat corresponding sections) for further assessment and surgery planning. Eventually it turned out to be a perforated appendicitis in an unusually high location

Additionally, a CT angiography (CTA) may be needed in some dedicated vascular queries or when preparing for transplantation. However, pediatric abdominal CTA can be challenging particularly in neonates and needs specifically adapted protocols and timing tricks (e.g., microbolus technique—see Fig. 3) (Dillman and Hernandez 2009; Hellinger et al. 2010; Sorantin et al. 2013b; Sorantin et al. 2015) (Fig. 5). The typical findings are similar as in adults; again, the paucity of visceral fat may make depiction (e.g., of the appendix), assessment, and interpretation particularly in unenhanced scans difficult. Vivid contrast enhancement is

seen in an acutely inflamed appendix—however, in a gangrenous or necrotic appendix or in lympho-­ nodular appendicitis this can be less or lacking. Mesenteric swelling and stranding as well as enhancement can be helpful signs. In perforation some free gas is usually seen in the highest abdominal aspects; however, it may also be entrapped—thus a thorough search for some small bubbles of free gas is mandatory, and sometimes even no free gas can be depicted. Also abscesses appear as one is used from adults, with a complex central fluid collections and an enhancing peripheral membrane.

M. Riccabona and A. Pilhatsch

1046

References AAMP (American Association of Physicists in Medicine) Task Force. Size-specific dose estimates (ssde) in pediatric and adult body CT examinations. http://www. aapm.org/pubs/reports/. Last accessed 11 Feb 2017 Amerstorfer EE, Haberlik A, Riccabona M (2015) Imaging assessment of renal injuries in children and adolescents: CT or ultrasound? Eur J Pediatr Surg 50:448–445 Brenner DJ, Hall EJ (2007) Computed tomography—an increasing source of radiation exposure. N Engl J Med 357:2277–2284 Brenner DJ, Elliston C, Hall EJ, Berdon WE (2001) Estimated risks of radiation-induced fatal cancer from pediatric CT. AJR Am J Roentgenol 176:289–296 Brody AS, Frush DP, Huda W, Brent RL, American Academy of Pediatrics Section on Radiology (2007) Radiation risk to children from computed tomography. Pediatrics 120:677–682 Chang HL, Jin Mo G, Hyun JY, Sung-Joon Y, Chang MP, Eun JC, Jung-Gi I (2008) Radiation dose modulation techniques in the multidetector CT era: from basics to practice. Radiographics 28:1451–1459 Chodick G, Ronckers CM, Shalev V, Ron E (2007) Excess lifetime cancer mortality risk attributable to radiation Fig. 5  CTA in a neonatal mesenteric arteriovenous vascuexposure from computed tomography examinations in lar malformation children. Israel Med Assoc J 9:584–587 da Costa e Silva EJ, da Silva GA (2007) Eliminating unenhanced CT when evaluating abdominal neoplasms in 4 Summary children. AJR Am J Roentgenol 189:1211–1214 Damasio B, Darge K, Riccabona M (2013) Multi-detector CT in the paediatric urinary tract. EJR Eur J Radiol Pediatric abdominal CT is a powerful, fast, read82:1118–1125 ily available, and comprehensive imaging tool— Dillman JR, Hernandez RJ (2009) Role of CT in the evaluhowever, it comes with some considerable ation of congenital cardiovascular disease in children. radiation burden. Therefore the indication for an Am J Roentgenol 192:1219–1231 abdominal CT study in children has to be particu- Frush DP, Donnelly LF, Rosen NS (2003) Computed tomography and radiation risks: what pediatric health larly well selected and justified based on the care providers should know. Pediatrics 112:951–957 ALARA principle. As children offer unique US Garcia Peña BM, Cook EF, Mandl KD (2004) Selective scanning conditions many queries that may need imaging strategies for the diagnosis of appendicitis in children. Pediatrics 113:24–28 a CT in adults can be sufficiently addressed by sonography; furthermore MRI has become avail- Goske MJ, Applegate KE, Boylan J et al (2008) The ‘Image Gently’ campaign: increasing CT radiation able for many pediatric queries too and thus has dose awareness through a national education and helped to replace in CT in some scenarios. If a awareness program. Pediatr Radiol 38:265–269 CT is performed, the study must be planned and Granata C, Magnano G (2013) CT in pediatric oncology. adapted individually using dedicated, size-­ EJR Eur J Radiol 82:1098–1107 Greess H, Lutze J, Nömayr A, Wolf H, Hothorn T, Kalender adapted pediatric protocols, and should be taiWA, Bautz W (2004) Dose reduction in subsecond lored to the individual query, with utmost multislice spiral CT examination of children by online tube current modulation. Eur Radiol 14:995–999 precautions concerning radiation protection such as avoiding multiple phases and proper age- and Hellinger JC, Pena A, Poon M, Chan FP, Epelman M (2010) Pediatric computed tomographic angiography: organ-adapted contrast timing. And finally never imaging the cardiovascular system gently. Radiol Clin “try a CT”—make sure that it will be diagnostic N Am 48:439–467 even if restrainments or sedation is necessary to Herzog C, Mulvihill DM, Nguyen SA et al (2008) Pediatric cardiovascular CT angiography: radiation grant a successful scan.

CT of the Pediatric Abdomen dose reduction using automatic tube current modulation. AJR Am J Roentgenol 190:1232–1240 Huda W, Vance A (2007) Patient radiation doses from adult and pediatric CT. AJR Am J Roentgenol 188:540–546 Image Gently—The Alliance for Radiation Safety in Pediatric Imaging. Development of Pediatric CT Protocols 2014. http://www.imagegently.org/Portals/6/ Procedures/IG. CT Protocols, last accessed 13 Jan 2017 Khawaja RD, Singh S, Otrakji A, Padole A, Lim R, Nimkin K, Westra S, Kalra MK, Gee MS (2015) Dose reduction in pediatric abdominal CT: use of iterative reconstruction techniques across different CT platforms. Pediatr Radiol 45:1046–1055 Kleinerman RA (2006) Cancer risks following diagnostic and therapeutic radiation exposure in children. Pediatr Radiol 36(Suppl 2):S121–S125 MacDougal RD, Kleinman PL, Callahan MJ (2016) Size based protocol optimization using automated tube current modulation and automated KV-selection in computed tomography. J Appl Clin Med Phys 17:5756 Mathews JD, Forsythe AV, Brady Z, Butler MW, Goergen SK, Byrnes GB, Giles GG, Wallace AB, Anderson PR, Guiver TA, McGale P, Cain TM, Dowty JG, Bickerstaffe AC, Darby SC (2013) Cancer risk in 680,000 people exposed to CTs in childhood or adolescence: data linkage study of 11 million Australians. BMJ 346:f2360 Nagy E, Marterer R, Tschauner S, Lindbichler F, Sorantin E (2016) Adaptierung des Schwellenwertes für das Bolus-Tracking bei der CT Angiographie unter Berücksichtigung einer verringerten Röhrenspannung. Radiologe 56:754. (abstract) National Cancer Institute (USA). Radiation risks and pediatric computed tomography (ct): a guide for health care providers. http://www.cancer.gov/cancertopics/ causes/radiation-risks-pediatric-CT. Last accessed 12 Feb 2017 Nievelstein RAJ, van Dam IM, van der Molen AJ (2010) Multidetector CT in children: current concepts and dose reduction strategies. Pediatr Radiol 40:1324–1344 Paterson A, Frush DP, Donnelly LF (2001) Helical CT of the body: are settings adjusted for pediatric patients? AJR Am J Roentgenol 176:297–301 Pearce MS, Salotti JA, Little MP, McHugh K, Lee C, Kim KP, Howe NL, Ronckers CM, Rajaraman P, Craft AW, Parker L, Berrington de González A (2012) Radiation exposure from CTs in childhood and subsequent risk of leukaemia & brain tumours: retrospective cohort study. Lancet 380:499–505 Pilhatsch A, Riccabona M (2011) Role and potential of modern ultrasound in pediatric abdominal imaging. Imaging Med 3:393–410. (18) Rajaraman P, Simpson J, Neta G, Berrington de Gonzalez A, Ansell P, Linet MS, Ron E, Roman E (2011) Early life exposure to diagnostic radiation and ultrasound scans and risk of childhood cancer: case–control study. BMJ 342:d472

1047 Riccabona M (2013) CT in children: why and what to consider for CT in children (editorial). EJR Eur J Radiol 82:1041–1042 Riccabona M (2014a) Diagnostic Ultrasonography in neonates, infants and children—why, when and how. EJR Eur J Radiol 83:1485–1486 Riccabona M (2014b) Basics, principles, techniques and modern methods in paediatric ultrasonography. EJR Eur J Radiol 83:1487–1494 Riccabona M, Avni FE, Blickman JG, Dacher JN, Darge K, Lobo ML, Willi U (Members of the ESUR paediatric paediatric recommendation work group and ESPR paediatric uroradiology work group). Imaging recommendations in paediatric uroradiology, part II: urolithiasis and haematuria in children, paediatric obstructive uropathy, and postnatal work-up of foetally diagnosed high grade hydronephrosis. Minutes of a mini-symposium at the ESPR annual meeting, Edinburg, June. Pediatr Radiol 2009; 39: 891–898 Riccabona M, Avni FE, Dacher JN, Damasio B, Darge K, Lobo ML, Ording-Müller LS, Papadopolos F, Willi U (2010) ESPR uroradiology task force and ESUR ­paediatric working group: imaging and procedural recommendations in paediatric uroradiology, part III. Minutes of the ESPR uroradiology task force mini-symposium on intravenous urography, uro-­CT and MR-urography in childhood. Pediatr Radiol 40:1315–1320 Rogers LF (2001) Taking care of children: check out the parameters used for helical CT. AJR Am J Roentgenol 176:287–287 Shah NB, Platt SL (2008) Alara: is there a cause for alarm? Reducing radiation risks from computed tomography scanning in children. Curr Opin Pediatr 20:243–247 Shah R, Gupta AK, Rehani MM, Pandey AK, Mukhopadhyay S (2005) Effect of reduction in tube current on reader confidence in paediatric computed tomography. Clin Radiol 60:224–231 Singh S, Kalra MK, Shenoy-Bhangle AS, Saini A, Gervais DA, Westra SJ, Thrall JH (2012) Strahlendosisreduktion mit hybrider iterativer Rekonstruktion für pädiatrische CT. (radiation dose reduction with hybrid iterative reconstruction for pediatric CT). Radiologe 263:537–546 Siripornpitak S, Pornkul R, Khowsathit P, Layangool T, Promphan W, Pongpanich B (2013) Cardiac CT Angiography in children with congenital heart disease. EJR Eur J Radiol 82:1067–1082 Slovis TL (2002) The ALARA (as low as reasonably achievable) concept in pediatric CT intelligent dose reduction. Multidisciplinary conference organized by the Society of Pediatric Radiology. ALARA conference proceedings. Pediatr Radiol 32:217–317 Sorantin E (2008) Soft-copy display and reading: what the radiologist should know in the digital era. Pediatr Radiol 38:1276–1284 Sorantin E, Wießpeiner U (2010) Dose savings in Computed Tomography due to a new dedicated Kernel image reconstruction—influence on image quality. Pediatr Radiol 40(Suppl):S1100. (abstract)

1048 Sorantin E, Zsivcsec B, Zebedin D, Fotter R (2002) Optimierung von i.V. Kontrastmittel-applikatonen für pädiatrische spiral-CT Untersuchungen [optimisation of i.V.Contrast application for pediatric spiral-CT]. Radiologe 45:683–684 Sorantin E, Hasenburger G, Weissensteiner S, Riccabona M (2013a) CT in Children—Dose protection and general considerations when planning a CT in a Child. EJR Eur J Radiol 82:1043–1049 Sorantin E, Riccabona M, Stücklschweiger G, Guss H, Fotter R (2013b) First experience with volumetric (320 row) pediatric CT. EJR Eur J Radiol 82: 1091–1097 Sorantin E, Weissensteiner S, Oppelt B, Bompoit A, Fister N (2015) Paediatric CT: contrast agent application & modern development. Pediatr Radiol 45(Suppl 2):S247–S368 Stöver B, Rogalla P (2008) CT-Untersuchungen bei Kindern [CT examinations in children]. Radiologe 48:243–248

M. Riccabona and A. Pilhatsch Strauss KJ (2014) Developing patient-specific dose protocols for a CT scanner and exam using diagnostic reference levels. Pediatr Radiol 44(Suppl 3):S479–S488 Strauss KJ, Goske MJ (2011) Estimated pediatric radiation dose during CT. Pediatr Radiol 41(Suppl 2):S472–S482 Strauss KJ, Goske MJ, Kaste SC, Bulas D, Frush DP, Butler P, Morrison G, Callahan MJ, Applegate KE (2010) Image gently: ten steps you can take to optimize image quality and lower CT dose for pediatric patients. AJR Am J Roentgenol 194:868–873 Thomas KE, Wang B (2008) Age-specific effective doses for pediatric MSCT examinations at a large children’s hospital using DLP conversion coefficients: a simple estimation method. Pediatr Radiol 38:645–656 Vock P (2005) CT dose reduction in children. Eur Radiol 15:2330–2340 Zhu X, McCullough WP, Mecca P, Servaes S, Darge K (2016) Dual energy compared to single energy CT n pediatric imaging: a phantom study for DECT clinical guidance. Pediatr Radiol 46:1671–1679

Part VIII Miscellaneous Topics

Emergency CT Samad Shah, Sunil Jeph, and Savvas Nicolaou

Contents 1    Introduction

 1052

2    Trauma Imaging Technique 2.1  Use of Intravenous Contrast 2.2  Use of Oral Contrast 2.3  Radiation Dose 2.4  Total-Body Protocol 2.5  Role of Computed Tomography Angiography (CTA) 2.6  Total-Body CT Versus Standard Workup 2.7  Split-Bolus Single-Pass CT 2.8  Dual-Energy CT (DECT)

 1052  1053  1053  1053  1053

3    Role of Other Modalities 3.1  Conventional Radiographs 3.2  Ultrasound (US) 3.3  MRI

 1055  1055  1055  1055

4    Neurological Trauma 4.1  Skull Fracture 4.2  Temporal Bone Fracture 4.3  Facial Fractures 4.4  Extra-axial Hemorrhage 4.5  Intra-axial Injuries 4.6  Secondary Traumatic Injury 4.7  Cervical Trauma

 1055  1056  1056  1057  1059  1061  1062  1063

5    Chest Trauma 5.1  Pneumothorax (PTX) 5.2  Hemothorax 5.3  Pulmonary Contusion

 1065  1065  1066  1066

 1054  1054  1055  1055

S. Shah, MD (*) • S. Jeph, MD Department of Radiology, Geisinger Medical Center, Danville, PA, USA e-mail: [email protected] S. Nicolaou, MD Associate Professor of UBC, Director of Emergency/ Trauma Imaging Vancouver General Hospital, University of British Columbia, Vancouver, BC, Canada

5.4  Traumatic Aortic Injury

 1067

6    Body Trauma 6.1  Spleen 6.2  Liver 6.3  Pancreas

 1068  1068  1070  1071

7    Genitourinary Trauma 7.1  Renal 7.2  Bladder

 1072  1072  1073

References

 1074

Abstract

Trauma is one of the leading causes of death in people under the age of 44 years (Rivara et al. 2015). Trauma also has a substantial financial impact on the healthcare system, accounting for over one-third of all emergency department visits and resulting in over $80 billion per year in direct medical care costs in the USA (WISQARS 2016). A significant cause of preventable deaths in trauma is delay in proper surgical care. Significant injuries are overlooked at a high rate in patients with major trauma (Rivara et al. 2015). In particular, the physical examination is unreliable in patients with a reduced level of consciousness (Rivara et al. 2015). With the marked decrease in the use of diagnostic peritoneal lavage, diagnosis of abdominal injuries now relies almost exclusively on the interpretation of computed tomography (CT) examinations (Catre 1995). As a member of the trauma team, the radiologist plays a significant role by contributing

Med Radiol Diagn Imaging (2016) DOI 10.1007/174_2016_88, © Springer International Publishing Switzerland Published Online: 26 November 2016

1051

S. Shah et al.

1052

to the rapid diagnosis of critical and emergent diagnoses (Clarke et al. 2002; Schueller et al. 2015). Until a few years ago, CT scanning was performed at the end of the clinical evaluation of the patient. However, with the advance of high-speed assessment by multidetector CT (MDCT), scanning can now be performed immediately upon patient arrival in the trauma bay. As a result, the mortality rate has significantly decreased, especially in cases of severe trauma (Schueller et al. 2015; Jiang et al. 2014).

1

Introduction

Trauma is one of the leading causes of death in people under the age of 44 years (Rivara et al. 2015). Trauma also has a substantial financial impact on the healthcare system, accounting for over one-third of all emergency department visits and resulting in over $80 billion per year in direct medical care costs in the USA (WISQARS 2016). A significant cause of preventable deaths in trauma is delay in proper surgical care. Significant injuries are overlooked at a high rate in patients with major trauma (Rivara et al. 2015). In particular, the physical examination is unreliable in patients with a reduced level of consciousness (Rivara et al. 2015). With the marked decrease in the use of diagnostic peritoneal lavage, diagnosis of abdominal injuries now relies almost exclusively on the interpretation of computed tomography (CT) examinations (Catre 1995). As a member of the trauma team, the radiologist plays a significant role by contributing to the rapid diagnosis of critical and emergent diagnoses (Clarke et al. 2002; Schueller et al. 2015). Until a few years ago, CT scanning was performed at the end of the clinical evaluation of the patient. However, with the advance of high-speed assessment by multi-detector CT (MDCT), scanning can now be performed immediately upon patient arrival in the trauma bay. As a result, the mortality rate has significantly decreased, especially in cases of severe trauma (Schueller et al. 2015; Jiang et al. 2014).

Dedicated organ injury scores (OIS) have been developed for CT imaging which helps the trauma team predict whether the patient will need surgery or be managed conservatively (Schueller et al. 2015). These will be discussed separately in their respective sections.

2

Trauma Imaging Technique

MDCT is considered the imaging modality of choice for the initial evaluation of patients with multiple injuries (Rieger et al. 2002; Topp et al. 2015) and is usually physically located either within or close to trauma resuscitation room. CT is widely available and can quickly identify a wide array of injuries, making it ideal for the emergency department setting (Kelleher et al. 2015). In patients with polytrauma, whole-body CT (aka pan-CT or total-body CT), which includes CT of the head, neck, chest, abdomen, and pelvis, has become the standard of care to diagnose and ascertain the severity of the injuries. Whole-body CT has been shown to have a negative predictive value of up to 99 % in penetrating abdominal trauma (Ramirez et al. 2009). Possible disadvantages are that it results in higher radiation doses and impairs the concurrent performance of lifesaving procedures such as chest tube insertion, resuscitation with cardiac massage, or laparotomy/thoracotomy (Topp et al. 2015). CT is the initial imaging modality of choice for evaluation of traumatic brain injury. CT has a high sensitivity for detection of intra-/extra-axial blood, fracture, ventricle size, and mass effect (Davis and Expert Panel on Neurologic Imaging 2007). However, it has a low sensitivity for detection of nonhemorrhagic lesions, such as contusions, diffuse axonal injury, and hypoxic ischemic encephalopathy. CT plays an important role in detecting secondary signs of trauma, such as edema, ischemia, and herniation (Lolli et al. 2016). There is a consensus about early non-­ enhanced contrast CT of the head for moderate and severe traumatic brain injuries. However, the role of CT in minor head trauma is controversial. In children less than 2 years of age, given the

Emergency CT

d­ifficulty of assessment of neurological status, early CT is the appropriate step (Lolli et al. 2016). Although MRI is better than CT in detecting focal traumatic brain lesions, it is not the initial modality of choice due to logistics of patient transport, long scan time, and patient motion artifacts. It is used for problem-solving if the patient has neurological symptoms not explained by CT (Provenzale 2007). The technologist has to search and remove items, such as necklaces, that could result in unwanted artifacts in the imaging study. Prior to the injection of iodinated contrast material, every intravenous (IV) access should be checked with an injection of saline. Patient instruction should be clearly communicated, if feasible, to avoid motion during the scan.

2.1

Use of Oral Contrast

Oral contrast material is not indicated as it delays the imaging study and potentially lifesaving procedure. Thus, oral contrast is no longer administered at most large trauma centers in the setting of blunt trauma (Stuhlfaut et al. 2004).

2.3

injuries seen in the portal venous phase. Automated dose modulation is used routinely with modern CT scanners to decrease radiation exposure (Geyer et al. 2016). Lastly, newer ­techniques, such as split-bolus single-pass CT, dramatically reduce dose to the patient.

2.4

Total-Body Protocol

A scan of the head and cervical spine is performed first followed by the thorax, abdomen, and pelvis with contrast. The body protocol includes two phases: a bolus-tracking arterial phase acquisition from the lung apices to the pubic symphysis followed by a manually triggered venous phase from the diaphragm to the pubic symphysis acquired 50 s after completion of the arterial acquisition (Leung et al. 2015).

Use of Intravenous Contrast

Before scanning the body, all trauma patients should receive a bolus of intravenous contrast material, typically 100–150 mL ideally injected at a rate of 3–5 mL/s through an 18- or 20-gauge cannula located in a large peripheral vein (Soto and Anderson 2012).

2.2

1053

Radiation Dose

Since the trauma population often involves young patients, it is important to balance the risk of radiation exposure with the benefits of an examination. Several techniques are employed to achieve this task. The number of phases acquired, beyond the routine arterial and portal venous phases, should be carefully selected. Delayed phase series should be limited to patients with visceral

Head and Cervical Spine In a few seconds, CT allows us to detect the most traumatic cranioencephalic conditions and complications as well as guide the surgical treatment (Smith et al. 2007). Non-contrast head CT is acquired at 120 kv and 200–400 mA in adults. Two sets of axial data must be acquired using the soft tissue and bone algorithm, respectively. The scan is done with arms positioned down. Thin slices (usually 1 mm) reconstructed at 0.5-mm increment are obtained for the bone window using a very sharp kernel. Thicker slices (usually 5.0 mm) are typically reconstructed at 5-mm increment using the soft tissue kernel (Eichler et al. 2015). Thin-section (1-mm) MDCT with the bone algorithm and 3D reconstruction is recommended for evaluation of the skull base, orbit, and facial bone fractures (Soto et al. 2004). 3D reformats are mostly used in difficult cases, especially in complex facial temporal bone fractures. There are two ways of 3D reconstructions: surface-­ shaded display and volume rendering; both improve detection of fracture difficult to appreciate on axial images. Chest, Abdomen, and Pelvis CT acquisition is acquired from the middle of the seventh cervical spinal column to the level of the

S. Shah et al.

1054

proximal femur. The arms are positioned over the head (Eichler et al. 2015). An automated pump injector is usually used to inject a fixed quantity of iodinated contrast. A typical multi-trauma CT protocol includes an arterial phase through the chest, acquired at 25–35 s, followed by portal venous phase images of the abdomen and pelvis, acquired at 65–80 s. A growing number of institutions acquire arterial phase images through the abdomen and pelvis in addition to the chest (Uyeda et al. 2010; Boscak and Shanmuganathan 2012). Arterial phase images enable detection of trauma to the major vessels and demonstrate visceral vascular injuries that may not be apparent on portal venous images. Delayed imaging can be used to detect injury to the renal collecting system. It is usually only performed in patients with evidence of solid organ injury in the portal venous phase images (Eichler et al. 2015) to limit the amount of radiation delivered (Stuhlfaut et al. 2006). Finally, patients suspected of having a bladder injury should undergo CT cystography. Full (at least 300–400 mL of diluted water-­ soluble contrast) distention of the bladder through a Foley catheter is required for higher sensitivity (Peng et al. 1999). For soft tissue and lung windows, images are reconstructed using 5-mm slice thickness using smooth and sharp kernels, respectively (Eichler et al. 2015). The bone window reconstructions are generated using thinner (1- or 2-mm) slice thickness using very sharp kernel. Subsequently, sagittal and coronal reconstructions are obtained (Eichler et al. 2015).

2.5

 ole of Computed Tomography R Angiography (CTA)

Traumatic neck vascular injury is a common cause of morbidity accounting for 10 % of all trauma injuries with a mortality of 3–10 % (Morales-Uribe et al. 2016). Patients with active bleeding, massive hemoptysis, stridor, and growing neck hematoma require urgent surgery. Hemodynamically stable patients need additional evaluation (Uyeda et al. 2010; Morales-Uribe

et al. 2016). Conventional angiography was considered as the gold standard for the diagnosis of traumatic vascular injuries. However, it is an invasive procedure that is not widely available and has potential for iatrogenic vessel injury (Morales-Uribe et al. 2016). CT angiography is a noninvasive, safe, fast, and highly accurate method for detecting vascular trauma with a sensitivity and specificity higher than 97 % (Morales-­ Uribe et al. 2016). It can detect both direct and indirect signs of vascular injury with simultaneous evaluation of other organs.

2.6

Total-Body CT Versus Standard Workup

Until a decade ago, standard workup for hemodynamically stable polytrauma patients consisted of plain radiographs of the chest, cervical spine, pelvis, and an abdominal ultrasound. Only those with positive results in one of these tests got an organ-focused CT or total-body CT (Kelleher et al. 2015). This approach is now considered ineffective and inefficient in the assessment of high deceleration injury (Sierink et al. 2012). Furthermore, in the past few years, multiple retrospective studies have shown improved survival in trauma patients undergoing total-body CT scanning (van Vugt et al. 2012; Healy et al. 2014). An increasing number of trauma centers have incorporated total-body CT imaging into their daily practice for the assessment of trauma patients (Topp et al. 2015; Sierink et al. 2016). However, some studies have failed to show a significant survival benefit of total-body CT scan compared with standard radiological workup except in severely injured patients or high-­ velocity trauma (Kelleher et al. 2015; Sierink et al. 2016). Therefore, in patients where CT examinations of several body regions were expected, total-body CT scanning was beneficial due to time benefit. Furthermore, such patients received a similar or lower radiation dose with total-body CT scanning without an increase in medical costs (Sierink et al. 2016). Currently, there are no clear guidelines as to who gets a pan­CT as opposed to traditional workup.

1055

Emergency CT

2.7

Split-Bolus Single-Pass CT

The split-bolus technique provides both arterial and venous enhancement within a single body scan usually from the circle of Willis to the pubic symphysis. The contrast medium is injected in two phases with an initial slow injection providing portal venous enhancement and a second faster bolus giving arterial enhancement (Leung et al. 2015; Eichler et al. 2015). The key benefit from using the split-bolus protocol is the dramatic, almost 50 %, dose reduction achieved while maintaining both arterial and venous enhancement (Leung et al. 2015). Some radiologists favor the traditional two-­ phase protocol due to its familiarity; whereas others feel that split-bolus protocol is advantageous as it prevents image overload and reduces reporting time (Leung et al. 2015).

2.8

Dual-Energy CT (DECT)

DECT imaging has recently become available and is based on the principle that different materials have different attenuation coefficients when exposed to X-ray beams of different mean energy (McLaughlin et al. 2015). Therefore, DECT can distinguish different materials that would have comparable attenuation on conventional CT. For cervical CTA, it provides faster and more accurate bone subtraction (Morhard et al. 2009) and better visualization of intraosseous segments of carotid and vertebral vessels (Deng et al. 2009). DECT with non-calcium images can be used to distinguish between acute and chronic compression fractures (Karaca et al. 2016). DECT, compared to MRI, requires less time, has better spatial resolution, and has comparable accuracy for evaluation of acute vertebral fractures. DECT can be used as an alternative in patients with contraindications to MRI (Karaca et al. 2016). However, MRI is superior for fractures near end plates (Karaca et al. 2016). DECT has superior diagnostic value in differentiating contrast and blood compared to regular CT (Tan et al. 2016). It has shown to differentiate hematoma and contrast extravasation which may

both appear as hyperdense collections with high sensitivity, specificity, and positive predictive value (Tan et al. 2016). DECT, therefore, can be utilized in evaluation for hemorrhage in patients who have recently received contrast.

3

Role of Other Modalities

3.1

Conventional Radiographs

Compared to MDCT imaging, plain radiography of the cervical spine, the thorax, and the pelvis has substantially lower sensitivity (Schueller et al. 2015). However, one obvious advantage of initial X-ray diagnosis of the chest in the resuscitation room is the possibility to perform lifesaving procedures like chest tube insertion, which can be performed immediately and simultaneously.

3.2

Ultrasound (US)

Some centers advocate for focused assessment with sonography for trauma (FAST) examinations in the resuscitation room, especially in unstable patients. Abundant free fluid in the abdomen on a FAST study often leads to the decision to perform an emergency laparotomy. Recently, US is being used to diagnose cardiac injuries from penetrating trauma (Topp et al. 2015; Körner et al. 2008).

3.3

MRI

MRI has essentially no role in trauma imaging. However, in a case of diffuse axonal injury (DAI), CT is accurate in only 30 % of cases. MRI of the head should be performed to rule out DAI in patients with a normal CT and low Glasgow Coma Scale (Deunk et al. 2007).

4

Neurological Trauma

The traumatic brain injury is a serious health concern with a high incidence of approximately 2 million per year in the USA, causing 50,000–

S. Shah et al.

1056

75,000 deaths and 70,000–90,000 permanent disabilities (Lolli et al. 2016; Udstuen and Claar 2001). Approximately 2000 are rendered in persistent vegetative state. The financial burden due to traumatic brain injury is 60 billion dollars in the USA alone (Udstuen and Claar 2001; Bodanapally et al. 2015). Motor vehicle accidents are the most common cause of head trauma followed by falls (Udstuen and Claar 2001). Emergency imaging plays a critical role in early detection of trauma as immediate neurosurgical intervention significantly improves outcome. Early imaging also plays an important role explaining the neurological symptoms and determining prognosis (Lolli et al. 2016; Provenzale 2007). Mortality in acute head injury is strongly correlated with GCS score at initial presentation. A high GCS score of 13–14 (normal is 15) indicates mild traumatic injury with 15 % of patients having persistent symptoms up to 1 year. ­ Whereas, a GCS of less than 8 indicates severe injury with up to 36 % mortality (Udstuen and Claar 2001). Since GCS and neurological examination are unreliable under sedation, CT scan scoring systems are utilized to determine early mortality after moderate to severe traumatic brain injury. Rotterdam scoring system (Table 1) includes additional variables like subarachnoid hemorrhage (SAH) and intraventricular hemorrhage (Lolli et al. 2016). The score from 0 to 6 Table 1  Rotterdam CT score Rotterdam score element Basal cisterns  Normal  Compressed  Absent Midline shift  No shift or shift ≤5 mm  Shift >5 mm Epidural mass lesion  Present  Absent Intraventricular blood or tSAH  Absent  Present Sum score tSAH traumatic subarachnoid hemorrhage

progressively portends worse prognosis (Deepika et al. 2015). Primary, traumatic brain lesions typically progress in the first 24 h with an increase in the size of parenchymal contusions in 25–45 % and development of new mass lesions in approximately 16 % patients with diffuse injuries. Hence, follow-up CT scan may be recommended (Lolli et al. 2016). Head injury can be divided into primary and secondary lesions (Table 2).

4.1

Skull Fracture

Skull fracture is seen in only 25 % of fatal head traumas (Hardman and Manoukian 2002) and can be closed or open, simple or comminuted, linear, depressed, elevated, or diastatic. Skull base fracture is associated with major complications including CSF leak, infection, cranial nerve injury, dural sinus, and internal jugular and internal carotid artery injury (Baugnon and Hudgins 2014). However, the presence of skull fracture itself does not indicate the severity of traumatic brain injury (Hardman and Manoukian 2002).

4.2

Temporal Bone Fracture

Temporal bone fractures have a higher risk of injury to the middle meningeal artery, causing an epidural hematoma (Provenzale 2007). New

Score 0 1 2 0 1 0 1 0 1 +1

Table 2  Classification of head injury (Lolli et al. 2016) Primary lesions: due to direct trauma to the head 1. Skull fractures 2. Extra-axial hemorrhage (a) Epidural hematoma (b) Subdural hematoma (c) Subarachnoid hematoma 3. Intra-axial injury (a) DAI (b) Cortical contusion (c) Intraparenchymal hematoma Secondary lesions: complication of primary lesions 1. Cerebral edema 2. Brain herniation

1057

Emergency CT Table 3  New temporal bone fracture classification Petrous 1. Involving otic capsule 2. Involving petrous apex with/without otic capsule

Non-petrous 1. Involving the middle ear 2. Involving mastoid with/without the middle ear

Fig. 2  CT face axial – bone algorithm. Comminuted, displaced fracture of bilateral nasal bones (arrows)

to thin slice acquisition. MDCT is sensitive for subtle non-displaced facial fractures and is useful to generate multiplanar and 3D reconstructions (Winegar et al. 2013). Fig. 1  CT face axial – bone algorithm. Linear, non-­ 4.3.1 Nasal Bone petrous longitudinal fracture involving the right mastoid Nasal bone fracture is the most common fracture extending toward the middle ear (arrows) due to it being most prominent projection on the

t­ emporal bone fracture classification which divides the fractures into petrous and non-petrous categories (Table 3) was found to be more clinically relevant compared to the old system of anatomic classification into transverse, longitudinal, and mixed (Little and Kesser 2006). Petrous fractures are more commonly associated with complications such as facial nerve injury, carotid injury, CSF leaks, and sensorineural hearing loss (Little and Kesser 2006) compared to non-petrous (Fig. 1).

4.3

Facial Fractures

MDCT is the modality of choice for evaluation of facial fractures because of its high resolution due

face, superficial location, and thin cortex (Fig. 2). It constitutes roughly half of all facial bone fractures (Salvolini 2002).

4.3.2 Le Fort Fractures Complex fractures involve multiple facial buttresses. There are three common patterns, each involving pterygoid plates. Depending on the mechanism and intensity of traumatic forces, multiple combinations of Le Fort fracture patterns may be seen in the same patient (Winegar et al. 2013). Type 1: It is also called as “floating palate” fracture because it involves separation of the hard palate from the remaining face. The fracture line is horizontally oriented and involves anterior, lateral, and medial maxillary walls. It

S. Shah et al.

1058

a

b

Fig. 3 (a) CT face coronal – bone algorithm. Le Fort type 1: linear right and mildly displaced left fracture of the nasal aperture (yellow arrows). (b) Fracture extends pos-

teriorly causing comminuted fractures of the bilateral maxilla and pterygoid plates (red arrows)

also traverses through the inferior margins of piriform aperture and nasal septum anteriorly (Fig. 3) (Winegar et al. 2013). Type 2: It is also called as “pyramidal” fracture since it forms a pyramidal shaped maxillary fracture segment that may move independently from the rest of the face. The oblique fracture line passes through the medial orbital wall, orbital floor, and zygomaticomaxillary suture. The zygomatic bone is spared (Winegar et al. 2013). Type 3: It is also called “craniofacial dissociation” since there is complete dissociation of the skull base and face. It involves fracture line extending from nasofrontal suture anteriorly toward medial and lateral orbital walls and zygomatic arch. Unlike type 1 and type 2, type 3 involves zygomatic bone (Winegar et al. 2013).

4.3.3 Zygomaticomaxillary Fracture Complex It is also called tripod, tetrapod, or malar fracture. It is caused by direct trauma to malar eminence causing fracture of the zygomatic arch and inferior and lateral orbital rims (Fig. 4) (Schuknecht and Graetz 2005).

Apart from Le Fort fractures, there are two other major fracture types that involve facial bones.

4.3.4 Orbital Blowout Fracture It is caused by blunt trauma to the eyeball causing fracture of the orbital floor (Fig. 5). Sometimes “inferior blowout fractures” are associated with “medial blowout fracture,” displacing the medial orbital wall (lamina papyracea) in the ethmoid sinus (Winegar et al. 2013). 4.3.5 Management Disruption of facial buttress can change the orientation of facial bones and cause problems with normal function. The treatment is typically fixation with rigid titanium plates and screws

Emergency CT

1059

anchored into buttress, with or without bone graft. The aim is to restore and stabilize the 3D structural anatomy with acceptable mastication function and appearance of overlying soft tissue.

4.4

Fig. 4  Non-displaced fractures involving the lateral (yellow arrow) and inferior (red arrow) orbital walls. This patient also had a zygomatic arch fracture (not shown). There is also a fracture of the lateral maxillary wall (blue arrow)

Extra-axial Hemorrhage

4.4.1 Epidural Hemorrhage (EDH) EDH occurs at the site of impact and is usually associated with fracture in >90 % cases. They have an overall mortality rate of 5 %. They are mostly supratentorial in location, >95 %, and commonly involve parietal and temporal bones causing disruption of the middle meningeal artery. Infratentorial EDHs are rare and mostly venous in origin, caused by disruption of dural veins and sinuses. EDHs do not cross cranial suture lines although they freely extend across the dural reflections (Lolli et al. 2016; Bodanapally et al. 2015). Imaging  On acute presentation, EDH typically appears as hyperattenuating, biconvex, extra-axial collection on an NECT (Fig. 6). The biconvex shape is due to the firm attachment of the dura to the skull. It may also appear heterogeneous with areas of low attenuation due to unclotted blood, suggesting active extravasation (swirl sign) (Zimmerman and Bilaniuk 1982). Management  EDH is a neurosurgical emergency, which requires immediate evacuation to prevent or relieve mass effect (Provenzale 2007).

Fig. 5  CT face coronal. Depressed, comminuted left orbital floor blowout fracture (yellow arrow). The left inferior rectus muscle along with small amount of extraconal fat is inferiorly displaced (red arrow)

4.4.2 Subdural Hemorrhage (SDH) SDH is seen in 12–29 % of patients with severe traumatic brain injury (TBI). They have a high mortality of up to 60 % (Zumkeller et al. 1996). Most common causes of SDH include motor vehicle accidents (MVA), falls, and assaults. They are mostly venous in origin and caused by rupture of bridging cortical veins due to shear forces (Lolli et al. 2016; Gennarelli and Thibault 1982). Imaging  Acute SDH (first week) appears as hyperattenuating, crescent-shaped collection

1060

Fig. 6  CT head axial. Biconvex, hyperdense epidural hemorrhage (red arrows) in a patient with comminuted left-sided calvarial fractures (not shown). Also seen is countercoup cerebral contusion in right superior temporal lobe (yellow arrow)

between the inner table of the skull and cerebral cortex (Fig. 7). The loose attachment between the dura and arachnoid allows the blood to flow freely and assume a crescentic configuration over the cerebral convexity. SDHs are mostly supratentorial in location and are frequently seen along the falx and tentorium overlying the cerebral convexity. Unlike EDH, SDHs cross suture lines but never cross the dural reflections. In subacute phase (2nd and 3rd week), SDH becomes isoattenuating to gray matter making it difficult to identify. Careful inspection of CT may reveal medial displacement of gray–white matter junction, effacement of cortical sulci, and an inability of surface sulci to reach the inner table of the skull (Bodanapally et al. 2015). Management  SDH can cause significant mass effect resulting in diffuse cerebral edema of underlying brain parenchyma, midline shift, and subfalcine herniation (Lolli et al. 2016).

S. Shah et al.

Fig. 7  CT head axial – blood window. Hyperdense blood layering over the left cerebral convexity sulci (red arrows). There is also a small crescent-shaped subdural hematoma overlying the left frontal lobe (yellow arrows)

4.4.3 Subarachnoid Hemorrhage (SAH) Traumatic SAH (tSAH) is seen in 40 % of patients with moderate to severe TBI. Small tSAHs are caused by disruption of blood vessels in subarachnoid space (Murray et al. 1999). tSAH on hospital admission in patients with TBI is associated with increased morbidity and mortality as high as twofold (Armin et al. 2006; Eisenberg et al. 1990). Imaging  Large SAHs are usually caused by direct extension of hemorrhagic contusion and aneurysmal SAH. Contrary to aneurysmal SAH, traumatic SAHs are usually peripheral, overlying cerebral cortex (Fig. 7). The blood is seen as curvilinear foci of increased attenuation in sulci, Sylvian fissure, and cisterns (Lolli et al. 2016). Management  Calcium channel blockers (CCB) are classically thought to interfere with the influx of calcium into the smooth muscle cell, therefore preventing blood vessel constriction and

Emergency CT

theoretically reducing the risk of vasospasm and secondary ischemia after TBI. As tSAH may primarily be an early indicator of brain damage, some institutions favor vigilant diagnostic surveillance including serial head CT and prevention of secondary brain injury owing to hypotension, hypoxia, and intracranial hypertension than attempting to treat vasospasm associated with tSAH (Armin et al. 2006).

1061

The most common site of the cerebral contusion is at the inferior frontal lobe and anterior temporal lobe due to gliding of the brain over the irregular surface of anterior and middle cranial fossa.

Imaging Findings  Contusions are typically multifocal. Cerebral contusions can be hemor­ rhagic or nonhemorrhagic. Nonhemorrhagic contusions are difficult to detect on CT and may appear as small foci of hypoattenuation caused by 4.5 Intra-axial Injuries vasogenic edema. Hence, MR brain is the preferred modality. Hemorrhagic contusions 4.5.1 Cerebral Contusion appear as small foci of high attenuation in the Cerebral contusions are seen in 43 % of blunt gyral crest mostly associated with an ill-defined head traumas and constitute 50 % of all intra-­ area of hypoattenuation caused by vasogenic axial traumatic injuries (Gentry et al. 1988). edema (Lolli et al. 2016). Cerebral contusions are bruises to the brain involving the crowns of gyri causing necrosis and Management  25–45 % of contusions evolve and hemorrhage (Fig. 6) (Hardman and Manoukian manifest as hemorrhages, which progressively 2002). Contusions can occur in a coup or coun- increase over time, also known as the hemorrhagic tercoup fashion and can increase in size on subse- progression of contusion (Beaumont and quent imaging (Fig. 8). Coup contusions occur at Gennarelli 2006). Fifteen percent of patients the site of direct trauma and are usually smaller have delayed parenchymal hemorrhage (within than the countercoup contusion that occurs on the 6–9 h) in areas of the brain that were initially opposite side of the trauma. nonhemorrhagic. Therefore, follow-up CT should be timed accordingly (Narayan et al. 2008).

4.5.2 Intraparenchymal Hematoma (IPH) IPHs and cerebral contusions are part of the same spectrum of injuries. Like cerebral contusion, IPHs are multifocal and predominantly involve frontal and temporal lobes. IPHs, however, are less common than contusions and are seen in 15 % of acute head injuries (Fig. 9). IPHs may be observed without associated cerebral contusions in patients with penetrating trauma (Lolli et al. 2016).

Fig. 8  CT head axial – blood window. This is the same patient as in Fig. 6. There has been an increase in the size of right superior temporal lobe contusion

4.5.3 Diffuse Axonal Injury (DAI) DAI, also called “axonal stretch injury,” results from acceleration forces in coronal and sagittal planes causing shear forces on axons. CT has a very poor sensitivity in detecting DAI, especially the nonhemorrhagic type. For the hemorrhagic type, the sensitivity of CT further decreases after acute phase as degrading blood products become isoattenuating to brain parenchyma (Lolli et al.

S. Shah et al.

1062

2016; Bodanapally et al. 2015). The severity of trauma can be correlated to the anatomic localization of lesion, with more severe trauma involving deeper brain structures (Bodanapally et al. 2015; Adams et al. 1989). Imaging  DAI lesions are typically small (1–15 mm), ovoid with long axis parallel to axonal tracts, multiple, small focal hemorrhagic and nonhemorrhagic lesions. They involve gray– white matter junction, corpus callosum, deep periventricular white matter, the dorsolateral aspect of the midbrain, and upper pons (Bodanapally et al. 2015). Nonhemorrhagic DAIs shows small foci of hypodensity due to associated edema secondary to underlying trauma. Hemorrhagic DAIs show tiny foci of hyperdensity due to blood products (Lolli et al. 2016; Bodanapally et al. 2015). Management  DAI has a very low mortality, but it is the most frequent cause of persistent vegetative

Fig. 9  CT head axial – blood window. Large left frontoparietal intraparenchymal hematoma with surrounding hypodense vasogenic edema (yellow arrows). Also seen is mildly displaced left frontal bone fracture with overlying scalp hematoma (red arrow). There is near-complete effacement of the left lateral ventricle and rightward midline shift (blue arrow)

state due to trauma (Bodanapally et al. 2015). MRI is the investigation of choice for evaluating DAIs and is typically performed in unconscious patients with head CT showing no evidence of traumatic injury.

4.6

Secondary Traumatic Injury

4.6.1 Cerebral Edema Post-traumatic brain edema may be focal or diffuse and is seen in 10–20 % of patients with TBI. Post-traumatic cerebral edema is found to be both intracellular (cytotoxic) and extracellular (vasogenic). It typically has a delayed onset, appearing approximately 24–48 h after initial trauma (Osborn 2016). Imaging  Early on, the attenuation and gray–white matter differentiation are preserved. As edema progresses, the gray–white matter differentiation is lost, and the attenuation of the brain parenchyma decreases (Fig. 10). The falx and cerebral blood vessels appear relatively dense relative to the swollen brain parenchyma (Osborn 2016).

Fig. 10  CT head axial. Diffuse hypodense appearance of the cerebrum with foci of tiny intraparenchymal hemorrhage in the left frontal lobe. Edema causes effacement of the suprasellar cistern consistent with uncal herniation. There is associated depressed fracture of the left frontal bone

Emergency CT

Management  Decompressive craniotomy is the treatment of choice (Honeybul et al. 2011).

4.6.2 Cerebral Herniation Herniation is the displacement of the brain from one cranial compartment to another due to the mass effect caused by intracranial injury. Subfalcine Herniation  It is the most common type of herniation, also called as “midline shift” caused by displacement of anterior cingulate gyrus under the falx cerebri (Fig. 9). CT findings include compression of ipsilateral lateral ventricle due to mass effect and dilation of contralateral ventricle due to obstruction of the foramen of Monro (Lolli et al. 2016). In more severe cases, compression of pericallosal arteries may result in anterior cerebral artery (ACA) territory infarct (Lolli et al. 2016). Descending Transtentorial Herniation It is the second most common type of herniation, caused by the caudal descent of the brain tissue through tentorial incisura. The uncus of temporal lobe descends over the free margin of incisura into the ipsilateral suprasellar cistern. As the mass effect progresses, ipsilateral hippocampus herniates into quadrigeminal cistern causing mass effect on the midbrain. Displacement of the midbrain can also cause compression of contralateral cerebral peduncle against the tentorium (Derakhshan 2009). Mass effect on posterior cerebral artery may lead to infarction. Compression of the oculomotor nerve can cause ipsilateral nerve palsy. Compression or stretching of basilar artery perforators causes hemorrhagic midbrain infarct also known as Duret hemorrhages. Depending on the distortion caused by cerebral aqueduct, herniation can cause obstructive hydrocephalus (Lolli et al. 2016). Tonsillar Herniation  It is caused by downward herniation of cerebellar tonsil and medial aspect of cerebellar hemisphere through foramen magnum into the cervical canal. CT findings include effacement of cisterna magna and CSF space around the medulla oblongata. Complications caused by the mass effect include obstructive hydrocephalus and compression of the posterior inferior cerebellar artery (PICA)

1063

causing cerebellar infarct. Mass effect on ascending reticular activating system in the lower brain stem can result in impaired consciousness. Further progression causes mass effect on respiratory and cardiac centers in the brain stem leading to death (Johnson et al. 2002).

4.7

Cervical Trauma

4.7.1 Vascular Trauma Blunt ICA trauma is mostly seen in patients with motor vehicle accidents caused by stretching of the artery by rapid deceleration (Larsen 2002). The ICA typically stretches over lateral masses of the third and fourth cervical vertebral body and causes intimal tear which further progresses as dissection (Kraus et al. 1999). Cervical vertebral artery injury has various causes, including chiropractic manipulation, tennis, seat belt use, yoga, head banging, and kickboxing. Blunt carotid and vertebral artery injuries typically have a delayed onset of symptoms varying from hours to weeks after the trauma. Hence, it is imperative to evaluate for vascular injury initially to decrease mortality and mortality. Cervical CTA is the modality of choice (Lolli et al. 2016). Imaging Findings On CTA, pseudoaneurysms are contrast-filled saccular projections outside the arterial wall. Dissections appear as severe, irregular luminal narrowing of the vessel, also known as “string sign.” The external vessel diameter, however, is increased (Lolli et al. 2016; Pappas 2002). Intramural hematoma on NECT appears as crescent-­shaped hyperattenuating area in the vessel wall (Rodallec et al. 2008).

4.7.2 Vertebral Trauma The incidence of spine injuries is 5–10 % in patients with blunt polytrauma. Among all the traumatic spinal cord injuries, cervical spinal cord is involved in 55 %. Yearly cost of treating quadriplegic patients in the USA is $5.6 billion (Dreizin et al. 2014). MDCT is the investigation of choice for initial evaluation of patients with suspected spine

S. Shah et al.

1064

trauma. Identification of cervical spine injury should prompt evaluation of entire spine due to 10–15 % incidence of noncontiguous injury (Dreizin et al. 2014). Conventional radiography has a low sensitivity for detection of spine fractures compared to CT, that is, 43 % and 98.5 %, respectively (Nuñez et al. 1994). MR imaging has a higher sensitivity compared to CT for evaluation of disk and ligament injury but also has a high false-positive rate. The American College of Radiology (ACR) recommends cervical spine CT for patients with distracting injuries or other positive findings (National Guideline Clearinghouse 2016). Craniocervical Dissociation  Craniocervical dis­ sociation refers to atlantooccipital dissociation caused by severe ligamentous disruption, leading to unstable injury. Neurological deficits accompany it. Early identification and treatment are essential. It is often missed because it is a very subtle finding. Powers ratio must be used, which is a ratio of the distance from posterior margin of foramen magnum to anterior arch of atlas divided by the distance from the tip of basion to the posterior arch of C1. It has a sensitivity of 74 % if the upper limit of normal is considered to be 1 (Dreizin et al. 2014). Atlas Fracture Atlas is involved in 25 % of craniocervical injuries. Atlas fracture classification categorizes fractures into five types (Table 3) (Kakarla et al. 2010). Atlas fractures are mechanically stable unless they are associated with other cervical spine or ligamentous injuries (Kakarla et al. 2010). Axis Fracture Three main types of fracture patterns are seen, including odontoid fracture, Hangman fracture, and fracture of the body of the axis (Pryputniewicz and Hadley 2010). Odontoid fracture is the c­ ommonest fracture of axis and is involved in 59 % of axis fractures (Table 4). It is more common in elderly due to increased transmission of force through the dens caused by spine stiffness (Dreizin et al. 2014). There are three types of odontoid fractures (Table 5). Type 2 is the most common fracture (Fig. 11). It is

Table 4  Atlas fracture classification Type 1 Type 2 Type 3 (classic Jefferson burst fracture) Type 4 Type 5

Isolated fracture of posterior arch Isolated fracture of anterior arch Bilateral posterior arch fracture with unilateral or bilateral anterior arch fractures Lateral mass fracture Transverse anterior arch fracture caused by avulsion of longus colli or atlantoaxial ligament

Table 5  Classification of odontoid fractures Type 1 (least common) Type 2 (most common) Type 3

Oblique fracture through the tip of odontoid At the junction of dens and body of axis Fracture through the cancellous portion of C2

Fig. 11  CT cervical spine sagittal – bone algorithm. Linear, midly displaced fracture involving the base of the odontoid process (arrow)

associated with increased chances of nonunion with displacement >6 mm, age >50 years, and comminuted fracture (Dreizin et al. 2014). Hangman Fracture  It is also called bilateral pars interarticularis fracture. It is caused by compressive hyperextension and distractive hyperflexion. Hangman fracture is associated with neurological sequelae in 26 % of cases

Emergency CT

1065

(Mirvis et al. 1987). It is mostly treated with immobilization (Dreizin et al. 2014).

5

Chest Trauma

Thoracic injury is the third most common trauma, after head and limbs injuries, responsible for 25 % of death in polytrauma patients (Schueller et al. 2015). The chest wall is composed of bony, cartilaginous, and muscular structures that serve to protect vital internal organs and structures, such as the heart and great vessels. Traumatic lung injury can occur from a direct force, barotrauma, or rapid deceleration of the organs against the inner chest wall (Feden 2013). Such injuries include pneumothorax, hemothorax, and pulmonary contusion. In most centers, the chest radiograph remains an integral part of the quick imaging assessment of trauma patient. However, the ability of MDCT to detect many types of pathology that are not usually diagnosed radiographically has increased use of this modality.

5.1

Pneumothorax (PTX)

Introduction  Traumatic pneumothoraces have been found to occur in up to half of all cases of chest trauma, with only rib fractures being more common (Zarogoulidis et al. 2014). It occurs when the chest wall is pierced allowing air to enter the pleural space (Zarogoulidis et al. 2014). Dyspnea with pleuritic chest pain and diminished lung sounds on the affected side is the classic presentation of PTX. Conventional chest radiography (CXR) has long been the initial screening test after trauma (Feden 2013). Imaging  A hypodense appearance of the affected hemithorax is usually observed, with variable degrees of lung collapse (Fig. 12). Management  Guidelines for the management of traumatic PTX are lacking, but treatment generally parallels that of spontaneous PTX. Clinically stable patients with a small PTX (3-cm apex-to-pleural dome distance) in clinically stable patients should be re-expanded by insertion of a percutaneous catheter or chest tube, and these patients are typically admitted to the hospital (Baumann et al. 2001). The chest tube is removed after the absence of an air leak with the tube to water seal is documented (Feden 2013).

5.1.1 Tension Pneumothorax (TPTX) TPTX is a life-threatening condition that is rapidly fatal if untreated (Leigh-Smith and ­ Harris 2005). TPTX occurs when the injury creates a “one-way valve” like mechanism that communicates with the pleural space resulting in accumulation of air within the pleural cavity with each inspiration (Kong et al. 2016). This results in a high intrathoracic pressure reducing cardiac filling, compressing the ipsilateral lung, and usually displacing the mediastinum contralaterally. Cardiovascular collapse can eventually cause the death of the patient. It remains a clinical diagnosis that must be treated immediately once suspected. The Advanced Trauma Life Support (ATLS) course recommends immediate needle decompression (Kong et al. 2016).

S. Shah et al.

1066

5.1.2 Occult Pneumothorax With the growth of CT use in trauma, the entity known as occult PTX has emerged. It is identified only by CT scan and otherwise may go unrecognized due to its small size (Yadav et al. 2010). Observation is a reasonably safe strategy in the management of traumatic occult PTX (Mowery et al. 2011).

5.2

Hemothorax

Introduction  Hemothorax is defined as a collection of blood within the pleural cavity and is present in 50 % of major trauma patients (Mirvis 2004). It usually results from lung laceration from rib fractures or vessel injury (Feden 2013). Clinical presentation resembles that of pneumothorax, except that dullness to percussion may be evident on the affected side rather than hyperresonance. Imaging  Fluid collections greater than approximately 300 mL can be identified as a fluid that blunts the costophrenic angle on upright CXR. As much as 1000 mL of blood may be missed on supine CXR because it produces only a mild diffuse hazy appearance (Feden 2013). If a hemothorax is under pressure, the mediastinum can shift contralaterally. CT scan has excellent sensitivity and complements CXR. On CT,

Fig. 13  CT thorax – axial. A small amount of dependent high-density fluid is seen in the right hemithorax (orange arrow). Also seen are displaced right rib fractures (red arrow)

hemorrhage can be diagnosed by its density of 30–45 HU (Fig. 13), depending on the hematocrit, the admixture of other types of fluid, and the state of clot formation (Mirvis 2004). Management  Occult or small hemothorax (measured as 1 cm (orange arrow) in thickness and linear laceration >1 cm in depth (red arrow). (b) Grade III. Laceration >3 cm in depth and reaches the renal hilum (arrow) without evidence of vascular injury. (c) CT

with IV contrast (arterial phase). Grade IVA. Loss of splenic shape and contour with a perisplenic hematoma. There is evidence of extravasation of high-density contrast centered within splenic parenchyma (arrow)

S. Shah et al.

1070

Management  Management decisions in cases of acute splenic injury are based on clinical signs and symptoms and injury grade. Grades I–III are usually managed conservatively; whereas grade IV injuries often require surgical or endovascular treatment.

6.2

Liver

Introduction  The liver injury scale is based on the presence, location, and size of liver lacerations and hematomas, as well as the presence of more extensive tissue maceration or devascularization in higher-­grade injuries. Injuries affecting the posterior aspect of superomedial liver, region that is not covered by peritoneum, may cause large retroperi­ toneal hematomas. However, no correlation was observed between the degree of hemoperitoneum and the need for surgery (Poletti et al. 2000). Imaging  Similar in approach to splenic trauma, AAST liver injury scale (Table 8) is applied when assessing the severity of injury (Moore et al. 1995). Additional imaging findings have been found to be useful in guiding clinical management decisions. These include (a) extension of the injury to involve the major hepatic veins, which requires laparotomy, and (b) the presence of active bleeding or pseudoaneurysm, which requires endovascular intervention (Fig. 18) (Becker et al. 1996; Taourel et al. 2007).

a

Fig. 18 (a) CT with IV contrast – PV phase. Grade I: linear hypodense laceration 3 Couinaud segments within a single lobe Vascular Juxtahepatic venous injuries (i.e., retrohepatic vena cava and/or central major hepatic veins) VI Vascular Hepatic avulsion

b

hypodense laceration runs from the periphery to the porta hepatis involving segments IVa and IVb of left hepatic lobe

Emergency CT

Management  Majority of blunt hepatic injuries are now successfully managed conservatively (Croce et al. 1995). The growing trend toward nonsurgical management of hepatic injuries has increased the incidence of delayed complications such as bile leaks, biliary strictures, and hepatic abscesses (Taourel et al. 2007). Lacerations that extend to the porta hepatis are commonly associated with bile duct injury and may result in the development of biloma (Hassan et al. 2010). Therefore, CT plays an important role in detecting delayed complications of conservative management.

6.3

1071

superficial (involving 50 % parenchymal tissue (Rekhi et al. 2010; Lahiri and Bhattacharya 2013). Grades I and II injuries indicate contusion and/or superficial laceration without ductal injury (Fig. 19). Grade III indicates deep laceration or transection with duct injury to the left of SMV. Grade IV injury indicates deep laceration or transection with duct injury to the right of SMV (Fig. 20). Grade V injury is the most severe and indicates pancreatic head disruption and is frequently associated with concomitant duodenal injury (Fig. 21). Pancreatic duct injury is the main prognostic indicator of clinical outcome in patients with a blunt traumatic injury. However, pancreatic duct

Pancreas

Introduction  Incidence of pancreatic injury in blunt abdominal trauma is rare ranging from 0.2 to 12 % with mortality rates as high as 30 % (Cirillo and Koniaris 2002). Duodenum and pancreas are commonly injured simultaneously with an incidence of 50–98 % (Daly et al. 2008; Linsenmaier et al. 2008; Rekhi et al. 2010). Mortality and morbidity significantly increase when the diagnosis is not recognized at admission. Thus, a prompt diagnosis of pancreatic injury is critical in delivering appropriate and timely interventions and for minimizing complications (Wolf et al. 2005). Blunt pancreatic injury (BPI) occurs with compression of pancreas against the first and second lumbar vertebrae secondary to upper/mid-­abdomen midline frontal impacts. Pancreatic injury is more common in children and adolescents secondary to the paucity of intra-abdominal fat and a thinner layer of peripancreatic fat, which provides protection from blunt trauma (Linsenmaier et al. 2008; Rekhi et al. 2010). Physical examination is generally nonspecific, but the pancreatic injury should always be ­considered in patients with epigastric tenderness (Lahiri and Bhattacharya 2013). Imaging  The direct CT findings include (1) laceration, (2) transection, and (3) focal pancreatic enlargement or hematoma. A laceration that extends through the entire pancreas is termed transection or fracture. Lacerations can be further divided into

Fig. 19  Grade 1 injury. Axial CT in a patient with a pancreatic contusion. There is irregularity of pancreatic tail and mild stranding in the anterior pararenal space (arrow)

Fig. 20  Grade IV injury. Axial CT shows complete transection of the head of the pancreas to the right of SMV (arrow)

S. Shah et al.

1072

7

Genitourinary Trauma

Wide-impact blunt abdominal trauma is responsible for most closed injuries of the genitourinary organs, with motor vehicle crashes most frequently associated with renal and bladder injuries (43 % and 16 % of cases, respectively) (Paparel et al. 2006).

7.1

Fig. 21  Grade V injury. Axial CT image shows massive disruption (arrows) of the pancreatic head

injury is poorly characterized on CT, and it is often detected during laparotomy (Lahiri and Bhattacharya 2013). Diagnosing a pancreatic duct injury remains a significant challenge in the management of pancreatic trauma. Grading of injury allows accurate description and localization and to determine surgical versus nonoperative treatment. Management  The location of injury (proximal versus distal), the involvement of the pancreatic duct, and the overall status of the patient are major determinants of appropriate management (Potoka et al. 2015). Nonoperative management has been proposed in the management of grade I or II pancreatic injuries. It consists of bowel rest with total parenteral nutrition, serial abdominal examination, and serial serum amylase determination with laparotomy or further imaging studies performed for worsening abdominal examination or persistent hyperamylasemia. However, nonoperative management in the setting of main pancreatic ductal injury leads to a high incidence of late complications, in particular pseudocysts and pancreatic fistula (Wind et al. 1999). On the other hand, proximal pancreatic ductal injuries (Grades IV and V) are difficult to manage with little data in the literature to suggest the optimal approach. Patient with grade IV and V injuries frequently get a pancreatic stent and Whipple procedure, respectively. Recent trends, however, favor more conservative management such as closed suction drainage alone (Potoka et al. 2015).

Renal

Introduction  Renal injuries are relatively common, involving 8–10 % of the patients admitted to an emergency department for blunt abdominal trauma (Körner et al. 2008; Alonso et al. 2009). When a renal laceration is detected on CT, a 10-min delayed scan should be obtained to assess the collecting system and evaluate for urinary extravasation (Stuhlfaut et al. 2006). Although gross hematuria is the typical presenting symptom of renal injury, it may be absent in about 5 % of renal injuries, including renal vascular injuries (Smith and Kenney 2003; Ismail 2013). Conversely, there is no absolute correlation between the presence or degree of hematuria and the amount of renal injury that is present (Ramchandani and Buckler 2009). Imaging  According to the American Association for the Surgery of Trauma (AAST), the severity of renal injuries is graded from 1 to 5 (Table 9), with an increasingly poorer prognosis (Shariat et al. 2007). Any injury that disrupts the collecting system is at least grade IV (Fig. 22a). Grade V injury that devascularizes the kidney (Fig. 22b) is most severe requiring urgent intervention. This grading system does not integrate some significant CT findings, such as arterial contrast extravasation, and is yet to be modified to better integrate abnormalities seen only on imaging (Santucci et al. 2004). Management  The indications for surgical treatment of renal trauma have progressively decreased with a consequent drastic reduction in nephrectomy rates. Currently, nonoperative management, which includes close observation, endovascular treatments, and endourological treatments, can be considered

1073

Emergency CT

the most appropriate first-line approach for about 90–95 % of renal injuries (Smith and Kenney 2003; Jacobs et al. 2012; Aragona et al. 2012). Conversely, renal exploration may be necessary in the case of hemodynamic instability following renal trauma, extensive devitalized tissue, active hemorrhage, a large injury to the collecting system, or ureteral disruption (Smith and Kenney 2003; Bonatti et al. 2015).

Table 9  Revised AAST classification of renal trauma Grade Type I Parenchyma

Hematoma II

Parenchyma

Hematoma

III

Parenchyma

IV

Collecting system Vascular Vascular

V

Description Microscopic or gross hematuria with normal urologic studies (contusion) Non-expanding subcapsular hematoma Laceration 1 cm in depth without collecting system rupture All collecting system injuries (including shattered kidney) Segmental vessel injury Main artery and/or vein injury (including laceration, avulsion, or thrombosis)

a

Fig. 22 (a) CT with IV contrast – delayed phase (bone window). Grade IV: leakage of high-density contrast from the left ureteropelvic junction (arrow). (b) CT with IV

7.2

Bladder

Introduction  The most frequent causes of bladder trauma are motor vehicle crashes, crush injuries, and blows to the lower abdomen (Gomez et al. 2004). Although most patients with blunt bladder injury have pelvic fractures (85–89 %) (Ramchandani and Buckler 2009; Gross et al. 2015), only a minority of patients with pelvic fractures will have some bladder injury (6–8 %) (Avey et al. 2006). Consequently, gross hematuria with pelvic fracture is an absolute indication for evaluation of the bladder in a patient with trauma (Quagliano et al. 2006) due to a high likelihood of injury (29 %) (Gross et al. 2015). Pelvic ­ fractures are usually associated with extraperitoneal ruptures; whereas intraperitoneal rupture occurs when there is a blow to the lower abdomen in a patient with a distended urinary bladder. It causes a sudden rise in the intraluminal pressure of the bladder and rupture of the dome, which is the weakest portion of the bladder. Imaging  Compared to the AAST scale, which takes into account the length or extent of the bladder wall laceration, classification system endorsed by a consensus panel of the Societe Internationale D’Urologie (Gomez et al. 2004) is directed toward determining whether there is a full-­thickness tear of the bladder as judged by contrast extravasation on CT cystography. It classifies bladder injury into four types: type 1 is b

contrast. Grade V: non-enhancement of the left kidney and non-opacification of the distal renal artery at the hilum (arrow) consistent with renal artery injury

S. Shah et al.

1074

bladder contusion; type 2, intraperitoneal rupture (Fig. 23); type 3, extraperitoneal rupture; and type 4, combined injury. Active distention of the urinary bladder with contrast material is essential for a high-quality CT cystogram that is reliable in excluding a bladder rupture. With extraperitoneal leaks, the contrast agent remains confined to the pelvis giving a “molar tooth” appearance. Whereas, with intraperitoneal leaks, contrast material may extend into the paracolic gutters and diffusely into the peritoneal cavity outlining bowel loops. Management  Type 2 and 4 require surgical repair. Type 2 injuries typically result in a large tear in the dome of the bladder that cannot be treated successfully with catheter drainage alone. Uncomplicated type 3 is managed conservatively with catheter drainage because 85 % resolve within 10 days and almost all resolve within 3 weeks (Corriere and Sandler 1999).

Fig. 23  CT cystogram – bone window. Type 2 injury: defect in the bladder dome (orange arrow) with leakage of high-density contrast into the peritoneal cavity (red arrow) which lines the small bowel loops

References Adams JH, Doyle D, Ford I, Gennarelli TA, Graham DI, McLellan DR (1989) Diffuse axonal injury in head injury: definition, diagnosis and grading. Histopathology 15(1):49–59 Alonso RC, Nacenta SB, Martinez PD, Guerrero AS, Fuentes CG (2009) Kidney in danger: CT findings of blunt and penetrating renal trauma. Radiogr Rev Publ Radiol Soc N Am Inc. 29(7):2033–2053 Aragona F, Pepe P, Patanè D, Malfa P, D’Arrigo L, Pennisi M (2012) Management of severe blunt renal trauma in adult patients: a 10-year retrospective review from an emergency hospital. BJU Int 110(5):744–748 Armin SS, Colohan ART, Zhang JH (2006) Traumatic subarachnoid hemorrhage: our current understanding and its evolution over the past half century. Neurol Res 28(4):445–452 Avey G, Blackmore CC, Wessells H, Wright JL, Talner LB (2006) Radiographic and clinical predictors of bladder rupture in blunt trauma patients with pelvic fracture. Acad Radiol 13(5):573–579 Baugnon KL, Hudgins PA (2014) Skull base fractures and their complications. Neuroimaging Clin N Am 24(3):439–465 , vii–viii Baumann MH, Strange C, Heffner JE, Light R, Kirby TJ, Klein J et al (2001) Management of spontaneous ­pneumothorax: an American College of Chest Physicians Delphi consensus statement. Chest 119(2):590–602 Beaumont A, Gennarelli T (2006) CT prediction of contusion evolution after closed head injury: the role of pericontusional edema. Acta Neurochir Suppl 96:30–32 Becker CD, Gal I, Baer HU, Vock P (1996) Blunt hepatic trauma in adults: correlation of CT injury grading with outcome. Radiology 201(1):215–220 Bilello JF, Davis JW, Lemaster DM (2005) Occult traumatic hemothorax: when can sleeping dogs lie? Am J Surg 190(6):841–844 Bodanapally UK, Sours C, Zhuo J, Shanmuganathan K (2015) Imaging of traumatic brain injury. Radiol Clin North Am 53(4):695–715 , viii Bonatti M, Lombardo F, Vezzali N, Zamboni G, Ferro F, Pernter P et al (2015) MDCT of blunt renal trauma: imaging findings and therapeutic implications. Insights Imaging 6(2):261–272 Boscak A, Shanmuganathan K (2012) Splenic trauma: what is new? Radiol Clin North Am 50(1):105–122 Catre MG (1995) Diagnostic peritoneal lavage versus abdominal computed tomography in blunt abdominal trauma: a review of prospective studies. Can J Surg J Can Chir 38(2):117–122 Cirillo RL, Koniaris LG (2002) Detecting blunt pancreatic injuries. J Gastrointest Surg Off J Soc Surg Aliment Tract 6(4):587–598 Clarke JR, Trooskin SZ, Doshi PJ, Greenwald L, Mode CJ (2002) Time to laparotomy for intra-abdominal bleeding from trauma does affect survival for delays up to 90 minutes. J Trauma 52(3):420–425

Emergency CT Cleverley JR, Barrie JR, Raymond GS, Primack SL, Mayo JR (2002) Direct findings of aortic injury on contrast-enhanced CT in surgically proven traumatic aortic injury: a multi-centre review. Clin Radiol 57(4):281–286 Corriere JN, Sandler CM (1999) Bladder rupture from external trauma: diagnosis and management. World J Urol 17(2):84–89 Cox EF (1984) Blunt abdominal trauma. A 5-year analysis of 870 patients requiring celiotomy. Ann Surg 199(4):467–474 Croce MA, Fabian TC, Menke PG, Waddle-Smith L, Minard G, Kudsk KA et al (1995) Nonoperative management of blunt hepatic trauma is the treatment of choice for hemodynamically stable patients. Results of a prospective trial. Ann Surg 221(6):744–753 , 755 Cullen EL, Lantz EJ, Johnson CM, Young PM (2014) Traumatic aortic injury: CT findings, mimics, and therapeutic options. Cardiovasc Diagn Ther 4(3):238 Daly KP, Ho CP, Persson DL, Gay SB (2008) Traumatic retroperitoneal injuries: review of multidetector CT findings. Radiogr Rev Publ Radiol Soc N Am Inc. 28(6):1571–1590 Davis PC, Expert Panel on Neurologic Imaging (2007) Head trauma. AJNR Am J Neuroradiol 28(8): 1619–1621 Deepika A, Prabhuraj AR, Saikia A, Shukla D (2015) Comparison of predictability of Marshall and Rotterdam CT scan scoring system in determining early mortality after traumatic brain injury. Acta Neurochir 157(11):2033–2038 Deng K, Liu C, Ma R, Sun C, Wang X, Ma Z et al (2009) Clinical evaluation of dual-energy bone removal in CT angiography of the head and neck: comparison with conventional bone-subtraction CT angiography. Clin Radiol 64(5):534–541 Derakhshan I (2009) The Kernohan-Woltman phenomenon and laterality of motor control: fresh analysis of data in the article “Incisura of the crus due to contralateral brain tumor”. J Neurol Sci 287(1–2):296 Deunk J, Dekker HM, Brink M, van Vugt R, Edwards MJ, van Vugt AB (2007) The value of indicated computed tomography scan of the chest and abdomen in addition to the conventional radiologic workup for blunt trauma patients. J Trauma 63(4): 757–763 Deunk J, Poels TC, Brink M, Dekker HM, Kool DR, Blickman JG et al (2010) The clinical outcome of occult pulmonary contusion on multidetector-row computed tomography in blunt trauma patients. J Trauma 68(2):387–394 Dreizin D, Letzing M, Sliker CW, Chokshi FH, Bodanapally U, Mirvis SE et al (2014) Multidetector CT of blunt cervical spine trauma in adults. Radiographics 34(7):1842–1865 Eichler K, Marzi I, Wyen H, Zangos S, Mack MG, Vogl TJ (2015) Multidetector computed tomography (MDCT): simple CT protocol for trauma patient. Clin Imaging 39(1):110–115

1075 Eisenberg HM, Gary HE, Aldrich EF, Saydjari C, Turner B, Foulkes MA et al (1990) Initial CT findings in 753 patients with severe head injury. A report from the NIH Traumatic Coma Data Bank. J Neurosurg 73(5):688–698 Feden JP (2013) Closed lung trauma. Clin Sports Med 32(2):255–265 Forman MJ, Mirvis SE, Hollander DS (2013) Blunt ­thoracic aortic injuries: CT characterisation and treatment outcomes of minor injury. Eur Radiol 23(11): 2988–2995 Gennarelli TA, Thibault LE (1982) Biomechanics of acute subdural hematoma. J Trauma 22(8):680–686 Gentry LR, Godersky JC, Thompson B (1988) MR imaging of head trauma: review of the distribution and radiopathologic features of traumatic lesions. AJR Am J Roentgenol 150(3):663–672 Geyer LL, Körner M, Harrieder A, Mueck FG, Deak Z, Wirth S et al (2016) Dose reduction in 64-row whole-­ body CT in multiple trauma: an optimized CT protocol with iterative image reconstruction on a gemstonebased scintillator. Br J Radiol 89(1061):20160003 Gomez RG, Ceballos L, Coburn M, Corriere JN, Dixon CM, Lobel B et al (2004) Consensus statement on bladder injuries. BJU Int 94(1):27–32 Gonzalez RP, Ickler J, Gachassin P (2001) Complementary roles of diagnostic peritoneal lavage and computed tomography in the evaluation of blunt abdominal trauma. J Trauma 51(6):1128–1134 ; discussion 1134–1136 Gross JA, Lehnert BE, Linnau KF, Voelzke BB, Sandstrom CK (2015) Imaging of urinary system trauma. Radiol Clin North Am 53(4):773–788 Hardman JM, Manoukian A (2002) Pathology of head trauma. Neuroimaging Clin N Am 12(2):175–187 , vii Hassan R, Ralib M, Raghib M, Razali A, Mohamed C, Kamariah S et al (2010) Computed tomography (CT) in blunt liver injury: a pictorial essay. Med J Malaysia 65(4):321–327 Healy DA, Hegarty A, Feeley I, Clarke-Moloney M, Grace PA, Walsh SR (2014) Systematic review and meta-analysis of routine total body CT compared with selective CT in trauma patients. Emerg Med J EMJ 31(2):101–108 Honeybul S, Ho KM, Lind CRP, Gillett GR (2011) Decompressive craniectomy for diffuse cerebral swelling after trauma: long-term outcome and ethical considerations. J Trauma 71(1):128–132 Hughes TMD, Elton C (2002) The pathophysiology and management of bowel and mesenteric injuries due to blunt trauma. Injury 33(4):295–302 Ismail M (2013) Renal trauma imaging: diagnosis and management. A pictorial review. Pol J Radiol 78(4): 27–35 Jacobs MA, Hotaling JM, Mueller BA, Koyle M, Rivara F, Voelzke BB (2012) Conservative management vs early surgery for high grade pediatric renal trauma– do nephrectomy rates differ? J Urol 187(5): 1817–1822

1076 Jiang L, Ma Y, Jiang S, et al (2014) Comparison of whole-body computed tomography vs selective radiological imaging on outcomes in major trauma patients: a meta-analysis. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 22:54. doi:10.1186/s13049-014-0054-2 Johnson PL, Eckard DA, Chason DP, Brecheisen MA, Batnitzky S (2002) Imaging of acquired cerebral herniations. Neuroimaging Clin N Am 12(2):217–228 Kakarla UK, Chang SW, Theodore N, Sonntag VKH (2010) Atlas fractures. Neurosurgery 66(3 Suppl): 60–67 Karaca L, Yuceler Z, Kantarci M, Çakır M, Sade R, Calıkoglu C et al (2016) The feasibility of dual-energy CT in differentiation of vertebral compression fractures. Br J Radiol 89(1057):20150300 Kelleher MS Jr, Gao G, Rolen MF, Bokhari SJ (2015) Completion CT of chest, abdomen, and pelvis after acute head and cervical spine trauma: incidence of acute traumatic findings in the setting of low-velocity trauma. Radiology 279(2):395–399 Kong V, Sartorius B, Clarke D (2016) Traumatic tension pneumothorax: experience from 115 consecutive patients in a trauma service in South Africa. Eur J Trauma Emerg Surg 42(1):55–59 Körner M, Krötz MM, Degenhart C, Pfeifer K-J, Reiser MF, Linsenmaier U (2008) Current role of emergency US in patients with major trauma. Radiogr Rev Publ Radiol Soc N Am Inc 28(1):225–242 Kraus RR, Bergstein JM, DeBord JR (1999) Diagnosis, treatment, and outcome of blunt carotid arterial injuries. Am J Surg 178(3):190–193 Lahiri R, Bhattacharya S (2013) Pancreatic trauma. Ann R Coll Surg Engl 95(4):241–245 Larsen DW (2002) Traumatic vascular injuries and their management. Neuroimaging Clin N Am 12(2):249–269 Leigh-Smith S, Harris T (2005) Tension pneumothorax– time for a re-think? Emerg Med J EMJ 22(1):8–16 Leung V, Sastry A, Woo TD, Jones HR (2015) Implementation of a split-bolus single-pass CT protocol at a UK major trauma centre to reduce excess radiation dose in trauma pan-CT. Clin Radiol 70(10): 1110–1115 Linsenmaier U, Wirth S, Reiser M, Körner M (2008) Diagnosis and classification of pancreatic and duodenal injuries in emergency radiology. Radiogr Rev Publ Radiol Soc N Am Inc. 28(6):1591–1602 Little SC, Kesser BW (2006) Radiographic classification of temporal bone fractures: clinical predictability using a new system. Arch Otolaryngol Neck Surg 132(12):1300–1304 Lively MW, Stone D (2006) Pulmonary contusion in football players. Clin J Sport Med Off J Can Acad Sport Med 16(2):177–178 Lolli V, Pezzullo M, Delpierre I, Sadeghi N (2016) MDCT imaging of traumatic brain injury. Br J Radiol 89(1061):20150849 Marmery H, Shanmuganathan K, Alexander MT, Mirvis SE (2007) Optimization of selection for nonoperative management of blunt splenic injury: comparison of

S. Shah et al. MDCT grading systems. AJR Am J Roentgenol 189(6):1421–1427 Marmery H, Shanmuganathan K, Mirvis SE, Richard H, Sliker C, Miller LA et al (2008) Correlation of multidetector CT findings with splenic arteriography and surgery: prospective study in 392 patients. J Am Coll Surg 206(4):685–693 McLaughlin PD, Mallinson P, Lourenco P, Nicolaou S (2015) Dual-energy computed tomography. Radiol Clin North Am 53(4):619–638 Miller PR, Croce MA, Bee TK, Qaisi WG, Smith CP, Collins GL et al (2001) ARDS after pulmonary contusion: accurate measurement of contusion volume identifies high-risk patients. J Trauma 51(2):223–228 ; discussion 229–230 Mirvis SE (2004) Diagnostic imaging of acute thoracic injury. In: Seminars in Ultrasound, CT and MRI [Internet]. Elsevier; 2004 [cited 2016 Aug 30], pp. 156–179. Available from: http://www.sciencedirect.com/science/article/pii/S0887217104000071 Mirvis SE, Shanmuganathan K (2007) Diagnosis of blunt traumatic aortic injury 2007: still a nemesis. Eur J Radiol 64(1):27–40 Mirvis SE, Young JW, Lim C, Greenberg J (1987) Hangman’s fracture: radiologic assessment in 27 cases. Radiology 163(3):713–717 Mokrane FZ, Revel-Mouroz P, Saint Lebes B, Rousseau H (2015) Traumatic injuries of the thoracic aorta: the role of imaging in diagnosis and treatment. Diagn Interv Imaging 96(7–8):693–706 Moore EE, Cogbill TH, Jurkovich GJ, Shackford SR, Malangoni MA, Champion HR (1995) Organ injury scaling: spleen and liver (1994 revision). J Trauma 38(3):323–324 Morales-Uribe C, Ramírez A, Suarez-Poveda T, Ortiz M, Sanabria A (2016) Diagnostic performance of CT angiography in neck vessel trauma: systematic review and meta-analysis. Emerg Radiol [Internet]. 2016 Jun 1 [cited 2016 Aug 30]; Available from: http://link. springer.com/10.1007/s10140-016-1412-3 Morhard D, Fink C, Graser A, Reiser MF, Becker C, Johnson TRC (2009) Cervical and cranial computed tomographic angiography with automated bone removal: dual energy computed tomography versus standard computed tomography. Invest Radiol 44(5):293–297 Mowery NT, Gunter OL, Collier BR, Diaz JJ, Haut E, Hildreth A et al (2011) Practice management guidelines for management of hemothorax and occult pneumothorax. J Trauma 70(2):510–518 Murray GD, Teasdale GM, Braakman R, Cohadon F, Dearden M, Iannotti F et al (1999) The European Brain Injury Consortium survey of head injuries. Acta Neurochir 141(3):223–236 Narayan RK, Maas AIR, Servadei F, Skolnick BE, Tillinger MN, Marshall LF et al (2008) Progression of traumatic intracerebral hemorrhage: a prospective observational study. J Neurotrauma 25(6):629–639 National Guideline Clearinghouse (2016) ACR Appropriateness Criteria® suspected spine trauma. National Guideline Clearinghouse [Internet]. [cited

Emergency CT 2016 Sep 10]. Available from: https://www.guideline. gov/summaries/summary/37931 Nuñez DB, Ahmad AA, Coin CG, LeBlang S, Becerra JL, Henry R et al (1994) Clearing the cervical spine in multiple trauma victims: a time-effective protocol using helical computed tomography. Emerg Radiol 1(6):273–278 Orwig D, Federle MP (1989) Localized clotted blood as evidence of visceral trauma on CT: the sentinel clot sign. AJR Am J Roentgenol 153(4):747–749 Osborn AG (2016) Osborn’s brain: imaging, pathology and anatomy [Internet]. [cited 2016 Sep 5]. Available from: https://www.lww.com/opencms/opencms/PEMR/content/WCA421/about_toc.html Paparel P, N’Diaye A, Laumon B, Caillot J-L, Perrin P, Ruffion A (2006) The epidemiology of trauma of the genitourinary system after traffic accidents: analysis of a register of over 43,000 victims. BJU Int 97(2):338–341 Pappas JN (2002) The angiographic string sign. Radiology 222(1):237–238 Parmley LF, Mattingly TW, Manion WC, Jahnke EJ (1958) Nonpenetrating traumatic injury of the aorta. Circulation 17(6):1086–1101 Peng MY, Parisky YR, Cornwell EE, Radin R, Bragin S (1999) CT cystography versus conventional cystography in evaluation of bladder injury. AJR Am J Roentgenol 173(5):1269–1272 Poletti PA, Mirvis SE, Shanmuganathan K, Killeen KL, Coldwell D (2000) CT criteria for management of blunt liver trauma: correlation with angiographic and surgical findings. Radiology 216(2):418–427 Potoka DA, Gaines BA, Leppäniemi A, Peitzman AB (2015) Management of blunt pancreatic trauma: what’s new? Eur J Trauma Emerg Surg 41(3): 239–250 Provenzale J (2007) CT and MR imaging of acute cranial trauma. Emerg Radiol 14(1):1–12 Pryputniewicz DM, Hadley MN (2010) Axis fractures. Neurosurgery 66(3 Suppl):68–82 Quagliano PV, Delair SM, Malhotra AK (2006) Diagnosis of blunt bladder injury: a prospective comparative study of computed tomography cystography and conventional retrograde cystography. J Trauma 61(2):410– 421 ; discussion 421–422 Ramchandani P, Buckler PM (2009) Imaging of genitourinary trauma. Am J Roentgenol 192(6):1514–1523 Ramirez RM, Cureton EL, Ereso AQ, Kwan RO, Dozier KC, Sadjadi J et al (2009) Single-contrast computed tomography for the triage of patients with penetrating torso trauma. J Trauma 67(3):583–588 Rekhi S, Anderson SW, Rhea JT, Soto JA (2010) Imaging of blunt pancreatic trauma. Emerg Radiol 17(1): 13–19 Renzulli P, Gross T, Schnüriger B, Schoepfer AM, Inderbitzin D, Exadaktylos AK et al (2010) Management of blunt injuries to the spleen. Br J Surg 97(11):1696–1703 Rieger M, Sparr H, Esterhammer R, Fink C, Bale R, Czermak B et al (2002) Modern CT diagnosis of acute

1077 thoracic and abdominal trauma. Anaesthesist 51(10): 835–842 Rivara FP, Kuppermann N, Ellenbogen RG (2015) Use of clinical prediction rules for guiding use of computed tomography in adults with head trauma. JAMA 314(24):2629–2631 Rodallec MH, Marteau V, Gerber S, Desmottes L, Zins M (2008) Craniocervical arterial dissection: spectrum of imaging findings and differential diagnosis. Radiogr Rev Publ Radiol Soc N Am Inc. 28(6):1711–1728 Salvolini U (2002) Traumatic injuries: imaging of facial injuries. Eur Radiol 12(6):1253–1261 Santucci RA, Wessells H, Bartsch G, Descotes J, Heyns CF, McAninch JW et al (2004) Evaluation and management of renal injuries: consensus statement of the renal trauma subcommittee. BJU Int 93(7):937–954 Schueller G, Scaglione M, Linsenmaier U, Schueller-­ Weidekamm C, Andreoli C, De Vargas Macciucca M et al (2015) The key role of the radiologist in the management of polytrauma patients: indications for MDCT imaging in emergency radiology. Radiol Med (Torino) 120(7):641–654 Schuknecht B, Graetz K (2005) Radiologic assessment of maxillofacial, mandibular, and skull base trauma. Eur Radiol 15(3):560–568 Shariat SF, Roehrborn CG, Karakiewicz PI, Dhami G, Stage KH (2007) Evidence-based validation of the predictive value of the American Association for the Surgery of Trauma kidney injury scale. J Trauma 62(4):933–939 Sierink JC, Saltzherr TP, Reitsma JB, Van Delden OM, Luitse JSK, Goslings JC (2012) Systematic review and meta-analysis of immediate total-body computed tomography compared with selective radiological imaging of injured patients. Br J Surg 99(Suppl 1): 52–58 Sierink JC, Treskes K, Edwards MJR, Beuker BJA, den Hartog D, Hohmann J et al (2016) Immediate total-­ body CT scanning versus conventional imaging and selective CT scanning in patients with severe trauma (REACT-2): a randomised controlled trial. The Lancet 388(10045):673–683 Smith JK, Kenney PJ (2003) Imaging of renal trauma. Radiol Clin North Am 41(5):1019–1035 Smith JS, Chang EF, Rosenthal G, Meeker M, von Koch C, Manley GT et al (2007) The role of early follow-up computed tomography imaging in the management of traumatic brain injury patients with intracranial hemorrhage. J Trauma 63(1):75–82 Soto JA, Anderson SW (2012) Multidetector CT of blunt abdominal trauma. Radiology 265(3):678–693 Soto JA, Lucey BC, Stuhlfaut JW, Varghese JC (2004) Use of 3D imaging in CT of the acute trauma patient: impact of a PACS-based software package. Emerg Radiol 11(3):173–176 Starnes BW, Lundgren RS, Gunn M, Quade S, Hatsukami TS, Tran NT et al (2012) A new classification scheme for treating blunt aortic injury. J Vasc Surg 55(1):47–54 Stuhlfaut JW, Soto JA, Lucey BC, Ulrich A, Rathlev NK, Burke PA et al (2004) Blunt abdominal trauma:

1078 p­erformance of CT without oral contrast material. Radiology 233(3):689–694 Stuhlfaut JW, Lucey BC, Varghese JC, Soto JA (2006) Blunt abdominal trauma: utility of 5-minute delayed CT with a reduced radiation dose. Radiology 238(2): 473–479 Tan LA, Chen M, Muñoz LF (2016) Letter to the Editor: utility of dual-energy CT in differentiating contrast extravasation from intracranial hematoma. J Neurosurg 124(1):279–280 Taourel P, Vernhet H, Suau A, Granier C, Lopez FM, Aufort S (2007) Vascular emergencies in liver trauma. Eur J Radiol 64(1):73–82 Topp T, Lefering R, Lopez CL, Ruchholtz S, Ertel W, Kühne CA (2015) Radiologic diagnostic procedures in severely injured patients – is only whole-body multislice computed tomography the answer? Int J Emerg Med [Internet]. 2015 Dec [cited 2016 Aug 30];8(1) Available from: http://www.intjem.com/content/8/1/3 Tyburski JG, Collinge JD, Wilson RF, Eachempati SR (1999) Pulmonary contusions: quantifying the lesions on chest X-ray films and the factors affecting prognosis. J Trauma 46(5):833–838 Udstuen GJ, Claar JM (2001) Imaging of acute head injury in the adult. Semin Ultrasound CT MR 22(2):135–147 Uyeda J, Anderson SW, Kertesz J, Soto JA (2010) Pelvic CT angiography: application to blunt trauma using 64MDCT. Emerg Radiol 17(2):131–137 van Vugt R, Kool DR, Deunk J, Edwards MJR (2012) Effects on mortality, treatment, and time management as a result of routine use of total body computed tomography in blunt high-energy trauma patients. J Trauma Acute Care Surg 72(3):553–559 Velmahos GC, Tabbara M, Gross R, Willette P, Hirsch E, Burke P et al (2009) Blunt pancreaticoduodenal injury: a multicenter study of the Research Consortium of

S. Shah et al. New England Centers for Trauma (ReCONECT). Arch Surg Chic Ill 1960 144(5):413–419 ; discussion 419–420 Wanek S, Mayberry JC (2004) Blunt thoracic trauma: flail chest, pulmonary contusion, and blast injury. Crit Care Clin 20(1):71–81 Wind P, Tiret E, Cunningham C, Frileux P, Cugnenc PH, Parc R (1999) Contribution of endoscopic retrograde pancreatography in management of complications following distal pancreatic trauma. Am Surg 65(8): 777–783 Winegar BA, Murillo H, Tantiwongkosi B (2013) Spectrum of critical imaging findings in complex facial skeletal trauma. Radiographics 33(1):3–19 WISQARS (Web-based Injury Statistics Query and Reporting System) (2016) Injury Center CDC [Internet]. [cited 2016 Sep 4]. Available from: https:// www.cdc.gov/injury/wisqars/ Wolf A, Bernhardt J, Patrzyk M, Heidecke C-D (2005) The value of endoscopic diagnosis and the treatment of pancreas injuries following blunt abdominal trauma. Surg Endosc 19(5):665–669 Yadav K, Jalili M, Zehtabchi S (2010) Management of traumatic occult pneumothorax. Resuscitation 81(9): 1063–1068 Zarogoulidis P, Kioumis I, Pitsiou G, Porpodis K, Lampaki S, Papaiwannou A et al (2014) Pneumothorax: from definition to diagnosis and treatment. J Thorac Dis 6(4):S372–S376 Zimmerman RA, Bilaniuk LT (1982) Computed tomographic staging of traumatic epidural bleeding. Radiology 144(4):809–812 Zumkeller M, Behrmann R, Heissler HE, Dietz H (1996) Computed tomographic criteria and survival rate for patients with acute subdural hematoma. Neurosurgery 39(4):708–712 ; discussion 712–713

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery Pedro Augusto Gondim Teixeira and Alain Blum

Contents

Abstract

1    Introduction

 1079

2    Trauma 2.1  Polytrauma 2.2  Acute Trauma (Preoperative Assessment) 2.3  Subacute or Chronic Injuries

 1080  1081

3    Tumors 3.1  Acquisition Technique 3.2  Clinical Applications

 1085  1087  1089

4    Postsurgery 4.1  Acquisition Techniques 4.2  Clinical Applications

 1092  1093  1098

Conclusion

 1102

Bibliography

 1102

Soon after its introduction to the clinical ­practice in the late 1970s, CT began to be used for the evaluation of musculoskeletal disorders (Wilson et al. 1978). As with other organs and systems, modern MSK imaging strategy uses a multimodality approach, taking advantage of the strengths of various imaging methods (radiographs, ultrasound, nuclear medicine, CT and MR imaging) (Cotten 2013). For a variety of reasons, CT is frequently part of the diagnostic workup and posttreatment follow-up of patients with MSK disorders. Due to its capacity to depict bony structures in great detail without superimposition, CT offers considerable advantages over conventional radiographs. Bone abnormalities are involved in the physiopathology of several types of MSK diseases, such as acute trauma, overstress syndromes, osteoarthritis, neoplasia, and inflammatory diseases. Finally, contrast-­enhanced CT also has multiple applications in MSK imaging, allowing better visualization of soft tissue anomalies and further characterization of bony lesions.

 1081  1084

1

P.A. Gondim Teixeira • A. Blum (*) Service d’Imagerie Guilloz, CHRU Nancy, Av de Lattre de Tassigny, Nancy 54 000, France e-mail: [email protected]

Introduction

Soon after its introduction to the clinical practice in the late 1970s, CT began to be used for the evaluation of musculoskeletal disorders (Wilson et al. 1978). As with other organs and systems, modern MSK imaging strategy uses a m ­ ultimodality

Med Radiol Diagn Imaging (2017) DOI 10.1007/174_2017_25, © Springer International Publishing AG Published Online: 02 March 2017

1079

P.A. Gondim Teixeira and A. Blum

1080

approach, taking advantage of the strengths of various imaging methods (radiographs, ultrasound, nuclear medicine, CT and MR imaging) (Cotten 2013). For a variety of reasons, CT is frequently part of the diagnostic workup and posttreatment follow-up of patients with MSK disorders. Due to its capacity to depict bony structures in great detail without superimposition, CT offers considerable advantages over conventional radiographs. Bone abnormalities are involved in the physiopathology of several types of MSK diseases, such as acute trauma, overstress syndromes, osteoarthritis, neoplasia, and inflammatory diseases. Finally, contrast-­enhanced CT also has multiple applications in MSK imaging, allowing better visualization of soft tissue anomalies and further characterization of bony lesions. The number of CT studies performed and its indications are ever increasing, supported by the wide availability of CT scanners, fast image acquisition, and high temporal and spatial resolution (Brenner and Hall 2007). Moreover, novel CT-based imaging techniques (such as perfusion, bone subtraction, kinematic imaging) are available for clinical use, further increasing CT use (Gondim Teixeira et al. 2015). The final element that ensures the paramount role of CT in the evaluation of MSK diseases is the progressive and significant reduction in radiation exposure dose seen after the increased awareness of low-dose radiation induced neoplasia risk early in the twenty-first century (Lee and Chhem 2010; Gervaise et al. 2013). Due to advances in hardware (superior detector sensitivity, data transfer, and processing power) and software (iterative reconstruction algorithms), high-quality CT studies can be performed with 20–30% of the dose used in the classic scanner models, supporting the clinical application of this method. Another trend in modern CT imaging is a progressive increase in the width of detector systems (Parrish 2007). Wide detector CT scanners can image whole joints with a single gantry rotation, which can be as fast as 0.27 s in the fastest scanner models. Wide detector CT scanners (more than 80 mm z-axis length) offer various advantages for MSK imaging, greatly reduced acquisition time, and fewer motion artifacts. Wide z-axis coverage led to renewed interest in clinical protocols using

sequential acquisitions, which do not require data interpolation for image reconstruction and minimize dose by eliminating ­over-­ranging and reducing over-beaming effects (Gondim Teixeira et al. 2015). In addition, with sequential acquisitions, all the pixels of a single volume are acquired at exactly the same time (temporal uniformity), which facilitates the use of multiphasic acquisition protocols. Further advantages of CT are the acquisition of isotropic voxels and high-resolution images, which will be further increased in the next generation of CT scanners. Current post-processing options, however, go far beyond classic multiplanar and volume-­ rendered 3D reformats. Tools allowing highly realistic volume-rendered reformats, bone subtraction, motion quantification, and metal artifact reduction represent a few of the novel post-processing techniques available, further increasing the diagnostic power of CT (Dappa et al. 2016; Gondim Teixeira 2013; Gondim Teixeira et al. 2014a). In this chapter, the clinical application of CT for the evaluation of patients after trauma, with musculoskeletal tumors, and after surgery will be discussed. Knowledge of the CT imaging aspects of these common types of pathology is paramount for the clinical practice of both musculoskeletal and general radiologists. Guidelines for protocol optimization focused on wide detector scanner models will be provided, along with information on the indications and use of advanced CT techniques. This information can help radiologists and technicians to fully exploit the potential of modern CT scanners, increasing diagnostic performance and widening the list of CT indications for the evaluation of MSK disorders.

2

Trauma

Precise study of bone architecture and the morphology of calcified tissue is key for the diagnosis of traumatic bone injury and allows for accurate preoperative evaluation and posttreatment follow-up. Although unenhanced CT imaging can already provide valuable diagnostic information, intra-articular contrast injection (CT arthrography) also allows for detection of fine morphologic anomalies of the hyaline cartilage

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery

related to trauma or degenerative disease, which can be hard to identify with other imaging methods such as MRI (Omoumi et al. 2015). CT is particularly useful in the acute posttraumatic setting due to its fast acquisition speed and its wide gantry bore, which facilitates the positioning of patients with pain and limited joint movement. Furthermore, kinematic CT sheds light on the biomechanical impact of bony and soft tissue traumatic injuries (Teixeira et al. 2015a). Traumatized patients constitute a heterogeneous group, with different diagnostic needs. In an effort to provide a comprehensive analysis of CT imaging in cases of trauma, three clinical scenarios will be considered: polytrauma, acute trauma, and subacute/chronic injury.

2.1

Polytrauma

Polytrauma can be defined as the presence of two or more bone or soft tissue traumatic injuries, at least one of them representing a serious threat to the patient’s life. In these cases, a critical but stable patient will frequently undergo a full-body CT scan in order to identify MSK, neurologic, and internal organ injuries that could have a significant impact on management (Thomas et al. 2008). Discussion of the imaging aspects of neurologic/spinal and thoracoabdominal traumatic lesions is beyond the scope of this chapter. In the setting of a polytrauma, the aim of CT examination is to identify and grade all traumatic injuries that may have a prognostic impact.

2.1.1 Acquisition Technique Although there are no clear guidelines, helical acquisitions of the head, thorax, and abdomen are almost always obtained (Geyer et al. 2016; Sedlic et al. 2013; Foster et al. 2011). For optimal analysis of MSK injuries, images should be reconstructed with a bone kernel. Specific additional acquisitions may be required for the evaluation of injuries in the extremities, according to clinical findings (Foster et al. 2011). Post-contrast data sets in the arterial and venous phases are useful for the evaluation of secondary vascular injuries associated with MSK lesions.

1081

2.1.2 Clinical Applications Rib fractures are frequent in polytrauma cases and if numerous and multifocal may lead to respiratory distress. Isolated rib fractures have little clinical impact but may serve as a harbinger for the identification of more serious injuries of the internal organs. Rib fractures are usually searched for in a rib-by-rib fashion using true axial images or multiplanar reformats, which can be time-consuming and prone to error. This is especially true because full-body CT scans with a few thousand or more images are interpreted in an emergency setting, frequently under close surveillance from emergency physicians eager to provide prompt and optimized care. Recently, complex reformatting algorithms that allow unfolding all ribs in a single plane have become available, greatly facilitating and speeding up thoracic wall evaluation and potentially reducing diagnostic errors (Fig. 1). Displaced bone fragments and shear stress during trauma may lead to vascular damage, which can increase mortality and change patient management. Trauma can lead to arterial occlusion with peripheral acute ischemia, pseudoaneurysm formation, and active bleeding. CT angiography can accurately depict arterial occlusions and pseudoaneurysms that should be sought adjacently to fractures (Fig. 2). The diagnosis of occlusion is usually straightforward with a complete lack of opacification in the injured portion of the artery, while pseudoaneurysms usually present as a saccular structure with an arterial-­ enhancing pattern. The latter warrants close surveillance and/or early treatment, because, when ruptured, pseudoaneurysms can lead to massive bleeding. The diagnosis of active bleeding is usually more challenging, requiring a comparative analysis of pre-contrast, arterial, and venous phases to confirm the presence of an enlarging extravascular contrast extravasation.

2.2

Acute Trauma (Preoperative Assessment)

Imaging patients with acute trauma can be challenging. Victims of serious acute trauma are frequently under stress, in pain, and unable to move

P.A. Gondim Teixeira and A. Blum

1082

a

b

Fig. 1  A 44-year-old male victim of a high-energy motor vehicle accident. (a) Global illumination of an unfolded view of the rib cage demonstrating thin non-displaced fractures of the right fifth, sixth and seventh ribs (red

ellipse). (b) Multiplanar reformat demonstrating the same fractures (upper image). The image series on the bottom show the aspect to the rib fracture marked by the green crosshair in the upper image in all orthogonal planes

the affected body part freely either due to pain or to the presence of medical constraint devices. Additionally, the patients have difficulties in keeping still, increasing the frequency of motion artifacts. In cases of acute trauma, radiographs are frequently of low quality due to suboptimal positioning. Ultrasound assessment is also made difficult by pain under probe compression and the presence of bandages and constraining devices as well as positioning issues. Even though MRI could provide a global evaluation of both bony and soft tissue injuries in patients with acute trauma, its application is greatly hampered by examination length, limited coverage, difficulties in patient positioning due to coil and gantry size, as well as susceptibility to motion artifacts. CT is

an excellent modality for imaging patients with acute trauma because acquisitions can be performed very fast (a few 100 ms) in a wide gantry bore (70 cm in diameter or wider) without a need for compression or dedicated coils. In this context, the main objective of CT imaging is to acquire diagnostic-quality images and perform a global evaluation of the fracture, providing all the information required for surgical planning.

2.2.1 Acquisition Technique Since acquisition time should be kept to a minimum, volumetric sequential acquisition available with wide detector systems, also known as “snapshot imaging,” can be recommended in this setting. In this acquisition mode, a single volume is

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery

a

b

1083

c

Fig. 2  A 48-year-old male with a thigh injury following a motor vehicle accident. No sign of arterial injury was found during clinical examination (palpable distal pulse). Maximum intensity projection image in the coronal plane (a) and volume-rendered reformat (b) of a CT angiogram showing a three-fragment fracture of the distal femoral

diastasis (fat arrows). An area of fusiform vascular dilation can be seen adjacent to the displaced femoral fragment (arrow heads). (c) Axial image of the same patient at the level of the vascular thickening demonstrating a small pseudoaneurysmal formation (thin arrow) adjacent to the tip of the bone fragment (arrow head)

acquired in less than 500 ms providing a significant reduction in motion artifacts. Dose is also reduced by the use of volumetric imaging, which is of obvious importance for children and young adults. For regions longer than 16 cm (z-axis), snapshot imaging can still be used by joining together two or more overlapping volumes (stitching mode). However, if more than three volumes are required to cover the region of interest, a helical acquisition is preferable because there is no more gain in acquisition time and dose reduction. If a vascular injury is suspected, intravenous contrast administration can be useful. 3D volume-rendered and multiplanar reformats created from isotropic datasets can help overcome suboptimal patient positioning. Recently, new algorithms that incorporate factors such as light source and tissue light reflec-

tion in volume-rendered reformats have become available for clinical use. This technique, called global illumination, provides highly realistic 3D reconstructions that are superior in quality to conventional density-based volume-­ rendering techniques (Dappa et al. 2016; Ebert et al. 2016) (Fig. 3).

2.2.2 Clinical Application CT can provide accurate assessment of fracture line orientation, number of fragments, and bone displacement, allowing optimal fracture classification (Gibson et al. 2016). In addition, 3D volume-­rendered images can be useful for surgical planning (Mendel et al. 2013). Although less accurate than MRI, CT can also detect tendon and ligament avulsions, especially when a bone avulsion is also present. Hematic collections

P.A. Gondim Teixeira and A. Blum

1084

Fig. 3 A 28-year-old female presenting to the emergency department after falling from a horse. A CT scan was performed using volumetric snapshot mode yielding this global illuminated reconstruction which provides a realistic visualization of the fracture of the proximal humerus involving the greater and the lesser tuberosities

adjacent to fractures can also be identified on CT images reconstructed with soft tissue kernels. Finally, as described previously for polytrauma, various associated vascular injuries can be diagnosed with contrast-enhanced CT.

2.3

Subacute or Chronic Injuries

In this setting, conventional CT can be used to identify occult fractures or intra-articular calcified foreign bodies, which can account for chronic patient symptoms and change management. These findings can be hard to identify on conventional radiographs and sometimes in MR imaging. More recently, kinematic CT became available for clinical use. This method, based on the acquisition of multiple volumes of the same anatomic region during motion or stress maneuvers, allows identification of fine bone positional anomalies which can be useful for the diagnosis of instability and impingement. With the current technology, kinematic CT of joints can be performed with limited radiation dose and high temporal and spatial resolutions (Gondim Teixeira et al. 2015). Current post-processing tools allow semiautomatic quantitative analysis of dynamic CT data, increasing the diagnostic possibilities and the scientific interest of this method (Gondim Teixeira et al. 2017a).

2.3.1 Acquisition Technique Static images are acquired using a standard protocol, if possible with volume acquisition, reconstructed with both bone and soft tissue kernels. CT arthrography can be useful for the evaluation of intra-articular ligaments, cartilage, and ­osteochondral lesions. Although this method is not frequently performed outside Europe, the high contrast between iodine, ligaments, and hyaline cartilage coupled with the high spatial resolution and isotropic voxels of CT images provides an elegant way of demonstrating fine and superficial lesions to the ligaments and cartilage. Kinematic CT has more complex protocol requirements, and various clinical maneuvers can be performed for the evaluation of a given joint. Volume acquisition speed (VAS) is the most important protocol parameter affecting image quality in kinematic CT studies (Beeres et al. 2015; FarshadAmacker et al. 2013). VAS is influenced by the gantry rotation speed and the image reconstruction technique (e.g., full versus partial reconstruction). As VAS is increased, motion artifacts decrease exponentially (Gondim Teixeira et al. 2017b). Kinematic CT studies should therefore be performed with the highest gantry rotation speeds available. Half reconstruction techniques use only a part of the data available obtained with a 360 gantry rotation for volume reconstruction. This technique significantly increases VAS at the expense of image noise (Bardo and Brown 2008; Yu and Pan 2003). The use of half reconstruction is warranted in kinematic CT, as the gain in artifact reduction largely outweighs the increase in noise. Various maneuvers can be performed for the evaluation of different joints. The most frequent recommendations for kinematic acquisitions are provided in Table 1. 2.3.2 Clinical Application A high level of knowledge of musculoskeletal anatomy is useful for the CT evaluation of subacute/chronic trauma. A multimodality approach with CT identifying calcification and bony anomalies and MR or ultrasound providing information on soft tissue structure (usually ligaments and tendons) can be useful to identify which structures have been injured. CT is also useful to confirm occult fractures or bone stress changes in cases in which MRI reveals only a nonspecific bone marrow edema pattern.

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery

1085

Table 1  Clinical recommendations Tube rotation Half reconstruction Tube output

Patient training

Patient immobilization

Motion axis Angular speed estimation

Recommendation As fast as possible Yes Adapted to the anatomic region. It should compensate for half reconstruction-related noise increase Should always be performed if possible (motion under volitional control). Motion speed should be controlled to be constant, with the least variation possible Limit motion to a single axis

Ascertain the predominant axis of motion with respect to the gantry rotation axis Evaluate prior to scanning the amplitude of displacement in degrees and the acquisition length

Goal Increase volume acquisition speed improve image temporal resolution Radiation dose optimization ALARAa principle Homogenous image quality throughout acquisition Allow estimation of the mean angular speed Allow determination of the fulcrum of motion Limit linear speeds at the examination zone Anticipate the influence of motion artifacts Assess if the protocol used is compatible with the motion speed being evaluated

Modified from Gondim Teixeira et al. (2017b) a ALARA as low as reasonably achievable

Kinematic CT has three main applications: Entrapment, impingement, and snapping. The advantage of kinematic studies is that the zone of impingement is directly demonstrated, helping confirm the diagnosis and guide surgical therapy. Kinematic CT can be useful for the evaluation of bony impingement in various joints such as the wrist, hip, and scapulothoracic joints (Fig. 4). Evaluation of the sufficiency of intra-articular ligaments. Kinematic CT arthrography allows in vivo visualization of intra-articular ligaments under stress. This technique can be used in the wrist to evaluate the sufficiency of the intrinsic ligaments of the first carpal row (Fig. 5). Further studies are needed to explore the application of this technique in other clinical scenarios. Analysis of complex motion. In some joints, such as the wrist and the subtalar joint, motion is complex with a strong rotary component, and dynamic pathology may be difficult to identify with static imaging or imaging in extremes of joint motion (van de Giessen et al. 2012). Kinematic CT is able to identify and measure small changes in bone relationships during the course of a given motion, making it particularly interesting for the evaluation of complex movements (Gondim Teixeira et al. 2017a) (Fig. 6).

Although kinematic CT is available for clinical use, more research is still necessary to identify generalizable diagnostic criteria and determine diagnostic performance.

3

Tumors

Detection of vascular enhancement is important for the characterization of bony and soft tissue tumors, and contrast-enhanced CT may help identify occult lesions (Gruber et al. 2016). In the MSK system, the prevalence of benign and pseudotumors is much higher than that of malignant tumors (Jo and Fletcher 2014). Thus, noninvasive tumor characterization is of great importance to reduce the morbidity of patient treatment and to better plan surgical interventions. Contrast-­ enhanced CT is frequently performed for the workup of patients with bone and soft tissue masses, either alone or in association with other imaging methods, particularly MRI (Gondim Teixeira et al. 2014b). It can provide information on lesion aggressiveness, vascular involvement, bone invasion, and the morphology of intra-­ tumoral calcifications. Due to its ability to demonstrate bone architectural changes and the morphology of tissue calcifications, contrast-­ enhanced CT can be considered as the gold

Fig. 4  A 26-year-old patient with grinding and snapping of the scapula during shoulder rotation and elevation. No morphologic anomaly could be found in static imaging (not shown). (1–3) Sequential volume-rendered 3D images of a kinematic CT study of

the shoulder with a tangential view to the scapulothoracic joint, showing a zone of scapulocostal impingement at the level of the superomedial angle of the scapula (red ellipse)

1086 P.A. Gondim Teixeira and A. Blum

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery

a

1087

b

1

2

3

4

Fig. 5  A 31-year-old female with wrist pain after trauma. CT arthrogram of the wrist in the coronal (a) and axial planes (b) demonstrate a partial rupture of the scapholunate ligament (thin arrow) with an intact dorsal portion (fat arrow) and a discrete irregularity at the insertion of

the ventral portion (arrowhead). (1–4) Sequential image series from a kinematic CT examination of the wrist during radioulnar deviation demonstrating a dynamic increase in the scapholunate gap (red circle), which is evocative of scapholunate ligament insufficiency

s­tandard for the characterization of bone neoplasms. Although MRI remains the method of choice for the characterization of soft tissue masses, these lesions are frequently first detected on CT. Moreover, contrast-enhanced CT can be considered for tumor characterization in patients with MR contraindications and for CT-guided interventions such as biopsies and percutaneous ablation. Novel imaging techniques such as CT perfusion and digital subtraction angiography (DSA) – like bone subtraction – can help further characterize MSK masses (Gondim Teixeira 2013; Teixeira et al. 2015b). Perfusion benefits from the high resolution provided by wide detector CT scanners and has the potential to improve MSK tumor characterization and posttreatment follow-up because some tumors have a known enhancement pattern, and it also provides an objective criteria for follow-up (Teixeira et al. 2015b). Non-lytic bone lesions and heavily calcified masses can seriously hinder visual analysis of contrast enhancement. DSA-like bone s­ ubtraction

allows clear visualization of iodinated contrast enhancement in a high-density background facilitating interpretation of contrast-enhanced CT images.

3.1

Acquisition Technique

The standard CT evaluation of an MSK mass requires a pre-contrast acquisition reconstructed with both bone and soft tissue kernels. Arterial phase acquisition is recommended for analysis of the relationship between the mass and neighboring vessels. A late acquisition 2–3 min after injection (the optimal delay will vary depending on the location of the mass with respect to the heart) can help identify areas of tumor necrosis and late enhancement. CT perfusion uses a low-dose intermittent acquisition of a target area to allow tissue enhancement analysis during the first passage of the contrast medium bolus. Contrast medium must be injected with a power injector, and the

1088

P.A. Gondim Teixeira and A. Blum

Fig. 6  Quantitative dynamic CT analysis of the subtalar joint during a prono-supination cycle in a 25-year-old female after an ankle sprain. Two markers are placed in the volume that shows the largest distance between the points to be evaluated (red square). The exact location of the selected points is displayed on the multiplanar

images (arrowheads). After post-processing, the markers are automatically projected in all volumes of the dynamic acquisition, and the graph on the lower right shows the distance variation during prono-supination of the foot

use of a bolus tracking technique is warranted to determine the optimal delay between injection and starting the acquisition. A 5-s inter-volume delay can be recommended for the arterial phase of the acquisition (first 45 s of the acquisition). For the venous phase, a 10-s inter-volume delay can be used in order to minimize radiation dose. The full acquisition length should be around 3 min. A standard contrast dose can be used (2 ml per kilogram) with a high injection rate (4–5 ml/s). General protocol guidelines for MSK CT perfusion are provided in Table 2.

DSA-like bone subtraction uses a pre-contrast acquisition phase (mask) that is registered and subtracted from one or more post-contrast acquisition phases (Fig. 7). Although the tube output parameters may vary between pre- and post-­ contrast acquisitions (the mask is usually acquired with a lower dose), all other acquisition parameters (kV, FOV, tube rotation speed, pitch, acquisition mode, etc.) must be identical. This technique can be used with not only conventional pre- and post-contrast acquisitions but also perfusion studies. For bone imaging, algorithms using

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery

a combination of rigid and nonregistration are preferred because they are more sensitive to contrast enhancement than those using nonrigid registration only. Finally, optimal patient immobilization is paramount to obtain good quality subtraction, particularly when using rigid registration. Motion between the mask and the other acquisition phases will lead to subtraction artifacts. Table 2  CT perfusion acquisition protocol recommendations extracted with permission from Gondim Teixeira et al. (2015) CT perfusion MSK Tube output Z-axis coverage Slice thickness Injection rate Contrast volume Bolus tracking Number of phases Inter-volume delay – arterial phase Inter-volume delay – venous phase

Adapted to anatomy and patient body habitus 40–160 mma 0.5 mm 5 ml/s 2 ml/kgb Yes 18 5 s 10 s

Z-axis coverage should be kept as short as possible to avoid unnecessary radiation exposure b Up to a maximum of 150 ml a

Fig. 7  Schematic representation of the DSA-like bone subtraction procedure in four steps. In the first two steps, the mask volume and the post-contrast volume are acquired. Images 1, 2, axial CT of the fifth lumbar vertebra. Osseous enhancement is not seen. Step 3 can performed any time after acquisition with a dedicated workstation. Image (4), axial bone-subtracted CT image

3.2

1089

Clinical Applications

CT allows the study of various aggressiveness criteria of bone tumors with high resolution and without the superimposition issues seen with conventional radiographs. The type of bone lysis and the presence of a sclerotic reaction and ­periosteal reaction are among the many features that can be assessed in great detail on CT. Moreover, this technique can shed additional light on the morphology of tissue calcification, which is helpful in determining the matrix of a bone tumor (e.g., chondroid versus osteoid). Helical acquisition mode can be used to image whole bones, which can be difficult and timeconsuming with MRI, especially in long bones such as the femur or humerus, which are susceptible to magnetic field inhomogeneity artifacts at the extremities. Whole bone imaging is useful for staging and preoperative planning to locate the lesion with respect to the proximal and distal joints (Fig. 8). This information is particularly important for limb salvage surgery planning (Ruggieri et al. 2011). Whole bone imaging is also warranted in the search for skip metastasis in cases of aggressive bone lesions such as osteosarcoma or Ewing’s sarcoma (Leavey et al. 2003; Davies et al. 1997).

of the same anatomic region. Following bone subtraction, a focal zone of contrast enhancement is seen at the right superior articular facet (arrow). Retrospectively, a discrete bone irregularity is visible in the non-subtracted volumes. Percutaneous bone biopsy confirmed the diagnosis of an osteoid osteoma (Gondim Teixeira et al. 2015)

P.A. Gondim Teixeira and A. Blum

1090

a

b

Fig. 8  A 17-year-old male with a distal femur sarcoma. (a) Coronal CT of the whole femur demonstrating the relationship between the lesion and the proximal and distal articular surfaces. Note also that there are no skip metastases to be identified. Moreover a Codman triangle

can be seen (arrowhead). (b) Axial CT image demonstrating an aggressive-type sunburst periosteal reaction (arrows) along with irregular bone lysis without marginal sclerosis (fat arrow), all of which are signs of tumor aggressiveness

CT perfusion can provide useful hemodynamic information in cases of vascular tumor or micro-traumatic vascular injury (e.g., hypothenar hammer syndrome, vascular malformations). Similar to MRI perfusion, CT perfusion can be useful for tumor characterization, differentiation between residual and recurrent tumor, and postoperative fibrosis and biopsy targeting (Teixeira

et al. 2013, 2014b). Some bone and soft tissue masses have suggestive enhancement patterns. This is the case for osteoid osteomas, osteoblastomas, giant cell tumors, and solitary fibrous tumors that usually present with intense and early enhancement (Fig. 9). Other lesions such as typical enchondroma, myxoid tumors, and fibrous lesions usually present with weak and p­ rogressive

Fig. 9  A 20-year-old male with a typical osteoid osteoma of the mid diaphysis of the right radius before and after treatment. (a) Coronal CT image showing a well-defined nidus with a central calcification (white arrow) and a marked periosteal reaction. (b) Pretreatment time-to-­ density curve demonstrating a type IV enhancement curve of the nidus (blue curve), with an early enhancement with respect to the ulnar artery (yellow curve). (c) Arterial flow colored map in the sagittal plane demonstrating the high arterial flow in the nidus (white arrowheads) and the cor-

responding CT image. (d) Volume-rendered CT image depicting the nidus cavity (fat black arrow) and an adequate needle trajectory (thin black arrow) after percutaneous ablation therapy. (e) Posttreatment CT perfusion time-todensity curve demonstrating the perfusion change in the nidus, which does not present any detectable enhancement (blue curve). (f) Postoperative arterial flow colored map in the sagittal plane demonstrating a marked drop in the nidus arterial flow (white arrowheads) with respect to the preoperative study and the corresponding CT image

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery

a

c

b

d

e

f

1091

P.A. Gondim Teixeira and A. Blum

1092

enhancement. This information may be useful for lesion characterization in specific cases. Perfusion, however, is not a good method for differentiation of benign versus malignant tumors due to the important overlap in the perfusion characteristics of MSK tumors (van Rijswijk et al. 2004; van der Woude et al. 1998). CT perfusion could also be used for intraoperative assessment of the effectiveness of percutaneous ablation in highly vascular tumors such as osteoid osteomas, which tend to show little or no enhancement after successful treatment (Gondim Teixeira et al. 2014b; Teixeira et al. 2013) (Fig. 9). It is important to keep in mind that MRI remains the method of choice for perfusion studies due to its higher sensitivity in contrast detection and the absence of ionizing radiation. CT perfusion is, however, a valuable option when MRI is contraindicated or unavailable. Despite these advantages of MRI over CT, in clinical practice, CT and MRI perfusion have a similar diagnostic performance (Otton et al. 2013; De Simone et al. 2013). The main advantage of CT over MRI relates to the linear association between iodine concentration and CT number, which makes density-related perfusion parameters easier to calculate and compare with the former. Radiation dose remains an important issue with CT perfusion, and this technique is better suited for extremities, where it can be performed with effective doses under 0.5 mSv. Bone tumors frequently have a nonspecific appearance on imaging. The diagnosis in these cases relies on invasive procedures (percutaneous or surgical biopsy) for the histologic confirmation of the lesion’s nature. Relatively few lytic bone lesions are known to be associated with bone marrow edema pattern (BMEP), which is a distinguishing feature in such cases, including osteomas, osteoblastomas, chondroblastomas, and eosinophilic granulomas (James et al. 2008). In addition, the size of the edematous bone marrow reaction adjacent to a bone tumor may be used as a marker of chemotherapeutic response (Fletcher 1991). DSA-like bone subtraction makes it possible to identify BMEP with CT, which in some cases may help reach a conclusive

diagnosis and reduce the need of additional imaging investigations (Fig. 10). The presence of BMEP is also a key feature for the diagnosis of stress fracture, bone-on-bone impingement, and osteomyelitis. The use of DSA-like bone subtraction may help identify BMEP areas directly, which can expedite the diagnosis, since in many of these cases, CT is the first evaluation method employed. Identification of an intraosseous-enhancing lesion may be of significant clinical importance because it may be related to early metastatic disease or aggressive bone tumors.

4

Postsurgery

Imaging plays a key role in monitoring the outcomes of orthopedic surgery, follow-up of patients, and investigation of complications. Standard radiography is generally the principal method of monitoring patients and the first diagnostic step in search of complications, but improving the performance of CT scan and MRI modifies the traditional diagnostic strategy (Blum et al. 2016a). For the majority of imaging methods used in postoperative checks, the most striking technical development in the recent years is undoubtedly the reduction of metal artifacts (MAR). However, the various imaging methods differ in their ability to analyze the surgical material and its environment (Blum et al. 2016a, b). These methods can now be classified according to the analyzable structures: metallic material, its interface with cement or bone, cement, bone, and soft tissue. Thus, MRI with MARS sequences provides good visibility of the structures close to prostheses but does not permit analysis of the implants themselves (Koch et al. 2010). Tomosynthesis improves the performance of standard radiography but does not show the soft tissue (Machida et al. 2016). Ultrasound performs well in the study of periprosthetic soft tissue, but analysis of the bone and the material is insufficient. Only the CT can provide satisfactory visualization of all the structures (Blum et al. 2016a, b).

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery

a

b

c

d

1093

Fig. 10  A 71-year-old patient with a spindle cell sarcoma of the popliteal fossa, which completely regressed after chemotherapy. The patient presents 1 year later with posterior knee pain. (a) Axial CT image with a bone kernel showing no signs of bone lysis. (b) Axial CT perfusion image in the late arterial phase demonstrating an early enhancing popliteal mass (thin arrows) evocative of a tumor recurrence and no clear abnormality in the bone

marrow. (c) DSA-like bone subtraction of the same CT perfusion study demonstrating clear enhancement in the bone marrow (fat arrow) adjacent to the soft tissue mass (thin arrows), which indicates bone invasion. (d) Axial T2-weighted fat-saturated MR image confirming marrow invasion seen at CT with DSA-like bone subtraction (wide arrow)

4.1

intensity also depends on the metal alloy used, being greater with chromium-cobalt than stainless steel, which in turn causes more artifacts than titanium. Technically, high exposure settings (kV and mAs), pitch less than 1, thin-section acquisition, and images reconstructed from thin sections with a standard convolution filter can partially reduce metal artifacts. Based on these findings, Roth et al. recommended acquisition parameters of 140 kV and 350–450 mAs in patients with a

Acquisition Techniques

4.1.1 Metal Artifact Reduction Conventional Technique CT has long been used to investigate metal implants, but until recently, it did not perform well owing to the presence of metal artifacts, which were all the more pronounced for prostheses with large metallic components and in patients with bilateral hip prostheses. Artifact

P.A. Gondim Teixeira and A. Blum

1094

s­ingle hip prosthesis and 450–650 mAs in patients with bilateral hip prostheses (Roth et al. 2012). Nonetheless, other authors recommend using 120 kV as the ability of the photon beam to penetrate metal is not improved with higher energies (Lee et al. 2007). Iterative Reconstruction Iterative reconstruction is mainly used to lower image noise and ultimately reduce exposure dose. Several algorithms have been developed. There are two main types: hybrid algorithms that combine analytical and iterative methods and model-based iterative reconstruction (MBIR) algorithms that take into account system geometry, the scanning process, and the statistical model describing the interaction of photons with matter (Geyer et al. 2015). The effects of these algorithms are variable. Generally, they improve the quality of the image by reducing noise, X-ray beam-hardening artifacts, and to some extent metal artifacts. They can also alter the noise/texture characteristics of the image. Regardless of the algorithm used, they all allow the use of reasonable exposure settings that are adapted to patient morphology and are the same as those used for patients without prostheses. These algorithms have now completely replaced the use of filtered back projection, and in some conditions, they make it possible to a

b

Fig. 11  Follow-up after bone grafting for scaphoid nonunion with ultra-low-dose technique in a 28-year-old man. (a) Standard radiography shows a suspicion of bone graft avascular necrosis (black arrow). (b) Coronal MPR with hybrid iterative reconstruction algorithm and low-dose technique (DLP = 15.9 mGy.cm, ED = 3.5 μSv) demonstrates the non-incorporation of the graft (white arrow).

p­erform sub-microSv CT scans (Fig. 11) (Gervaise et al. 2013; Morsbach et al. 2013). Finally, for very small metal objects, MBIR reduces the distortion of the hardware and improves the visibility of structures in its vicinity compared to all the other algorithms (Fig. 12). Monochromatic Reconstruction of Dual-­ Energy CT Scans Dual-energy scans are used to simulate high-­ energy monochromatic beam acquisitions that reduce both X-ray beam-hardening artifacts and metal artifacts (Bamberg et al. 2011; Zhou et al. 2011; Bongers et al. 2015). This approach is interesting but clearly insufficient. It reduces beam-hardening artifacts but cannot compensate for insufficient numbers of photons reaching the detectors – the main cause of metal artifacts. It therefore performs less well for imaging the prosthetic hip than for smaller metallic objects. In addition, simulation of a high-energy beam leads to decreased contrast resolution. Finally, various factors affect the optimal energy level, such as the type of CT scanner, and implant properties. Consequently, alone, this technique is of relatively little interest for imaging the prosthetic hip. To be of use, it generally needs to be combined with a metal artifact reduction algorithm (Lee et al. 2012; Pessis et al. 2013). c

(c) Coronal MPR with model-based iterative reconstruction algorithm and ultra-low-dose technique (DLP = 0.6 mGy.cm, ED = 0.13 μSv). Although the image quality is decreased, the quality of the examination is sufficient to provide the necessary information. DLP dose length product, ED effective dose

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery

a

c

b

d

1095

Fig. 12  Follow-up after fixation of an osteochondral fragment in a 16-year-old man. (a, b) Axial slice and sagittal MPR obtained with a hybrid iterative reconstruction algorithm. (c, d) Same slices obtained with a model-based iterative reconstruction algorithm. With the hybrid itera-

tive reconstruction algorithm, a metal artifact obscures the bony structures surrounding the screw. This artifact is not present with the model-based iterative reconstruction (MBIR) algorithm yielding more precision about the healing status of the bone

Metal Artifact Reduction Algorithms Metal artifact reduction (MAR) algorithms are intended to compensate for photon starvation caused by hyperattenuating material along the X-ray path, which cannot be achieved with filtered back projection. Several methods have been described (Gondim Teixeira et al. 2014a; Verburg and Seco 2012; Liu et al. 2009; Kalender et al. 1987). Projection-interpolation-based methods are implemented to remove corrupted projections from the raw data and then compensate for the missing data by interpolation of non-missing neighboring projections. However, although these methods do not require large computational capacities, they tend to generate artifacts and are being abandoned.

In iterative methods, corrupted projections are removed and followed by iterative reconstruction using the whole set of remaining projections. An alternative method consists of segmenting the metal in the image and reconstructing its sinogram, metal subtraction from the original sinogram, interpolation of missing data, image reconstruction with iterative reconstruction, and then addition of the initially extracted metal (Fig. 13). The combined use of these algorithms with conventional scanning techniques performs better than those based on dual-energy acquisitions (Bongers et al. 2015; Higashigaito et al. 2015; Andersson et al. 2015). However, they require greater computational power and longer reconstruction times than conventional iterative reconstruction (Jeong et al. 2015).

P.A. Gondim Teixeira and A. Blum

1096

Interpolated sinogram

Original projection data 4

Metal trace in sinogram

1

Elimination of metal objects

FBP 3

Original image

Forward projection

Interpolated image

Metal image

2

5

Extraction of metal objects Image correction

8

Final Image

Addition of metal objects

6

7 Iterative Reconstruction

Corrected sinogram

Fig. 13  Principle of the SEMAR (Single Energy Metal Artifact Reduction) algorithm. This algorithm comprises eight steps that is divided into three phases. The first phase (steps 1–3) is aimed at metal deletion. During the second phase (steps 4–7), the data missing after metal deletion are

interpolated to reconstruct the image without metal artifacts. In the third phase (step 8), the metal image is blended to the reconstructed image. FBP filter back projection

MAR techniques are the most efficient way of removing metal artifacts, and their performances are not reduced in patients with bilateral implants (Gondim Teixeira et al. 2014a). On the other hand, they do alter the visualization of metal components and sometimes their immediate vicinity. Although mainly intended to improve the visualization of the soft tissue, MAR algorithms also have a positive effect on viewing periprosthetic bone (Fig. 14).

to brighten contrast-enhanced regions (increased signal-to-background ratio) by mask subtraction as previously presented (Fig. 15).

4.1.2 Intravenous Injection of Contrast Medium and Subtraction Techniques Intravenous injection of contrast medium may be indicated in three circumstances: potential abscess localization, vascular damage, and recurrent tumor activity. The previously described MAR techniques improve lesion visualization, but some scanners also include features intended

4.1.3 Visualization and Post-processing It is important to keep in mind that images must always be reviewed in at least three different window settings specifically adjusted for viewing the different tissue types and implant components: soft tissue window, bone and cement window, and implant component window (W = 8,000 HU; L = 1,000 HU). Due to its isotropic capabilities, CT provides high-resolution multiplanar reconstructions (MPR), the plane of which can be perfectly adjusted to the alignment of the metallic material. 3D reconstruction is of limited diagnostic interest and is not routinely performed. It p­ rovides a global view of the prosthesis which can

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery

a

b

1097

d

e

c

Fig. 14  Septic loosening of a right total hip arthroplasty in a 69-year-old. man. (a) Standard radiography shows a wide radiolucency at the cement/bone interface, cortical bone destruction at the medial aspect of the femur, and a periosteal reaction. (b, c) Axial slices without and with MAR showing a joint effusion (* ) and a collection adja-

a

b

c

cent to the joint (arrow head), better depicted with MAR, (d, e) coronal MPR with bone and soft tissue kernels and MAR showing the bone destruction and a soft tissue abscess (arrow). Note that some portion of the ball of the femoral implant is erased

d

e

Fig. 15  Asymptomatic muscle metastasis in a 69-year-­ old man with a massive implant for a grade III pleomorphic sarcoma of the right femur. CT was used for follow-up. (a, b) Coronal MPR with MAR and 3D (VRT)

image showing a massive prosthesis. (c–e) Axial slices without MAR, with MAR, and with MAR and subtraction depicting a small nodule (arrow) close to the metal implant

P.A. Gondim Teixeira and A. Blum

1098

s­ ometimes be useful when assessing component fracture or implant migration. 3D reconstructions are generally used following reconstruction with a standard filter and iterative reconstruction, as MAR algorithms may modify how metal components are visualized (Fig. 15).

4.2

Clinical Applications

The indications for postoperative scanning are varied, and its use has increased considerably since the development of techniques for the reduction of metallic artifacts. The most common applications are the evaluation of bone structures to assess their level of consolidation in the presence of metallic material, evaluation of arthroplasty complications, and monitoring of patients treated for bone sarcoma.

4.2.1 Bone Healing Assessment of fracture union is a critical concept in clinical orthopedics. There is no universally accepted method of determining fracture healing; therefore, orthopedic surgeons must rely on a range of tools that may include radiographic assessment, mechanical assessment, serologic markers, and clinical evaluation (including functional outcomes). Despite its limitations, radiographic assessment has remained a crucial tool in determining fracture healing. CT is superior to plain radiography in the assessment of union and visualization of fracture in the presence of an abundant callus or overlaying cast or in fractures involving the metaphyseal bone or the carpal bones, which tend to heal with less callus (Cook et al. 2015; Morshed 2014). CT is also a very efficient technique to detect a sequestrum. CT is a useful technique to visualize the interface between the bone graft materials and bone itself. Bone graft materials are widely used in reconstructive orthopedic procedures to promote new bone formation and bone healing and provide a substrate and scaffolding for the development of bone structure. The imaging appearance of an autograft depends on its type, composition, and age (Beaman et al. 2006). The CT appearance of autografts is similar to the adjacent bone.

The primary sources of bone autografts are the iliac crest and the fibula. An allograft (cadaveric bone transplants) has an appearance similar to that of the cortical bone. The discrete boundary between host and graft is initially identifiable; however, as union progresses, the graft-host junction is obliterated as a result of trabecular ingrowth. Synthetic bone substitutes are much more variable in imaging appearance. Bone grafts are often associated with metal hardware, and the new MAR algorithms are very helpful to minimize metal artifacts and verify graft integration.

4.2.2 Prosthesis Complications Over the last 10 years, the frequency of joint replacement surgery has steadily increased due to the aging population and the effectiveness of such implants in alleviating pain and restoring joint function (Kurtz et al. 2007). Despite the improved life span of prosthetic implants, the growing life expectancy of the population is resulting in an increased number of prosthesis-­ related complications and hence a global increase in the number of revision surgeries. Prosthesis-­ related complications are diverse and depend largely on the type of prosthesis used. They can involve the implant itself, the bone in which the implant is placed, as well as the joint and surrounding soft tissue (Figs. 16, 17, and 18). More accurate, earlier detection of some of these complications has recently become possible owing to the progress in imaging techniques. The use of such techniques has quickly spread due to concern raised by the high rate of complications associated with resurfacing implants and has most probably led to an excessive number of diagnostic imaging examinations. Many articles and consensus conferences emphasize the prominent role of MRI in depicting soft tissue pseudo-tumors related to metal-­ on-­ metal implants (Günther et al. 2013; Hannemann et al. 2013; Kwon et al. 2014). However, most reports in the literature focus on a single modality and do not attempt to compare its best performance with other imaging techniques. Therefore, the best strategy for the diagnosis of implant complications remains unclear. Based

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery

a

b

c

d

1099

Fig. 16  Dislocation of the polyethylene of the glenoid component in a 72-year-old woman with an anatomical shoulder prosthesis. (a, b) Axial slice and sagittal MPR images fail to show any anomaly. (c, d) Axial slice and

sagittal MPR reconstructed with MAR algorithm clearly depict the polyethylene insert located at the posteroinferior aspect of the joint (arrow)

upon a review of the recent literature and their experience with all imaging techniques at their best technological level, a panel of experts in fields related to prosthesis imaging (radiology, nuclear medicine, orthopedic surgery) found CT to be the most versatile and cost-effective ­imaging

solution and therefore a key tool for diagnosing the complications of hip replacement surgery (Blum et al. 2016a, b) (Fig. 19). With the same semiotic as previously described in the literature with CT, metal artifact reduction techniques make the diagnosis of most complications

P.A. Gondim Teixeira and A. Blum

1100

a

b

d

c

e

d

Fig. 17  Septic loosening of a right total hip arthroplasty in a 93-year-old man. (a) Radiography showing the loosening of the polyethylene cup. (b) Axial slice without

a

b

c

Fig. 18  Total hip arthroplasty dislocation in a 69-year-­ old woman with pelvic migration of the acetabular component. (a) Radiography showing the breakage and dislocation of the implant. (b, c) 3D (VRT) highlighting the migration of each individual metal component. (d)

MAR. (c–e) Axial slice and coronal MPR showing a joint effusion with gas bubbles (arrows) as well as gas bubbles in the great tuberosity (arrow head) indicating a sepsis

d

Axial slice with MAR showing the metal back (black arrow) and the polyethylene insert (white arrow) migration in the pelvis embedded in a pelvic collection with dense materials indicating a metallosis (arrow heads)

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery

1101

Clinical history and physical examination

Standard radiography and blood tests

Hardware breakage periprosthetic fracture prosthesis malposition

Aseptic loosening

questionable

CT-scan

CT-scan or SPECT-CT

Particle disease

Infection

Iliopsoas impingement

Muscular or tendinous lesion

Osseous sarcoma follow-up

definite osteolysis

CT-scan

CT-scan

MRI or CT-scan or PET-scan and joint puncture

US and CT-scan

US

CT-scan

Fig. 19  Algorithm for the evaluation of hip prosthesis depending upon history of the patient, physical examination, radiography evaluation, and blood tests according to

the Nancy Association for Prosthesis Exploration (NAPE) group (Blum et al. 2016b)

straightforward. CT becomes a one-stop shop technique to evaluate the hardware, the metal-­ cement-­bone interfaces, the osseous structures, and the surrounding soft tissues. In Addition, the intra-articular anomalies can now be depicted.

therapy and the advent of accurate diagnostic imaging techniques for the precise delineation of the extent of tumors. LSS techniques include tumor excision with wide surgical margins followed by reconstruction of the resulting surgical defect. Small resected segments of the bone may be reconstructed using autograft, but in most cases, defects are too large to use autografts. In such cases, the options for reconstruction include massive prosthesis or allograft prosthetic composite (APC). Massive prostheses are nowadays modular replacement systems which allow a customization of the implant to perfectly fit the bone and compensate for the defect. An APC is usually the combination of a fresh-frozen bone allograft and a metallic prosthesis for segmental and articular reconstruction, respectively. Pelvic reconstruction techniques include homolateral, proximal, femoral autograft and total hip prosthesis, saddle prosthesis, auto- or allograft, modular prosthesis or custom-made prosthesis, and femoro-iliac arthrodesis (Anract et al. 2014). Three types of complications may occur: infection, mechanical complications, and tumor

4.2.3 F  ollow-Up of Limb Salvage Surgery or Pelvic Reconstruction The follow-up of limb salvage surgery (LSS) or pelvic reconstruction is critical and highly demanding. Indeed, the rate of complication or recurrence is high, and the large amount of metal associated with reconstruction techniques limits the performances of the different imaging techniques (Beaman et al. 2006; Fritz et al. 2012; Tan et al. 2015; Chakarun et al. 2013; Bancroft 2011). Surgery remains the cornerstone of the management of primary bone sarcomas. Indications for such surgical procedures are anticipated tumor removal with adequate margins, along with anticipated acceptable functional and cosmetic results. Seventy to ninety percent of extremity osteogenic sarcomas are treated with LSS. Factors that have advanced the field of LSS include the development of effective c­hemotherapy and radiation

P.A. Gondim Teixeira and A. Blum

1102

recurrence. Mechanical complications include wearing and loosening of the implant, dislocation, hardware fracture, and periprosthetic osseous fracture (Fritz et al. 2012; Chakarun et al. 2013). The rate and type of complication depend on many factors (surgical margins, tumor grade, type of surgery, type of implant, anatomic site, irradiation of the allograft, and adjuvant treatment) (Aurégan et al. 2016). Radiography is the standard imaging modality for postoperative assessment after LSS. This technique is quite useful for the detection of mechanical complications, but its inability to evaluate the soft tissues and provide cross-­ sectional imaging limits its use in complex cases (Bancroft 2011). According to Fritz et al., CT is one of the most effective techniques for the assessment of suspected postoperative complications (Fritz et al. 2012). In our experience, MAR techniques increase the performance of CT, making this technique the standard for the evaluation of LSS or pelvic reconstruction. CT is used without contrast medium injection for the evaluation of mechanical complications. Contrast-enhanced CT is helpful in cases of septic complication or recurrence. The subtraction technique which highlights the enhancing tissues further improves the performance of this technique (Fig. 15). Conclusion

This chapter discusses the most important technical and practical aspects of the application of CT for the assessment of traumatic, oncologic, and postsurgery MSK conditions. Wide detector CT scanners offer great advantages in the evaluation of MSK lesions, facilitating the use of the low-­ dose multiphasic protocols that are the basis for advanced techniques such as perfusion and kinematic imaging. Additionally, novel CT-based techniques related to advances in imaging post-­processing such as DSA-like bone subtraction and MAR have been presented. This information is important to allow radiologists and technicians to use the full diagnostic potential of modern CT scanners in the evaluation of MSK pathology.

Bibliography Andersson KM, Nowik P, Persliden J, Thunberg P, Norrman E (2015) Metal artefact reduction in CT imaging of hip prostheses – an evaluation of commercial techniques provided by four vendors. Br J Radiol 88(1052):20140473 Anract P, Biau D, Babinet A, Tomeno B (2014) Pelvic reconstructions after bone tumor resection. Bull Cancer 101(2):184–194 Aurégan J-C, Pietton R, Bégué T, Anract P, Biau D (2016) Effect of anatomic site and irradiation on the rates of revision and infection of allograft-prosthesis composites after resection of a primary bone tumor: a meta-­analysis. Arch Orthop Trauma Surg 136(10):1371–1380 Bamberg F, Dierks A, Nikolaou K, Reiser MF, Becker CR, Johnson TRC (2011) Metal artifact reduction by dual energy computed tomography using monoenergetic extrapolation. Eur Radiol 21(7):1424–1429 Bancroft LW (2011) Postoperative tumor imaging. Semin Musculoskelet Radiol 15(4):425–438 Bardo DME, Brown P (2008) Cardiac multidetector computed tomography: basic physics of image acquisition and clinical applications. Curr Cardiol Rev 4(3):231–243 Beaman FD, Bancroft LW, Peterson JJ, Kransdorf MJ, Menke DM, DeOrio JK (2006) Imaging characteristics of bone graft materials. Radiographics 26(2):373–388 Beeres M, Wichmann JL, Paul J, Mbalisike E, Elsabaie M, Vogl TJ et al (2015) CT chest and gantry rotation time: does the rotation time influence image quality? Acta Radiol 56(8):950–954 Blum A, Meyer J-B, Raymond A, Louis M, Bakour O, Kechidi R et al (2016a) CT of hip prosthesis: new techniques and new paradigms. Diagn Interv Imaging 97(7–8):725–733 Blum A, Gondim-Teixeira P, Gabiache E, Roche O, Sirveaux F, Olivier P et al (2016b) Developments in imaging methods used in hip arthroplasty: a diagnostic algorithm. Diagn Interv Imaging 97(7–8):735–747 Bongers MN, Schabel C, Thomas C, Raupach R, Notohamiprodjo M, Nikolaou K et al (2015) Comparison and combination of dual-energy- and iterative-based metal artefact reduction on hip prosthesis and dental implants. PLoS One 10(11):e0143584 Brenner DJ, Hall EJ (2007) Computed tomography – an increasing source of radiation exposure. N Engl J Med 357(22):2277–2284 Chakarun CJ, Learch TJ, White EA, Menendez LR, Brien EW, Matcuk GR et al (2013) Limb-sparing surgery for distal femoral and proximal tibial bone lesions: imaging findings with intraoperative correlation. AJR Am J Roentgenol 200(2):W193–W203 Cook GE, Bates BD, Tornetta P, McKee MD, Morshed S, Slobogean GP et al (2015) Assessment of fracture repair. J Orthop Trauma 29(Suppl 12):S57–S61 Cotten A (2013) Imagerie musculosquelettique: pathologies générales, 2nd ed. Elsevier Masson, p 1064

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery Dappa E, Higashigaito K, Fornaro J, Leschka S, Wildermuth S, Alkadhi H (2016) Cinematic rendering – an alternative to volume rendering for 3D computed tomography imaging. Insights Imaging 7(6):849–856 Davies AM, Makwana NK, Grimer RJ, Carter SR (1997) Skip metastases in Ewing’s sarcoma: a report of three cases. Skelet Radiol 26(6):379–384 De Simone M, Muccio CF, Pagnotta SM, Esposito G, Cianfoni A (2013) Comparison between CT and MR in perfusion imaging assessment of high-grade gliomas. Radiol Med 118(1):140–151 Ebert LC, Schweitzer W, Gascho D, Ruder TD, Flach PM, Thali MJ et al (2016) Forensic 3D visualization of CT data using cinematic volume rendering: a preliminary study. AJR Am J Roentgenol 8:1–8 Farshad-Amacker NA, Alkadhi H, Leschka S, Frauenfelder T (2013) Effect of high-pitch dual-source CT to compensate motion artifacts: a phantom study. Acad Radiol 20(10):1234–1239 Fletcher BD (1991) Response of osteosarcoma and Ewing sarcoma to chemotherapy: imaging evaluation. AJR Am J Roentgenol 157(4):825–833 Foster BR, Anderson SW, Uyeda JW, Brooks JG, Soto JA (2011) Integration of 64-detector lower extremity CT angiography into whole-body trauma imaging: feasibility and early experience. Radiology 261(3):787–795 Fritz J, Fishman EK, Corl F, Carrino JA, Weber KL, Fayad LM (2012) Imaging of limb salvage surgery. AJR Am J Roentgenol 198(3):647–660 Gervaise A, Teixeira P, Villani N, Lecocq S, Louis M, Blum A (2013) CT dose optimisation and reduction in osteoarticular disease. Diagn Interv Imaging 94(4):371–388 Geyer LL, Schoepf UJ, Meinel FG, Nance JW, Bastarrika G, Leipsic JA et al (2015) State of the art: iterative CT reconstruction techniques. Radiology 276(2):339–357 Geyer LL, Körner M, Harrieder A, Mueck FG, Deak Z, Wirth S et al (2016) Dose reduction in 64-row whole-­ body CT in multiple trauma: an optimized CT protocol with iterative image reconstruction on a gemstonebased scintillator. Br J Radiol 89(1061):20160003 Gibson PD, Bercik MJ, Ippolito JA, Didesch J, Hwang JS, Koury KL et al (2016) The role of computed tomography scans in surgical planning for trimalleolar fracture. A survey of OTA members. J Orthop Trauma Gondim Teixeira PA (2013) Musculoskeletal disorders: DSA-like bone subtraction with 320 detector row CT. Visions magazine 21 Gondim Teixeira PA, Meyer J-B, Baumann C, Raymond A, Sirveaux F, Coudane H et al (2014a) Total hip prosthesis CT with single-energy projection-based metallic artifact reduction: impact on the visualization of specific periprosthetic soft tissue structures. Skelet Radiol 43(9):1237–1246 Gondim Teixeira PA, Lecocq S, Louis M, Aptel S, Raymond A, Sirveaux F et al (2014b) Wide area detector CT perfusion: can it differentiate osteoid osteomas

1103

from other lytic bone lesions? Diagn Interv Imaging 95:587–594 Gondim Teixeira PA, Gervaise A, Louis M, Lecocq S, Raymond A, Aptel S et al (2015) Musculoskeletal wide detector CT: principles, techniques and applications in clinical practice and research. Eur J Radiol 84(5):892–900 Gondim Teixeira PA, Formery A-S, Jacquot A, Lux G, Loiret I, Perez M et al (2017a) Quantitative analysis of subtalar joint motion with 4D CT: proof of concept with cadaveric and healthy subject evaluation. AJR Am J Roentgenol 208(1):150–158 Gondim Teixeira PA, Formery A-S, Hossu G, Winninger D, Batch T, Gervaise A et al (2017b) Evidence-based recommendations for musculoskeletal kinematic 4D-CT studies using wide area-detector scanners: a phantom study with cadaveric correlation. Eur Radiol 27(2):437–446 Gruber L, Loizides A, Luger AK, Glodny B, Moser P, Henninger B et al (2016) Soft-tissue tumor contrast enhancement patterns: diagnostic value and comparison between ultrasound and MRI. AJR Am J Roentgenol 13:1–9 Günther K-P, Schmitt J, Campbell P, Delaunay CP, Drexler H, Ettema HB et al (2013) Consensus statement “current evidence on the management of metal-­ on-­ metal bearings” – April 16, 2012. Hip Int 23(1):2–5 Hannemann F, Hartmann A, Schmitt J, Lützner J, Seidler A, Campbell P et al (2013) European multidisciplinary consensus statement on the use and monitoring of metal-on-metal bearings for total hip replacement and hip resurfacing. Orthop Traumatol Surg Res 99(3):263–271 Higashigaito K, Angst F, Runge VM, Alkadhi H, OF D (2015) Metal artifact reduction in pelvic computed tomography with hip prostheses: comparison of virtual monoenergetic extrapolations from dual-energy computed tomography and an iterative metal artifact reduction algorithm in a phantom study. Investig Radiol 50(12):828–834 James SLJ, Panicek DM, Davies AM (2008) Bone marrow oedema associated with benign and malignant bone tumours. Eur J Radiol 67(1):11–21 Jeong S, Kim SH, Hwang EJ, Shin C-I, Han JK, Choi BI (2015) Usefulness of a metal artifact reduction algorithm for orthopedic implants in abdominal CT: phantom and clinical study results. AJR Am J Roentgenol 204(2):307–317 Jo VY, Fletcher CDM (2014) WHO classification of soft tissue tumours: an update based on the 2013 (4th) edition. Pathology 46(2):95–104 Kalender WA, Hebel R, Ebersberger J (1987) Reduction of CT artifacts caused by metallic implants. Radiology 164(2):576–577 Koch KM, Hargreaves BA, Pauly KB, Chen W, Gold GE, King KF (2010) Magnetic resonance imaging near metal implants. J Magn Reson Imaging 32(4):773–787

1104 Kurtz S, Ong K, Lau E, Mowat F, Halpern M (2007) Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 89(4):780–785 Kwon Y-M, Lombardi AV, Jacobs JJ, Fehring TK, Lewis CG, Cabanela ME (2014) Risk stratification algorithm for management of patients with metal-on-metal hip arthroplasty: consensus statement of the American Association of Hip and Knee Surgeons, the American Academy of Orthopaedic surgeons, and the hip society. J Bone Joint Surg Am 96(1):e4 Leavey PJ, Day MD, Booth T, Maale G (2003) Skip metastasis in osteosarcoma. J Pediatr Hematol Oncol 25(10):806–808 Lee T-Y, Chhem RK (2010) Impact of new technologies on dose reduction in CT. Eur J Radiol 76(1):28–35 Lee M-J, Kim S, Lee S-A, Song H-T, Huh Y-M, Kim D-H et al (2007) Overcoming artifacts from metallic orthopedic implants at high-field-strength MR imaging and multi-detector CT. Radiographics 27(3):791–803 Lee YH, Park KK, Song H-T, Kim S, Suh J-S (2012) Metal artefact reduction in gemstone spectral imaging dual-energy CT with and without metal artefact reduction software. Eur Radiol 22(6):1331–1340 Liu PT, Pavlicek WP, Peter MB, Spangehl MJ, Roberts CC, Paden RG (2009) Metal artifact reduction image reconstruction algorithm for CT of implanted metal orthopedic devices: a work in progress. Skelet Radiol 38(8):797–802 Machida H, Yuhara T, Tamura M, Ishikawa T, Tate E, Ueno E et al (2016) Whole-body clinical applications of digital tomosynthesis. Radiographics 36(3):735–750 Mendel T, Radetzki F, Wohlrab D, Stock K, Hofmann GO, Noser H (2013) CT-based 3-D visualisation of secure bone corridors and optimal trajectories for sacroiliac screws. Injury 44(7):957–963 Morsbach F, Bickelhaupt S, Wanner GA, Krauss A, Schmidt B, Alkadhi H (2013) Reduction of metal artifacts from hip prostheses on CT images of the pelvis: value of iterative reconstructions. Radiology 268(1):237–244 Morshed S (2014) Current options for determining fracture union. Adv Med 2014:708574 Omoumi P, Rubini A, Dubuc J-E, Vande Berg BC, Lecouvet FE (2015) Diagnostic performance of CT-arthrography and 1.5 T MR-arthrography for the assessment of glenohumeral joint cartilage: a comparative study with arthroscopic correlation. Eur Radiol 25(4):961–969 Otton J, Morton G, Schuster A, Bigalke B, Marano R, Olivotti L et al (2013) A direct comparison of the sensitivity of CT and MR cardiac perfusion using a myocardial perfusion phantom. J Cardiovasc Comput Tomogr 7(2):117–124 Parrish FJ (2007) Volume CT: state-of-the-art reporting. AJR Am J Roentgenol 189(3):528–534

P.A. Gondim Teixeira and A. Blum Pessis E, Campagna R, Sverzut J-M, Bach F, Rodallec M, Guerini H et al (2013) Virtual monochromatic spectral imaging with fast kilovoltage switching: reduction of metal artifacts at CT. Radiographics 33(2):573–583 Roth TD, Maertz NA, Parr JA, Buckwalter KA, Choplin RH (2012) CT of the hip prosthesis: appearance of components, fixation, and complications. Radiographics 32(4):1089–1107 Ruggieri P, Mavrogenis AF, Mercuri M (2011) Quality of life following limb-salvage surgery for bone sarcomas. Expert Rev Pharmacoecon Outcomes Res 11(1):59–73 Sedlic A, Chingkoe CM, Tso DK, Galea-Soler S, Nicolaou S (2013) Rapid imaging protocol in trauma: a whole-­ body dual-source CT scan. Emerg Radiol 20(5):401–408 Tan TJ, Aljefri AM, Clarkson PW, Masri BA, Ouellette HA, Munk PL et al (2015) Imaging of limb salvage surgery and pelvic reconstruction following resection of malignant bone tumours. Eur J Radiol 84(9):1782–1790 Teixeira PAG, Chanson A, Beaumont M, Lecocq S, Louis M, Marie B et al (2013) Dynamic MR imaging of osteoid osteomas: correlation of semiquantitative and quantitative perfusion parameters with patient symptoms and treatment outcome. Eur Radiol 23:2602–2611 Teixeira PAG, Gervaise A, Louis M, Raymond A, Formery A-S, Lecocq S et al (2015a) Musculoskeletal wide-­ detector CT kinematic evaluation: from motion to image. Semin Musculoskelet Radiol 19(5):456–462 Teixeira PAG, Beaumont M, Gabriela H, Bailiang C, Verhaeghe J-L, Sirveaux F et al (2015b) Advanced techniques in musculoskeletal oncology: perfusion, diffusion, and spectroscopy. Semin Musculoskelet Radiol 19(5):463–474 Thomas J, Rideau AM, Paulson EK, Bisset GS (2008) Emergency department imaging: current practice. J Am Coll Radiol 5(7):811–6e2 van de Giessen M, Foumani M, Vos FM, Strackee SD, Maas M, Van Vliet LJ et al (2012) A 4D statistical model of wrist bone motion patterns. IEEE Trans Med Imaging 31(3):613–625 van der Woude HJ, Verstraete KL, Hogendoorn PC, Taminiau AH, Hermans J, Bloem JL (1998) Musculoskeletal tumors: does fast dynamic contrast-­ enhanced subtraction MR imaging contribute to the characterization? Radiology 208(3):821–828 van Rijswijk CSP, Geirnaerdt MJA, Hogendoorn PCW, Taminiau AHM, van Coevorden F, Zwinderman AH et al (2004) Soft-tissue tumors: value of static and dynamic gadopentetate dimeglumine-enhanced MR imaging in prediction of malignancy. Radiology 233(2):493–502 Verburg JM, Seco J (2012) CT metal artifact reduction method correcting for beam hardening and missing projections. Phys Med Biol 57(9):2803–2818

Clinical Application of Musculoskeletal CT: Trauma, Oncology, and Postsurgery Wilson JS, Korobkin M, Genant HK, Bovill EG (1978) Computed tomography of musculoskeletal disorders. AJR Am J Roentgenol 131(1):55–61 Yu L, Pan X (2003) Half-scan fan-beam computed tomography with improved noise and resolution properties. Med Phys 30(10):2629–2637

1105

Zhou C, Zhao YE, Luo S, Shi H, Li L, Zheng L et al (2011) Monoenergetic imaging of dual-energy CT reduces artifacts from implanted metal orthopedic devices in patients with factures. Acad Radiol 18(10):1252–1257

Incidental Findings in Multislice CT of the Body Mikael Hellström

Contents 1    Introduction

 1108

2    Definition and Misunderstandings Regarding Incidental Findings

 1108

3    Frequency and Spectrum of Incidental Findings on Multislice CT of the Abdomen 3.1  Incidental Renal Tumors and Cysts 3.2  Incidental Adrenal Lesions 3.3  Incidental Liver Lesions 3.4  Incidental Lesions of the Gallbladder and Biliary Tree 3.5  Incidental Lymphadenopathy 3.6  Incidental Pancreatic Lesions 3.7  Incidental Abdominal Vascular Findings 3.8  Incidental Adnexal and Uterine Lesions

 1129  1129  1129

8    The Problem of False-Positive Findings, Overdiagnosis, and Indolent Cancers  1131  1109  1111  1114  1116  1117  1118  1118  1120  1120

4    Frequency and Spectrum of Incidental Findings on Multislice CT of the Chest  1123 4.1  Incidental Pulmonary Nodules  1123 4.2  Incidental Thoracic Vascular Calcifications  1125 4.3  Incidental Thyroid Lesions  1127 5    How Extensively Should Incidental Findings Be Searched for on Multislice CT?

 1127

6    Technical Factors Limiting Detection and Characterization of Incidental CT Findings

 1128

M. Hellström, MD Department of Radiology, Sahlgrenska University Hospital, Gothenburg, Sweden e-mail: [email protected]

7    Reporting of Incidental Findings 7.1  Incidental Findings Are Not Always Reported 7.2  Reasons for Reporting or Not Reporting Incidental CT Findings

9    Do the Patients Want to Know About Incidental Findings? 9.1  Who Should Decide Which Information to Convey to the Referring Physician and to the Patient? 9.2  Potential Impact of e-Medicine

 1131  1132  1133

Conclusion

 1133

References

 1134

Abstract

In radiology, an incidental finding (“incidentaloma”) is an incidentally discovered mass or lesion, detected by CT or other modality performed for an unrelated reason. With high-­ resolution cross-sectional imaging, such findings are very frequent, and an everyday challenge for radiologists. In this chapter on incidental CT findings, the use and misuse of the term is discussed, as well as the frequency, characteristics, workup, and importance of common incidental findings in the abdomen, pelvis, and chest, with reference to guidelines for management. Emphasis is, e.g., on controversies over management of adrenal incidentalomas, as management has changed with increasing knowledge about their clinical importance.

Med Radiol Diagn Imaging (2018) https://doi.org/10.1007/174_2018_186, © Springer International Publishing AG Published Online: 07 June 2018

1107

M. Hellström

1108

Incidental renal tumors and complex cysts are discussed in depth, as over half of renal cancers are detected incidentally, potentially ­providing better prognosis. On the other hand, early detection may lead to overdiagnosis of indolent tumors with little clinical risk, also discussed in this chapter. Common incidental findings in the liver, pancreas, and reproductive organs are also discussed. In the chest, focus is on detection, management, and clinical importance of incidental pulmonary nodules, thyroid lesions, and vascular (including coronary) calcifications, with reference to updated international guidelines. There is also focus on technical CT factors that may limit detection of incidental findings, on reasons for not reporting incidental findings, and on patients’ opinions of incidental findings. In conclusion, this chapter may deepen the understanding of incidental findings, and aid in the delicate task for the radiologist to balance benefits and risks when reporting such findings and recommending certain actions.

1

Introduction

Detecting and managing incidental findings, i.e., findings not intentionally searched for, has always been part of diagnostic imaging. However, in the era of conventional radiography, such findings were very limited, due to the limitations in spatial and contrast resolution, and lack of crosssectional image reconstructions. With the development of high-resolution cross-sectional imaging, especially multislice CT, incidental findings have become very frequent, and an everyday challenge for the radiologists. Multislice CT examinations usually cover not only the organs of interest but also neighboring organs and tissues. For example, CT of the abdomen and pelvis for an abdominal complaint includes the intra-, retro-, and extraperitoneal spaces, as well as extra-­abdominal soft tissues, bony structures of the spine, sacrum, pelvis and hips, and the lower part of the chest including parts of the lungs and pleural spaces. The multi-

tude of organs and tissues involved often makes CT reading complex and allows for a number of incidental findings that may be of degenerative, neoplastic, or other etiology. The following chapter does not intend to cover every aspect of incidental body CT findings or systematically cover specific organs, but concentrates on general aspects and highlights some relevant or controversial organ-­ specific incidental findings in adults. A systematic organ-oriented overview of abdominal incidental CT findings has recently been published (Hellström M. Incidental findings on abdominal CT.  In: Incidental radiological findings. Editor: Weckbach S, Springer International Publishing 2017, pp. 127–168).

2

Definition and Misunderstandings Regarding Incidental Findings

In radiology, an incidental finding, sometimes called incidentaloma, can be described as an imaging finding not intentionally searched for, or an incidentally discovered mass or lesion, detected by CT or other imaging modality performed for an unrelated reason. The terms incidental finding and incidentaloma are therefore inappropriate when the radiological finding is related to the clinical question, or to the clinical symptoms or signs that motivated the CT examination. Thus, the terms are inappropriate when a tumor is detected in a patient with high suspicion of a malignant process. In such a case the organs and tissues are intentionally scrutinized for masses at any location, and therefore the finding of a tumorous lesion may not be entirely incidental. Nevertheless, such a finding may still be benign, and thereby “incidental” in relation to what was expected or searched for (i.e., metastases or malignant disease). The term incidental finding, as used in diagnostic imaging, can also be discussed from other aspects. From the radiologist’s viewpoint, the meaning and use of the term incidental finding or incidentaloma depend on how much, and how specific, clinical information is given on the radiology request form. This in turn may depend on

Incidental Findings in Multislice CT of the Body

the clinical situation and on the individual referring doctor formulating the request form. With a very specific clinical question, the likelihood of classifying other “non-targeted” radiological findings as incidental may be high, while the same radiological findings may be covered by a broader, more unspecific clinical question and thereby less likely to be called incidental. The term incidental finding is therefore a relative term. Incidental radiological findings also need to be related to previous radiological and other information. A CT finding that appears incidental in relation to the clinical question may already be known from previous studies, and thereby not truly incidental, although it may be incidental to the reporting radiologist if he or she does not have access to previous examinations. The term incidental finding or incidentaloma is therefore best applied to findings that are not previously shown on radiological examinations. The usually non-standardized text summarizing the patient history and clinical questions on radiological request forms, and variations in interpretation of the clinical question by different radiologists, in addition to variations in diagnostic interpretation of the actual radiological images, means that comparisons of frequencies of incidental findings in different studies are, to be modest, uncertain. One may also argue that if the frequency of a certain diagnosis in a defined population is known from previous studies, such as the frequency at CT of abdominal aortic aneurysm (AAA) in 65-year-old men, the identification of such an aneurysm in a 65-year-old male patient is not entirely unexpected, even if not asked for by the referring doctor. On a population basis, such a finding is thereby not entirely incidental. However, the finding in the individual patient may still be incidental if not covered by the clinical question. The term incidental finding is therefore best applied on an individual patient basis. Incidental findings that are masses or tumorlike are often called incidentalomas, such as adrenal incidentaloma or thyroid incidentaloma. It is important to understand that the term incidentaloma is not a diagnosis, but only a description of how a lesion was identified, i.e.,

1109

incidentally. Not uncommonly, the term is incorrectly used by radiologists and clinicians to denote a benign finding. In fact, the term incidentaloma says nothing about the character or etiology of the lesion found. Thus, an incidentaloma may be benign or malignant—and it may be clinically unimportant or important. The frequency of incidental findings in multislice CT is strongly related to the age, sex, and clinical background of the studied population, and it also depends on the criteria used for definition of incidental findings.

3

Frequency and Spectrum of Incidental Findings on Multislice CT of the Abdomen

In a recent retrospective study of 1040 consecutive abdominal contrast-enhanced CT examinations, performed for a variety of reasons (mean age 66 years), “relevant incidental findings,” i.e., findings leading to further imaging, clinical evaluation, or follow-up, were found in 19% of the examinations (Sconfienza et al. 2015). Such incidental findings were slightly more common in inpatients (23%) than in outpatients (15%), and there was an increase with patient age. The relative distribution among the involved organs was kidneys (14%), gallbladder (14%), lungs (12%), uterus (10%), adrenal (10%), and vessels (10%). In total, 39 different types of relevant incidental findings were made on the 1040 contrastenhanced abdominal CT examinations. It is notable that the frequency figures were based on review of the radiology reports, and not on review of the CT images, which may have revealed even more findings. Therefore, these figures should be considered minimum figures. Incidental findings have been extensively studied in CT colonography and it therefore provides an example of what incidental findings can be expected on abdominal-pelvic CT examinations. The clinical question in CT colonography is focused on the rectum and colon itself. However, a CT colonography examination covers the entire abdomen and pelvis, from the dia-

1110

phragm to the symphysis pubis, and thereby allows full assessment of colonic as well as extracolonic organs and tissues. It may be argued that by using 3D endoluminal “virtual colonoscopy” image reconstructions and 2D images zoomed-in on the colon and rectum with wide window-settings, it is theoretically possible to fully assess the colon and rectum without proper visualization of, and attention to, the extracolonic tissues. There is, however, general agreement that evaluation of extracolonic organs and tissues should be an integral part of CT colonography. Thus, the ESGAR CT Colonography Working Group states that “the extracolonic organs should be interrogated and abnormalities reported, noting the limitations if an unenhanced and/or low-dose technique was used” (Neri et  al. 2013). The ESGAR recommendation for CT colonography can be extended to all CT of the abdomen and chest, meaning that organs and tissues outside the primary target of the CT examination should be scrutinized for clinically relevant incidental findings. Extracolonic findings are very common on CT colonography, and the majority of these can be considered as incidental findings, although the terms are not entirely interchangeable. Extracolonic findings are commonly categorized as being of minor, moderate, or major importance. Findings of major importance are usually defined as those that potentially may lead to further imaging, surgical procedures, or clinical follow-up. In a CT colonography study, mainly including screening subjects, at least one extracolonic finding was made in 55% of those aged 41–64  years, and in 74% of those aged 65–92  years (Macari et  al. 2011). More importantly, clinically significant findings leading to a recommendation for further radiological imaging were made in 4–6% of the same population. This suggests that the vast majority of incidental findings are of minor clinical importance, but also that relevant findings are made in a smaller proportion of those screened. In two other large CT colonography screening studies in asymptomatic individuals (over 10,000 and 2000 participants, respectively), unsuspected extracolonic cancers were identified with similar (Veerappan et  al.

M. Hellström

2010), or even higher frequency (Pickhardt et al. 2010) than in the colon-rectum itself. In a more recent publication, 2.5% of an asymptomatic CT colonography screening population had extracolonic findings of potentially major clinical importance, and in nearly 70% of these, significant pathology was proven at follow-up (Pooler et al. 2016a).The findings primarily involved the vascular system (26% of the incidental findings, including aortic, iliac and other aneurysms), the urogenital system (18%), the liver (15%), the gastrointestinal system (10%), the lungs (9%), and the gynecological organs (7%). Considering that screening for abdominal aortic aneurysms can be performed simultaneously, it has been suggested that CT colonography is a highly costeffective screening method (Pickhardt et  al. 2009). Nevertheless, the question about the potential and real impact of extracolonic incidental findings on long-term morbidity and mortality, cost-effectiveness and acceptance of CT colonography for screening on a population basis remains a major issue, not least for decisionmakers regarding general societal imbursement. In symptomatic patients investigated with CT colonography for suspected colorectal pathology, previously unknown extracolonic findings of major importance have been found in 7–13% of the cases (Hellstrom et  al. 2004; Badiani et  al. 2013), and in the symptomatic elderly in up to 24% (Tolan et  al. 2007). In the large SIGGAR study on CT colonography in symptomatic patients, extracolonic findings were made in 59%, and further investigated in 8.3% of the population (Halligan et  al. 2015). Extracolonic findings are more common in older, as compared to younger patients (Macari et al. 2011; Khan et al. 2007) and in females, due mainly to findings in the female reproductive organs (Khan et al. 2007). It is obvious that incidental findings may constitute important medical information in both asymptomatic and symptomatic patients. Despite this, it has sometimes been suggested that extracolonic findings (incidental findings) on CT colonography should be reported by the radiologist only if specifically asked for. However, the high frequency of significant extracolonic (incidental) findings implies that extracolonic findings should

Incidental Findings in Multislice CT of the Body

always be looked for, and reported when of potential clinical significance. Most studies on incidental findings classify the importance of the extracolonic findings as minor, moderate, or major, exemplified in a recent systematic review (Lumbreras et al. 2010). In order to standardize and facilitate reporting of extracolonic findings on CT colonography, classification within the CRAD CT colonography categorization system has been proposed (Zalis et  al. 2005). Extracolonic findings are categorized as E0–E4: • E0: “Limited examination. Compromised by artifact; evaluation of extracolonic soft tissues is severely limited.” • E1: “Normal examination or anatomic variant. No extracolonic abnormalities visible.” Example: retroaortic left renal vein. • E2: “Clinically unimportant finding. No workup indicated.” Examples: renal or hepatic cysts, gall stone without cholecystitis, or vertebral hemangioma. • E3: “Likely unimportant finding, incompletely characterized. Subject to local practice and patient preference, workup may be indicated.” Example: minimally complex or homogeneously hyperattenuating kidney cyst. • E4: “Potentially important finding. Communicate to referring physician as per accepted practice guidelines.” Examples: solid renal mass, lymphadenopathy, aortic aneurysm, non-uniformly calcified parenchymal lung nodule ≥1 cm. • This system of structured classification of incidental findings is particularly useful for comparison between studies, but can also be applied to any abdominal CT examination, providing a rough guideline for the radiologist and the referring clinicians.

3.1

Incidental Renal Tumors and Cysts

3.1.1 Solid Renal Tumors Incidental renal tumors are of particular clinical importance and may serve as an example of issues raised when evaluating incidental findings.

1111

An increasing proportion of renal cancers are detected incidentally on imaging examinations performed for unrelated reasons. In 2016, 60% of newly diagnosed renal cancers in Sweden were detected incidentally, an increase from 43% in 2005 (The Swedish National Quality Registry for Kidney Cancer 2017 http://www.cancercentrum. se/samverkan/cancerdiagnoser/urinvagar/njurcancer/kvalitetsregister/). Most of these cancers were detected on CT examinations of the abdomen and sometimes CT of the chest, while MRI of the abdomen and spine and abdominal ultrasonography contributed to a lesser extent. The incidentally detected renal cancers were smaller (mean 50 mm) than those presenting with symptoms (mean 76 mm), and thereby of lower stage with potentially better prognosis. This is reflected in statistics on the mean size of all newly detected renal cancers over time, decreasing from 71 mm in 2005 to 59 mm in 2016 (The Swedish National Quality Registry for Kidney Cancer). The proportion of newly diagnosed renal cancers of stage 1a (40  years old, with no known malignancy, hepatic dysfunction, abnormal liver function tests, or hepatic malignant risk factors or symptoms attributable to the liver.” High-risk individuals are defined as those “with known primary malignancy with a propensity to metastasize to the liver, cirrhosis, and/or other hepatic risk factors. Hepatic risk factors include hepatitis, chronic active hepatitis, sclerosing cholangitis, primary biliary cirrhosis, hemochro-

Incidental Findings in Multislice CT of the Body

matosis, hemosiderosis, oral contraceptive use, and anabolic steroid use” (Berland et al. 2010). Although multidetector CT with thin slices may sometimes reveal focal liver lesions measuring only 2–3  mm in size, characterization of lesions measuring 0.5  cm or even 1  cm in size may be difficult and uncertain. The ACR suggests that incidental liver lesions  20  HU, or heterogeneous appearance, should have follow-up (6  months or closer) in all risk groups. Lesions 0.5–1.5  cm with “flash filling” (“robustly enhancing”) such as typical hemangioma or FNH in patients with low or average risk need no further follow-up. If “flash filling” or robustly enhancing lesion occurs in high-risk patient, evaluation with MRI or follow-up in 6  months should be considered. For high-risk patients, comprehensive guidelines for the identification of hepatocellular carcinoma have been published by EASL-EORTC (European Association for Study of Liver and European Organisation for Research and Treatment of Cancer 2012). For lesions  >  1.5  cm with low attenuation and benign appearance, no further follow-up is needed. For lesions  >  1.5  cm with low attenuation but suspicious imaging features (as above), low-risk patients should have followup in 6 months, average risk patients should have prompt evaluation, preferably with MRI, and for high-risk patients biopsy should be considered. For lesions > 1.5 cm with “flash filling” (robustly enhancing) and benign imaging features, hemangioma, FNH, or other benign etiology should be confirmed, if not confidently diagnosed with

1117

CT. If the CT shows robust enhancement but no benign diagnostic features, multiphasic MRI and possibly biopsy should be performed to confirm or rule out hepatocellular carcinoma and metastatic liver disease. A structured approach to incidentally detected liver lesion on CT examinations as described above (Berland et al. 2010) is certainly valuable and helpful. In the era of patient-centered care, also the preferences of the patient need to be taken into account. Structured guidelines should therefore be seen as—guidelines—for obtaining reasonably safe and adequate patient care.

3.3.2 Steatosis Steatosis of the liver parenchyma is a very common finding on abdominal CT, if actively looked for. Using a threshold of 40 HU, Boyce et al. (2010) found steatosis in 6.2% of 3357 asymptomatic individuals undergoing screening CT colonography at a mean age of 57  years (Boyce et  al. 2010). Steatosis may vary in degree over time, as measured on abdominal CT (Hahn et al. 2015). When marked, steatosis may be apparent for to the naked eye when the hepatic vasculature has a higher density than the surrounding liver parenchyma on nonenhanced CT. Considering the potential relationship between liver steatosis and the metabolic syndrome and other metabolic and hormonal disorders, it seems reasonable to routinely scrutinize the liver for steatosis on abdominal CT, and to report it to the referring physician, although there is no immediate therapeutic action or patient benefit coupled to such a finding, at present.

3.4

I ncidental Lesions of the Gallbladder and Biliary Tree

Asymptomatic gall stones are one of the most common incidental findings on abdominal CT. In the study of Sconfienza et  al. (2015) of about 1000 abdominal CT examinations, gall stones were the most frequent incidental finding. In most cases, this is a trivial finding, but it should be mentioned in the radiology report for clinical correlation. CT is very sensitive to calcium deposits, meaning that most calcified gall stones

M. Hellström

1118

are identified, but many gall stones are only faintly or not at all calcified, and are easily missed on CT, while they are apparent on ultrasonography. When gallstones are encountered, the gall bladder wall should be scrutinized to reveal inflammatory or chronic general wall thickening. Similarly, widening of the extra- and intrahepatic biliary tree should be searched for. A common bile duct >7 mm in a patient with the gall bladder present and > 10 mm after cholecystectomy can be considered as dilated and indicative of obstruction (Sebastian et al. 2013) Gall bladder wall calcification (porcelain gall bladder) has been claimed to be associated with gall bladder cancer, but the association appears weak, and the ACR Incidental Findings Committee does not generally recommend follow-up for calcified gall bladder wall without an associated softtissue mass (Sebastian et al. 2013). Uniform gall bladder wall thickening over 3  mm without a mass lesion can be associated with previous inflammation (chronic cholecystitis) but, importantly, also with, e.g., congestive heart failure and hypoproteinemia. Although seen more commonly on ultrasonography, gall bladder polyps and cancer may occasionally be detected incidentally on CT (Mellnick et al. 2015). Soft-tissue filling defects with contrast enhancement is suggestive of polyps. If 10  mm (Sebastian et  al. 2013). Irregular focal gall bladder wall thickening with contrast enhancement can be indicative of gall bladder cancer, which is the most common biliary tract cancer. It is frequently incidental, but only in the meaning that it is unsuspected until detected at laparoscopic or open gall stone surgery in a symptomatic patient (Cavallaro et al. 2014).

3.5

Incidental Lymphadenopathy

Incidental detection of single, clustered, or generalized lymph node enlargement is an important finding, which may indicate lymphoma or other malignancy. If not generalized, however, it is dif-

ficult to determine the clinical importance of the finding, considering the normal variation in size, and the overlap in appearance of inflammatory, reactive, and malignant nodes. Lymph nodes in the abdomen and pelvis tend to have different size in different compartments, and there is a variation in the number of visible nodes on CT.  Short-axis node diameter provides stronger correlation to malignancy than long axis, and is recommended for assessment. Short axis of 1 cm or more can be considered as abnormal in the retroperitoneum (Heller et al. 2013), although nodes in, e.g., the retrocrural space normally are smaller. In patients with malignancy, enlarged nodes on CT are likely to be malignant, but may also be reactive and benign. Conversely, normal node size does not exclude malignant involvement. An increased number of normal sized nodes may be indicative of a pathological process. It has been suggested that a cluster of three or more nodes in a single node station or a cluster of two or more nodes in two nodal stations is suspicious. If encountered in the absence of clinical explanation, a 3-month follow-up for growth may then be motivated (Heller et al. 2013). Isolated enlargement of mesenteric lymph nodes is sometimes detected incidentally, combined with infiltrated, encapsulated fatty mesenteric tissue and a perivascular fatty rim. These findings are indicative of sclerosing mesenteritis (panniculitis) (Sabate et al. 1999), which may be asymptomatic or present with vague abdominal symptoms.

3.6

Incidental Pancreatic Lesions

3.6.1 Solid Tumors Solid tumors of the pancreas usually represent ductal adenocarcinoma or neuroendocrine neoplasms. Incidental detection of solid pancreatic adenocarcinoma is uncommon and probably contributes only marginally to the overall survival for this patient group at large. Neuroendocrine tumors may be functional, i.e., hormone producing, named after the hormones produced, e.g., insulinomas and gastrinomas. If incidentally detected they are likely to be nonfunctional and

Incidental Findings in Multislice CT of the Body

symptom-free. In a retrospective review of cases referred for assessment of solid pancreatic masses, 24 (7%) of 321 cases were detected incidentally (Goodman et  al. 2012). Of these, 14 were adenocarcinomas and 10 were neuroendocrine tumors, initially identified at CT performed for various unrelated reasons. Eleven of the 24 patients had metastases already at the time of incidental detection, and the overall survival in those with adenocarcinoma was only 22 months, reflecting its dismal prognosis, despite pre-symptomatic detection. Incidental detection of a hyperdense contrast-enhancing pancreatic mass suggests neuroendocrine etiology with a slightly better prognosis (mean survival 42 months, range 16–82 months).

3.6.2 Cystic Lesions As compared to solid pancreatic tumors, cystic pancreatic lesions are more common as incidental findings at CT, and much more likely to be benign. Over the last decades, there has been a marked increase of incidentally detected cystic pancreatic lesions, due to the increased use and improved resolution and overall image quality of multidetector CT, and due to increased awareness of their existence. In an analysis of consecutive cystic pancreatic lesions subjected to surgery over a 33-year time period, there was an increase of incidental detection from 22% in 1978–89 to 50% in 2005–2011 (Valsangkar et  al. 2012). Laffan et al. (2008) retrospectively re-examined 2832 contrast-enhanced abdominal outpatient CT examinations, excluding those with symptoms or history of pancreatic disorders. In that population with a mean age of 58  years, they found cystic pancreatic lesions in 73 cases (2.6%). No pancreatic cysts were found in those under 40 years of age, while the frequency in the age group 80–89  years was 8.7%. It should be noted that only contrast-enhanced CT examinations were evaluated. In non-contrast-enhanced CT examinations, the incidental detection rate may be lower, due to less conspicuity of the lesions in the absence of intravenous contrast injection. On the other hand, the real frequency of cystic pancreatic lesions may be considerably higher than that found at CT, as autopsy studies

1119

revealed cystic pancreatic lesions in up to 24% of the studied population (Kimura et al. 1995). In a recent, large, retrospective analysis of predominantly men (88%), including all cyst etiologies, patients with pancreatic cysts had a nineteen times higher risk of developing pancreatic cancer over 8 years observation, compared to those without a diagnosis of pancreatic cysts (Munigala et al. 2016). When the radiologist encounters an incidental cystic pancreatic lesion, the first question to be asked is if it could represent a pseudocyst associated with previous acute pancreatitis or chronic pancreatitis. This may be apparent from available earlier radiological examinations or from medical files, and may also be indicated by CT findings such as parenchymal calcifications, necrotic areas, dilatation of the main duct and side branches, parenchymal atrophy, and extra-pancreatic location of the pseudocyst. In other cases, the differentiation between a pseudocyst and a mucinous cystic neoplasm may be difficult and of concern, as the clinical handling and prognosis are different. If a pseudocyst and cyst-like necrosis in a solid pancreatic cancer can be ruled out, the cyst is likely to represent a serous cystadenoma (SCA), mucinous cystic neoplasm (MCN), or intraductal papillary mucinous neoplasm (IPMN). Comprehensive guidelines on the management of MCN and IPMN have been published (Tanaka et  al. 2012). Serous cystadenomas are benign tumors with female preponderance, occurring in elderly women (median age 68  years), therefore sometimes called “grandmother tumor” (Zaheer et al. 2013). On CT, they may occur as a mass consisting of small, multiple cysts with multiple septations, and sometimes a characteristic central scar with or without calcification. Further investigation of incidentally detected cystic pancreatic lesions include multiphase CT, with native, arterial as well as venous phase imaging. MRI has a similar, or better, accuracy in differentiating benign from malignant cystic pancreatic lesions, and together with MRCP allows visualization of the pancreatic duct, and in case of branch duct-IPMN, the connection to the main

M. Hellström

1120

pancreatic duct (Tanaka et  al. 2012). Although not performed as first-line investigation, PET-CT has the highest accuracy in this respect (Kauhanen et  al. 2015). If uncertainty remains, endoscopic ultrasonography with fine-needle aspiration is a recommended option (Muthusamy et al. 2016). In a very recent update of a White Paper on management of incidental pancreatic cysts, ACR provides a comprehensive algorithm, based on pancreatic cyst size (2.5  cm) and age at presentation (5  cm should have follow-up with ultrasonography at 6–12  weeks. In postmenopausal women, a similar benignappearing cyst needs no follow-up if 3  cm or smaller, while larger cysts should have prompt follow-up with ultrasonography (Patel et  al. 2013). However, based on results from combined autopsy and ultrasound studies, benign cysts are very frequent and merely a normal finding in postmenopausal women (Valentin et  al. 2003), and it is therefore suggested that unilocular, benign-appearing cysts