Extracorporeal Life Support: The ELSO Red Book [5 ed.] 0965675653, 9780965675659

Table of contents :
Dedication
Preface to the 5th Edition


I: Extracorporeal Life Support: General Principles






1. The History and Development of Extracorporeal Support




Extracorporeal Support: Earliest Beginnings
Global Spread of ECMO
Perseverance: Experience and Growing Indications in Adult ECMO
References
2. The History of ECMO: First Hand Accounts

The History of ECMO: First Hand Accounts
A Conversation with Dr. Robert H. Bartlett
A Conversation with Dr. Jay Zwischenberger
3. Evolution and History of Global ELSO

EuroELSO
Asia-Pacific ELSO
Latin American ELSO
South and West Asian ELSO
References
4. The Physiology of Extracorporeal Life Support

Cardiopulmonary Physiology
Cardiopulmonary Pathophysiology
Cardiopulmonary Pathophysiology during ECMO
The ECMO Circuit
Modes of Vascular Access and Perfusion
CO2 Removal
ECMO Management when the Native Lung is Recovering
Summary
5. The Circuit

Introduction
Historical Background
General Principles of Circuit Design
Blood Tubing Plasticizers
ECLS Circuit Components
Blood Pumps
Pulsatility
Roller Pumps
Roller Pump Servo Regulation
Centrifugal Blood Pumps
Centrifugal Pump Inlet Pressure Monitoring Techniques
Blood Pump Performance
Regulatory Status
Pump Related Complications
Commercial Centrifugal Pumps
Gas Exchange Devices
Commercial Gas Exchange Devices
Gas Exchanger Related Complications
Heat Exchange and Heat Regulation
Heat Exchanger Related Complications
Circuit Priming
Circuit Monitoring
Summary
References
6. Adverse Effects of ECLS: The Blood Biomaterial Interaction

Introduction
Normal Hemostasis
Activation of the Coagulation Pathway
Activation of the Fibrinolytic Pathway
Developmental Hemostasis
Coagulation Pathway Activation and Inflammatory Response
Activation of the Coagulation System during ECLS
Activation of the Innate Immune System
Endothelial Function in Hemostasis and Circuit Modifications
References
7. Anticoagulation and Disorders of Hemostasis

Introduction
Anticoagulation
Anticoagulation Monitoring
Anticoagulation Laboratory Schedule and Blood Product Replacement
Hemorrhagic and Thrombotic Complications
Conclusions
References
8. Transfusion Management during Extracorporeal Support

Blood Product Transfusion during Extracorporeal Support
Use of Coagulation Factor Replacement
References

II. Extracorporeal Life Support: Neonatal Respiratory Disease

9. Neonatal Respiratory Diseases



Introduction
Congenital Diaphragmatic Hernia (CDH)
Meconium Aspiration Syndrome (MAS)
Persistent Pulmonary Hypertension of the Newborn
Pneumonia/Sepsis
Surfactant Deficiency – Hyaline Membrane Disease
Surfactant Deficiency – Term Infants in Respiratory Failure
References
10. Congenital Diaphragmatic Hernia and ECMO



IntroductionDiagnosis
Management
Mechanical Ventilation
Inhaled Nitric Oxide
Sildenafil
Fetal Interventions
Management Adjuncts
ECMO
Summary
References
11. Indications and Contraindications in Neonates with Respiratory Failure

Introduction
Patient Selection Criteria
Indications
Contraindications
Weight

Citation preview

Table of Contents Dedication .............

1$ • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 8 ••••

V

th

Preface to the 5 Edition ............................................................................................... xvii

I. Extracorporeal Life Support: General Principles 1. The ffistory and

of Extracorporea] Support..•.....•••..............••••....•.•..1

Support: Earliest Beginnings ......................................................... 1

.,nr",," of ECMO ~~~~ .. ~ .... ~~ .... ~~ .... ~ .....,. . ~ ......."."....... ~ ........... *._.".. * ••• ~"~ .. ** •• ~~ .... 6

Perseverance: and Growing Indications in Adult ECMO ................... 9

¥ . . . . . ,, . . . . . . .

* ••

2. The

A Conversation with Dr. Jay Zwischen berger....................................................... 22

3. Evolution and ffistory Global ELSO .................................................................. 25

................................................................................................ 26 ELSO .......................................................................................... 27 South and West Asian ELSO .............................................................................. 28

References ............................................................................................................ 29

4. The Physiology of Extracorporeal Life Support .....................................................31

Cardiopulmonary Physiology ............................................................................... 31

Cardiopulmonary Pathophysiology ......................................................................33

Cardiopulmonary Pathophysiology during ECMO

The ECMO Circuit. ............................................................................................. 34

Modes and .............................................................36

5. The Circuit........................................................................................

t> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . '"

49

49

lastl.cmlrs ..................................................................................... 51 ECLS Circuit Components ...................................................................................52

Table ofContents

Commercial Centrifugal Pumps .......................................................................... 60

Gas Exchange Devices .........................................................................................63

Commercial Gas Exchange Devices ..................................................................... 65

Gas Exchanger Related Complications ........................................................ ,...... 70

Heat Exchange and Heat Regulation ........................................................ ............ 71

Heat Exchanger Related Complications ............................................................... 72

Circuit Priming .....................................................................................................73

Circuit Monitoring ....... ......................................................................................... 73

Summary ................................................................ ........................ ....................... 75

References .................... .............. ............................ ...............................................76

6. Adverse Effects ofECLS: The Blood Biomaterial Interaction ................•.............81

Introduction ........................... ............................................... ................................. 81

Nonnal Hemostasis ........... ............................................................. .. ................... 82

Activation of the Coagulation Pathway ................ ............................. ................... 82

Activation of the Fibrinolytic Pathway ................................................................. 83

Developmental Hemostasis .................. .. ............. .. ................................................ 83

Coagulation Pathway Activation and Inflammatory Response ........................... 84

Activation of the Coagulation System during ECLS ........ .................................... 87

Activation of the Innate Immune System .............................................................. 87

Endothelial Function in Hemostasis and Circuit Modifications .................... ...... 88

References ........................................... .. ............................................................... 90

7. Anticoagulation and Disorders of Hemostasis........................................................ 93

Introduction ....................................... .................................................................... 93

Anticoagulation .. ................................................................ ................................... 93

Anticoagulation Monitoring ..................................................... .. .. .........................95

Anticoagulation Laboratory Schedule and Blood Product Replacement... .......... 98

Hemorrhagic and Thrombotic Complications ..................................................... 99

Conclusions ................................. ......................................................................... 99

References .......................................................................................................... 100

8. Transfusion Management during Extracorporeal Support ................................ 105

Blood Product Transfusion during Extracorporeal Support ............................... 105

Use of Coagulation Factor Replacement ............................................................ III

References ........................................... ........... .. .. ................................................. 117

ll. Extracorporeal Life Support: Neonatal Respiratory Disease 9. Neonatal Respiratory Diseases .............................................................................. 123

Introduction......................................................................................................... 123

Congenital Diaphragmatic Hernia (CDH) ............. ... .......................................... 123

Meconium Aspiration Syndrome (MAS) ............................................................ 123

Persistent Pulmonary Hypertension of the Newborn .......................................... 126

Pneumonia/Sepsis ................ .......................... ................................ ..................... 127

Surfactant Deficiency - Hyaline Membrane Disease ........................................ 128

Surfactant Deficiency - Tenn Infants in Respiratory Failure ........... ........... ...... 129

References ..................... ..................................................................................... 130

xx

Table o/Contents

10. Congenital Diaphragmatic Hernia

ECMO .................................................. 133

133

Diagnosis ........................................................................................................... .

134

Mechanical Ventilation ..................................................................................... .

Oxide ........................................................................................... 1

Inhaled Sildenafil ............................................................................................................. 135

lmervelluons .............................................................................................. 136

Management Adjuncts ........................................................................................ 137

ECMO ................................................................................................................. 137

Summary ............................................................................................................ .

Neonates with Respiratory Failure ........ 151 151 u","",,,,,,,.. Selection Criteria ,..~~~**~ ...,.*~~ .... ~,. ......... *~ ~~~~.~.~~* .......... 6~ ~.~~~ "~~"'.e .... ~ ..... ~*~ 151 ....

...

* ...............

.. ,. ..

B ..

1 Contraindications Weight kg ...................................................................................................... 154

Gestational Age wks ................................................................................... 154

iemorrtlage ..................................................................................... 155

Chromosome Abnormalities ............................................................................... 156

Pre-ECMO 157 12. ECLS Cannulation for Neonates with Respiratory Failure ............................... 159

.................................................................................................................. 159

vv- or VA­ ........................................................................................... 160

Choice of Cannula and Vessel ........................................................................... 160

Cannulation Technique ....................................................................................... 161

Post Cannulation liJ.,u"i'>H15 167 13. Congenital Comorbidities among Respiratory Neonatal ECLS Patients •......•. 169

Background ........................................................................................................ 169

Congenital Anomalies ........................................................................................ 169

Malignancies ...................................................................................................... .

llU'",", . .Ull" ............................................................................................................ 173

.......................................................................................................... 175 Errors of Metabolism .............................................................................. 176 Abnormalities ........................................................................................ 176 Conclusion .......................................................................................................... 177

178

14. Medical Management ofthe Neonate with Respiratory on 183

and ...................................................................... 183

Respiratory......................................................................................................... 184

xxi

Table ofContents

Cardiovascular .................................................................................................... 186

Infection .............................................................................................................. 187

ECMO Circuit Considerations in the Neonatal Patient .................................... .

Hematologic ....................................................................................................... 190

Neurologic 191

Summary ............................................................................................................ 194

195

15.

Cardiovascular .................................................................................................. . Neurologica1. ...................................................................................................... 204

Fluid JJu,Jt.....""v.L'-VJUu,u

XXIl

Table ojContents

ExtracorporeaJ

Support: Pediatric Respiratory Disease

18. Pediatric Respiratory Diseases Predisposing to £.t'L,JU;:h........'. .........'...................." .. .!,.:1 ... for KeSPlratC)ry

"'1T~TV,rt

Failure Asthmaticus The Inununocompromised ..................................................................................233

MediastmallH'L;);)

I-'Pf·llU,pr,.1"n,p

. .AU'LI

Cardiac

,"-"WlHILUH Charles Stol ~ MD Warren Zap< I MD Joseph (Jay) ~wischenberger MD

-

11

Chapter 1

References 1. Bartlett RH. Extracorporeal Life Support: Gibbon fulfilled. J Am Coil Surg 2014; 218:317-327. 2. Richardson BW. An inquiry into the possi­ bility of restoring the life of warm blooded animals in certain cases where the respira­ tion, circulation and the ordinary manifes­ tations of organic motion are exhausted or have ceased. Proceedings of the Royal Sociey of London, 13:358-371, 1865. 3. Konstantinov I, Alexi-Meshkishvili V. Sergei S. Brukhonenko: the development of the first heart-lung machine for total body perfusion. Ann Thoracic Surg 2000; 69:962-6. 4. Lillehei CWO History of the development of extracorporeal circulation, in Arensman RM, Cornish JD, eds. Extracorporeal Life Support in Critical Care (1 st edition), Bos­ ton: Blackwell Publications, 1993. 5. Kanto WP, Shapiro MB. The development of prolonged extracorporeal circulation, in Zwischenberger JB, Bartlett RH, eds. ECMO: Extracorporeal Cardiopulmonary Support in Critical Care, Ann Arbor, MI: ELSO, 1995. 6. Lee WH, Krumhaar D, Fonkalsrud EW, et al. Denaturation of plasma proteins as a cause of morbidity and death after intracardiac operations. Surgery. 1961; 50:29-39 7. Custer JR. The evolution ofpatient selection criteria and indications for extracorporeal life support in pediatric cardiopulmonary failure: next time let's not eat the bones. Organogenesis. 2011 ;7(1): 13-22. 8. Kammermeyer K. Silicone rubber as a selec­ tive barrier. Ind Eng Chern. 1957;49:1685. 9. Kolff WJ, Effler DB, Groves LK, Peere­ boom G, Moraca PP. (1956). Disposable membrane oxygenator (heart-lung ma­ chine) and its use in experimental surgery. Cleve Clin Q. 156;23(2):69-97.

12

10. Ko10bow T, Za 01 W, Pierce JE et al. Par­ tial extracorporta1 gas exchange in alert newborn lambs tith a membrane artificial lung perfused v~. an AV shunt for periods up to 96 hours. rans Am Soc Artif Intern Organs. 1968; 1 :238. II. Bartlett RH, Is~erwood J, Moss RA, Ol­ szewski WL, Po~et H, Drinker P. A toroidal flow membrane ygenator: four day partial bypass in dogs. S gForum.1969;20:152-3. 12. KolobowT, Zap I WM, Sigmon RL, Pierce J. Partial cardiop lmonary bypass lasting up to seven days in ert lambs with membrane lung blood oxygrnation. J Thorac Cardio­ vasc Surg. 1970;fO(6):781-788. 13. Baffes TG, FridmF, JL, Bicoff JP, Whitehill JL. Extracorporerl circulation for support of palliative card ·ac surgery in infants. Ann Thorac Surg 197 '; 10(4):354-363. 14. Bartlett RH, G aniga AB, F ong SW, Jef­ HV, Haiduc N. Extra­ feries MR, Roo corporeal memb ane oxygenator support for cardiopulmo ' ary failure. Experience in 28 cases. J horac Cardiovc Surg. 1977;73(3):375-386. 15. Hill JD, O'Brie TG, Murray JJ et al. Prolonged extra orporeal oxygenation for acute post-tr umatic respiratory fail­ ure (shock-lung syndrome): use of the Bramson Membr e Lung. N Engl J Med. 1972;286(12):629 634. 16. Dorson W Jr, Biker E, Cohen ML, et al. (1969). A P rfusion system for in­ fants. Trans Am oc Artif Intern Organs. 1969;15: 155-60. 17. White JJ, Andre~S HG, Risemberg H, et al. (1969). Prolo ged respiratory support in newborn infan s with a membrane oxy­ genator. Surgery 969;70(2):288-296. 18. BartlettRH. Esper . Trans Am SocArtif Intern Organs. 19 5; 31:723-735. 19. Bartlett RH. Artifi ial organs: basic science meets critical car . J Am CoIl Surg. 2003; 196(2): 171-179.

The History and Developm nt ofE'xtracorporeal Support

20. Wolfson PJ. The development and use ofex­ tracorporeal membrane oxygenation in neo­ nates.Ann Thorac Surg. 2003;76(6):S2224­ 2229 21. Bartlett RH. Extracorporeallife support in the management of severe respiratory fail­ ure. Clin Chest Med. 2000;21(3):555-56l. 22. Bartlett RH, Gazzaniga AB, Jefferies R, Huxtable RF, Haiduc RF, Fong SW. (1976). Extracorporeal membrane oxygenation (ECMO) cardiopulmonary support in infan­ cy. Trans Am Soc ArtifOrgan. 1976;22:80­ 93. 23. Bartlett RH, Andrews AF, Toomasian JM, Haiduc NJ, Gazzaniga AB (1982). Ex­ tracorporeal membrane oxygenation for newborn respiratory failure: forty-five cases. Surgery. 1982;92(2):452-433. 24. Bartlett RH, Roloff DW, Cornell RG, Andrews AF, Dillon PW, Zwischenberger JB. Extracorporeal circulation in neonatal respiratory failure: a prospective random­ ized study. Pediatrics. 1985;76(4):479-487. 25. O'Rourke PP, Crone RK, Vacanti JP, et al. Extracorporeal membrane oxygenation and conventional medical therapy in neonates with persistent pulmonary hypertension of the newborn: a prospective randomized study. Pediatrics. 1989;8(6):957-963. 26. Knox RA. A Harvard study on newborns draws fire. Boston globe, August 7, 1989:25. 27. Marwick C. NIH Research Risks Office reprimands hospital institutional review board. JAMA. 1990;263:2420. 28. UK Collaborative ECMO Trial Group. UK collaborative randomized trial of neonatal extracorporeal membrane oxygenation. Lancet. 1996;348(9020):75-82. 29. Wright L, Ed. Report of the Workshop on Diffusion ofECMO Technology; National Institutes of Health, 1993 30. Custer JR, Bartlett RH. Recent research in extracorporeal life support for respiratory failure. ASAIO J. 1992;38(4):754-771.

31. Paden ML, Co ad SA, Rycus PT, Thiaga­ rajan RR; ELS Registry. Extracorporeal Life Support ~ganization Registry report 2012. ASAIO .2013;59(3):202-10 32. Moler FW, Pa isano J, Custer JR. Extra­ corporeal life s PPort for pediatric respira­ tory failure: pretlictors ofsurvival from 220 patients. Critcj.reMed. 1993;21(10):1604­ 1611. 33. Pettignano R, Frtenberry JD, Heard M, et al. Primary use ffthe venovenous approach for extracorpor al membrane oxygenation in pediatric acu e respiratory failure. Pediatr Crit Care Med. 2003; 4(3):291-298. 34. Fackler J, Bo D, Gl·een T, et al. ECMO for ARDS; stop ing a RCT. Am J Resp Crit Care Med. 199 ;155:A504. 35. Green TP, Tons OJ), Fackler JC, Moler FW, Thompso AB, Sweeney MF. The impact of extr corporeal membrane oxy­ genation on s ivai in pediatric patients with acute res iratory failure. Pediatric Critical Care S dy Group. Crit Care Med. 1996;24(2):323 329. 36. Zabrocki LA, Bogan TV, Statler KD, Poss WB, Rollins , Bratton SL. Extracorpo­ real membrane xygenation for pediatric re­ spiratory failur : survival and predictors of mortality. Crit are Med. 2011 :39(2):364­ 370. 37. Gow KW, Hei s K, Wulkan ML, et al. Extracorporeal ife support for support of children with alignancy and respiratory or cardiac failur . The Extracorporeal Life Support Organ· tion Experience. Crit Care Med. 2009: 37( ):1308-1316 38. MacLaren G, ' utt W, Best D, Donath S, Taylor A. E tracorporeal membrane oxygenation fo refractory septic shock in children: on institution's experience. PediatrCritCar Med. 2007; 8(5):447-451. 39. Barbaro RP, 0 etola FO, Kidwell, K, et al. Association ~fhospital-Ievel volume of extracorporeal ~embrane oxygenation case and mortality. A1nalysis of the Extracorpo­ 1

I

13

Chapter I

real Life Support Organization registry. Am J Respir Crit Care Med. 2015; 191(8):894­ 901. 40. Zapol WM, Snider Mr, Hill JD et al. Extra­ corporeal membrane oxygenation in severe acute respiratory failure. A randomized pro­ spective study. JAMA. 1979;242(20):2193­ 6. 41. Morris AH, Wallace CJ, Menlove RL, et al. Randomized clinical trial of pressure­ controlled inverse ratio ventilation and extracorporeal C02 removal for adult respiratory distress syndrome. Am J Resp Crit Care Med. 1994;149(2 Pt 1):295-305. 42. Gattinoni L, Pesenti A, Bombino M et al. Role ofextracorporeal oxygenation in adult respiratory distress syndrome management. New Horiz. 1993;1(4):603-612. 43. Peek GJ, Mugford M, Tiruvoipati R et aL Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet. 2009;374(9698):1351-1363. 44. Davies A, ANZI C ECMO Investigators et al. Extracorporeal membrane oxygenation for 2009 influenza A (H 1N 1) acute respiratory distress syndrome. JAMA. 2009: 304: 1888­ 1895. 45. Noah MA, Peek G, Finney S, et al. Referral to an extracorporeal membrane oxygen­ ation center and mortality among patients with severe 2009 influenza A (HINI). JAMA. 2011 ;306(15): 1659-68. 46. Thiagarajan R, Barbaro R, Rycus P, et al. ELSO International Registry Report 2016. ASAIO J. 2017, in press. 47. Karagiannisidis C, Brocie D, Strassman S, et al. ECMO: evolving epidemiology and mortality. Intensive Care Med. 2016; 42(5):889-96. 48. Wang D, Zhou X, Liu X, Sidor B, Lynch J, Zwischenberger JB (2008). Wang-Zische double lumen cannula. Toward percutane­ 14

ous and ambula ry paracorporeal artificial lung. ASAIO J. 4(6):606-11. 49. Javidfar J, Wan D, Zwischenberger JB, et al. Insertion ofb~ caval dual lumen extracor­ poreal membran oxygenation catheter with image guidance. SA!O J. 2011 ;57(3):203­ 205. 50. Harig F, Feyrer Mahmoud FO, Blum D, von der Emde J. Reducing the post-pump syndrome by us' g heparin-coated circuits, steroids, or apr~inin. Thorac Cardiovasc Surg. 1999; 47(.~p: 111-118. 51. Lim MW. The hibtOry ofextracorporeal ox­ ygenators. Anae thesia. 2006;61 (10):984­ 994. 52. Frenckner B, P a er K, Linden V. Extra­ corporeal respirat ry support and minimally invasive ventila on in severe ARDS. Mi­ nerva Anestesiol gica 2002; 68:381 53. Turner DA, Chei etz IN!, Rehder KJ, et al. Active rehabilita .on and physical therapy during ECMO hile awaiting lung trans­ plantation: a pra9 ical approach. Crit Care Med. 2011; 39(1j):2593-2598. 54. Agerstrand CL, ~acchetta MD, Brodie D. ECMO for adult respiratory failure: current use and evolvina applieations. ASAIO J. 2014; 60(3):255- 62. 55. Abrams DC, Bre er K, Burkart KM, et al. Pilot study 0 extracorporeal carbon dioxide removal to facilitate extubation and ambulation' exacerbations ofchronic obstructive pulm nary disease. Ann Am Thoracic Soc. 20 3; 10(4):307-14. 56. Gow KW, Heiss KF, Wulkan ML, et a1. Extracorporeal Ii support for support of children with ma ignancy and respiratory or cardiac failur~: the ELSO experience. Crit Care Med. 2 09; 37(4):1308-1316. 57. MacLaren G, Bu W, Best D, Donath B. Central extracorp real membrane oxygen­ ation for refracto pediatric septic shock. Pediatr Crit Care ed. 2011; 12(2): 133-136. 58. Thiagarajan RR, Laussen P, Rycus PT, Bartlett RH, Bra on SL. Extracorporeal

t

The History and Develop ent ofE.xtracorporeal Support

membrane oxygenation to aid cardiopul­ monary resuscitation in infants and children. Circulation. 2007; 116(15): 1693-1700. 59. Raymond IT, Cunnyngham CB, Thompson MT, Thomas JA, Dalton HJ, NAdkarni VM. Outcomes among neonates, infants, and children after extracorporeal cardiopulmo­ nary resuscitation for refractory in hospital pediatric cardiac arrest: a report from the National Registry of Cardiopulmonary Resuscitation. Pediatr Crit Care Med. 2010; 11(3):362-371. 60. Sakamoto T, Morimura N, Nagao K et al. Extracorporeal cardiopulmonary resuscita­ tion versus conventional cardiopulmonary resuscitation in adults with out-of-hospital cardiac arrest: a prospective observational study. Resuscitation. 2014;85(6): 762-8. 61. Fan E, Gattinoni L, Combes A, et aI. Vena­ venous extracorporeal membrane oxygen­ ation for acute respiratory failure: a clinical review from an international group of ex­ perts. Intensive Care Med. 2016;4 2( 5): 712­ 724. 62. Churchill, W (1958), cited in Winston S, Langsworth RM. Churchill: In His Own Words, Ebury Press, 2008.

15

2 The History of ECMO: First Hand Accounts Giles Peek, MD, James D. Fortenberry, MD

The History ofECMO: First Hand Accounts

The words of those directly iJllvolved in the development and early days of translating extracorporeal support to the bedside provide compelling insights that facts alone cannot ex­ press. Dr. Giles Peek (Figure 2-1) sat down with two of those pioneers, Dr. Robert Bartlett and Dr. Joseph (Jay) Zwischenberger (Figure 2-2) to hear their stories. What follows is an edited tran­ script of their conversations at the EuroELSO Congress in Glasgow Scotland in May, 2016.

Dr. Robert H. Bartlett: The way we got started with this tl~ing that became ECMO was through cardiac swgery. I was a resident at the Boston Children's, Hospital in the 1960s, and the mortality at th~ time for complex congenital heart lesions was ~bout 50%. We knew ifthese babies could live fqr a day or two, they would be alright. In retrosp~ct it was what we would call we asked about whether myocardial stun. $0, I we could use the l1Fart lung machine from the operating room to, keep these babies on the

A Conversation with Dr. Robert R. Bartlett

Dr. Giles Peek: What were your first memo­ ries of ECMO?

Figure 2-1. Dr. Giles Peek headed towards

another peak.

Figure 2-2. Drs. Robert Bartlett and Jay Zwischenberger trading surgical instruments for musical versions.

17

Chapter 2

pump for a few days. My surgical chief and extracorporeal memb ane circulation in animals. mentor, Dr. Robert E. Gross said, "Of course Studies remained in t e laboratory. that is impossible because of problems with At that point, loved to the University the heart lung machine. It is very damaging to of California at Irv~· e Medical Center, with blood, and causes lots of problems after an hour a brand new medica school that opened the or two, but why don't you work on it?" That is day I arrived, togeth with Al Gazzaniga, my why we always toast Dr. Gross at ELSO func­ surgical partner. In jetrospect, it was a great tions, "To Dr. Robert E. Gross, without whom opportunity because here was no one around none of this would be possible!". "Why don't to say, "We don't do it that way." So we did you work on that problem?" he said, so we set what we thought woulr be good for our patients. out to work on it. There was an empty fildin g where we built a We knew that the problem with the heart laboratory and began . ore studies ofprolonged lung machine was the lung, which was simply extracorporeal suppo . We switched our model bubbling oxygen through blood and we and to sheep because the 1were much more suited other people had demonstrated that that was the to chronic awake eXPiriments, will stand there limiting factor. At about that time silicone rub­ passively for days at l:j- time as long as you feed ber became available with the unique properties them, unlike the dog.l So we got to the point of transferring respiratory gases. With another where we could do t!iS successfully without resident named Lou Plzak we built some mem­ any deleterious side ffects in sheep for days brane envelopes out of silicone rubber, hooked at a time. In 1971, we actua ly got a call from a surgi­ it up to a dog, and ran venous blood through it. It worked beautifully for about a minute and cal friend in Santa B bara who said, "Is your then it stopped working altogether. That was machine ready? I hav a kid here who is dying interesting. We presented this at Dr. Francis of respiratory failure." We said we were not, but Moore's research conference and at the same let's call Don Hill he is up there in San Francisco. conference an engineer named Phil Drinker was Don had, at that time treated six patients, all presenting his work. He was trying to minimize unsuccessfully, using he Bramson oxygenator­ hemolysis and had a machine that mixed blood -a huge device, but it orked. So Don came to up ferociously, but as long as no oxygen was in­ Santa Barbara and pu this young man on. We volved, no hemolysis. I spoke with Dr. Drinker went up to watch what e was doing. After about to see what we could do with gases. We got a two days of support t is young man recovered. membrane oxygenator with efficient externally His problem was tra a and ARDS following induced mixing, and were able to exchange trauma with a rupture thoracic aorta which had oxygen and CO 2 • That oxygenator allowed us been repaired but was · respiratory failure. So to begin the study of prolonged extracorporeal he recovered and it w s really the first success­ circulation. So we developed models in the dog ful case. The case wa, published in 1972. and other animals and got to the point where we In 1972, Al and were doing cardiac sur­ could, in fact, maintain extracorporeal circula­ gery in addition to ev rything else in this little tion for days on end--unprecedented at the time. hospital. We had a Ii e boy thatAI had done a There were two other labs working on the same Mustard procedure on~d he was failing postop. project. One was that ofDonald Hill, a cardiac This was a 2-year-ol boy, so we brought our surgeon in San Francisco, and the other was Ted oxygenator equipmen from the laboratory and Kolobow, a researcher at the NIH who both de­ hooked it up in the h spital and put this little veloped membrane oxygenators and prolonged boy on. He was on r a couple of days and recovered. Those e Jy cases demonstrated

f

i

18

The Histo

ofEeMO: First Hand Accounts

what we suspected but no one really knew: if and to actually come to understand that that you could support the circulation and respira­ was what was h ppening; that must have been tion for a short time, the native heart or lung amazing. could recover. A few other people started doing Dr. Bartlett: e didn't understand it, but we similar things and by 1975 there were about 25 learned about it. e little girl was the first child successful cases reported in the literature--some of a girl herself om Mexico who had crossed in children, some in adults with respiratory into the country illegally to deliver her baby. The mother was t ld that her baby was certainly failure or with cardiac failure. In 1975, we treated our first newborn in­ going to die and 1jhat the doctors were going to fant with respiratory failure. Why did we do try something that has never worked before but that? Well, as I said, we were the surgeons for they don't have mtCh hope for it. So the mother, scared off by all he forms and questions, and the hospital. We did the pediatric surgery and the cardiac surgery and the traumas and the running the risk f being deported, kissed the hernias and the hemorrhoids and everything baby goodbye an disappeared. We never saw in between, so it was not unusual for the neo­ her mother again. And so the nurses named this natologists to call us. At the time, left to right baby Esperanza w 'ch means "hope" in Spanish. She was well upported on VA-BCMO, but shunting through the ductus arteriosus was a every time we trie to turn down the pump flow common problem in newborn infants, and we would be doing 3 to 4 ductus ligations a week. she got very hypo emic and looked terrible. In the mid-1970s pa ent ductus arteriosus was a So that is why they would ask us to see these patients. We went to see this beautiful full term common problem. This was before prostaglan­ baby whose P02 s were in the teens, clearly dy­ din inhibitors and other agents and we ligated ing of hypoxemia and respiratory failure. She the ductus in man;babies. So, that must be the had some meconium aspiration, but that didn't problem here, but we wanted to demonstrate explain it. Based on our experience with that that before just ch ging in--after all we had this cardiac case a few years before and trials with baby on bypass. ~~Visualize the patent ductus, other newborn cases which were unsuccessful, this was before ec 0 , we did a cineangiogram, we said, "Sure, we will try it again because we injecting dye into t e arterial perfusion catheter, have demonstrated that it will work." We can­ expecting to see it go from the aorta into the lungs. What we sa was just the opposite. In nulated that baby on VA-BCMO through the vessels in the neck, a technique which we had fact, no blood was getting into the lungs and developed in the laboratory--not something we the dye was being ashed out at the level ofthe aorta so the flow as going right to left, fetal just did on the spur of the moment. That child ultimately recovered and was the first neonatal circulation. I mus admit, we didn't believe respiratory failure patient to survive. We later that so we did wh · t any other Gross-trained learned that there is a syndrome called persistent surgeon would do ith unusual cath data. We fetal circulation or pulmonary hypertension of just decided to go t the operating room to see the newborn in which normal pulmonary vaso­ what we could see. constriction, which exists in the fetus continues Well, she had a huge ductus alright, but it appeared that blood as flowing right to left. We after birth with profound right to left shunts through the ductus and the foramen ovale, with thought, "That can' be right, but it's a ductus, let's ligate it." So e did and of course if she secondary hypoxemia. Dr. Peek: We understand that phenomenon had not been on VA ypass she would have died very well now. But at the time there was no on the spot, because in retrospect she had very concept of persistent pulmonary hypertension, high pulmonary vas ' ular resistance. So, when 19

Chapter 2

we ligated the ductus, her pulmonary artery expanded with high pressures. I didn't know what was going on so we put a catheter in her pulmonary artery through a little pursestring suture, using a handy piece of silicone rubber tubing. The PA pressure was twice systemic­ -why should this be in a newborn with perfectly normal cardiac anatomy? We assumed she had some kind of malformation of her distal pul­ monary arteries down in the lungs somewhere, and a fatal condition. So, we left the monitoring catheter in the chest. I planned to just turn her ECMO off but I didn't want to do it in the operating room. I wanted to wait until the middle ofthe night, turn her off in the ICU and see what the pathologist told us was wrong with her lungs. These are the stories that aren't published. I went in about 12 hours later to tum her off, considering it another failure, but two things had happened. First ofall, her pC02 had dropped precipitously with new exhaled CO 2 gas. Secondly, the PA pressures had dropped to less than systemic levels. So, we decided to continue ECMO. She continued to improve and we were able to wean offofECMO, then quickly wean off the ventilator after that. So, after we are off ECMO and off the ven­ tilator then we had to take the catheter out ofthe pulmonary artery. As I pulled it out it was about 4cm shorter that what I had put in. So we took a chest x-ray and in her right lower lobe is a piece of silicone rubber. Otherwise, she was fine. We had a dilemma, what should we do? What we decided was to just leave it there. She was fine and she lived with it and, in fact, we lost track of her altogether. She was eventually adopted by the foster family that took her in. That family moved to Missouri a long way from California and I lost track of her. I moved to Michigan, she was in Missouri. Fourteen years later I got a call from a pediatrician in Missouri saying there is a girl here in my office and she is telling me a story and I can't believe it. He said she has a left thoracotomy, she has an incision in her neck and I took a chest x-ray and she has a piece of 20

catheter in her right lower lung. I said, "Oh, thank you very much I have been looking for that little girl!" So th t is how we reestablished contact with her. Esperanza is now 45 years old and has a daughter of her own and a piece of silicone rubber tubin I in her right lower lung. In retrospect, he problem was PPHN. It turns out that there w one paper in the pediat­ ric literature that was ublished the year before. We went on to treat 0 e newborn and another and another. They w~ Id get better in a day or two. The use of ECMO for neonatal respira­ tory failure proved to tie remarkably successful mostly because thes1babies had completely normal lungs. They j st needed a day or two to accommodate their ulmonary circulation. Dr. Peek: So, ob liously ECMO with the early circuits was morr challenging than what we have today. It w~ ideal for a short-lived disease like PPHN. It las the perfect treatment. But how did you get frfm those patients to the older children and the\adults? What were the challenges you had to ?vercome? Dr. Bartlett: WeI '" the biggest challenge was not the mechanic of the system. It was the perception ofthe p blic. In 1975, the year we treated Esperanza, and based on the very early successes, the sponsored a prospec­ tive randomized trial 0 extracorporeal support in adult respiratory fail reo It went on for three years and it was ultim,tely published in 1979. It was through that stu~y group the technique got to be called ECM . Somebody made up that term because mem rane oxygenation was the focus of it. In retrospect, it was a poorly designed trial, done a the wrong time with immature technology. verything about it was badly done. I can say hat because I was one of the designers of the ial and one of the par­ ticipants in that study. here were nine centers, only three of whom h d ever done a clinical ECMO case before that. There was no break-in period. There was no y to learn the technol­ ogy. There was no sta dard anticoagulation regimen. Most people bl todeath. Therewere

The Histo

lots ofproblems with the trial and it was stopped after 90 patients because of futility. There was 10 percent survival in both the ECMO and the control groups. The paper was published with the conclu­ sion that ECMO offers no advantage in the treatment of ARDS. This was a correct inter­ pretation of results from a badly done study. We cared for several adult patients that were in that trial, in fact, one ofthe only survivors. That study essentially stopped research on ECMO on adult patients for the next 20 years. But we continued to do it in adults, and Luciano Gat­ tinoni continued to do it in Italy, but until 1990 there was basically no other research going on in adults. We did expand it to older children and to cardiac surgery children where it worked very well, but it was the results ofthat Nlli sponsored trial that slowed everything down for a couple of decades. Dr. Peek: So, when you were beginning to do these adults in Michigan, do you think it was just gradually learning each aspect of the treatment and improving rather than a "eureka moment?" Dr. Bartlett: No, there was definitely no eu­ reka moment. It was just one step after another. All the children as well as adults with respiratory failure were placed on VA bypass. In fact, most ofthe adults were on VA through the neck using and ligating the carotid artery and the jugular vein. We had learned that it was safe to do it in babies. We learned after about 50 cases that if you ligate the carotid artery in an adult patient who is in cardiac arrest, stroke resulted about 15 percent of the time. So we eventually stopped doing that and used the femoral artery in prefer­ ence. The carotid actually works perfectly well for adults. So, we treated a lot ofadults. About a hundred adult patients up until 2005, something like that, and had about a 50-60 percent survival depending on the primary diagnosis. So, we knew that it would work well in those patients. We eventually switched to W access us­ ing two separate venous catheters (lVC and

ofECMO: First Hand Accounts

SVC). It was s fer because it didn't involve access to the art rial circulation and supported the patients pe ectly well. So, other groups gradually started eating adults with respiratory failure. In theeantime, the major growth in this procedure w in cardiac support. Initially it was used in patie ts who were immediate post­ cardiotomy, pati nts whose heart did not work well or perhaps eaned directly from cardio­ pulmonary bypas onto ECMO for a while. The results were quit good, with about 50 percent survival compare to death otherwise. It then grew into other ty~es ofsupport for myocarditis and myocardial ~arction, things like that. Dr. Peek: Ob~.OUSlY' one of the problems with ECMO was at you couldn't move the technology very asily in the early days. The patients were very ick and would have to come to you in the hosp ital. But you had the concept that you could tak ECMO to the patient. Dr. Bartlett: Yi s, we started with transports, actually. Our first ansports were in 1973, 1974 in the California d ys. We continued that when we went to Michig in 1980, but developed a much bigger team and had a much bigger op­ eration including a I ery sophisticated transport system. Many ofth patients who were referred to us were too sick t transport by conventional means so we learne to take the ECMO machine to the patient to c ulate the patient wherever they were and brin them back. In fact, that got to be a major s urce of our patients. We much preferred to ansport the patient not on ECMO and put the on once they got to our hospital, but there w re many patients that were just too sick to do th t. Over the first ten years or so at Michigan e transported about 100 patients. The results rere actually better in the transported patients and, looking back on it, the reason was that the~ere usually patients that got very sick very q ckly. We would find them on the first day or so of illness rather than 5 to 7 days later after som outside centers trying to recruit their lung in t e process.

21

Chapter 2

Dr. Peek: For sure. And obviously when you take it out of hospital and on the road it becomes exponentially more complex, so you have all kinds of issues with the technology. Dr. Bartlett: Well, there are lots ofproblems. It isjust a matter of planning ahead and learning to do that so we, in two suitcases, could take the entire EeMO system plus all of the operat­ ing equipment. We went with two nurses and two surgeons and the pilot to go to a place and we developed a technique if we went for 60 miles we would go by ground. We would get the patient, bring them back. If it was from 60 to about 200 miles we would go by helicopter. If it was more than 200 miles we would go by fixed-wing aircraft. Dr. Peek: Thank you, Bob.

A Conversation with Dr. Jay Zwischenberger Dr. Peek: Jay, thank you very much for talking with us. I wonder if you can tell us what it was like coming up with the concept and the idea of building the double lumen can­ nula and basically changing the paradigm from VA-ECMO to VV-ECMO. Dr. Zwischenberger: Well, like all good ideas, it isn't a lightning bolt or an "aha" mo­ ment, it evolves from a single idea in a discus­ sion session to the point to where you actually try to make it work. My first exposure to VV­ ECMO was with a pediatric surgeon named Mike Klein who was at the University ofMichi­ gan working with Bob Bartlett. Mike had be­ come very enamored ofthe idea that VV-ECMO would be less invasive than VA because back in 1983 Dr. Bartlett had just come to Michigan and he was trying to talk everybody into doing ECMO. So it took a couple of the young sur­ geons to embrace this, otherwise, Bartlett would have to do this all by himself. Mike Klein was one ofthe early adopters and he became a major contributor to ECMO. I was a resident at the time and I scrubbed in with Mike on a patient. We were doing VV

22

cannulation and the fI moral vein fell apart and then it fell apart som more, and eventually it dissected and the pa ient died. It turned out the patient had Ehler- anlos syndrome but we didn't know it at the . e. The combination of trying to do VV-EC (and in this syndrome) and watching a femo al vein in a child tear all the way up the vena 9ava and knowing that ty­ ing up the carotid wa~ quite a challenge. Back in those days we wer4 highly frightened by all that. We were so wo~. ed about the stroke rate. We hadn't quite work d out what was the con­ tributor to the inciden e of strokes. We didn't know if it was the anticoagulation, we didn't know if it was premathrity. We didn't know if it was tying off the c~otid so we lost a lot of sleep over the idea of~e procedure. So when I becafe the ~cond ECMO surgeon at Michigan January of 1984 Bob Bartlett came up to meEd said, "Zwisch, I just want you to do two th gs in the lab. Number one, I want you to C me up with a heparin bonded circuit that all 0 s us not to have to use heparin and I want you jO come up with a double lumen catheter." Well, actually I worked hard on both, and history ~as written the heparin bonded circuit went t~ough several iterations. We did an ionic bondi g which then evolved into covalent bonding, hich then evolved into multi layering and com lex layering like what you see now. We sit h re in 2016. That is not resolved. But in 1984, wha I was able to do was partner with an engin ering resident named Ken Drake on the cathfter. Ken was a "good 'ole boy" who loved t,~ work with his hands and was sort of a mec9anical engineer. I had done work on a fann an? I had restored antique cars so I was a tinkere and a fixer upper and a hands-on kind of gU as a surgery resident. The two of us got toge~her and we decided to design a double lumen c~theter. Well, because of our background, we bott went to the hardware store and we bought a hole array of stainless steel tubing, ranging fro: small to fairly large.

!p I

1

l 1

The Histo

ofECMO: First Hand Accounts

He had a machine shop in his basement and the was my mentor and I loved him dearly, his two of us would get together and we came up contribution w just that. And I have to credit with different combinations and permutations John with my firs pUblication. So I presented at of reinfusion cannulas and drainage cannulas. ASAlO and peo e came up and said, "Wow--a Our thought at the time was to have a concentric double lumen ca heter that really works!" So, configuration. That is a small cannula inside we were then in pired by that to keep doing a big cannula. We tried at first a double barrel experiments. This was bac in the late 1980s when the approach like a double barrel shotgun but that was hard to put in and always resulted in a fig­ FDA went throu h a period where virtually ure eight configuration. We hadn't yet learned nothing was bein approved [for clinical use]. to make eccentric tubing so we decided to do There was about two year period where there was a very risk- verse atmosphere, virtually concentric tubing of a tube in a tube. Well, believe it or not, over a month we had nothing new hap ened and virtually all new tried virtually every combination and permuta­ technology was e ported to Europe--Giles you tion of cannula that you could buy or construct lived through that. This period was when the ex­ by soldering stainless steel and having one plosion ofresearc and development took place drain and one reinfuse, and we came up with in Germany and ' urope. The United States became more of t e prototyping site and then the empiric ratio of two to one. Tfthe venous catheter was twice the internal diameter of the Europe would dev lop it in the first patients and back to the United States. reinfusion catheter, that was a sweet spot for then it would co drainage and reinfusion. And, as history is That paradigm exis today--at least in my world written, that sweet spot is the same today. All of cardiothoracic s gery. double lumen catheters basically use that two We went throu a 3-year period trying to to one configuration. We did it empirically find a company tha would make the double lu­ and then Bartlett's group went on to create the men catheter. We J,ere very unsuccessful with mathematics to create the "M" number. But the major companies ai d we virtually almost gave M number was generated off' of all that data of up on the idea that ~ WOUld ever have a doub Ie drainage catheters and reinfusion catheters that lumen catheter that e would have to continue we did with the stainless steel tubing. with the old cutting the chest tubes and putting Then we constructed a soldered machined them in the jugular ein and the carotid artery double lumen catheter out of stainless steel: the until a small comp~ called Kendall came to first properly oriented, properly proportioned us and offered to m e the catheter. The only drainage and reinfusion catheter. We made a caveat was that we ouldn't have any intellec­ beautiful soldered catheter and then we worked tual property rights. 0 Dr. Bartlett, Ken Drake, with John Toomasian, who was the director of and I made the deCiSion that we would give the the lab at that time, and Al Petkus, a veterinarian. intellectual property 0 the Kendall Company if And we did a whole series ofanimals where we they would just mak -Ithe catheter. They agreed. did VV support. We supported those animals for That first catheter we t through several compa­ 8 hours without a ventilator on total VV support. nies acquiring the te hnology and changed its name several times. But that catheter was the It was my first presentation at the American Society of Artificial Internal Organs (ASAIO), standard of care for ai ost 20 years. And I am and John helped me with the first presenta­ pleased to see it hap en. As we tried to e ucate people on double tion and paper. After all this work, Dr. Bartlett looked at it and said, "Well Zwisch, that looks lumen catheters the ost difficult part was that simple. That's no big deal." So, although he it was designed with significant (20 percent) 23

Chapter 2

recirculation. We knew that, but when you used it on patients and it started to recirculate it would make the critical care physicians crazy because they couldn't get the kind of gas exchange that they wanted to get in order to have "perfect" numbers. They couldn't get an oxygen satura­ tion of greater than 90 percent. They couldn't get the numbers they had grown accustomed to look for. So, we had to educate people that a saturation of80% and a pC0 2 and a pH that was normal was a perfectly viable blood gas for a newborn with respiratory failure. But that was a huge educational process in those days. Dr. Peek: Still ongoing ... Dr. Zwischenberger (laughing): ~s. ~ natural tendency if the oxygen saturation was 't ~c WE to turn up the flow which creajed two problems; first a further increase in recircu­ lation and -;;cond, hemolysis. So it would, by definition, frustrate the caregiver because higher VV-ECMO flow was more recirculation and less support than lower VV flow without recircula­ tion. We gave lots of talks and demonstrations labs and we would give simulations [to help with understanding]. VV-ECMO with a single double lumen cannula was slow to be adopted, but those who learned how to position the catheters and use it correctly were wild enthusiasts. There was a period when about half ofthe ECMO programs supported ECMO with a double lumen catheter and about half wouldn't even try it. And, the better results were initially being obtained with VA. But over time, the W enthusiasts showed that the outcomes were better in respiratory fail­ ure. There were less bleeding complications and cerebral vascular problems. There was a general feeling that the W patients got better quicker and 35 years later, it is hard to show that. But that was the general feeling at the time and that is what allowed W use to continue. Dr. Peek: Is that what led you to want to develop the bicaval cannula--the thought ofget­ ting through the recirculation and giving people a better cannula? 24

ZWiSChenber~1

Dr. er: The bicaval cannula that evolved 20 years. ater was based on a lot of those early drawings here we thought through all the combinations d permutations. I was a cardiac surgeon by traoe so I was used to bicaval cannulation anytime i,e had to open the atrium or do major valvular ~6-gery on the tricuspid or mitral valve. So, that was an accepted technol­ ogy to me. We also fel that the Kendall catheter, no matter how big yo~ make it, would never be useful for adults with rrspiratory failure because of the recirculation an~ because of the possibil­ ity of developing pu~onary hypertension. So, we felt strongly that we had to get a cannula that allowed total pulmonary support. Now, there was a lot new evidence up to that point. We had tried the intravenous oxygenator (IVOX). We knew t1at partial support even though it was enticing didn't really work that well. We had already ~ed arteriovenous CO 2 removal. We knew trat it would work in a limited patient pOPulatf·n but wasn't universally applicable. So, we w ted to come up with a universal catheter tha would work on adults and we also wanted it ~o potentially be ambula­ tory. Even then, back· the late I 990s, we felt strongly that ambulati n was how the human had evolved. We had eJolved as hunter gather­ ers. We evolved as erl ct, ambulatory animals and the idea of Criticalicare with the patient se­ dated, flat on their back intubated, with a Swan Ganz catheter, an art I· e and a Foley catheter and sedated and paral~zed and aspirating we just didn't think that ~at was going to be the future. And with my c?l1aborator, Dong-Fang Wang, we created the bi~aval cannula, a big next step in allowing patien~ mobility. Dr. Peek: We are ~nning out of time, but not out of stories. Thank you Jay.

Of

3 Evolution and History of Global ELSO Graeme MacLaren, MBBS, FRACp, FRCp, FCICM, FCCM, Rodri Diaz, MD, MEd, Malaika Mendonca, MD, Giles Peek, MD, FRCS, CTh, FFICM, Suneel Poobo i, MBBS, MD, DCH, FRCp, FRCPCH, FCCp, Peter Rycus, MPH

Following the influenza A(H1N1) pan­ demic and publication ofthe CESAR study! in 2009, the use of ECMO for adult respiratory failure increased exponentially.2 In 2012-2013, the annual number ofadult ECMO cases world­ wide reported to ELSO overtook pediatric and neonatal cases for the first time. As ECMO use increased in adults, many ECMO-naIve centers outside North America expressed an interest in adopting this technology. This attention helped rekindle interest in forming regional ELSO chapters across the globe in order to better ad­ dress local needs. The inaugural chairs ofthese chapters recount how they evolved.

EuroELSO Giles Peek I don't recall exactly where Dr. Bartlett sandbagged me, but indeed it was a "great idea," as he put it. As a long-time member ofthe ELSO steering committee, I had often vocally complained to my North American colleagues that ELSO was (mainly) a North American organization, even though it had a global remit. "Giles, we need a local ELSO organization in Europe," said Dr. Bartlett, "It's a great idea," he continued. I had to agree. So in 2011 we had a preliminary meeting at La Piti6-Salpetriere Hospital in Paris and the EuroELSO steering

committee was du~ elected. As the inaugural chair, I quickly re.tlized that this was a BIG job and I really ha~ ~o idea how to achieve our objectives, or inde,d exactly which objectives of education, re~SirCh' communication, and the Registry were e most important. But in the words of the so g "With a little help from my friends," the ist spectacular annual Eura­ ELSO conference "fas organized by Roberto Lorusso in Rome 12012, with 878 delegates. This was followed by a sophisticated meeting in Stockholm orgark:ed by Bjorn Freckner in 2013 with 910 delegr.tes. In 2014, Alain Combes welcomed us to Pa~s in the springtime where 1200 colleagues attepded. 2015 saw EuroELSO in the historic city ofReg ens burg with Thomas Mueller hosting ov+ 1300 ECMOlogists. The attendance of so m~ny delegates far exceeded our expectations an? more than vindicated Dr Bartlett's inspirati~nal directive. This does make each successive meeting more stressful for the conference org~izers as they try to attract more delegates than, Ithe previous year. We gradually drveloped our "European" way of working w?fch was perhaps more de­ volved than that of~e parent organization. The annual conference c~air was given absolute au­ thority over the conti rence and this has allowed each chair to stamp is mark on the successful but subtly individu I annual meetings which followed in Stoc oIm, Paris, Regensburg,

25

Chapter 3

and Glasgow. We included our nursing and perfusion colleagues as full voting members of the steering committee and we established the semi-permanent nonvoting officers of the steering committee, the treasurer, and secretary, to keep the administrative and financial mat­ ters of EuroELSO on an even keel. Each and every member of the committee has played an instrumental part in making EuroELSO the suc­ cessful "organisation" it is. As British English is the official language ofEuroELSO, this word is spelled with an ' s', not a ' z'. The growth of ECMO within Europe has paralleled the global experience. By the end of 2014, there were 65 EuroELSO member centers, 20% of the world total, reporting 1488 cases for the year (21 % of the total). Comparing Eu­ ropean with global data from the January 2016 ELSO report shows some interesting trends (Table 3-1). Note that all groups of patients are somewhat under-represented compared to the number of centers but the proportion of adult respiratory cases is much higher. It will be interesting to watch this develop over the following years.

Table 3-1. EuroELSO

VS.

Global ELSO

January 2016 ~urope

iNeonatal

n (%)

rrotal

26

~8723

Respiratory

~274

L-ardiac

~73 (11%)

6269

ECPR

83 (7%)

1254

(11%)

Pediatric lRespiratory 1417(20%)

Adult

~Iobal

7210

L-ardiac

1128 (14%)

~021

ECPR

1287 (10%)

12788

iRespiratory 12.759 (30%)

9102

\"'ardiac

1129 (14%)

7850

ECPR

463 (18%)

2379

11213 (15%)

73596

n

I am thankful to Dr. Bartlett for giving me the inspiration and opportunity to build the EuroELSO team. N w, having stepped down as Chair, I am imm nsely proud to see them go forwards under t e leadership of Roberto Lorusso to achieve s much more than we could have dreamed of in 011.

Asia-Pacific ELSO

Graeme MacLaren A group of EC 0 Directors from the Asia-Pacific region et in Rome 2012 for the first EuroELSO con erence. Given how well attended and succes ful the meeting was, it seemed worthwhile t set up a similar chapter in the Far East and 0 eania. In July 2012, a preliminary meeting in Melbourne, Australi was held to develop the concept further, host d by Dr. Yin Pellegrino. Steve Conrad, then th Chair ofELSO, attended. From the beginning, p ople seemed prepared to go to great lengths to ake sure it worked. For example, four doctor from Japan flew down just for the meeting d left the next day. There was unanimous agre ent about the potential value ofthe chapter d a ballot was conducted across the Asia-Pac' c ELSO centers to elect the inaugural steerin committee. Rather than use the same struc as the parent organiza­ tion, the APELSO ste ring committee decided to have a different fi rmat, with an emphasis on education. The fir t meeting in Beijing in October, 2013, was ery successful and was followed by an equal y successful conference in Kyoto in July, 201, . Dr. Cun Long stepped down as conference c air and was replaced by Dr. Ryoichi Ochiai. er the Kyoto meeting, Dr. John Fraser became th conference chair for the third meeting, to be h ld on the Gold Coast in Australia in October 2 17. From the beginn', g, the primary aim of APELSO has been e ucation. From 2014 on­ wards, Simon Sin, P er La~ and colleagues

!

Evoluti nand History a/Global ELSa

from Queen Mary Hospital, Hong Kong, have run annual ECMO workshops, all selling out months in advance. At the inaugural workshop, a steering committee meeting was held and again demonstrated the lengths some members were prepared to go to: two doctors flew down from Beijing for the meeting and left the next day, while a Korean doctor flew to Hong Kong to attend the two-hour meeting, arrived just as we started, then returned to the airport and flew home. Over the last few years, APELSO has laid the foundations to help promote ECMO and educate clinicians in the region about how best it may be performed. In the future, we hope to provide more resources to clinicians from lower income countries.

Latin American ELSO Rodrigo Diaz Latin America started to do ECMO in the early 1990s.3 These were just sporadic cases until 2003, when the first hospital in South America joined ELSO, Pontificia Universidad Cat61ica de Chile in Santiago, led by the neona­ tologist Javier Kattan.4 After training in the U.S., his group started a successful and regionalized experience in pediatric and neonatal care. Five years after that, another center in Santiago be­ came the second hospital reporting to ELSO. In 2009, hospitals in Colombia, Brazil, and Mexico all joined ELSO. Over the last five years, programs have grown quickly and centers contributing to the ELSO Registry now come from several coun­ tries, inc1udingArgentina, Costa Rica, and Peru. In December 2012, the Latin American chapter was founded with the assistance ofPeter Rycus, William Lynch, Steve Conrad, and Mike Hines. More than 200 people attended the first Latin American Symposium on ECMO and started the international collaboration based on ELSO's vision and mission. Two co-chairs

were elected, Javi r Kattan and Rodrigo Dfaz. 5 Their aim was to establish the chapter steer­ ing committee an together institute its legal foundations, enaC~1 the bylaws, start mutual collaboration, and ring international standards and education in CMO care to the region. Now there are mo e than 25 ELSO centers in the chapter that h e reported more than 600 cases, more pape s are being published by local authors, and collaboration in education is growing. The F· st ECMO Specialist Train­ ing Course was h ld in 2013 in Brazil at the Heart Institute, Un versity of Sao Paulo Medi­ cal School. 6 Since 2014, the Latin American Chapter team, tog her with different national and international te chers, have done an annual VV Adult ECMO ourse in Santiago. In 2015, an "MCS in Cardi genic Shock Course" was started in Bucaram gao Efforts are being made to establish trainin courses in other countries ofthe region. In December 2 14, the first Latin American ELSO Chapter Co gress was held and more than 500 attendees shared experiences at this meeting. Luiz Can 0 from Brazil was elected the Chapter's chao an. Latin America as unique characteristics, and we have to be able to share experiences, adapt techniques to rur needs, and maintain the best standards ofcare focused in education, data registry, and pUblicftion of our experience. In 2015, a first center ip the region got the Center of Excellence Awai and three centers in the region are regularly oing mobile ECMO: Mon­ terrey in Mexico, ucaramanga in Colombia, and one center in antiago, Chile, 6 and more cases are being rep rted each year. In Chile, a joint commission 0 the Intensive Care Society and the government established formal criteria for ECMO and mad it obligatory to report the results to a national atabase.

27

Chapter 3

South and West Asian ELSO Suneel Pooboni and Malaika Mendonca The suggestion of establishing a global chapter covering West Asia and Gulf Coun­ tries, which were already cooperating in terms of training, came early in discussions between ELsa representatives and the ECMa Society of India, which was established in 20 I o. Early in 2013, the South and West Asian Chapter of ELSO (SWAC ELSO) was fonnally created dur­ ing the 4th ECMa Society ofIndia conference in New Dehli, with the help of Peter Rycus and Steve Conrad. The executive body was elected during the same meeting, with Suneel Pooboni being elected as Chairperson. The first two SWAC ELSO annual confer­ ences were held in India (Hyderabad in 2014, Bangalore in 20 IS). The increasing presence and engagement from people in the Gulf region then brought the third annual conference to Abu Dhabi, UAE, in 2016, which led to sharpened awareness of ELSO as an organisational body, and an increase in centers registering from outside India. During this steering committee meeting, people from different countries were assigned to be members ofsubcommittees, and Malaika Mendonca was elected as co-Chair with Suneel Pooboni. The already vast region was further extended to include Africa, in par­ ticular the Republic of South Africa. As a young chapter, SWAC is still in the transition phase of defining roles and respon­ sibilities of its subcommittees, its bylaws and financial alignments. The chapter covers a diverse area, with ~­ tential catchment population ofnearly 2 bil.lion people. Although centers of only 11 countries have registered membership in ELSa by the end of2016, their participation is developing rapidly. The nature and diversity ofthe SWAC area leads to specific challenges and indications for ECMO rarely seen in other regions, such.QS ml'0carditis due to scorpion stings or snake 28

pneumonias. The region inelu es countries where access to sophisticated hea hcare is possible without any limitations to c st, and others where the provision of basic h althcare remains the main priority. The area al 0 includes countries torn apart by war. Healthc re funding is very diverse, with South Asia co sisting of populations of mixed income gro s with healthcare fund­ ing coming from a ix of private companies, government sponsor hip, and individual health insurance, while hea~hcare in West Asian coun­ tries is mostly based n government funding. In this context, our focu has been on the exchange ofexperience and 1m ledge among the centers already performimg CMO and those who wish to start. Courses, w rkshops, and continous education are the pri I rities of the chapter. SWAC ELSO, as ith all the other chapters, has the important m· ssion to knit together a community of divers but motivated specialists. Exciting times lie ah ad.

Evoluti n and History a/Global ELSa

References

1. Peek GJ, Mugford M, Tiruviopati R, et al. Efficacy and economic assessment of conventional ventilator support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet 2009; 374: 13 51-63 2. Extracorporeal Life Support Organization ECLS Registry Report. lnternationaJ Sum­ mary, Ann Arbor, 2016 3. Castillo L, Bugedo G, Hernandez G, Mon­ tes lM, llic JP, Labarca E et al. Soporte res­ piratorio extracorp6reo: nuestra experiencia. Rev Moo Chile 1996; 124: 45-56. 4. Kattan J, Gonzalez A, Castillo A. Oxigenacion con membrana extracorporea neonatal-pediatrica. Rev Chil Pediatr 2013; 84 (4): 367-378. 5. Diaz R. ECMO and ECMO Mobile. Me­ chanical Cardiopulmonary Support. Rev. Med. Clin. Condes - 2011; 22(3) 377-387. 6. Caneo LF ; Neirotti R. ECMO: Improving our Results by Chasing the Rabbits. Braz J Cardiovasc Surg 2015;30(6):657-9 7. Kattan J, Diaz R. Latin ELSO Chapter: the Latino American experience. Presented at 25th ELSO meeting, September 2014. Ann Arbor, MI, USA

29

4 The Physiology of Extracorporeal Life Support Robert H Bartlett, MD, StevenA. Conrad, MD, PhD

EeLS (BeMO) is the use of mechanical devices to support heart andlor lung function in severe heart or lung failure, unresponsive to optimal conventional care. With circulation and respiration supported by EeMO, damaging heart and lung treatment can be stopped (eg, vasopressors, high ventilator settings) while the failing organs are treated, recover, or can be replaced. Managing patients with EeMO differs actively from conventional care, and requires a thorough understanding of cardiopulmonary physiology, pathophysiology, and EeMO physi­ ology. This chapter includes a brief review of normal and abnormal cardiopulmonary physiol­ ogy, and the physiology related to mechanical replacement of circulation and respiration.

Cardiopulmonary Physiology Figure 4-1 summarizes the essentials ofnor­ mal cardiopulmonary physiology. All tissues of the body function by combining substrates (food) with oxygen, producing heat, energy, eoz' and water in the process called metabolism. Metabolism is related most closely to oxygen consumption, and is determined by measuring the amount of oxygen consumed per minute, which is called VOz' The rate of metabolism for adults at rest is approximately 3 cc/kg/min or 120 cc/minlmz for a typical adult (children 4 cc/kg/min, infants 5 cc/kg/min). Metabolism

is controlled by a center in the brain, and in­ creases or decreaseISdepending on activity and other factors. Metabolic rate increases with activity, fever, and drugs and horlnones, and decreases with sleep, paralysis, arld cooling. Metabolic rate increases as muctl as five times in extreme exercise. When thelmetabolic rate changes, the delivery of substr~te and oxygen changes in proportion, accomplished by a change in cardiac output. The amounf. of oxygen available in the bloodstream for metabolism is normally five times the amount a~tuallY used by the tissues. A complex system ofreflexes and hormones keeps this all in balance, teferred to as homeostasis. Oxygen gets ilito the blood through the lungs and arrives tissues via perfusion of

F

r-ve-ntCi-l.-tion:-- -O, XYGE. ~_::~I ::~:'E:~~~:~~~~N co, cl.aranc~/

----+-- .' -----'" r '

/

--.L___. VENOUS 0 CONTENT' Delivery.Conlon'

!

.

.

Let:

;;, ~~ a:~~~~~~T ! i \ "' IT'l l "

-~ ~

HEMOGlOBIN

({Tm/JOO c()

.

VfN OUS

RETURN

,

y.~ ..

.

OM .

SE-FSIS

,--' -. CO2 ProductioO!

I

WORK

- 1

r --;,_ R'_, ~ 1 o~ Cb N SUMPTIO N L_ l a/m,. ) _.__ _,J

Figure 4-1. A

s'~ of oxygen consump­

tion, delivery, andi:e~olism.

31

Chapter 4

the capillaries. About 20-25% of the oxygen is removed for metabolism (although the oxygen extraction ratio varies from organ to organ) so 75-80% of the oxygen is still in the venous blood on the way back to the heart and lungs. Carbon dioxide is produced during metabolism; the amount (VC02) is essentially the same as the amount ofoxygen consumed (3 cc/kg/min). CO 2 comes out ofthe blood in the lungs and into the exhaled air. The amount of oxygen consumed and CO2 produced is different for each organ but the average for all organs is measured by 02 and CO2 exchange in the lungs (Figure 4-2). These principles apply to all ages and sizes, and size specific parameters are normalized to weight or BSA. Typical adult values are used in the examples in this chapter.

tant (and as a Singlf value, perhaps the only important) measure ent of oxygen in blood. In clinical practice the amount of dissolved oxygen is less than % of the content, so the second half of the equation involving P02 is usually ignored. Fi~[ re 4-3 shows the relation among these meas ements. Notice that there is twice as much ox gen in arterial blood at a normal hemoglobin content 20 mlldl) than in anemic blood (contept 10 mlldl), even though the oxygen saturatio and P02 are the same in both samples. The amount of 0 gen delivered to metabo­ lizing tissue is the a gen content in arterial blood times the blo d flow (cardiac output), called the oxygen de ivery (DO).

t

D02 (cc/min) = CO2 cc/dl) x CO (llmin) x 10 (dlll)

Oxygen in Blood The oxygen content is the amount ofoxygen bound to hemoglobin plus the amount dissolved in plasma related to P02:

CO2(cc/dl) = Hb (gmJdl) X S02 x 1.34 (cc/gm) + P02 (mmHg) x .003 (cc/dllmmHg) Oxygen content is difficult to measure di­ rectly so it is typically calculated and reported by blood gas analyzers. It is the most impor­

For an adult the rmal D02 is 600 cc/minl m2 (20 cc/dl x 3 11m' 1m 2 x 10). The normal amo t of oxygen consumed by tissues at rest is 1 0 cc/minlm 2 , abbreviated as V02 • Figure 4-4 isplays the relationship between these conce ts. The D02 is controlled

r

"'0 ......

c.)

c.)

..­

c

O2 KINETICS:

0.)

Cardiac Index = 3L@l Sala: 100% C. ~ =20 ccldl CO,

".

c/eftvSry - 002

cc@l Ii 120 cc@l ~ 1 - 002N~

----.."",,;oq 600

Hb 1 Hb 7.

....c

V0 2120cc@l

Hb 1

0

{I--­ (I-­

r;-­

0

0

HbO P0 2 : 25 50 75

100

150

S9

100

7, 600

.

Ratto

V02 ~

SAT: 5075

9

@l-I-

Figure 4-2. Oxygen delivery/consumption (D0 2NOJ Typical adult values are shown.

Figure 4-3. Oxygen in blood is measured as P0 2, oxyhemoglobin saturation, and oxygen content. Oxygen conent is the only measure­ ment of the amount o · oxygen in blood, hence the most important m asurement.

32

----------------------------------------------------------+-----------------------­

Va

1)0;..:

C 0 'J.. ( "

l-

OJ.- " C6 x (("D l- 5

~$

t

C'II 0..) The Physiology

:z.: ,

by homeostatic mechanisms to be 5 times V0 2, so in a resting adult 20% of the available oxygen is used for metabolism, leaving 80% in the venous blood. Therefore, the normal arterial oxygen values ofa patient breathing air are P02 90 mmHg, saturation 100%,°2 content 20 ccl dl. Normal venous oxygen values are P0 2 40 mmHg, saturation 80%, content 16 ccldl. The V0 2 increases with exercise, catecholamine release or administration, and sepsis. The D02 adjusts to V0 2, maintaining the ratio at 5: 1. D02 is limited primarily by cardiac output. If V02 increases relative to D02 (or ifD02is impaired), a higher fraction ofthe arterial oxygen content is removed by the tissues, so the content in the venous blood decreases from the normal 16 ccl di to lower levels. This is well tolerated until the DOjV02 ratio is below 2: I (50% extractigy). At that point there is not enough oxygen available to maintain oxygen-dependent (aerobic) metabolism, and metabolism switches to anaerobic processes which causes exhaustion and lactic .ilcidosis. The V0 2 below this level then becomes dependent o~ the supply of 02' Anaerobic metabolism is tolerated for a few hours at most, leading to cardiovascular and metabolic collapse if it persists. 6·

002 /

va, Ratio V sat

VOl

2:1

3:1

4:1

500/0

66%

75%

5:1 80%

NormaJ HypermetaboUc

(ml/kglmln) 3

Figure 4-4. DOjV02 relationships during normal and elevated metabolic rate. D0 2 adjusts to changes in VO~ ?ver a wide range, maintaining D0z. 5 times v0 2. IfD0 2 drops below 5: I, aerobiC metabolism continues, but ifDOjV02 is less than 2:1, anaerobic metabo­ lism and s1:iock occurs.

r

Exlr=mporeol Ufo Support

Cardiopulmonary Pathophysiology

I The relations~ip between D02 and V0 2 can be affected b~disease states, primarily those that affect 0 genation and cardiac out­ put. If the D02 is ecreased compared to V0 2 (eg, in low cardiac output states, anemia, or hypoxemia), V0 2 icontinues at the same rate, thus more oxygen ~s extracted per dl of blood, leaving less oxyger in venous blood. Normal aerobic metabolisf continues in this setting. However, when t~D02 is less than twice the V0 2, oxygen suppl is inadequate to maintain aerobic metabolis and anaerobic metabolism ensues, producing 1 ctic acid rather than CO2, A D0 2:V02ratio less than 2: I leads to supply-de­ pendency hypoxia nd systemic acidosis, with resultant organ fail e (Figure 4-4). A goal of managing any criti all ill atient is to maintain DO :VO close to n rmal (5: I ,or at least more tlian the critical 2: I So it is important to know the V02 and DO hen planning management. Cardiopulmona ECMO

Pathophysiology during

ECMO is used !hen heart or lung failure is so severethatD02 :Y02 is less than 2:1, or when the interventions ~eeded to keep D02 twice V02 are inherently~aging (eg, high airway pressure, high Fi 2' or vasoactive drugs at high doses). In its s plest form (venoarterial), ECMO maintains n rmal D02:V02by draining most of the venous lood, pumping it through a membrane oxyge~~tor, and into the systemic circulation. Most of~e blood bypasses the heart and lungs and the artificial organs replace the function of the disefsed heart and lungs. This is shown in Figure 4-5, in a neonate, as an ex­ ample. On ECMO,Isafe D02:V02 is restored and the damaging v~ntilator settings and drugs are discontinued. is provides time for the organ dysfunction t be diagnosed and treated, leading to organ re very in most cases.

33

Chapter 4

The ECMO Circuit

Centrifugal puml s create suction which can lead to hemolysis w~en the suction pressure is Cannulation high, so centrifugal rumps are operated under conditions to avoid high suction pressures. Gen.. I Blood flow through the extracorporeal cir­ erally, pressures no ore negative than about cuit is limited primarily by the size ofthe venous -50 mmHg are target d. drainage catheter. Resistance to blood flow var­ ies directly with the length of the catheter and inversely with the fourth power of the radius of the catheter. Consequently, the shortest and Modem membr~ne lungs achieve gas ex·· change by perfusing! venous blood through a largest internal diameter catheter that can be network of thousan4s of small hollow fibers. placed in the right atrium via the access vein will allow the highest rate of extracorporeal The tubes are filled tith continuously flowing blood flow. Blood drains through the venous gas (the "sweep gas'i), either 100% oxygen or an air/oxygen mix, hile blood flows exterior tubing to a pump that directs the blood through to the fibers. The h I How fibers are made of the membrane lung and back into the patient. a material that aHo s gases to diffuse across Blood Pump the membrane wall, ut prevent liquids from passing through. ygen and CO 2 diffuse Blood pumps are designed to direct the between the gas an the blood as a function venous drainage through the membrane lung, of the gradient betw en the partial pressures then return it into the patient. Pumps can be on each side. When i he gas is I 00% oxyg~n, centrifugal, servo-modified roller, or peristaltic. the gradient driving ~as transfer is from ®9 Centrifugal pumps modified for long-term use to 40 mmHg for 02' d 45 to 0 for CO 2 , Even are the most commonly used. Rupture of the though the gradient i much larger for oxygen, circuit can occur when the post pump pressure the solubility and d' , sivity of C02 is much exceeds 300 mmHg, so pumps are modified to greater, so the amount of02 and CO 2 exchanged prevent overpressure. is roughly equal whe I to ratio of blood flow to gas flow is I: 1.

°

Oxygen Transfer in ~embrane Lungs ,.,..._. .... ,:."'~ ,

r--:::''':~

:.:~:

>- .....

The maximal 0 2 transfer capacity of any membrane lung is d .termined by the gas ex­ change surface areatd the amount of disrup­ tion oflaminar flow blood passes through the device. Laminar flo ' allows equilibration of blood of lower P02 a the membrane interface, reducing the gradienr for diffusion. Lamin~ flow is disrupted by 4mall secondary flows as the blood moves through the irregular blood flow path, mixing saturated red cells with deoxygenated red ce Is and maintaining the gradient. The amount of 'xi flows is one of the most im ortant factors'

fully

Figure 4-5. A simple diagram ofVA·.£CMO, shown in a newborn.

34

The Physiology

determining the maximal 0 enatin ca acity. All these actors are summarized in the concept of"rated flow." When venous blood is perfused at a low flow through a membrane lung there is sufficient time for equilibration and the hemo­ globin saturation of the outlet blood is 100% saturated. As flow increases, a point is reached when the blood passes through so fast that all the red cells are not oxygenated, and the outlet saturation drops below I 00% saturation. ~ flow ofvenous blood which exits the membrane lung at 95% saturation is defined as the "rated flow" (standard venous blood is defined as :Hb12 grnIdl, and saturation 70%, Figure 4-6). Oxygenators for ECMO are chosen based on the rated flow for oxygenation. The size of the oxygenator is matched to the oxygen require­ ment of the patient. As long as a membrane lung is perfused at a rate below rated flow, the amount of oxygen delivered to the blood by any membrane lung is the outlet minus inlet 02 content difference (DO-I) times the flow. Normal DO-J differ­ ence is 5 cc/dl. Figure 4-7 shows the amount oT oxygen delivered related to blood flow for different DO-I.

if Extracorporeal Life Support

coz Transfer The amount 0 CO2 cleared by any mem­ brane lung is the· et minus outlet CO2 content difference (DI-O 02). At 1:1 gas to blood flow ratio this will be a out the same as oxygen. But when the sweep to flood flow ratio is increased to as high as 8:1, .r much larger DI-O can be achieved and muc~ more CO 2 can be removed. Therefore when a F embrane hmg is used pri­ marily for CO2 reIY,oval, high gas:blood ratios are used and CO2 olearance can be achieved at a much lower bloo flow than when the goal is oxygenation. The weep gas flow rate is set by the operator based n the desired PaC02. These phenomena are d onstrated in Figure 4-8.

Other componen1 The cannulas, pump, and membrane lung are connected by cdnduit tubing. It might seem desirable to have the circuit as close to the patient as Possible.~but usually the connection lines between the atient and the circuit are about 6 feet long ecause it is easiest to care for both the patien and the circuit when they

OO 200

100

75

Hb !lO '1.. 5Olur­ Olion

fOf6, A-VSvol. %

ITiriiiil Ihillt (venous)

JOturol iOfl

Maximum 02 dellvery__ "i' I I

I

I

I

41)0 -,-_ _ _......_ __

_

,.....".

I

150

02 100 Otli....ry

1 Outhn · Inlet Q2 CDnt!nl tecldL)

350

+-,.,,;..-~--!---~.£-'--7"'l

300

-l=..-=-----~

~250 ··~~~~~7,~J~~~

eel min



~

ci\

200 ··!:--':::~-c-~y:"",~-r-~

S15D +-____y/~~~~~--, 2

Z Flow L/mln

3

FIO'Iff L/min

Raled Flow : 3 L / min

Figure 4-6. The concept of "rated flow." Ve­ nous blood perfused through a membrane lung exists at 100% saturation until a limitation is reached and blood exists at less than 100% saturation. The capacity of membrane lungs is described as "rated flow."

40

50

60

Figure 4-7. The ~ount ofoxygen supplied by a membrane lung is he flow times the out-in 0 2 content difference. Bloooflow is in deciliters.

35

Chapter 4

are separated. One reason is because the pump and lung are mounted on a bulky cart which also carries the pump motor, a large battery, a water bath for circulating warm water through the heat exchanger, an oxygen tank and gas regulator for travelling, and the monitors and displays. Moni­ tors and alarms can include venous and arterial blood gases, pre and postpump pressure and flow, and blood temperature. There are access sites for infusion and blood sampling.

ECMO Circuit Physiology

with left ventricular which has traversed the lungs. Hence, CO 2 in the patient's combination of and the total of the extracorporeal plus the amount of blood passing the heart and lungs. Hemodynamics VA access are dem­ onstrated in Figure . As venous blood is drained from the' atrium and perfused into the aorta, the total flow remains constant but the pulse n.A,n..-n.. l1"i decreases since there is less blood ejected the left ventricle. When the extracorporeal is 100% of the venous return the systemic contour is flat. This is the situation in access for heart surgery (cardiopulmonary CPB). In CPB, the superior and inferior cavae are occluded proximal to the cannulas, so that all the venous return the coronary sinus) goes through the . In VA-ECMOtheflow is maintained at about ofvenous return, so 20% passes through heart and lungs. The reason is to avoid tq-t600 180 Autotransfusion Yes No Yes Arterial Filter NO Patient Asleep Awake Environment OR, Hours ICU Days ·Venoarterial bypass, same devices, physiology

37

Chapter 4

tion in the nonnal range of 5; 1. Therefore a very efficient heat exchanger and a large water bath are required for cardiopulmonary bypass for heart surgery. Aside from these differences in perfusion technology, the entire approach to management of extracorporea1 circulation differs markedly comparing CBP to EeLS. Cardiopulmonary bypass is conducted in the operating room with the sole intention of operating upon the heart. There is an appropriate sense of urgency to minimize the time on bypass. Complications including myocardial damage, renal failure, liver failure, hemolysis, and abnonnal bleeding increase proportionate to the amount of time on bypass. Unlimited amounts ofb1eeding in the operating field are tolerated and managed by autotransfusion, with the realization that the effect of heparin will be reversed by protamine at the end of the procedure. An hour or two of rewarming and attempts to come off bypass is considered an exceedingly long and tedious interval. Sometimes huge doses of catechol­ amines are given to encourage a sluggish heart simply in order to come off bypass. If the pa­ tient cannot be weaned offbypass in a few hours, a mechanical support system (ECMO or VAD)

-----cM""' o-cnito, --,1

V

i:"

-

--:'

p VO" 'lCO,

'-----­

~alcuJ~ 00,

' ........

Gas Exchange in JI: During VA-EC 0, fully saturated blood from the circuit is perfused into the aorta and mixes with blood fro the left ventricle. If the lungs are functioning well, the mixed blood is well oxygenated and has normal PC02 • The patient can be weane from the ventilator and managed awake. If e lungs are functioning I

i Ii ! -­ i BP, PAP, CO Monitor

Complia nce, . SVR PVR !

_.. ....._

must be instituted. he patient is anesthetized and paralyzed renderng neurologic evaluation impossible. Everyone caring for the patient measures success O~failure in hours of CPB. In contrast, Ee S is managed in the leU by a team expecting days or weeks of continu.. ous care. The patie t is maintained awake or awakened at regular tervals to evaluate neuro­ logic function. Feed g, ventilation, antibiotic management, renal unction are all-important aspects of ECLS ca e. The use of inotropic drugs and high vent lator settings is minimal, and weaning from bX-pass may proceed over a period of hours or da~s. The patient commonly lacks heart, lung, or r1nal function for days, and futility is conceded 0 ly after many days of vital organ failure.

I

! i

-- ­ - ---,,--,.,

,;.,0"

S,O"

t-temoglob m ' - " ~

70%. When the native heart function is adequate, conduct a trial off bypass. When heart function is satisfac­ tory, decannulate the patient.

Venovenous ECMO In venovenous CMO (VV), the perfus­ ate blood is returned to the venous circulation and mixes with venofs blood coming from the systemic organs, rais' g the oxygen content and lowering CO 2 conte t in the right atrial blood. This mixed blood, n w higher in oxygen con­ tent, passes into the ight ventricle, the lungs, and into the systemi , circulation. VV access is achieved by drain' g venous blood from the IVC via the femoral v in and reinfusing into the RA via the jugular ' igure 4-13), or by drain­ ing from the IVC an SVC and reinfusing into the RA via a separate lumen in a double lumen cannula (Figure 4-14, . Hemod namics urin VV access are not affected by the circ it. Since the volume of blood removed is ex ctly equal to the volume of blood reinfused; ere is no net effect on central venous volu e or pressure, right or left ventricle filling, r hemodynamics. The content of oxygen a d CO2 in the patient's arterial blood represe ts that of right ventricle blood modified by an pulmonary function that might exist. The sy~temic blood flow is the native cardiac outpu~ and is unrelated to the extracorporeal flow. ~i

BP. PAP. CO SVO]. Sa ~, Hemoglobin

I ; i

- -'

EXTRACORPOREAL LIFE SUPPORT !J:t>:ldY:l~V~Pl'

LUl x; Tim",;:!;.' .

tlJ-o.

Figure 4-13. VV-ECMO for respiratory sup­ port. 2-Cannula access.

40

Figure 4-14. VV-EC cannula.

o with a double lumen

The Physiology

Gas Exchange in W-ECMO In VV-ECMO, some of the systemic ve­ nous return is drained into the EeMO system, oxygenated, and returned to the right atrium. Some of the systemic venous return goes di­ rectly to the right atrium where it mixes with the ECMO perfusate blood. The mixed blood passes through the right ventricle, native lungs, left heart and into the systemic circulation. In severe respiratory failure, the native lungs contribute little or none to gas exchange, so the arterial oxygen and cO2 levels are the result of mixing the oxygenated EeMO blood with the deoxygenated native venous blood. As a result, the arterial saturation ranges from ~O% to 90%, depending on the relative amount of EeMO flow, native venous flow, lung function, and cardiac output. The desaturated arterial blood results in normal systemic oxygen delivery as long as the cardiac output and hemoglobin concentration (oxygen content) are adequate. These relationships are often confusing to leu staff, because the usual goal of management is to keep the arterial saturation over 90%.

r

Extracorporeal Life Support

. . I I outIet millUS ill et o,i! content. Because the outlet blood is typically 00% saturated and PO2 is over 500 mmHg t?e dissolved oxygen can be as much as 10% of the oxygen content. Blood flow is !limited by the resistance to flow in the drain±e cannula, the suction pro­ duced by the pump or siphon, and the geometry ofthe cannulated lessel (usually the vena cava or right atrium). Typical maximum flow at 100 cm/lIp suction common sizes of venous cannulas is 4-5 lit s per minute.

Ii

til

Relation of Extr lForporeal Oxygenation to Systemic Oxygen Delivery Assuming that lhere is no native lung func­ tion, the systemi~ larterial content, saturation, and P0 2 will result from mixing the flow of oxygenated blood ~om the membrane lung with the flow of venous blood which passes into the right ventricle, not Tto the EeMO drainage can­ nula, (hereafter ref~?"ed to as the native venous flow). The amount 9foxygen in systemic arterial blood is the result ff the mixture of these two flows (Figure 4-15).

Oxygenation To illustrate the principles, the discussion begins with the assumption that there is no na­ tive lung gas exchange (which is often the case in EeMO patients). In a membrane lung (as in the native lung) oxygenation is a much greater problem than CO2 removal, so the initial focus is on oxygenation. The circuit and blood flow are planned for total oxygen supply (VO) at rest or during moderate exercise. For adults this is 120 cc/minlm2 (3 cclkglmin), or~3~0 cc of oxygen/min for average adult~. The membrane lung must be large enough to transfer this amount ofoxygen. AU devices (;urrently on the market can achieve this (see "rated flow"). The oxygen supply from the membrane lung is dependent on the blood flow, the hemoglobin concentration, and the difference between the

Figure 4-15. Mix: g of perfusate with native venous flow during W-ECMO. I

41

Chapter 4

Calculations Related to Mixing Two Flows

SystemicArterial

tOl' Saturation, and Con­

tent During VV He 0

When two blood flows of different oxygen contents mix, the resultant oxygen content is the average of the amount of oxygen in each of the two flows (not the average of the partial pressures). The amount of oxygen contributed by each flow is the oxygen content fln cc/dL) in the blood times the flow rate (in dUrniN. The equation summarizing these events is in Figure 4-16. The same calculation can be done ----~--~----------using saturatiop rather than oxygen content. This calculation using saturation is done for simplicity and is not an exact representation of the amount of oxygen content but is useful at the bedside (Figure 4-17). The combinations of flow and oxygen (expressed as saturation) variations are shown in Figure 4-18. Of the variables in the equation, all are known except the flow of venous blood which does not go through the extracorporeal circuit (the native venous flow). The natjve venous flow can be back calculated from the systemic arterial oxygen content or saturation. The total venous return (cardiac output) is the sum of the native venous and circuit flow.

Use of these e uations in VV patient physiology are sh0:T in Figure 4-18. In these examples, one variaple is changed while oth­ ers are held constant fO illustrate the principles. Clinically, all thes~ variables may change simultaneously and t different rates. For sim­ plicity of the examp es, we assume no native lung function, and a : proximate the points on the graphs. We do ot account for dissolved oxygen in calculatio of 02 content, although it can be significant hen the P02 is over 300. Example 1: Typ cal VV physiology. Sup­ pose the extracorpore I flow for an adult with no lung function is 4 V~'n and the systemic P02 is 50 mmHg, saturation 88%, 02 content 12.3 ccl dl. The Hb is 10.5 dl and the venous blood saturation is 64%. . e patient's oxygen con­ sumption is 200 cC/mjn. The oxygen content of blood leaving the me brane lung is determined primarily by the con entration of hemoglobin. At hemoglobin conce!tration of 10.5 gmJdl and 64% saturation, the d ainage (inlet) 0 2 content is 9 ccldl and the outl t content at 100% saturaI

Mixing 2 Blood FIOWjin VV ECMO: Example 1 Mixing l Blood Flows of Different Ol Content (C1 and Cl)

The resultant 02 content (C3) is the sum of each content xflow divided by total flOW C1 X Flow1 Total Flow

C,

X

Flow,

+ ------Total Flow

=C3

In VV EeMO, assuming no native lung function :

Using Content: 9 · Fz --- + = 12.3 ccldL 4 + F2 Fl

14 x 4

4(14-9) = 6 L/min (12.3-9)

Native Venous F o

w 6 - 4 2 Llmin

Cardiac Output

=

C1 and F2 are extracorporeal content and flow

C2 and F2 are the venous content and the native venous flow The equation can be solved for cardiac output and native venous flow

C,) =Cardiac Output C3 - C, Flow, = Total Flow - FloWl = Native Venous Flow Total Flow

°

= Flow1

(C I -

Figure 4-16. The relationship among flow and content are shown in the equation. During ECMO all the variables are known except na­ tive venous flow (F2) and total flow (cardiac output). Native venous flow can be calculated. Cardiac output is F2 plus Fl. This asswnes no Jiiiig furictlOn and no recirculation.

42

=

Using O2 Saturation:

100 x4

4x F

4 + F2

+ Fl

______ + Cardiac Output

1

= 90%

4(100-64} (90-64)

= 6 L/min

Native Venous Fow = 6 - 4 = 2 L/min

Figure 4-17. The r ationships among flow and 02 content using the data in Example 1. The same calculation are shown using satura­ tion rather than conte ' t.

if Extracorporeal Life Support

The Physiology

tion is 14 cc/dl. The amount ofoxygen supplied to the patient is the outlet minus inlet content (which is 5 cc/dl), times the flow (40 dVmin) equals 200 cc oxygen supplied per minute. The native venous flow is calculated at 2 Llmin (per equation in Figure 4-17) so the cardiac output is 6 Llmin (native plus circuit venous flow). D02 is the arterial content (12.3 cc/dl) x 60 dl/min = 738 cc/Olmin. D0 2N0 2 ratio is 3.64. The 02 content of native venous blood is the same as the drainage content (9 cc/dl). The final com­ plete equation is 40 dllmin x 14 c/dl divided by 60 dl/min, plus 20 dllmin x 9 dl divided by 60 dVmin=12.3 cc/Oldl (corresponding to a P02 of approximately 50 mmfHg). The calculation using saturation is 4 L/min x 100% + 2 Llmin x 70% -:- 6 Llmin which yields a systemic arterial saturation of 88% (point A, Figur{: 4-18)

is lower. The syst mic oxygen delivery is 920

DOiY?2

is 4.6. There has b~n cc/min. The a gain in s stemic 0 en deli e because of the hi er cardiac u ut despite a decrease. in arterial saturation nd content. If the patient's systemic oxygen ~onsumption is 200 cc/min, systemic oxygen d livery is perfectly adequate and full aerobic m tabolism is supported, even though the arterial P0 2 is 45 mmHg and arte­ rial saturation is 8f' No changes are required but the ICU staff eed to understand thaLthe hypoxemia does n1t require interven,tiQn. Un­ derstanding this co ce t can be difficult when the Ian is to kee the arterial saturation over .2QYo. (point B, Fi re 4-18) Example 3: 1nemia. The patient in Ex­ ample 1 is moderatEly anemic (Hb 10, 5 gm%) but stable. Supposf the hemoglobin suddenly Example 2: Increased cardiac output at drops to 8 gm%. Tfe venous drainage is fixed fixed ECMO flow. If, in the same patient, the at 4 Umin by th~ Iresistance of the drainage cardiac output (venous return) increases to 8 U cannula, and carditc output is 6 Umin. The min and the circuit flow is fixed at 4 L/min there outlet content at 10P% sat is 10.7. The amount will more native venous return at 64% sat mix­ of oxygen sUPPlie~ by the membrane lung is ing with the fully saturated ECMO flow. The 10.7 minus 9 whic is 1.7 cc/dl, so the mem­ systemic arterial content will decrease to 11.5 brane lung is supp :ying only 68 cc/min. The and the saturation will decrease to 84% corre­ native venous flow is 20 dl/min and content is sponding to P02 of45 mmHg. The total amount 9 cc/dl. The arterial content has gone from 11.5 ofoxygen going to the patient is the same (200 to 9.8, the arterial sat to 80% and the D02 has cc/min), but the systemic saturation and P02 gone from 738 cc/rrlin to 588. This results in a DO~02 ratio of 2\9 (assuming no difference in metabolic rate). rowever, since only 68 cc of oxygen is being ~ded per minute, and the Mixing Two Flows with Different 502 oxygen consumptibn is 200 cc/min, venous (inlet) content and 1aturation decrease quickly. When the inlet con ent falls to 5.7 (saturation Systemic r-.. . ~ -?'*"''::'O'''';:::f==''"t 1m Venous 50%) the membrane lung 0-1 difference is 5 ccl Arterial ~"""'-..¥",o...L.+-~~,.,;"--=t 70 Saturation Saturation dl and the oxygen sypplied is 200 cc/min. The mixture of the full~ saturated blood at 40 dl/ min and the 50% saturated native venous flow results in arterial sarlrration of75% and arterial content 9 cc/dl. The\ systemic oxygen delivery Arterial saturation and venou,s saturation at is 540 cc/min and th~ VO/D0 2ratio is 2.7. The ratios of ECMO flow to native v enous flow that ECMO flow Is 100% saWrtlted) patient can remain ih steady state with arterial saturation 75% and I enous saturation 50%, but Figure 4-18. Mixing of flows in VV-ECMO. Data from Examples 1-4 are shown. any further decrease in hemoglobin or increase I

(Assvm\"~-s

43

Chapter 4

in metabolic rate will result in supply depen­ dency and lactic acidosis (point C, Figure 4-18). Example 4: Increased metabolic rate. Suppose the patient in Example 1 becomes hypermetabolic (V0 2=250 cc/min). The size of the venous cannula determines that the circuit flow is at maximum at 4 Llmin so the circuit oxygen delivery is limited to 200 cc/min. The cardiac output is 6 Lim in. Going through the same arithmetic, the patient will fall behind at the rate of 50 cc of oxygen per minute and venous content and saturation will steadily de­ crease (70% to 45%, for example). As venous saturation and content decrease, the oxygenator will still increase the outlet saturation to 100% and oxygen content to 14 cc/dl, so the circuit outlet minus inlet oxygen difference (oxygen supply) will go up as the venous saturation goes down (from 5 to 6 for example). Systemic saturation will decrease because the saturation and content of the native venous blood going through the heart and lungs will decrease. At venous content 7.5 the 0-1 content difference is 6.5 and the oxygen supplied is 260 cc/min. Steady state is reached with arterial saturation at about 75% and Pa0 2 35 mrnHg. The D0 2N0 2 is 2.1 and any increase in activity will lead to anaerobic metabolism which will produce lac­ tate rather than CO 2 , and lactic acidosis results. In time, this will lead to multiple organ failure and death (point D, Figure 4-18). How can systemic oxygen delivery be in­ creased in Examples 3 and 4? Turning up the ventilator Fi02 or airway pressure will not help. Furthermore, the whole objective is to avoid in­ creasing Fi02 and pressure from the mechanical ventilator. There are four alternatives. The first is to increase the hemoglobin concenJration to normal. When hemoglobin goes from 10.5 to 15, systemic oxygen delivery goes to 930 ccl min, arterial saturation returns to 95%, venous saturation goes to 80%, and the patient is stable and well supported. The D02N0 2 is 4.6. The second alternative is to increase the suction or add another cannula and increase the circui.!

flow from 4 to 5 LI . (maintaining Hb 10.5).

The D02 increases to 792 cc/min, and the DO/ V0 2 is 3.9. The third temative for example 4 is to aralyze and cool t e patient, decreasing the V0 2 back to 200 cc/m' . A fourth consideration is to add another m9mbrane lung to increase the gas exchange surface, but 02 supply is still limited by the blood row, so this will not help. The tradeoff between extracorporeal flow and hemoglobin is defonstrated in Figure 4-19. This example shows . 80 kg man with oxygen requirement of240 CCl min' but the relationships remain the same for any size patient. Under most circumstances, the risks of increasing extraco oreal flow re e te than the 18 of transfusiQ,n.

Oxygenation in VV­ CMO The combination~ venous access cannula, membrane lung size, and hemoglobin concen­ tration should be pi ed to match or exceed resting V0 2 (120 ccl 2/min for adults). The membrane lung will upply the most oxygen at a normal hemoglo in (15 gm/dl). All the important variables r lated to blood flow and is essential to know : e patient's oxygen con-

Hb and f

ow in EelS

Typica l 80 kg ad ~ lt: total 0 2 support

240 c/min

Hb gmid"l-

--_.

--

EC FlowL/min 5.3

50

15

4.6 4.3

--=~~'==F=---- 4.0

Ri sk of transfusion:m inimal

3.6 3.3 3.0 2.6 Ri sk of high flow: high

Figure 4-19. The adeoff between hemo­ globin and flow. Th I relationships to deliver 240 cc 0 2 per minute rre shown, but the same calculations can be do e for any oxygen supply.

44

- - -- - - - - - - - - - - - - - - - - - ---- - - - - - - -------+------------------­

The Physiology

sumption and systemic oxygen delivery to de­ cide the best way to manage the extracorporeal circuit. Hypoxemia (pa02 40-60, Sa02 70-90) always occurs with venovenous support and is adequate to maintain normal oxygen delivery. If systemic oxygen delivery falls to a critical level (near twice consumption), circuit oxygen supply must be increased by: I) transfusing to a higher hemoglobin or; 2) adding additional venous drainage access to increase the flOF. There is a'tradeoff of risk between transfusion and in­ creasing circuit flow which favors transfusion ofRBCs. Membrane lungs function optimally at a normal hematocrit.

Recirculation during VV-ECMO Since VV-ECMO reinfuses blood drained from the vena cavae or right atrium back into the central venous system, the potential exists for some of the returned blood to be drained by the circuit before it mixes with the native venous flow. Recirculation reduces the effec­ tive extracorporeal flow by the recirculated amount. For example, if total circuit flow is 5 Umin and recirculation is 20% (1 Umin), then only 4 Umin of oxygenated blood is available to the patient. Determinants of recirculation include the type and placement of cannulas, and the extra­ corporeal flow in relation to cardiac output and vena caval native blood flow. Two-site single lumen cannulation is always associated with some recirculation, since the vena caval flow is less than the circuit drainage, necessitating blood from the atrium containing oxygenated blood from the circuit being drained as well. Three-site single lumen cannulation, in which both vena cavae are drained, reduces recircula­ tion. Similarly, the dual lumen bicaval cannula provides drainage from both vena cavae, and is associated with low recirculation. Recirculation fraction is also related to overall circuit flow relative to cardiac output. As cardiac output decreases, the fraction io­

if Extracorporeal Life Support

creases, and beco es markedly elevated when cardiac output dro s below extracorporeal flow. Increases in recirc lation during extracorporeal support can result from dislodgement of can­ nulas, such as clo er approximation of single­ lumen cannulas fro inadvertent advancement, or accidental with awal or advancement of a dual lumen cannul in which the distal drainage port moves into th right atrium, or the reinfu­ sion port advance into the inferior vena cava, respectively.

Hef1Wtiynamics d ' ing W-ECMO Unlike' venoa erial support, VV-ECMO does not provide a y direct hemodynamic sup­ port since there is po reinfusion of blood into the arterial system. yenous return and pulsatility are unaffected since drainage and reinfusion are in equilibrium. -I Despite the lac~ ofdirect support, however, myocardial functi~n and cardiac output typi­ cally improve during VV support. Reduction in mechanical ventila~oty sUPPQrt that is allowed from improved ga exchange reduces ri.&.ht ventricular afterlo d and improves RV func­ tion. Improved 0 genation of the pulmonary vascular bed can itigate h oxic vasocon­ s:triction. Higher 0 gen saturation of left ven­ tricular blood resul in, improved myocardiaL o genation and rei ersal of h oxia-induced m ocardial depress on. Reduction in acidosis from correction of hypercarbia and rever§l of anaerobic meta,olism also contribute to improved myocardifl function. This indirect support is usually sufficient to reduce or elimin~te requirements for cardio­ vascular pharmacolo ic agents. Improved organ perfusion can help everse renal, hepatic, and other organ failure.

COzRemoval CO2 production s equal to 02 consumption (when the Respirat ry Quotient is 1), so the 45

Chapter 4

amount of CO2 exchanged per minute is essen­ tially the same as the amount of oxygen (120 cc/minlm z for adults). Because CO 2 is much more soluble and diffusible in blood than 0Z' CO 2 clearance will exceed oxygenation in any circumstance, so all the circuit management is based on oxygenation. If CO 2 clearance is the only or the major goal, much lower blood flow can be used and hemoglobin concentration is not important. The amount ofCOz elimination is a function of the membrane lung surface area and the gradient between the inlet PCO z (typically 50 rrunHg) and the sweep gas (0). The systemic PCOz is the result of mixing circuit outlet blood (pC0 2 typically 30 mmHg) with native flow (typically 45 mmHg). Like oxygen­ ation, the actual amount ofCO2 removed by the membrane lung is the inlet CO2 content minus the outlet content times the flow. However, CO2 content is difficult to measure or calculate, so actual CO 2 removal is measured as the % CO2 in the exhaust gas times the gas flow. Unlike oxygenation, measuring or calculating the ac­ tual amount of CO2 exchanged by the circuit is not critical; the sweep gas is simply adjusted to maintain the desired systemic PC02 (typically

of gas exchange in t e placenta and fetus. Be­ cause of the blood ow requirements for gas exchange support, th · arteriovenous route is not a reasonable approa h to total extracorporeal respiratory support, e cept when the patient can tolerate a large arteri1venous shunt and increase in cardiac output (s~ch as a premature infant). However, AV flow ough a membrane lung can provide significa t CO2 removal, decreasing the need for mechan cal ventilation.

ECMO Manageme t when the Native Lung is Recovering

All the precedin discussion describes a situation when there s no native lung function. As the native lung be ins to recover, some oxy­ gen and CO2 exchan e will occur. The effect will be to improve sys emic arterial oxygenation and PaC02 with no ch ge in the extracorporeal flow rate and hemo lobin. It is tempting to increase ventilator sefings and FiOz in order to take advantage ofthis ecovery, but this may add to lung injury and del y lung recovery. ECMO is continued during r9st ventilator settings, and when arterial PC0 2 drops below 40 the sweep gas to the mem brane I ng can be proportionally 40mmHg). decreased. When th systemic arterial satura­ One phenomenon unique to ECMO is the tion exceeds 95%, th extracorporeal flow can effect of water accumulation on the gas side of the membrane lung. This is the only cir­ be gradually decreas d (changing the ratio of circuit to native ven us flow). When the ex­ cumstance in which CO 2 clearance is less than oxygenation. Understanding the reason is a tracorporeal support as decreased from total good exercise in understanding how membrane support to approXimaf,IY 50%, extracorporeal support can be briefly , .scontinued (at moderate lungs work. ventilator settings) to test native lung function. When native lung fun~tion is sufficient for total Arteriovenous CO2 Removal patient support, EC~O can be discontinued. Arteriovenous (AV) extracorporeal circula­ Because reestablis~.ng vascular access in tion is commonly used for hemodialysis but not ECMO can be diffic It, it is wise to continue for cardiac or pulmonary support. The AV route ECMO support for a day or two beyond this can be used for gas exchange provided the arte­ point to allow more lupg recovery, unless there rial blood is de saturated , and the cardiovascular is a pressing reason td take the patient ECMO system can tolerate the arteriovenous fistula (systemic bleeding or IC NS complications). with a large enough flow to achieve adequate gas exchange. This is, after all, the mechanism

46

The Physiology if Extracorporeai Life Support

Managing VV-ECMO Based on These

Principles In VV access, the parameters described in Figures 4-13-15 are monitored and V02, D0 2 are calculated from these measurements. That information is used to adjust the ECMO vari­ ables and the patient variables to maintain DOi V0 2 at 3: 1 or higher. 1. As in VA access, plan the circuit based on the best estimation of the metabolic rate (adults, 3-4 cc/kg/min for both 02 and CO) and the drainage flow which can be achieved from the largest drainage cannula (or cannulas) which can be placed. Plan for total support, realizing that there may be some native lung function and total support may not be necessary. For a septic 80kg adult you will need 5 Llmin flow, and an oxygenator with rated flow over 5 Llmin to supply 300 cc O/min. 2. On ECMO go to the highest flow to deter­ mine the maximum drainage capacity, then tum down the ventilator to rest settings (Fi0 2 0.3, CPAP 10-15 em HP) and wean off the vasoactive drugs. The hypermetabo­ lism will decrease to baseline. The lungs may go to total consolidation. Adjust the sweep gas to keep the PaC02 40 mm Hg. 3. When the patient is stable (usually 6-12 hours) determine the variables of 02 kinet­ ics, using the formulas described above. If oxygen supply is adequate (D02:V02 over 3) no changes are necessary. If oxygen sup­ ply is inadequate (D02:V0 2 under 3) and the patient is anemic, transfuse to a higher hemoglobin (12-14 gmldl). If 002 is still inadequate, change the drainage cannula to a larger size and increase flow. 4. Manage the patient based on continuous venous and arterial saturation monitoring. Plot the position on Figure 4-4 frequently. Calculate the variables if oxygen supply seems excessive or inadequate.

5. When the nati I lung begins to recover (the arterial saturaJ'.on is >95%) turn down the flow, keeping he venous saturation >70%, and the sweep flow, keeping the PaC02 at 40. When nati Ie lung function is adequate, trial OffECM] then decannulate. Summary

Managing a p tient on ECMO requires a thorough under tanding of normal and ab­ normal cardiopul onary physiology, and a thorough unders ding of the ECMO circuit. Based on this unde tanding, the ECMO system is used to replace art or all of heart and lung function, maintain· g normal systemic physiol­ ogy while the dam ged organs can recover or be replaced.

47

5 The Circuit John M Toomasian, MS, CCp, Leen Vercaemst, RN, ECCp, Stephen ottrell, CCp, Stephen B. Horton, PhD, CCp, FACBS

Introduction

The extracorporeal life support (ECLS) circuit consists ofthe set ofartificial organs and equipment that provides mechanical support of the failing heart and/or lungs for days, weeks, or months. A proper circuit design provides a foot­ print to meet the expected metabolic demands of the patient including adequate oxygen (02) delivery and carbon dioxide (CO) removal. The circuit may also provide access for additional therapeutic modalities such as hemofiltration, continuous renal replacement therapy, and car­ diovascular intervention. The circuit requires management by trained specialists who are readily available to respond and troubleshoot any (;ircuit related irregularity. This chapter describes the fundamental components of an ECLS circuit, based on the ELSO Guidelines that served as an update from the material from Chapter 8 of the 4th edition of the Redbook.' Historical Background

EeLS involves several mechanical plat­ forms developed to support the circulation, allowing time for the affected organ system to recover. The traditional circuit was adapted from devices and techniques used during stan­ dard cardiopulmonary bypass (CPB). Unlike traditional CPB that used a direct contact gas-

blood contact gas echange device with an open venous cardiotomylservOir for return of shed blood from the oper tive field, the ECLS circuit required a closed c cuit with a gas exchange device that had no irect air-blood contact. A "true" membrane- pe gas exchange device • I would functton for ys, weeks, or months. Several membr e lungs were used during the early clinical his ory ofECLS including the Lande-Edwards and ramson membrane lungs. 2, 3 However, the gas xchanger that allowed for routine use was th spiral wound reinforced silicone rubber me brane lung developed by Theodor Kolobow " the 1960s.4 This device had a long blood pa which resulted in a high resistance to blood £I. , w, often generating a 200­ 300 mmHg device p essure gradient. Priming was complex and rrUired a carbon dioxide £lush and the use of vacuum. The membrane was made from silic ne rubber, a highly gas permeable barrier. : as transfer occurred by simple molecular d' sion. The device was manufactured in seve I sizes based on effective surface area by SciM d, Avecor, and Medtronic. Some devices had an tegrated heat exchanger. The device was used :or 50 years, until produc­ tion was discontinue in 2012. Despite its few shortcomings, the Kol bow membrane lung had an outstanding and reltable performance record that was the gold stan!:iard for gas exchangers.

49

Chapter 5 ----------------------------------------~-----------------

The earliest circuits used roller pumps, as centrifugal pumps were only design concepts. The classic roller pump was described by many but was often credited to 5 Roller pumps used for ECLS were produr.ed in many and manufactured by various compa­ nies including: American Optical, Travenol, Polystan, Stockert, Cobe, Rhone Poulenc, and Sarns. Roller pumps were driven by either belt or direct and presented a unique set of v'-l(;tH~'HI'.,V", especially related to management of the circuit pressure. A catastrophic blowout occurred if the positive pressure (outlet) side was or occluded. If pump inlet pressure became (negative), air could enter. Regulating the blood inflow to circuit to any servo control mechanisms, the venous return could be gauged visually and pump adjusted manu­ ally. This was and "nT,,,,>{',1"1 cal for use outside the operating room. When the roller pump flow the venous return, the blood of the roller pump would easily air into the circuit. To minimize this risk, a small com­ pliant silicone rubber chamber was inserted into the circuit proximal to the pump. Ifthe chamber was blood was not impaired and pump would operate at its set speed. However, when pump flow exceeded drainage (ie, kinked venous cannula malposition, or hypovolemia), the chamber would empty and partially collapse, tripping a micro-switch that stopped or slowed the roller pump.6 As expanded and grew in the 1980s, this was copicd, and throughout tbe world. This system has been the historic template for circuit configurations that have used since the mid 2000s and has been eloquently described in earlier editions tbe Redbook.1·9 In the United States, all medical devices are "cleared" for use by the U.S. Food and Drug Administration (FDA) indications. lbe majority of components are manufactured 50

for short-term intrar"''''''!l1"1 (not components are to exceed 6 hours). any in an "off-Iabe " application for extended use at discretio~ of pbysician, in a manner that m+, be outside indication for use. Experience witb off-label use has been applied to select p groups for extended periods of time. Go ernmental restrictions for by ountry. For instance, the Mark ay bave different labelled clearance n the used in the U.S. The des 'ptions in this chapter do not any spec manufacturer's recom­ mendations.

General Principles circuit is made from biomaterials OJ""""'_" common y used in traditional CPB. The patient and tbe c rcuit are typically antico­ agulated to prevent thtombosis. Each circuit can vary in SOPhisticationi'nd complexity; however, new level of omplexity knowledge to trouble oot and problems that may result. The CLS circuit may contain a number of access ites that can measure or monitor specific patrnt parameters such as Pvt..!lf'A....'Ar'·'> blood owes), circuit pressures, oxygen saturation(s, hemoglobin or inUne blood gases, and oth r metabolic parameters. of this des go exist and depend on patient physiolog c needs, and institutional philosopby. When des'gning the a would be to or additional blood components in I e prime solution. that connects t~e patient to the essential circuit components shbuld not be but adequate in length to allow for patient move­ ment Re to flow increases

The Circuit

I

connectors, and access sites risk more circuit Blood Tubing PI ticizers related complications. Each connector causes The majority of conduit tubing is made turbulent blood flow, potential blood element damage, and creates an area of stagnation that from a formulatio~'ofpolyvinylchloride (PVC) Cat1\ induce clot formation. The "ring thrombus" mixed with a plas ·cizer. The amount ofplasti­ is commonly observed at a tubing-connector cizer determines e tubing durometer or flex­ junction. Luer lock connections have a potential ibility. Known as Pi-2(ethyl hexyl) phthalate to entrain air or leak blood. The access port of (DEHP), DEHP is fsed to make PVC flexible. any L.2.er connector should be positioned point­ However, in recenl:,ears, concerns have been it.!K upwards to avoid thrombus formatiQn that expressed regard· , the release of DEHP into often occurs if the connector is positioned at a the circulation. e FDA has previously is­ sued a public noti cation regarding exposure lower place, allowing blood cells to collect and to DEHP.14 The lipi content, temperature, and clot. Double luer connectors increase access and can be considered when multiple access sites the duration of exp sure all affect the leaching of DEHP. Adverse effects have been reported are required. Luer connectors can be used at the patient "bridge" site, eliminating the need for in laboratory an· ' Is, with the greatest con­ V-connectors used with a bridge or "diamond" cern being its effec on the male reproductive configurations. This also eliminates stagnant system. 15 The FDA oncluded that health risks areas that are subject to potential thrombosis. from DEHP expos re exist and non-DEHP With simple design objectives, gas ex­ containing substitu products should be used change devices have been developed that in male neonates, p gnant women, and peripu­ maximize exterior hollow fiber technology with bertal boys when av ilable. Leaching ofDEHP does occur in ECM circuits, although coated modern compressed surface polymers such as polymethylpentene (PMP) or polyolefin (PO). circuits decrease or eliminate DEHP leaching These materials exchange gas as efficiently as from pre-primed cir , uits. 16 The European ommission on Food and their predecessors and can be designed to have less surface area resulting in a lower device Health also publish d opinions on DEHP in pressure gradient. The new generation centrifu­ 2002, 2008, and 20f4. Opinions in 2008 and gal pumps are much more efficient than their 2014 stated "there iF no conclusive evidence predecessors and when used in conjunction that DEHP exposur~ia medical treatments has harmful effects in hans." However, "the new with the new lower resistance gas exchange information indicate that there is still a reason devices provide support without the problems ofhemolysis and mechanical failure commonly for some concern f300 mmHg can c use damaging turbulence and shear force whid will result in hemolysis. 63

Commercial Centri ugal Pumps There are a numb r of commercial centrifu­ gal pumps that have been used for extended ECLS whose charac eristics are summarized in Figure 5-6. Indiv dual device availability and regulatory status varies from country to country and by certif ing agency. Indications for use may also vat) by country. The regula­ tory status ofeach dev ce described is not listed due to the wide varia ion and changing status of each device.

Maquet (HLSf .CarJiiohel.D 5-0

Maquet (HI.,S) --Cardiohelo 7.(}

32

240

273

57

40

3/8"

3/8"

3/8"

3/8"

Maximum flow (LPM)

10

Outlet port size (")

3/8" 'MGdtroriic •.

Sorin • Revolution

Medtronic

Afflnitv

Rolaflow

10

Medlronic

BP,50 .

Sl Jude Medical' ,St judeM~dIe I. pediVAS. . Centrlmaa

MOOos

. OP3.

Static volume (mL)

.BPX·80

86

48

31

14

Maximum flow (LPM)

10

1.S

10

1.5

8

318"

1/4"

3/8"

1/4"

318"

Outlet port size (")

17

Figure 5-6. Characterization and performance parameters of ( ommercial centrifugal blood pumps.

60

The Circuit

Maquet RotaJlow

by a I nun steel t (Figure 5-3). The blood gap between the· peller and housing is conical on either side. Th outflow exits into a circular recuperator and a spiral volume chamber. All blood contacting ~arts are made of acrylic and polycarbonate reslps. I

The Rotaflow centrifugal pump (Maquet Cardiopulmonary AG, Hirrlingen, Germany) (Figure 5-7) was introduced in 1995. This pump was able to reduce heat generation from the bearing and seal and improved hydraulic effi­ ciency with minimal blood damage. In testing, the Rotaflow pump showed strong hydraulic efficiency compared to other devices. 45 The disposable pumphead features a low friction one-point pivot bearing that supports a multi­ finned impeller. The pumphead is controlled by a remote drive motor tethered to a control console. An ECLS lCU console incorporates additional safety features that minimize inad­ vertent manipulation of the speed control or power switch. The unit is manually controlled and not servo regulated. The pumphead is com­ pact with a priming volume of 32 mL and has an outer diameter of 85.5 nun, a height of 48 ffill1, and a weight of 60 gm. It is constructed to exploit the potential of the radial magnetic drive in eliminating a central shaft and seal us­ ing a blood flushed bearing, avoiding stagnant zones, and reducing areas of high shear and turbulence. The drive magnets are embedded in a shrouded impeller with four blood channels. A single pivot bearing supports the impeller at its bottom, in which the sapphire ball bearing is held in the center ofthe completely open rotor

Figure 5-7. Rotafiow centrifugal pumphead (Maquet Cardiopulmonary, Hirrlingen, Ger­ many).

Livanova Revolu

n Centrifugal Pump

The Revoluti~centrifugal pump (Liva­ nova, Arvada, C I) features a sealless, low friction bearing (Fi e 5-8). The pumphead is controlled by the ~.vanova Centrifugal Pump (SCP) System that can be integrated with the safety devices of t e S5™ and C5™ systems. When certain con itions are detected (low level, bubble, or re~ograde flow), this feature allows for some sertvo-reguIatory controls that are unavailable W~ih other centrifugal pump technologies. The volution 5 version that has been qualified for days of use is available in some countries.

Medtronic BPX-80 and BP-50 The Medtronic PX pump design (Medtron­ ic, Minneapolis, features a patented verti­ cal outlet and a smo th vortex cone design that provides superiormi o air retention. This pump has been used widel during clinical open heart surgery, but is associ ted with higher hemolysis

Figure 5-8. Revolu ·bn centrifugal pumphead (Livanova, Arvada, 0, USA).

61

Chapter 5

with extended use. 41 The current BPX80 has 3/8"

CentriMag lY1£.r1!n,ellC Levitated Pump

inlet and size outlet connectors and has a prim­ ing volume of 80 ml with a recommended blood flow up to 8 LPM. The BP50 has 1/4" inlet and outlet connectors and has a priming volume of 48 mL with a recommended blood flow up to 1.5 LPM. Both pumphead versions are available with Carmeda heparin coating and may be used with any Biomedicus pump console.

Medtronic Affinity Centrifugal Pump The Medtromc Affinity centrifugal pump (Medtronic, Minneapolis, MN) has a smooth cone a.nd low profile fins enabling blood flow to 10 LPM with a ceramic heat resistant pivot shaft. The pumphead has a 40 mL priming volume and interfaces with the Medtronic Biomedicus console. The external motor driver rotates in the opposite direction than the traditional Bio­ medicus pump head. The Affinity pump head is offered with two surface coatings: Carmeda heparin coating and the nonheparin Balance Biosurface (UK) (Figure 5-9).

Figure 5-9. Medtronic Affinity centrifugal pumphead (Medtronic, Minneapolis, MN).

62

technology (St. Jude The CentriMag ) is based on the prin­ Medical, St. Paul, ciples of magnetic itatioll. The CentriMag which contains features bearingless is levitated within the no seals. The pump disposable Uluvw",a", housing by a magnetic field so that when motor is powered up, the aligns and maintains which no direct contact pumphead housing oc­ is engaged the magnetic the rotor to spin at console. The risk

The Circuit Medos Delta Stream The Medos Deltastream DP3 pump (Me­ dos Medizintechnik AG, Stolberg, Germany) (Figure 5-11) is a hybrid of centrifugal and axial technologies that utilizes higher rpm but shorter transit time to generate blood flow. The DP3 has 3/8" inlet and outlet connectors for adUllt use (a design with 114" size connectors for pediatric use is pending). The pumphead prim­ ing volume is 16 mL and generates a flow range of 0-8 LPM at a pump speed from 100-10000 rpm. The pumphead incorporates a ceramic bearing and magnetic coupling with an optional pulsatility mode (rate 40-90 beats/min). The MDC console is portable, lightweight (10 kg), and contains two batteries (90-minute power/ battery) that allow the pump to be operated independently from the driving console. The detachable touch screen monitor can display op­ erational parameters such as flow, rpm, pressure, and temperature. The controlling hardware has three integrated pressure channels with nega­ tive pressure regulation, zero flow mode, and air bubble detection, as well as a display ofthe charging condition of both batteries. The zero

flow mode contr Is the pump speed to avoid inadvertent retro de flow. The system cannot be manually hand cranked.

Rated Flow The gas excha ge device is designed to add oxygen and remov CO2 • The device must be of adequate s~ to pr vide for the total anticipated metabolic require ents of the patient. The gas exchange device hould be able to transfer the amount of oxy , en being consumed by the patient, as well as the amount of CO 2 being produced by the p tient. The surface area and blood path mixin determine the maximum oxygenation capa ity of any gas exchanger. A standard referre to as "rated flow" allows for direct comparis ns of all gas exchange de­ vices. The rated fl w is the blood flow rate at

exchange device wi h a saturation of95% 64

Oxygen Delivery Patient metab lic needs determine the amount of oxygen d livery required. The maxi­ mal oxygen deliveJ of a gas exchange device is the amount of 0X)('?en delivered per minute when running at rater flow. This is determined · in oxygen content by calculating the d*erence between the outlet b :ood and the inlet blood to the gas exchanger. I · s difference is ically 4-5 cc Q/~. Multi lying this by blood flow rate through the gaS~XChanger yields the oxy­ gen delivered by the ' exchanger in ccO/min. A gas exchange devi e with a rated flow of2.0 Llmin will offer am imal oxygen delivery of 100 cc O/min. A devi e with a rated flow of4.0 Llmin will offer a m~al oxygen delivery of 200 cc O/min. Based fn the anticipated oxygen delivery reqUiremen~. " the appropriate size gas exchange device can be chosen. I

Figure 5-11. Medos Deltastream pump sys­ tem «Medos Medizintechnik AG, Stolberg,

Germany).

.

63

Chapter 5

Sweep Gas

The gas ventilated through the gas exchange device is referred to as sweep gas. For most applications, sweep will be 100% oxygen. There are occasions during cardiac support or by institutional preference when the sweep will oxygen and compressed

sweep gas. some a gas flow rate equal to the blood flow rate (1: 1) is used to begin support. gas to blood flow ratio is then adjusted to maintain the systemic a desired range (ie, 36-44 sweep flow rate CO2 clearance but does not affect oxygenation. Decreasing the sweep flow rate the pC02 , but usually does not oxygenation. Water va­ por can within the gas compartment the membrane lung and may be cleared by intermittently increasing sweep flow to a higher rate. :.:...::::::.::~~~~~i:.t.W=::U:~d;" bolic support. most gas exchange devices, the amount of oxygen added and removed is the same at a gas to blood flow ratio of 1: 1. gas ciearC02 more efficiently than adding oxygen, CO2 can be re­ moved with blood flows as low as 0.75 Llminl m2 • In cases when CO2 regulation is required, the gas device size can be than the device required for full support. Ventilation to maximize CO2 removal is typically than 4: 1 on the to blood flow ratio rate. It must be noted that some manufacturers limit the their recommended gas to blood flow instructions for use.

64

Pressure Gradienti Early generati~n gas exchange devices were characterized Iby a long blood path and a high pressure graclient These early designs included flat sheet ilicone rubber membranes and the "blood in hollow fiber bundle configurations. D sign modifications lead to a hollow fiber oncept with and blood g the configuration resul cd in significantly lower pressure gradients i the blood path.65 This de­ been applie to all subsequent designs regardless 0 hollow fiber material. The first widely used bio aterial was microporous polypropylene. Alth~ugh this material could sheet, fiber technol­ be manufactured as ogy more ffective surface area and was relatively inex ensive to manufacture. Engineering designsl to secondary flows and in~'e blood path, gas exchange, and I wering the device pres­ sure gradient have' proved performance and allowed for smaller .ized devices. The lower resistance made thes1newer generation gas ex­ change more ractical for long duration use. Microporous po ypropylene hollow fiber gas exchange devicesl are limited in longevity because over time, plasma will leak across the fiber eventually caus~g device failure. 66•61 Gas transfer normally occ,rs across a protein layer ~cropores. Over time, ex­ covering posure to phospholipi s in the blood causes •...,"""j'H at the lood/gas in the micropores to decreaS~AS the surface tension decreases, plasma be's to weep through the ,,,,.,,,..,,.,."',, into the g s side of the membrane. As a result, gas exchan e efficiency falls. 68 This membrane failure has limited the widespread for use. use of polypropylene (PMP) fibers that 2000s showed promise as a Polymethylpentene polypropylene fibers

The Circuit

I is compressed it displays a more leak resistant property than the microporous polypropylene. Although PMP has been described as more ofa "sonid" membrane, in fact PMP is microporous. During manufacture, the fibers are compressed to form an outer surface or "skin" with proper­ ties similar to that of a solid membrane. Even though the PMP membranes are "solid-like," gas can still be entrained across the PMP ma­ terial into the blood path if a large negative pressure is applied. Therefore, it is important to maintain positive pressure on the blood path side:. The PMP fibers, produced by Membrana GmbH (Wuppertal, Germany), have similar gas exchange characteristics as polypropylene but it is more challenging to handle, so PMP is typically matted and either wound or stacked in relatively short blood path configurations. Current PMP hollow fiber devices have low pressure gradients and the fibers mimic a "solid hollow-fiber" technology with minimal plasma leakage. 69 Polymethylpentene devices were first described in two pilot clinical studies. The device was reported as showing feasibility and longevity in long-term ECLS support. The Me­ dos I-liLite LT was used in six patients in whose transfusion rates were significantly lower and plasma leakage was not observed. 12 Similarly, no plasma leakage occurred in 23 patients with MaqUlet diffusion membrane device that was used for up to 46 days in conjunction with a centrifugal pump.69 However, plasma leakage has been observed in clinical application. The initial experience with PMP devices demon­ strated promise in durability, inflammatory response attenuation, and decreased transfu­ sion requirements making these gas excbange devices well suited for long-term ECLS use. Since their introduction, other PMP devices of varying sizes have entered the marketplace.

Gas Exchanger Induced Air Embolism Gas bubbles can pass across the gas ex­ change device membrane into the blood path if

.

the swee as ressure exceed th pressure. This m occur if the device blood path pressure be omes negative, which can occur with an u egulated centrifugal pump. Air can also be p lled across the membrane surface when bloo~ flow ceases and the de .ce is positioned abo e the level of the atient's eart. s CITC stance, ood within the gas exchange device c n drain by gravity into the dependent blood bing, subsequently pulling air across the fibers nd "de-priming" the device. To minimize the ris ofair embolization through the gas eXChange~IOOd path, pressures must be maintained hig er than the gas side pres­ sure.70 Pressure po , off valves may be placed in the gas supply . e to insure the gas pressure remains low. Swee gas pressures may be servo regulated and the s exchan e device should be ke t below the I, atient's heart to minimize air embolism risk. As newer generation gas exchange devices d centrifugal pumps have allowed the transp ' rt of patients within and between hospitals, e gas exchange device may inadvertently be re ositioned above the level of the heart while enous drainage pressures are subatmospheric If blood flow were stops by either a line k· ,patient cough, or clot formation within ga exchange device, risk of air embolism exists. I

Commercial Gas E Figure 5-12 sho s the utilization of mem­ brane lung types fro 2005-2015, based on data submitted to the Ex corporeal Life Support Organization Regis . The data incorporates all patient groups d is expressed as a per­ centage of the total umber of cases reported per year. The trend sows the growth in PMP gas exchanger use d the gradual phase out of the silicone rubbe units worldwide. Many commercial membr ne lungs are available for clinical use and s me are described in the sections below. The r!egulatory status for each individual device vari~s from country to country I

I

65

Chapter 5

and by certifying agency. In the U.S., most gas exchange devices are cleared for up to 6 hours of use. In Europe, CE marking may be extended to days or weeks. The regulatory status of each device is not listed, due to that wide variation and changing status of each device. There are several commercial gas exchange devices used for extended EeLS whose technical character­ istics are summarized in Figure 5-13.

contains an integrated prevents and protects against cO[ltarninatIC)ll from the gas side. The pressure drop across designs for bundling the ox~{ge:na{lJI and the flow pattern key factors in decreasing llvIIUv'''~U effectively

Maquet Quadrox- iD

Medos Hilite LT

Ibe Maquet Quadrox- ill diffusion mem­ brane device (Maquet Cardiopulmonary AG, Hirrlingen, Germany) comes in adult and in­ fant/pediatric sizes (Figure 5-14). The device is constructed of plasma resistant hydrophobic

The Medos Hil LT (Medos Medizintech·· nile AG, Stolberg, is produced in three sizes with an .... t,Prrt'·.,t",rt heat exchanger. The device is by a low pressure inlet and outlet and

Gas Exchanger Use: 2005· 20t5

c=JOlher --- Adult 100

c=:J Cardiohelp

-Neonatal

~PMP

eo:::::::Jl POII"orclOvtl!n@ -

- ~ - Pediatric

•••• Total

Silicone

90

800!)

7000

80

0000

70

500'0 4000 3000

;:0 200f) 20

10

Figure 5-12. Data from the Extracorporeal Life Support Organization Ch{"\,U/'TIO gas exchange device use from 2005-2015. The number ofcases is reported bYIDe(matal, adult subgroups and the total number of cases. Gas exchanger use is based total number reported per year. The trend shows a transition from silicone polymethylpentene devices in all subgroups.

66

1000

The Circuit

."'''''

POP

Uaximom I... (ll'WJ

18

Pt1me,'101ulI\I(ml)

PIlI'

PMP

PIlI'

SUrfac4arulm')

"

0.'

'" 1.1

'" OJ

'" 1.1

0.32

~m,."kI'

PO

PI!

PU

PU

POIjHlir

HE swtact WN (m'l

0.15

0.'

al 318"

u

0.01"

1M'

ltti-1/("

rMt oudtt pod. (I~I

,''

O.S

318"

"

1

PIIP l.IulfMlm noi'l' (LPU)

I .'

Prilntvolutr»(ml) Sulfac. &l'&a (m l)

" us

HE-mafmal

.

1.0

40

m

...

12' 1.81

.,.""

D."

P.ofyesttr

PoIy..ttr

Stalnlm .toel

SUInltS'SstHl

SIaIltMa.lIOI

HE !uriKIlMfllmI)

6.O7(

0.500/. 80% (>0.5-0.8 uimL), consider ATre placement if on rnaximwn dose of ill iFH and unable to obtain thera cutic anti-factor Xa assay

Anticoagulati n and Disorders ofHemostasis

and underlying pathophysiology should dictate adjustments of laboratory determinations and blood product administration. Laboratory test­ ing can be decreased and blood product trans­ fusion reduced in patients who have reached a stable clinical status. Blood product transfusion carries with it infectious and non-infectious risks and will be covered in Chapter 8

Conclusions

The number and complexity of patients supported with CLS continues to increase. With this growth comes an even greater need for improved surface for blood-circuit interaction, safer anticoagula on, and enhanced anticoagu­ lation monitoring Consultation with hematol­ ogy specialists an laboratory medicine should Hemorrhagic and Thrombotic Complica­ be considered fa managing complex issues regarding anticoa . ulation for ECLS or for any tions ECLS patient wi unexplained hemorrhagic or Hemorrhagic complications produce sig­ thrombotic compl cations. nificant morbidity and mortality among ECLS patients. Their prevention requires correcting coagulopathy and thrombocytopenia, careful titration of anticoagulation therapy, and repair of surgical sources of bleeding. The ELSO Registry defines a significant clotting complication as requiring a change in a portion of the circuit or entire circuit. In the July 2015 ELSO International Registry Report across all age groups and indications, circuit clots have been reported in nearly 40% of all ECLS runs, with the oxygenator being the site with the greatest number of reported clots. 52 Inadequate anticoagulation, low flow states, and patient hypercoagulability lead to increased risk of thrombosis. Areas of stasis or turbulence in the ECLS circuit are prone to clot formation. Even when thrombotic complications are not evident during an ECLS run, thrombosis is commonly observed during postmortem ex­ aminations of ECLS patients. A single center report ofpostmortem examinations ofpediatric ECLS patients revealed significant thrombosis in 69% of patients overall and 85% of patients with a history of congenital heart disease. 53 In a report of autopsy findings of postcardiotomy adult ECLS patients, over 70% had previously unrecognized thromboembolic complications.54

99

Chapter 7

References

thromboproph laxis: a meta-analysis.

Blood. 2005;10 (8):2710-2715. 1. Verstraete M. Heparin and thrombosis: a seventy year long story. Haemostasis. 1990;20 Suppl 1:4-11. 2. Verstraete M. Pharmacotherapeutic aspects ofunfractionated and low molecular weight heparins. Drugs. 1990;40(4):498-530. 3. Sinauridze EI, Panteleev MA, Ataul­ lakbanov FI. Anticoagulant therapy: basic principles, classic approaches and recent developments. Blood Coagul Fibrinolysis. 20 12;23(6):482-493. 4. Newall F, Johnston L, Ignjatovic V, Monagle P. Unfractionated heparin therapy in infants and children. Pediatrics. 2009; 123(3):e5 I 0­ 518. 5. Pratt CW, Church FC. Antithrombin: structure and function. Semin Hematol. 1991 ;28(1 ):3-9. 6. Finley A, Greenberg C. Review article: heparin sensitivity and resistance: man­ agement during cardiopulmonary bypass. Anesth Analg. 2013; 116(6): 121 0-1222. 7. Sandset PM, Bendz B, Hansen JB. Physi­ ological function of tissue factor pathway inhibitor and interaction with heparins. Haemostasis. 2000;30 Suppl 2:48-56. 8. Warkentin TE, Greinacher A. Heparin­ induced thrombocytopenia: recognition, treatment, and prevention: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest. 2004;126(3 Suppl):311 S-337S. 9. Warkentin TE. Heparin-induced thrombo­ cytopenia: pathogenesis and management. Br J Haematol. 2003;121(4):535-555. 10. Jang IK, Hursting MJ. When heparins promote thrombosis: review of heparin­ induced thrombocytopenia. Circulation. 2005; 111(20):2671-2683. 11. Martel N, Lee J, Wells PS. Risk for heparin­ induced thrombocytopenia with unfraction­ ated and low-molecular-weight heparin

100

12. Warkentin TE, S eppard JA, Horsewood P, Simpson PJ, Mo re JC, Kelton JG. Impact of the patient p pulation on the risk for heparin-induced hrombocytopenia. Blood. 2000;96( 5): 170 -1708. 13. AvilaML, Shah ,BrandaoLR. Systematic review on hepar' -induced thrombocytope­ nia in children: call to action. J Thromb Haemost.2013; 1(4):660-669. 14. Obeng EA, Ha ey KM, Moniz T, Arnold A, Neufeld EJ, renor CC, 3rd. Pediatric heparin-induced thrombocytopenia: preva·· lence, thromboti risk, and application of the 4Ts scoring system. The Journal of pediatrics. 2015;166(1):144-150. l5. Warkentin TE. eparin-induced throm-· bocytopenia. urr Opin Crit Care. 2015;21(6):576- 85. 16. Hirsh J, O'Donn II M, Weitz n. New anti­ coagulants. BIO~O .2005;105(2):453-463. 17. GladwellID. Bi alirudin: a direct thrombin inhibitor. Clin er. 2002;24(1 ):3 8-5 8. 18. Jyoti A, Mahesh ari A, Daniel E, Motihar A, Bhathiwal , Sharma D. Bivalirudin in venovenous tracorporeal membrane oxygenation. J Extra Corpor Techno!. 2014;46(1):94-9 . 19. Nagle EL, Dage WE, Duby JJ, et al. Bi­ valirudin in pedia .c patients maintained on extracorporeal Ii e support. Pediatric criti­ cal care medic in : a journal of the Society of Critical Care Medicine and the World Federation of Pe iatric Intensive and Criti­ cal Care Societi s.2013;14(4):eI82-188. 20. Pieri M, Agrach!a N, Bonaveglio E, et al. Bivalirudin ver us heparin as an antico­ agulant during (l tracorporeal membrane oxygenation: a dase-control study. J Car­ diothorac Vasc4esth. 2013;27(1):30-34. 2l. Ranucci M, Bal~otta A, Kandil H, et al. Bivalirudin-bas d versus conventional heparin anticoa lation for postcardiotomy I

Anticoagu/atio and Disorders ofHemostasis

extracorporeal membrane oxygenation. Critical care. 2011;15(6):R275. 22. Koster A, Weng Y, Bottcher W, Gromann T, Kuppe H, Hetzer R. Successful use of bivalirudin as anticoagulant for ECMO in a patient with acute HIT. The Annals of thoracic surgery. 2007;83(5): 1865-1867. 23. Ranucci M. Bivalirudin and post-cardioto­ my ECMO: a word ofcaution. Critical care. 2012;16(3):427. 24. Phillips MR Khoury AI, Ashton RF, Cairns BA, Charles AG. The dosing and moni­ toring of argatroban for heparin-induced thrombocytopenia during extracorporeal membrane oxygenation: a word of cau­ tion. Anaesthesia and intensive care. 2014;42(1):97-98. 25. Dolch ME, Frey L, Hatz R, et al. Extracor­ poreal membrane oxygenation bridging to lung transplant complicated by heparin­ induced thrombocytopenia. Exp Clin Trans­ plant. 2010;8(4):329-332. 26. Beiderlinden M, Treschan T, Gorlinger K, Peters J. Argatroban in extracorporeal membrane oxygenation. Artificial organs. 2007;31(6):461-465. 27. Scott LK, Grier LR, Conrad SA. Heparin­ induced thrombocytopenia in a pediatric patient receiving extracorporeal membrane oxygenation managed with argatroban. Pe­ diatric critical care medicine: a journal of the Society of Critical Care Medicine and the World Federation ofPediatric Intensive and Critical Care Societies. 2006;7(5):473­ 475. 28. Young G, Yonekawa KE, Nakagawa P, Nugent DJ. Argatroban as an alterna­ tive to heparin in extracorporeal mem­ brane oxygenation circuits. Perfusion. 2004; 19(5):283-288. 29. Mejak B, Giacomuzzi C, Heller E, et al. Argatroban usage for anticoagulation for ECMO on a post-cardiac patient with heparin-induced thrombocytopenia. JExtra Corp or Technol. 2004;36(2): 178-181.

30. Johnston N, ait M, Huber L. Argatro- .

ban in adult extracorporeal membrane oxygenation. J Extra Corpor Technol. 2002;34(4):28 -284. 31. Bembea MM, Annlch G, Rycus P, Olden­ burg G, Berk witz I, Pronovost P. Vari­ ability in antrcoagulation management of patients 0 extracorporeal membrane oxygenation: international survey. Pedi­ atric critical c e medicine: a journal ofthe Society ofCri 'ca1 Care Medicine and the World Federati n ofPediatric Intensive and Critical Care Spcieties. 2013; 14(2):e77-84. 32. Uden DL, Pay e NR, Kriesmer P, Cipolle RJ. Procedura: variables which affect acti­ vated clotting . e test results during extra­ corporeal me brane oxygenation therapy. Critical care edicine. 1989; 17(10): 1048­ 1051. 33. SeayRE, UdenDL, KriesmerPJ,PayneNR Predictive pe10rmance of three methods of activated clrtting time measurement in neon~tal E~Mp patients. ASAIO journal. 1993,39(1).39- 2. 34. Bosch YP, G I ushchak YM, de long DS. Comparison 0 CT point-of-care measure­ ments: repeata ility and agreement. Perfu­ sion.2006;21( ):27-31. 35. Maul TM, Wo EL, Kuch BA, Rosendorff A, MoreII VO, Fearden PD. Activated par­ tial thrombopl~stin time is a better trending tool in pediatri extracorporeal membrane oxygenation. ediatric critical care medi­ cine : a jouma of the Society of Critical Care Medicine and the World Federation of Pediatric I tensive and Critical Care Societies. 2012 13(6):e363-371. 36. Chan AK, Bla' k L, Ing C, Brandao LR, Williams S. Ut~lity of aPTT in monitoring unfractionated ~eparin in children. Throm­ bosis researc::h008;122(1):135-136. edge J, Newall F, et al. 37. Ignjatovic V, F Age-related d' frences in heparin response. Thrombosis res arch. 2006;118(6):741-745.

101

Chapter 7

38. Khaja WA, Bilen 0, Lukner RB, Edwards R, Teruya J. Evaluation of heparin assay for coagulation management in newborns undergoing BCMO. American journal of clinical pathology. 20 10; 134( 6):950-954. 39. Liveris A, Bello RA, Friedmann P, et a\. Anti-factor Xa assay is a superior corre­ late of heparin dose than activated partial thromboplastin time or activated clotting time in pediatric extracorporeal membrane oxygenation*. Pediatric critical care medi­ cine : a journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.2014;15(2):e72-79. 40. Bembea MM, Schwartz 1M, Shah N, et al. Anticoagulation monitoring during pediat­ ric extracorporeal membrane oxygenation. ASAIO journal. 2013;59(1):63-68. 41. Teruya J. Coagulation Tests Affected by Acute Phase Reactants Such as CRP and Factor VIII. Paper presented at: Interna­ tional Conference on Hematology and Blood Disorders; September 23 - 25, 2013; Research Triangle Park, NC USA. 42. lrby K, Swearingen C, Byrnes J, Bryant J, Prodhan P, Fiser R. Unfractionated heparin activity measured by anti-factor Xa levels is associated with the need for extracorporeal membrane oxygenation circuit/membrane oxygenator change: a retrospective pediat­ ric study. Pediatric critical care medicine : a journal of the Society of Critical Care Medicine and the World Federation of Pe­ diatric Intensive and Critical Care Societies. 2014;15(4):e 175-182. 43. Northrop MS, Sidonio RF, Phillips SE, et al. The Use ofan Extracorporeal Membrane Oxygenation Anticoagulation Laboratory Protocol Is Associated With Decreased Blood Product Use, Decreased Hemor­ rhagic Complications, and Increased Circuit Life. Pediatric critical care medicine : a journal ofthe Society ofCritical Care Medi­

102

cine and the W31d Federation of Pediatric Intensive and Cr tical Care Societies. 2014. 44. Kostousov V, uyen K, Hundalani SG, Teruya J. The i uence of free hemoglobin and bilirubin 0 heparin monitoring by activated partial thromboplastin time and anti-Xa assay. rchives of pathology & laboratory medi ine. 2014;138(11):1503­ 1506. 45. Alexander DC, I utt WW, Best JD, Do­ nath SM, Mon gle PT, Shekerdemian LS. Correlation of thromboelastography with standard te ts of anticoagulation in paediatric patie ts receiving extracorpo­ real life suppo '. Thrombosis research. 2010;125(5):387- 92. 46. Andrew M, pae~B' Johnston M. Devel­ opment of the h mostatic system in the neonate and you g infant. The American journal of pedia!1.c hematology/oncology. 1990;12(1):95-10 . 47. Wong TE, Huan YS, Weiser J, Brogan TV, Shah SS, Wi er CM. Antithrombin concentrate use' children: a multicenter cohort study. Th Journal of pediatrics. 2013;163(5):13291334.e1321. 48. Niebler RA, Ch 'stensen M, Berens R, Wellner H, Mik ailov T, Tweddell IS. Antithrombin repl cement during extracor­ poreal membrane oxygenation. Artificial organs. 2011;35(1 ):1024-1028. 49. Byrnes JW, Swe . gen CJ, Prodhan P, Fiser R, Dyamenahalli . Antithrombin Tn sup­ plementation on e tracorporeal membrane oxygenation: imp ct on heparin dose and circuit life. ASAI journal. 2014;60(1):57­ 62. 50. Ryerson LM, BrucF AK, Lequier L, Kuhle S, Massicotte MPl Bauman ME. Admin­ istration of AntitIiombin Concentrate in Infants and Childrrn on ECLS Improves Anticoagulation EtcaCy. ASAIO journal. 2014. 51. Mintz PD, Blatt P ,Kuhns WJ, Roberts HR. Antithrombin in fresh frozen plasma,

Anticoagulati n and Disorders ofHemostasis

cryoprecipitate, and cryoprecipitate-deplet­ ed plasma. Transfusion. 1979; 19(5):597­ 598. 52. Extracorporeal Life Support Organization. Registry Report. Ann Arbor: University of Michigan; July 2015. 53. Reed RC, Rutledge JC. Laboratory and clinical predictors of thrombosis and hemorrhage in 29 pediatric extracorporeal membrane oxygenation nonsurvivors. Pe­ diatr Dev Pathol. 2010;13(5):385-392. 54. Rastan AJ, Lachmann N, Walther T, et al. Autopsy findings in patients on post­ cardiotomy extracorporeal membrane oxygenation (ECMO). lot J Artif Organs. 2006;29(12): 1121-1131.

103

8 Transfusion Management during Extracorporeal Support Anne Marie Winkler, MD

Blood Product Transfusion during Extracorporeal Support Blood transfusion during extracorporeal support typically occurs for several reasons, including circuit priming, restoration ofoxygen carrying capacity, maintenance ofa hemostatic balance, and treatment of hemorrhagic com­ plications. The variability of blood products included in the circuit prime was captured in a recent international cross-sectional survey with 119 of 121 ELSO-reporting centers responding. l Ofthose centers, 92% ofcircuit primes included packed red blood cells (RBC), 50% fresh frozen plasma (FFP), one center whole blood, and an­ other center including platelets, in addition to other solutions. As a result, the majority of pa­ tients are exposed to at least one blood product during extracorporeal support. Additionally, 91 of 117 (78%) respondents from the same survey had an institutional protocol for blood product management with transfusion thresholds differ­ ing for pediatric only programs as compared to adult only or mixed adult/pediatric programs. Of the 81 pediatric only program respondents, the median hematocrit threshold reported for RBC transfusion for a typical ECMO patient was 35% (range 25-40%) compared to 30% (range 20-40%) for adult and mixed adult/ pediatric programs (p5.5 x 10 10 platelets while 90% of apheresis platelets must contain >3.0 x their use in critically ill atients including those 1011 platelets. As a result, 4-6 units of whole receiving ECLS. blood derived platelets must be pooled to make 110

Transfusion Management uring Extracorporeal Support

The most common indications for plasma tions have local idelines based upon expert transfusion include reversal of coagulopathy opinion and a sys ematic review found insuf­ in bleeding patients or for prophylactic use in ficient evidence to make recommendations for patients undergoing procedures; however, a or against platelet transfusion in critically ill paucity ofwell-designed studies define plasma preterm infants, c ·ldren, and adults. 54 transfusion practice.47 Despite the lack ofdata, While many c~ters target the transfusion plasma transfusion is a common procedure oc­ thresholds describ d above, there is a lack of curring in approximately lO% of critically ill evidence to guide ansfusion thresholds in the setting ofextracorp real support and additional patients. A recent multicenter study identified studies are needed 0 establish evidence based that the majority of adult plasma transfusions occur in patients who have a mildly elevated thresholds for tra sfusion support and diag­ international normalized ratio (INR) with 22.5% nostics to guide tr nsfusion therapy to assess given for an INR of less than 1.6 and 33% for efficacy of transfus on in this population. an INR of 1.6-2.0. 48 In addition, patients who received plasma transfusion received low doses Use of Coagulati n Factor Replacement, with 58% not reaching a posttransfusion INR Blood Derivativ s, and Antifibrinolytic of less than 1.6, which has been confirmed in Therapy during E tracorporeal Supp0l1 other adult and pediatric studies. 49-51 In a similar study performed in pediatric patients across 101 Recombinant Fact r VIla pediatric lCUs in 21 countries, 34% of patients who received plasma were neither bleeding nor Recombinant fftor VIla (NovoSeven®RT, scheduled for a procedure; only 22% received Novo Nordisk, Bagsrrerd, Denmark) is a lyophi­ plasma for critical bleeding.52 Moreover, de­ lized powder ofreco binant human coagulation creases in INR or activated partial thromboplas­ factor VIla (rFVIla which, when complexed tin time (APTT) were seen for values greater with tissue factor, ca activate factor X to factor than 2.0 or 60 seconds, respectively. As a result, Xa, as well as factor I to factor IXa. The tenase plasma transfusions do not correct a laboratory complex then conver prothrombin to thrombin, coagulopathy in critically ill adults and children. which leads to the ,ormation of a hemostatic However, it should be mentioned that coagula­ plug by converSi0f.*ffibrinogen to fibrin and subsequent cross r . g by factor XIII. Throm­ tion tests may not be the best surrogate as they bin generation may Iso occur on the surface of fail to predict bleeding, but alternative tests with correlations with clinical outcomes remain . activated platelets e ancing platelet adhesion unavailable. A single study demonstrated that and aggregation. InifiallY approved in the U.S. by the FDA, rFVIIa is currently indicated for a fixed dose of FFP decreased syndecan-l and factor VIII suggesting that plasma may have an the treatment of bl~ding episodes and periop-· endothelial stabilizing effect in 33 critically ill erative management adults and children with adult patients.53 While promising, it is unknown hemophilia A or B ~ith inhibitors, congenital what other biologic effects plasma transfusion factor vn deficiency, and for patients refractory to platelet transfusio with Glanzmann's throm­ may have especially in the setting ofECLS. basthenia with or wi out antiplatelet antibodies, Thrombocytopenia occurs commonly in critically ill patients and is present in 8.3-67.6% in addition to treaa nt of bleeding episodes ofadult lCU patients, and 20-50% ofneonates. 54 and perioperative m nagement in adults with As a result, patients commonly receive platelet acquired hemophili . However, a retrospec­ transfusions, which remain the primary treat­ tive audit of the Pre~ier Perspectives database ment despite a lack of evidence. Most institu­ from 615 nonfederal .S. hospitals from 2000

r

111

Chapter 8

through 2008 demonstrated a l43-fold increase in use over the study period. 55 Specifically, off­ label use of rFVlla represented 96% of cases with adult cardiac surgery (12,086 cases) and trauma (J 1,689 cases) representing the most rapidly emerging indications despite the limited data to support its use. Pediatric cardiac surgery accounted for an estimated 1684 cases of the 73,737 total number of cases from 2004-2008. Off-label use of rFVI1a in pediatric patients has also been examined. A single center in the U.S. reported decreased blood product admin­ istration 24 hours after rFVIIa administration in 135 patients who received 997 doses of rFVI1a for a variety of indications, of which 3 patients also received ECMO. 56 In the second study from a 75 hospital registry in Australia, 388 patients less than 16 years of age received rFVlla mostly for active bleeding (97%); the remaining 11 patients received rFVI1a for sur­ gical prophylaxis. 57 Of note, 26 patients were receiving EeMO at the time of rFVI1a admin­ istration. From this cohort, it was reported that there was a statistically significant reduction in the volume ofRBC, FFP, platelets, and cryopre­ cipitate transfused 24 hours after administration ofrFVlla, consistent with previous reports from observational studies. However, 21 patients (5.4%) experienced a thromboembolic adverse event with a higher rate reported in those who were receiving ECMO support (19% vs. 4%, p=0.009). While use of rFVlIa during EeLS was described in case reports and case series, no randomized controlled trial has yet been re­ ported in this population. Despite the limited data, 63/94 (67%) ELSO centers reported use of rFVIIa when asked if £-aminocaproic acid, rFVIla, tranexamic acid, aprotinin, or other products where used to manage anticoagulation, hemorrhage, or thrombosis in ECMO patients.) Similar to previous studies, case series of off­ label use of rFVIIa during ECLS demonstrated reduced blood loss, most commonly measured by chest tube output, and reduced transfusion 112

volumes for a de ed period of time postad­ ministration. Table -3 presents a summary of published case seri .58-66 In clinical trials, thrombotic adverse events occurred in 0.2% of atients treated with rFVIIa with hemophilia A or Band 4% of patients with acquired hem philia; the risk of throm­ boembolic complic ions in patients receiving rFVIIa for off-Iabe uses remains unknown. Currently, NovoSev n®RT carries a warning of possible increase risk of development of thromboembolic eve ts due to circulating tissue factor or predisposin' coagulopathy in patients with disseminated i travascular coagulation, advanced atheroscle otic disease, crush iI1jury, septicemia, or conco itant treatment with ac­ tivated or nonactiva d prothrombin complex concentrates, and in ncontrolled postpartum hemorrhage. In a r view of the FDA's Ad­ verse Event Report ng System, there were 168 reports of 185 t omboembolic events, of which 151 occurred n patients who received rFVIIa for unlabeled ·ndications. 67 Moreover, thromboembolic com lication was the probable cause of death in 720 of cases. In addition, a metaanalysis of35 ran omized controlled trials including 2815 subje ts who received rFVIIa for off-label indicatio s demonstrated a 10.2% thromboembolic even rate, compared to 8.7% of subjects who rec ived placebo (p=0.16); however, the rate of rterial thromboembolic events was higher (5.5 0) when compared to the placebo group (3.2%, =0.003) and this effect was more pronounced in subjects 65 years of age and 01der.68 The hromboembolic risk of rFVIIa administration as not been established in the setting of ECL and case reports of fa­ tal thrombosis after a ministration of rFVIIa have been published. 6 -72 As a result, careful consideration must be iven to administration of rFVIIa in EeLS pa ients as they have mul­ tiple risk factors for thr mbosis_ In addition, it is unknown whether e disturbed hemostatic balance during ECLS ay enhance thrombo­

Transfusion Management iuring Extracorporeai Support

Table 8-3. Case series reporting pediatric and adult patients who receive( rFVIIa for refractory bleed­

ing while receiving EeLS. Median

Study

II!J

age (year.)

PeJjatric Patients Wittenstein ct at

4

Median

Median

rFVIIa DoSt:

Number ofDose5

(I'!lik2)

Received

0.625

2

Effect Od Blood Prdduct Requirements or Blcedine

• Blood loss was reduced by 83%

Adverse I tient Even

Adverse Circnit Events

Survived to Discharge

Non

None

4

I thromboel bolie

2 circuit

9

evont

Lhromboses

None

2 oxygenator milures, 2 circuit thromboses

4

4 thromboen otic events"

2 oxygenator milures, 2 circuit obstructions

5t

None

None

NR

1 thromboemlX>lic

1 oxygenator failure

3

None

2

2 circuits

6

(p=(J.025) 6 hour. after I" rFVUa

infusion • Transfusion requirements decreased by 81 .9% for RBe (JRl.045), 90"10 for FFP (JRl.062), 92% cryoprecipitate (JRl.OO5), and 79.7% platelets (JRl.024) 6 hours after 1st

rPVIla infusion Agarwal et a I

12

9.5

36.5

1.6 (1-4)

• Significant reduction (59.4%) in :1"I" until tion may be rapid. Once tricuspid regurgitation switch operation. has developed, the volume loading of the or can cause Awaiting Cardiac Transplantation accentuated any further annular dilatation, creating more ECMO has been used to improve pubno­ regurgitation and systemic lar subsequent heart The in those patients waiting transplantationS Chapter 59) but development RV failure without unknown if it improves overall previous surgical intervention is almost always it associated with dysplasia of the Right ven­ outcomes. 9 tricular dilatation, association with volume Indications for ECMO Support in ACHD loading from regurgitation, the septum morphologic LV, which leads to displacement com­ ECMO can provide and 1)Ollents of the Tv, accentuates support as a bridge to recovery or morphologic regurgitation. In to transplantation. ECMO can failure following closure ofVSD cardiac arrest Severe or relief ofPS, decrease the morphologic heart failure unresponsive to maximal medi­ septum cal therapy or the use of an intraaortic baUoon LV pressure and realigmnent of toward the pulmonary ventricle can have the pump (lABP) represent the indications to our same effect creating tricuspid regurgitation. consider ECMO. The following is thought to the the inclusion and RV elopment oftricuspid failure, which is supported the observation • Mean arterial blood pressure < 60 ,UJJJC>,J.,J.i!;;" that banding to train the morphologic LV • Elevated left pressure (>20 mmHg), • Low cardiac index « 2 Uminlm 2 ), tricuspid regurgitation by is the septum.5 • Rising with ofend organ "VIJLA"'L,UU'""

U-' ...'VLU"'­

• likeli­ or qualification term mechanical support or ","",'va"',UIV

Role ofIschemia The RV as a ~v~:l~l1n1C VP.,'t1"l,('1 at systemic pressures with resultant hy­ pertrophy of the myocardium may not receive sufficient coronary blood flow. This results in ischemia RV which,

patients, ECMO should considered with failure wean from car­ diopulmonary despite high inotropic support with or without placement. nn,,1't''',rt11,r.1'nrrn,

333

Chapter 28

Contraindications to ECMO Support inACHD Among patients with multisystem organ failure ECMO is not associated with good outcome, most likely due to the severe damage that prior to initiation. it is ECMO prior to lrrf~V~lr 7 days. systemic bleeding -Malignancy with fatal continuation of with active treatment prognosis within 5 years anticoagulation) -Patient moribund -Age -SAPS II > 90 -Futility: patients who are too sick -Non drug-induced coma (e.g., following cardiac arrest -Irreversible neurological immunosuppression) , have been on pathology -ECMO cannulation not conventional therapy possible too long (MV >7 days) BM!-body mass index; Ces20% of the Uai'vlJW'-' value despite treatment for at least 2 hours with noninvasive ventilation, associated with respiratory rate breaths/min and use res:pu"at'" Blood flow is limited by

venous return, drainage cannula size, and posi­ tion which optimizing by augmenting output, cannula size, and position. " " ' ,..... v ..., the fraction should be ,",UlUU''''

,;apacity to oxygenate blood subphysiologic oxygen saturations measured in range of 85-90%) are expected. An attempt to increase blood flow to improve this situation may recirculation, hemolysis, and fluid problems, overload, add to venous ""'£11""'" additional cannulae. Importantly!!.

oxygen as Air-oxygen mixture should saturate blood to 100% well-functioning oxygenators 1-h",..",1-,r...'" an increase oxygen concentra­ to an oxy­ 'safety toxicity is insufficient blood

465

Chapter 42

blood as bicarbonate. 4 For that reason, less than 500 ml ofblood contains all CO2 produc~d per minute ~d very little blood flow is needed to clear that amount. The gas flow should be adjusted to achieve a normal pH rather than a normal pCO? since many patients may have a chronic respiratory acidosis with metabolic compensation. A slight increase in gas flow beyond this may aid a patient suffering from tachypnea and respiratory distress.

Initiation ofVV-ECLS After cannulation, a significant amount of blood volume will be drained from the patient toward the extracorporeal circuit and will be replaced by the priming fluid of the circuit. This most often consists of a crystalloid solu­ tion, which will lead to hemodilution of the blood volume. Often patients experience he­ modynamic compromise as a result of severe hypoxia and consequently may suffer from brief but significant hypotension. It is impor­ tant to insure that hemodynamic support with vasopressors or a fluid bolus can be delivered. Preexisting intravascular hypovolemia may be exacerbated by ECLS and optimal pump function may be compromised by poor venous drainage. If the drainage cannula is confirmed to be in an optimal position further fluid boluses may be needed. Once instituted, ECLS support delivers ox­ ygenated blood to the patient and often resulting in improved hemodynamic function. Changes in ventilatory settings allowing lung protective or even ultra-lung protective ventilation result in reduction in intrathoracic pressure with im­ proved preload and RV afterload. Unloading of the RV by a combination of reduced ventilator pressure and reduced pulmonary resistance secondary to the correction of hypoxia and hy­ percarbia may play the main role in this observa­ tion. ECLS also very effectively corrects severe respiratory acidosis. Rapid reversal or over correction of respiratory acidosis may be detri­ 466

mental in some settings such as spontaneously breathing patients or suspected brain injury. A stepwise increase in gas flow in association with a reduction in minute ventilation may be necessary. Since bolus heparin doses are o~n aEministe~d during cannulation the bleeding risk may be increased at this point.

Maintaining VV-ECLS The most common problem during ECLS is poor venous drainage. Venous drainage de­ pends on the size and position of the cannula, venous filling (depending on venous return and preload), and ECLS pump speed. Any of these parameters can cause suboptimal drain­ age. If we ensure that the VV-ECLS cannula is optimally positioned, adding an additional drainage cannula may be needed to address poor venous drainage but this extra cannula often results onJy in moderate increases in blood flow (see Chapter 38). Increasing ECLS pump seeed will increase blood flow tmt may increase recirculation and hemolysis, although this is less likely with optimal venous drainage. 5 Ofcourse, this strategy relies on adequate venous return. Once this critical volume is reached the drained vein may collapse leading to a sudden drop of blood flow to zero. The acute hypoxia will be substantial and only a reduction in pump speed will free up the drainage cannula. Figure 42-1 illustrates the importance of not increasing ECLS pump flow beyond a sustainable drainage (area A). Plateauing of the relationship (area B) may not be observed since venous filling is a dynamic process and can change quickly, for example secondary to increased intrathoracic pressures (eg, coughing). Increasing venous filling by administration offluid boluses is a common reflex when blood flow is deemed to be inappropriate. However, -~ this benefit is usually temporary and ultimately can lead to fluid overload. A restrictive approach to fluid management in ARDS improves survival and therefore the

Weaning and Decannulation ofAdults with Respiratory Failure on EeLS

same strategy needs to be applied during ECLS. Trying to maintain high levels of blood flow by fluid administration ultimately reduces the chance of organ recovery. Reducing the blood flow targets or improving venous drainage by additional cannulation are preferred options. ECLS circuit failure occurs rarely in adults but early signs need to be addressed to avoid a catastrophic event. Modem oxygenator and pumps can last for weeks or in some cases months without exchange and signs of failure often develop slowly so that elective changes can be planned for (see Chapter 5). A sudden loss of function should be an uncommon prob­ lem in modem adult ECLS practice. Risk factors for component failure include prothrombotic conditions including low or no anticoagulation, trauma, or underlying thrombophilic disease. Routine care includes regular monitoring and documentation of clot formation and of changes in biochemical mark­ ers of hemolysis and thrombosis such as free hemoglobin, d-dimers, fibrinogen, and platelets. Thrombin fragment and thrombin-anti-thrombin complexes may increase while factor XIII may drop, but these are not part of standard monitor­ ing in most centers.6 Failing oxygenators may gradually loose function noted by a decrease in

' - - -'•.

I ,

B .

C .i

ECl..5

Pump-Sp~



Figure 42-1. Function of blood flow to pump speed. Area A - linear increase of blood flow with pump speed. Area B -Ineffective increase in pump speed secondary to poor venous filling. Area C - acute drop in blood flow secondary to venous collapse. Not to scale.

post oxygenator Pa02 or need for increases in sweep gas flows. There is no single marker or score for device failure and it remains a matter of clinical judgement to determine if an ex­ change is needed, keeping in mind that patieots who are completely dependent on ECLS should undergo an elective exchan,ge rather than an urgent procedure which usually results in severe respiratory and/or hemodynamic instability.

Recirculation Recirculation of already oxygenated blood into the drainage cannula complicates VV­ iECLS. Again, higher ECLS pump speed and blood flow may increase this problem. As a con­ sequence, oxygen delivery does not increase in proportion to flow while complications includ­ fig hemolysis and poor drainage occur more commonly. No simple bedside test measures the recirculation fraction nor can it be completely avoided during VV-ECLS. Optimal placement of drainage and return cannula, in a two-site VV-ECLS configuration, and proper position­ ing of bicaval dual-lumen VV-ECLS cannula, are both essential to minimize recirculation. 7 The most frequent clinical sign associated with recirculation is desaturation, which can be acute and life threatening. An experienced team of nurses, ECLS specialists, respiratory therapists, and medical staff at the bedside is essential to deal with this swiftly to avoid poor outcome. Addressing drainage problems, circuit failure, and recirculation early will avoid the majority ofthese episodes. Additionally, in the absence of oxygenation by the lungs desatura­ tion always occurs when native cardiac output significantly exceeds ECLS blood flow. Clinical situations where this may be accelerated include pain, delirium, and sepsis. These syndromes are not always easy to diagnose during ECLS and it takes experience to address.them expeditiously (see Chapter 40).

467

Chapter 42

Weaning VV-ECLS A successful stepwise weaning of sweep gas flow may therefore indicate the recovery of the ventilator function (C0 2 clearance) but not necessarily the recovery of oxygenation. Over prolonged ECLS support the ability of an oxy­ genator to clear CO 2 can decline and increasing amounts of as flow become necess . When s change is observed in association with other indicators ofoxygenator failure a system change may be indicated. For VV-ECLS the following procedure should be followed: once the level of ventilator support is deemed to be appropriate and safe, the extracorporeal gas flow can be reduced in a stepwise fashion to zero with close monitoring ofarterial oxygenation and CO2 lev­ els. It usually becomes necessary to increase the respiratory rate and Fi02 settings used to rest the lungs while on full ECLS support. Some centers wean extracorporeal blood flow parallel to the gas flow, to a set minimum flow; however, no changes in blood flow are absolutely necessary, since during the provision ofVV-ECLS without gas flow, the drained venous blood will return to the venous system without any change in oxygen content. Protocols that include wean­ ing of extracorporeal blood flow may add the risk ofthromboembolic complications at a time when anticoagulation may also be reduced in preparation for decannulation. Over the following hours, sometimes ex­ tending to days, the patient needs to be closely monitored for signs of weaning failure. ~e extracorporeal devices have a built in mixed venous saturation monitor. A steep drop in saturation may indicate weaning failur.e. When a pulmonary artery catheter is present, the same phenomenon can be observed as well as signs ofacute cor-pulmonale. Echocardiography may also be helpful. Signs ofpoor tissue oxygenation such as lacticemia or new organ impairment may indicate that further extracorporeal support is needed. Once a patient has stable hemody­ namic parameters and adequate oxygen delivery 468

wjth no gas flow on VV-ECLS, decannulation can be attempted.

Liberation and Decannulation The weaning of mechanical ventilation drives the timing for liberation from extracor­ porea! support. The targeted level of ventilator support deemed appropriate to allow for the removal ofthe extracorporeal system varies from patient to patient. However, the majority will be ventilated at the time of liberation, well within lung protective ventilation with aPaOjFi02 ratio above 100 mmHg without extracorporeal support. It should also be possible to achieve almost nor­ mal CO2 clearance (PH >7.3). Factors which may have negative impact on the respiratory function including infections or other organ failure should be well controlled before decannulation, since reinstitution ofECLS often proves more complex. In some situations, continuing with ECLS to the point of liberating a patient from mechanical ventilation first is appropriate. Little evidence exists in which group of patients, perhaps with the exception of patients waiting for lung trans­ plantation, would benefit from tills strategy and more trials on this subject are desperately needed. An elevated aPTT is a risk factor for post decannulation complications. 8 Heparin free ECLS in adult patients on modem extracorporeal devices has been reported for up to 25 days and many centers routinely stop anticoagulation be­ fore decannulation. 9 Percutaneously placed can­ nula can be removed after clamping the circuit with the help of local compression for 20 to 60 miinutes. The use of a pursestring suture is often recommended. After prolonged ECLS runs and in patients with minimal subcutaneous tissue or signs oftissue infections at the cannula site surgi­ cal consultation may be needed. Clot formations around venous cannula have been described and risk for thromboembolic complications during decannulation exjsts.1O In case of a sudden hemodynamic deterio­ ration the possibility of a p"ulmonary embolism

-

Weaning and Decannulation ofAdults with Respiratory Failure on EeLS

must be considered and venous ultrasound pos­ tremoval is indicated in any case. Ifa cannula was placed with an open approach or an arterial can­ nula is in place the decannulation strategy should be discussed with the team who performed the procedure as an open repair may be indicated. Patients on VA support and respiratory failure need to follow the VA weaning protocol with special consideration for systemic oxygenation and differential hypoxia (see Chapter 51).

Other Modes of Support In the recent past low flow extracorporeal systems have been introduced for extracorporeal CO 2 removal, (ECCOR, see Chapter 63).11 These systems are either pump driven venovenous or pumpless arteriovenous in configuration. During weaning and liberation no change in blood flow is needed. Reduction in gas flow leads to less effective CO2 removal and ceasing it altogether mimics decannulation. The relation between CO2 removal and gas flow is not linear and attention needs to be taken to monitornative CO2 clearance during weaning. The Patient Who Cannot be Weaned According to the 2015 ELSO report, respi­ ratory support averaged approximately 12 days, which exceeds that needed for cardiac support. Much longer support times ofover 100 days have been reported and indeed according to a recent analysis of the database, duration of support alone does not predict futility.12 Full lung recov­ ery has been demonstrated after many weeks on support. 13 However, complications related to intensive care and extracorporeal support itself can accumulate and can make prolonged runs very difficult. ELSO reports suggest that about 2/3 of all adult runs wean successfully. In the CESAR trial only 9% ofall patients died of persistent respira­ tory failure while the majority of deceased pa­ tients developed extrapulmonary organ failure. 4

A small proportion of patients may qualify

for lung transplantation and early contact with the appropriate transplant center is indicated (see Chapter 58). In a recent review of the interna­ tionallung transplantation database ~4% of all transplants received ECLSJin;t. 14 Two scenarios common. Patients with chronic disease who have already been placed on the transplant wait­ ing list may have an acute deterioration and need ECLS support. These patients are unlikely to be weaned from ECLS successfully and ventila­ tor weaning and extubation may be desirable once it has been confirmed that transplantation remains an option. On the other hand, patients on extracorporeal support with the intention to bridge to recovery may fail to show significant improvement. No clear cutoff time exists after which a referral for lung transplantation should be considered. The appropriateness of this op­ ltion depends on the patient's condition, age, and lpremorbid state as well as on local availability of lung transplantation and wait times. Since full recovery after prolonged ECLS runs has been observed and long-term survival times after transplantation may be shorter the deci­ sion for transplantation can only be made on an individual basis in discussion with the local transplant center.13

are

Conclusion Since up to a third of adult patients on re­ spiratory support may not be able to be weaned, the appropriateness ofECLS should be reviewed regularly and similar to other invasive life sup­ port including mechanical ventilation or renal replacement therapy, should be continued only if IlL reasonable chance ofa positive outcome exists. The withdrawal of extracorporeal life support for futility needs to follow local ethical guide­ lines with ongoing discussions with family and! or patient. Organ donation post withdrawal of extracorporeallife support should be considered.

469

References

of Critical Care Medicine. 2014;16 (1):69­

72. 1.

2.

3.

4.

5.

6.

7.

Guidelines for respiratory support Vers 1.3 January 2016) https:11 www.elso.org/Resources/Guidelines.aspx Schmid Philipp A, Hilker M, et al. Ve­ novenous extracorporeal membrane oxyln+ ""'"""'-"". More recently, the randomized ... "~''''n'-'' with severe, ARDS reported _ __

of mmHg, a lowest pH a modified acute lung injury score of 3.8,3 other studies have reported outcomes of A(H1N1) influenza-related cornplllred to other ARDS etio\ogies. 4,5 "n.·LT••....., outcome has been (,""'ArT" by Schmidt et 5

were studied rp.Tlrn" a tion, with bleeding the most dreaded, and still widely impacts short-term outcome and the overall cost. The main mecha­ involved anticoagulation, throm­ bocytopenia, and consumption coagulation factors. In platelet function II and acquired von Wille brand increase patients. 12 In a the risk of bleeding for review ECLS complications among 1763 13 serious bleeding occurred in leedlfl,g n.r>('l1'rt"",.t1 in 29% on the A-(HjN I ) influenza 3 demic. ll'~'JU\JU~,~ofillnmclerebra bleeding for adult patients on ELSO database reports Aubron that blood cells transfusion et at

is­ policy on seems possible with implementation of a lower objective for systemic anticoagula­ tion, a transfusion threshold of 7 g/dL, and an auto-transfusion during decannulation. 15

Thromboembolic Events ';"CUll""u,c,, morbidity and mor­ to bleeding, anticoagulation tality targets have generally declined over 5-10 years, which may the risk of vu..LJ\.J",;;U'LJ\JIUV events. In 2006, Raslan et al. reported autopsy results performed on 78 out of 1 patients, died postcardiotomy major clinical and postmortem true incidence of thromboembolic events, which proached 50%, was highly underestimated by

Outcomes and Complications

clinical evaluation. 16 The risk such events increased with duration~ especially n"'I,{I"{I 6 days was still frequent despite IS systemic anticoagulation. Presently, predominantly provided miniaturized cen­ trifugal pumps, low and heparin-bonded components. All have contributed to lower anticoagulation require­ The prevalence of post-decannulation thrombosis in the VU'JLl'UUI."'"

Respiratory ECLS

a total of 146 ECMa procedures were performed on patients. Thirty-six patients had a of 46 infections (30.1 infectious 1,000 days of ECLSV1 cannula, and bloodstream pneu­ mtlectlons, as well as ventilator monia (yAP) were all collected recent studies. Although frequency between the most commonly infected site was lung. the frequency ofVAP ECLS was higher of Schmidt et (55% underlying discrepancies surveillance ECLS centers and the contro­ system versial use ofantibiotics prophylaxis to nrp,Vp.l!lT Nls with patients. Lastly, from a clinical view, the time to NI occurrence is imFor to the VAP, and ,nT.Pl'1"lnn were 8±11, 20 NI occurrence was consistently independently associated with death in ICU. Also risk of NIs increases with patient atICU and with longer duration ECLS specific "t'1"'~h>,rr, ass:ocllatf:d infections on studied. Research on preventive such as prophylaxis, routine surveillance or the use chlorhexi­ gluconate-impregnated sponges in carmula dressing is warranted.

rU:UJIJ.

·t.Ull'JilF;

Infectious Complications

outcomes.

reported infections and in complications in adults receiving most of them, indication for was both respiratory cardiac Hemolysis ELsa "''-.'"'F;''''' based on adult rate was than newborns (30.6 vs. 10.1 per 1000 ECLS especially Minor hemolysis commonly occurs during patients ECLS cardiogenic VV-ECLS. A study of 207 patients 19 Schmidt et shock (37 per 1000 ECLS one episode with EeLS reported at al. a of Nls (75.5 per hemolysis 66% patients.22 Hemolysis was 1000 day) among patients who un­ classified as mild gIL), (0.5­ derwent >48 hours support 1.0 or severe .0 gIL) in 47, and 7% car'dlC'j.l;emC shock. prevalence was higher ,_"''','V1.'''', respectively. Patients with hemolysis than that ofthe ELSa Registry and other studies experienced duration run and from single centers. For instance, in a mixed required more blood products. After controlling Australian population ofcardiogenic shock and for age, weight, pediatric index of mortality, 473

Chapter 43

and diagnosis, patients with severe "'",n",""" were more likely to die in the ICU and (odds ratio, 6.34; 95% CI, 1 p=0.006). A of 154 adult VA and VV-ECLS, showed patients that patients who demonstrated ";:;::::.:.::..:::..~:..~

study to investigate causes of hemolysis on and to influence on morbidity and mortality are

while maintaining Plateau pressure J1ncirome. N J Med. 2003;348(8):683-693. SchellingG, Stoll HallerM, etal. """"UH'­ related quality of and posttraumatic stress disorder in survivors of the acute respiratory distress syndrome. Crit Care Med.1998;26(4):651 Luyt Combes Becquemin MH, et al. Long-term outcomes of pandemic 2009 in­ fluenza A(H 1N 1)-associated severe ARDS. Chest. 2012; Cheung PY, Sawicki G, Salas Etches PC, Schulz Radomski MW. The mechanisms platelet dysfunction during extracorpo­ in critically ill real membrane neonates. Med.2000;28(7):2584­ 2590. U, Beyersdorf von Willebrand syn­ drome in patients with extracorporeal life support (ECLS). Intensive Care Med. 2012;3 B( 1):62-68. Landoni Zangrillo et aI. A meta-analysis of complications and mortality of extracorporeal membrane oxygenation. Critical care and resuscitation: journal of the Care Medicine. 2013;15(3):172­ 178. Aubron Cheng AC, Pilcher D, et al. Fac­ tors associated with outcomes of patients on extracorporeal membrane oxygenation support: a study. Crit Care. 201 . Agerstrand CL, Burkart KM, Abrams DC, Bacchetta MD, Brodie D. Blood conserva­ tion in extracorporeal syn­ ation for acute respiratory drome. Ann. Thorac. Surg. 2015;99(2):590­ 595. Rastan AJ, Lachmann N, Walther T, et al. Autopsy in patients on post­ cardiotomy extracorporeal membrane oxygenation (ECMO). Int. 1. Organs. 2006;29(1 1121-1 1.

Outcomes and Complications ofAdult rfP,'TJlr,m",,,, ECLS

17.

E, lence of Venovenous

J, Retter A, et aL Preva­ Thrombosis Following Membrane

Oxygenation in Patients With Severe Care 2015;43(12):e581-584. A, Lubnow M, et al. 18. Camboni D, Support time-dependent outcome analysis for veno-venous extracorporeal membrane M""''''"'''VU' Eur. J.

11 ;40( 6): 1341-1

SA,

19.

L'-UUUIlJ.UU

and outcome. Infect. Control Hosp. EpidemioL 2013;34(1 ):24-30. 22. Lou S, Maclaren G, Best Delzoppo C, Butt W. patients TO"",....."

and outcotliles 2014;42(5):121 Omar Mirsaedi M, S, et al. Plasma Hemoglobin is an Independent Predictor Mortality among Patients on ECMO Support. 2015 Hodgson CL, et Long-term with

for 2012; 16(5):R202. Hemmila SA,

Crit Care. Oct et aL

Extracorporeallife support for severe acute respiratory distress adults. Ann. 2004;240(4):595-605; discussion 605-597. Schmidt M, M, Sheldrake J, et "'....',,"-.,'M Survival after for Respiratory the "'''''''''''1"''''''' ECMO :.. lIn"",,. ''''''''''''''''AmJ 189(11): 1374-1 Roch A, S, Masson et Out­ come ofacute respiratory distress syndrome patients with extracorporeal mem­ brane a referral center. Intensive Care 2014;40(1):74­ 83. Enger TB, Philipp A, Videm Pre­ diction mortality in adult with severe acute lung failure veno­ venous extracorporeal membrane oxygena prospective observational 2014;18(2):R67. Pappalardo F, et aL Pre­ dicting mortality in patients nnrl"'''CH'Im.fT venovenous ECMO due to influ­ enzaA(HINl) pneumonia: the score. Intensive Care Med. 2013;39(2):275­ 281. 30. Campbell TM, RE, Bartlett RB. Impact ofECMO on neonatal mortality in Michigan (1980­ 1999). J Pediatr Surg. 2003;38(3):290-295; discussion 290-295. 31. Jen HC, Shew SB. Hospital readmissions and nonneonatal ECMO. 2010;125(6): AM, Karamlou T, Vafaeezadeh M, et aL Increased extracorporeal membrane O"'.HUllVU center case volume is aswith urn'u·n'"",,., membrane OX~{14enatlOn

477

pediatric patients. J. Thorac. Cardiovasc. 2013;145(2):470-475. NHS NICE GUIDANCE, last accessed March 14th 2014, http://www.nice.org.uk/ guidancelIPG3911. 34. last accessed March 14th 2014, http://www.ecmonet.org/. NSW Critical Tertiary Referral Netand Transfer ofCare (ADULTS), last accessed 14th 2014, http://wwwO. health.nsw.gov.au/policies/pd!20 1O/pdf! PD20 .pdf. Combes A, D, Bartlett et al. Position paper for the organization membrane oxygenation failure in Crit. Care L'\.V""f.'ll.

37. Beurtheret Mordant P, Paoletti Emergency circulatory support in ren:"aClJorv cardiogenic shock patients in remote tutions: a pilot study (the cardiac-RESCUE program). HeartJ.2013;34(2):112-120. 38. Forrest P, Ratchford J, Burns B, et al. Retrieval of critically ill adults using ex­ tracorporeal experience. 2011 ;37(5):824-830.

Linden V, Palmer K, Reinhard J, et survival iu adult patients tory syndrome by extracor­ membrane oxygenation, minimal se­ dation, and pressure ventilation. Care Med. 2000;26(11): 1637.

44 Adult Cardiovascular Defects, Diseases, Roberto Lorusso, ,VlirUll'O urt;tJ,VffHft!/.

Procedures that Predispose to ECLS

MD, Patrick PhD, Mirko MD, PhD, Ma;ess€~n, MD, PhD

Acute Myocardial Infarction Currently, the therapy for acute myocardial (AMI) is primary neous intervention (PCI), despite revascularization, of patients with AMI with ST-segment

treatment is not sufficient is imminent, use ofmechanical support should be considered, such as intraaortic balloon pump (IABP), percutaneous or more or extracated ventricular assist devices membrane oxygenation (ECMO).7-lo are no contraindications and ifthe time are respected, venoarterial ECMO (VA­ ECMO) should be rapidly implanted because it can concomitantly and effectively support heart and function, interrupt the refractory (\1"~'l'l".n oforgan prevent the oc­ ultimately the periprocedural

PhD,

course. ll - 14 The updated 2015 American ColofCardiology Foundation/American Heart Association/Society for Cardiovascular ography Interventions PCI rec:orrLm. Diseases, and Procedures that Predispose to ECLS

tnClep,enldeIlt predictors of

death at multivariab1e analysis. 62 a of patients, a to more sophisticated devices or to heart transplant is due to lack mac recovery of patients in these circumstances is still less than optimal,71,75 probably because patients surgical treatment are often in suboptimal condition with end-organ are usually eXlpm,ea

to prolonged ECMO therefore higher complication rates.60.71.75 How­ ever, favourable patients supported with transplantation have also 68 Following heart recovery and hospital dis­ AM recurrence is uncommon and can eITleCIIIVelV be treated with medical therapy.68,70,71 Mid and long-term submitted to for AM or AFM are unless myocardial in per-

Table 44-2. Published series of extracorporeal membrane OJ(l{genatlon in acute LlUI"UjJLru. . mvoc~lIditis.

ECMO Author

Ref Pts

Pts (%)

ECMO Duration (mean±SD, hI'S, if available)

Pts Switched to LV AD or Heart Transplant

Survival to Postop Survival Hospital Discbarge (%)(mean followup) Pts(%)

na

31 (59.6%) na

Aoyama 61

52

42 (80.7%)

186±134

Diddle

147

101 (69%)

138

62

90 (61%)

312±96

1*

80% (1 year)

129±50

2*

na

Chen

Hsu

66

51

na

171.5±121 (inel 6* 3/\ (inel pediatric pt'l) pediatric pt'l) 31 (61 %)

Ishida

67

20

12 (60%)

na

o

12 (60%)

8

35

na

240

4/\

24 (69%)

130

1*

na

o

Mirabel

na

na

na

100% (2.6±2.1 years)

100% (1.5 years)

na 100% 6 (75%)

(range 1.4 - 5.9

years)

Lorusso

71

57

43

Su

75

14

9

237±456

(64.2%)

I

ents; LVAD=left ventricular assist iched to heart trans !ant

2* - 3/\

65.9% 41 (71.9%) (5 years)

2/\

10

SD=Standard

Chapter 44

manent LV dysfunction. In these cases, a less favourable to other chronic eXlJlectc~d foIl owing

combination with coronary ischemia. Because mt":". and Procedures that

extracorporeal membrane oh1f1:~enatlon port. J Card Surg. 2011;26:666-8. Eudailey KW, Yi SY, Mongero gener G, Guarrera JV, 1. Transdiaphragmatic left ventricular during peripheral venous-arterial ex:tracor­ poreal membrane oxygenation. 2015;30:701-3 35. Hoefer D, Ruttmann Outcome evaluation ofthe hM,!1""''''_1"l'I,_nr,t1 concept in patients with car'Q1()gemc Ann Thorac Kouraki ~ ","'JLU,",_'''''''''

Characteristics

group. Clin 37. Alozie A, '",,,,',",,,,", ECMO as mam coronary ising concept of na~~m()LL1. membrane oxygenation. 2014;61: 1093-7 101. Flam Broome M, Frenckner Brans­ trom Bell M. Pheocromocytoma-induced Takotsubo-like cardiomyopathy leading to successfully oxygenation. J 2015;30:365-72 cases 102. Hu W, ehen Liu et aL electrical storm in fulminant myocarditis Med. treated with EeMO. Am J 2015;33:606.e3-8 103. PS, Lilly Nicolosi Use of ECMO to temporize circulatory instability severe Brugada electrical storm. Ann Thomc Surg. 2009;88:982-3 104. Tomasello SD, Boukhris M, Ganyukov V, etal. LUVUAUA«.U'"

ell, You et al. lighting indications ofECMO in endocrine emierglenCles. Sci Rep. 2015;5: 13361 et Juthier F, Ennezat trat life support in chromocytoma crisis. Ann Thorac Surg. 201

498

1

Seco M, Forrest P, Extracorporeal very

SA, et al. aortic

Adult Cardiovascular Defects, Diseases, and Procedures that Predispose to ECLS

valve implantation. Heart Lung & CircuI.

2014;23:957-62 106. Artl M, Phillip A, Voelkel S, et al. Early ex­ periences with miniaturized extracorporeal life-support in the catheterization laboratory. Eur J Cardio-Thorac Surg. 2012;42:858-63 107. Husser 0, Holzamer A, Philipp A, et al. Emergency and prophylactic use of minia­ turized veno-arterial extracorporeal mem­ brane oxygenation in transcatheter aortic valve implantation. Cath Cardiovasc Interv. 2013;82:E542-E551 108. Staudacher DL, Bode C, Wengenmayer T. Severe mitral regurgitation requiring ECMO therapy treated by interventional valve reconstruction using the MitraClip. Catheter Cardiovasc Interv. 2015;85:170-5 109. Ucer E, Fredersdorf S, Jungbauer C, et al. A unique acces for the ablation catheter to treat elecrical storm in a patient with extracorporeal life support. Europace. 2014;16:299-302 110. Komminemi M, Lang RM, Russo MJ, Shah AP. Percutaneous closure of infacrt related ventricular septal defects assisted with portable miniaturized extracorporeal membrane oxygenation: a case-series. Car­ diovasc Revasc Med. 2013;14:241-5 111. Rohn V, Spacek M, Belohlavek J, Tososki J. Cardiogenic shock in a patient with pos­ terior postinfarction ventricular septal rupture: successful treatment with extra­ corporeal membrane oxygenation (ECMO) as ventricular assist device. J Card Surg. 2009;24:435-6 112. Gregoric ill, Mesar T, Kar B, et al. Percu­ taneous ventricular assist device and extra­ corporeal membrane oxygenation support in a patient with postinfarction ventricular septal defect and free wall rupture. Heart Surg Forum. 2013;16:150-1 113. Hobbs R, Korutla V, Suzuki Y, Acker M, Vallabhajosyula P. Mechanical circula­ tory support as a bridge to definitive sur­ gical repair after post-myocardial infarct

ventricular septal defect. J Card Surg.

2015;30:535-40 114. Tsai MT, Wu HY, Chan SH, Luo CY. Ex­ tracorporeal membrane oxygenation as a bridge to definite surgery in recurrent postinfarction ventricular septal defect. ASAIO 1. 2012;58:88-9 115. ObadiaB, Theron A, Gariboldi V, CollartF. Extracorporeal membrane oxygenation as a bridge to surgery for ischemic papillary muscle rupture. J Thorac Cardiovasc Surg. 2014;147:82-4 116. Anastasiadis K, Antonitsis P, Hadjimiltia­ des S, et al. Management ofleft ventricular free wall rupture under extracorporeal membrane oxygenation. Int J Artif Organs. 2009;32:756-8 117. Formica F, Corti F, Avalli L, Paolini G. ECMO support for the treatment of cardio­ genic shock due to left ventricular free wall rupture. Interact Cardiovasc Thorac Surg. 2005;4:30-32 118. Abedi-Valugerdi G, Gabrielsen A, Fux T, Hillebrant CG, Lund LH, Corbascio M. Management ofleft ventricular rupture after myocardial infarction solely with ECMO. Circul Heart Fail. 2012;5:e65-e67 119. Noyes AM, Ramu B, Parker MW, Underhill D, Gluck JA. Extracorporeal membrane oxygenation as a bridge. Tex Heart Inst J. 2015;42:471-3 120. VohraHA, Jones C, VolaN, HawMP. Use ofextracorporeal membrane oxygenation in the management of sepsis secondary to an infected right ventricle-to-pulmonary artery Contegra conduit in an adult patient. Inter­ act Thorac Cardiovasc Surg. 2009;8:272-4 121. Santhosh JG, Preethi W, Sanghetaa M, Bijn T. Outcome of patients with infec­ tive endocarditis who were treated with extracorporeal membrane oxygenation and continuous renal replacement therapy. Clinics Pract. 2014;4:66-70 122. Elliott P, Andersson B, Arbustini E, et al. Classification of the cardiomyopathies: 499

Chapter 44

a position statement from the European Society Of Cardiology Working Group on Myocardial and Pericardial Diseases. Heart J. 2008;29:270-6 MERIT-HF study group. Effect of metoprolol CRIXL chronic heart failure: Metoprolol tion Trial in Heart (MERIT-HF). Lancet. 1999;353:2001-7 IPnl"lPf'u SM, Kavinsky Parrillo JE. Intern Med. 1999; UIVi"-"'''.''' shock. 125. Demondion p. Niculescu M, of 30-day and outcome in cases of myocardial infarction with cardiogenic shock treated by extracorpo­ life support. J Cardiothorac Surg. 14;45:47-54.

ogy on outcome. J 2015;150:333-40 Bermudez CA, Rocha RV, Y, et aL ECMO for advanced refractory shock in acute and chronic cardiomyopathy. Ann Thorac Surg. 2011 1 128. M, Bacchetta M. The thc"\4"""HH 129. D, Ruttmann Outcome evaluation ofthe OniJge:-to-orIa concept in patients with cardiogenic Ann Surg. 130. Marasco Lo Extra­ ('(WnA1CP!> I life assist device: the double ASAIO J. 2016;40:100-6. 131. A, Brodie D. new in membrane oxygen­ ation for cardiac failure and cardiac arrest in adults? Intensive Care 2014;40:609­ 12

500

45 Extracorporeal Cardiopulmonary Resuscitation Jan Belohlavek, MD,

Yih-Shamg

MD, PhD, Naoto Morimura, MD, PhD

Introduction Survival after cardiac arrest (CA) poor. Worldwide, neurologically

wlltnesse:Q cardiac arrest, shockab­ le rhythms, quality post cardiopulmonary resuscitation (CPR) care, targeted

compression depth compression to ventilation ratios, oxygenation, use oftargeted temperature management, catecholamines to improve cardiac output, thrombolysis and early cutaneous coronary intervention (PCl) adopted into care and an aim of improving cardiac arrest outcomes? Large studies employed mechanical chest compression devices. 4•5 The aim of res us-

citation, however, remains the early return of spontaneous circulation (ROSC). Prolonged CA and without ROSC dismal the chance noses. 6 In patients without when transport to hospital occurs 4%.7,8 CPR falls

initially patients with challenging survival. term survivors out patients cannu­ lated at bedside within 15 minutes ofCAwere reported by Mattox in 1976.10 four with 675 in CA treated percutaneous cardiopulmonary bypass between survival to of and 2005 (IQR 15.4%, 11 However, mostprob­ ably these ll.I/derreported poor outcomes. repqrts and advances availability ECLS devices useful for transporting pal•• ",,"..,,, compatible "v+...,,,,,,,,,,,,,,,r•.,,, taneous cannulas suj'tabJle and ongoing to improve outcome ofCA became the prerequisites implementation ofextracorporeal mechanical support technique to CPR, development of extracorporeal cardiopulmonary resuscitation (ECPR). been recognized in European • ...,...,ALU'J1V/';y,

501

Chapter 45

Guidelines 2015 as a rescue therapy for those

patient with venoarterial ECMO. Alternatively,

patients in whom initial advanced life support

for ECMO or ECLS, other terms including

measures are unsuccessful or to facilitate specif­ ic interventions (eg, coronary angiography and PCI or pulmonary thrombectomy for massive pulmonary embolism).12 The American Heart Association (AHA) guidelines are more cau­ tious regarding the recommendation for ECPR because of the ongoing paucity of available data.13 A recent metaanalysis compared ECPR to conventional CPR and showed a favorable effect on 3-6 month survival with good neuro­ logical outcome in ECPR subjects. However, the effect of ECPR on survival to discharge in OHCA has not been clearly shown indicating the need for strict criteria for implementation of ECPR in such a setting. 14 This chapter aims to provide an overview on the current role ofECPR in clinical practice, serving as a practical guide for performing ECPR.

percutaneous or emergency CPB and eventu­ ally portable or percutaneous cardiopulmonary support, are used interchangeably.

Definition of ECPR ECPR applies ECLS in patients who remain in CA despite conventional CPR, or when inter­ mittent ROSC occurs, but repetitive CA reoc­ curs. In other words, ECPR is used in refractory CA, when chances of obtaining and sustaining ROSC are poor. Unfortunately, definitions ofre­ fractory CA vary from 10 to 30 minutes ofCPR without ROSC. 15,16 The optimal duration of refractory CA prior to ECPR remains unknown, but both for mCA and OHCA, maximal time of approximate,!): 15 minutes of CPR seems to be reasonable. 17,ls Cannulation to ECPR ensures complete mechanical support suitable to provide tem­ porary circulation and gas exchange. Cannu­ lation usually occurs via the femoro-femoral venoarterial route, because ofeasy accessibility, (Figures 45-1 and 45-2). Alternative approaches include the femoro-subclavian, jugulo-subcla­ vian or jugulo-femoral routes. The term ECPR is used to describe an intention to treat a CA

502

Organizational Issues Related to ECPR Implementation ECPR usually occurs in tertiary care insti­ tutions provided by a dedicated ECMO team. The ECMO team consists ofprofessionals from intensive care, cardiology, cardiac surgery, and other specializations collaborating with special­ ized nurses and perfusionists. In-hospital coor­ dination between the ECMO and CPR teams must include rules to activate the ECPR team in a 2417 coverage model. ECMO team members must be well trained in venoarterial cannulation, circuit management, and early postresuscita­ tion care (see Chapter 67). An ECLS-trained physician and perfusionistor specialist provides 24-hour coverage for ECLS patients. Physicians trained in vascular ultrasound for the insertion, maintenance, and surveillance of ECMO can­ nulae and distal perfusion cannula should also be available. Intensive care unit (lCU) care

Figure 45-1. A typical scheme of a femoro­ femoral VAl ECMO circuit setting including distal perfusion cannula for prevention oflimb ischemia used for ECPR. NC=inferior vena cava; RA=right atrium; arrows depict blood flow direction.

Extracorporeal Cardiopulmonary Resuscitation

must be adjusted to ECMO use, ie, intensive care unit nurses should be specifically trained in ECLS and circuit management. The nurse-to­ patient ratio during the postresuscitation phase should be 1: 1, depending on the local regula­ tions or resource availability. The ECLS team should be self sufficient and able to set up the ECLS circuit quickly. The specialist team may also be responsible for managing equipment and supplies, regular service administration, troubleshooting, daily rounds, education, pa­ tient database administration, quality control, 17 and coordination of neurological, psychlatric, psychological and rehabilitation care. Adjust­ ments may be implemented depending on the site where ECPR is provided. ECPR for mCA is usually provided in the rcu or operating room. Experiences mainly from France and Germany show th18 and < 65 years Witnessed OHCA of presumed cardiac causc Minimum of 5 minutes of ACLS performed by emergency medical service team without sustained ROSC Unconscionsness" ECLS team and ICU bed capacity in cardiac center available

EIciusion Criteria OHCA of presumed noncardiac cause Unwitnessed collapsc Suspected or confirmed pregnancy

ROSC within 5 minutes of ACLS performed by EMS team Conscious patient Known bleeding diathesis or suspected or confirmed acute or recent intracranial bl~ Su,-pected or confirmed acute stroke Known severe chronic organ dysfunction or other limitations to therapy "'Do not resuscitate" order or other circumstances that make 180 day survival unlikely Known pre-arrest cerebral performance cat~'Sory CPC > 3

OHCA=out-of-hospital cardi.a.c: arrest; ACl...S=acIvanced cardiac life SUPPOrt:. ROSC=return ofspoataDcous cifCUla:tion; EMS=emergeocy medical service; CPC---cerebral performance ca1Cgory. 'Defined as no response 10 vcrl>al or painful stimuli during ACLS

504

Extracorporeal Cardiopulmonary Resuscitation

vice), and neurological status. Small pupils are considered to be a favorable prognostic factor/o similarly lactate values below 5 mmoLIL,31 however, lactate values are usually higher in ECPR patients. Alternatively, some reports in­ dicate brain near infrared spectroscopy (NIRS) can be a potentially valuable tool in assessing patients under CPR. 32 When care includes EPCR, a perfusionist or specialist primes the ECLS pump, femoral vessels are visualized by ultrasound, and the patient is draped in a sterile fashion to begin cannulation. Generally, cannulation during ongoing CPR is challenging, puncture ofthe vessels may not be easy, and ad­ vancement of the wire must be done cautiously but rapidly. Occasionally, it may be difficult to distinguish between the artery and the vein, therefore cannulation under fluoroscopy may be required. The direction the wires advances helps assess whether artery or vein has been punctured. Another alternative includes insert­ ing a regular catheterization sheath to perform brief angiography, which not onJy distinguishes artery from vein but helps to choose the cannula diameter and excludes possible peripheral arte­ rial disease (Figure 45-3). Position ofthe venous cannula can be confirmed under fluoroscopy or by echocardiography. Later positional changes without an inserted wire should be avoided. After cannulas are safely inserted and retrograde flow from the cannulas is brisk (an easy aid to confirm correct intraluminal position), the exTable 45-4. Inclusion and exclusion criteria for initiation of ECLS in a randomized study on ECPR.18 Inclusion Criteria No ROSe or ROSe with ongoing shock (defined as sustained hypotension less than 90 mmHg of systolic pressure or the need for bolus doses ofvasopressors to maintain circulation)

E1clu3ioD Criteria Signs of death or irreversible organ damage

Admission to catheteriza1ion lab not Known bleeding diathesis more than 60 minures after the collapse/initial call to EMS" Consensus ofECMO team members Inadequate arterial andlor venous on ECLS initiation access for femorcrfemoral cannulation ROSC=rctwn or sponllmeoWl eircul.8lion -If the exact couaPse time was oot known. an initial call to EMS was considctod

Panel A:

Panel B:

Figure 45-3. ECPR patient: previously healthy 55-year old man with ongoing CPR for ORCA from ventricle fibrillation resulting in CPR. Panel A: femoral artery cannulation, with two sheaths accidentally inserted into the same artery (arrows). One was subsequently removed, the other one exchanged for arterial ECLS cannula. No bleeding occurred. Panel B: angiography ofthe left coronary artery after institution of ECLS flow, non-beating heart, proximal LAD occlusion (upper arrow), signifi­ cant atherosclerotic lesion on distal circumflex artery (lower arrowhead).

505

Chapter 45

tracorporeal circuit is attached. Patient receives full anticoagulation, unless contraindicated (suspicion of bleeding), usually with a bolus ofunfractionated heparin of75-100 IUlkg fol­ lowed by continuous infusion to keep activated clotting time close to 200 seconds, or an aPTI at 50-70 seconds (see Chapter 7). In some centers or instances, ECPR with surgical cutdown and exposure of the femoral vessels is used, and might be preferrable in case of difficult echo visualization or lack of fluoroscopy imaging.

Optimal Cannula Diameter Choosing cannula diameter size can prove difficult but important because changing them after cannulation is stressful and potentially dangerous. When considering the appropriate cannula diameter, an important factor is the presence ofperipheral arterial disease. A smaller diameter cannula to prevent damage or tearing of the femoral or iliac artery can prevent this potentially fatal complication. Consequently, smaller cannula may not need a distal perfusion to prevent cannulated limb ischemia and some centers, including the Regensburg group, prefer this approachY Target ECLS flows relate to cannula diameter. For a targeted flow of 3.5-4 Llmin an arterial cannula of 15-17 Fr should suffice. For higher flows, larger cannulas are needed. The authors prefer larger cannulas to reach a flow of more than 60-70 mL/kg/min, 5 Llmin or more in larger patients, however no firm data support any single approach.

Reperfusion Technique and Initial ECLS Flow Setting The ECLS circuit, unless wet primed be­ forehand, is usually primed with cold crystal­ loid for a cold flush reperfhsjou techniQ)J.e,34-36 although evidence is scarce and based on experimental settings. Intravenous cold fluid boluses during ECLS initiation may help cor­ rect hypovolemia to assure ECLS flow and 506

institute target hypothermia quickly. However, this intervention is not based on established evi­ dence. Starting with a lower sweep gas oxygen concentration is based on studies suggesting the harmful effects of hyperoxja after ROSC.37,38 Titration to target peripheral saturation of 90­ 95% seems reasonable, bearing in mind that in a majority ofECPR patients pulse oximetry may be inaccurate. Thus brain tissue regional satu.ra­ tions by NIRS provide better in-line (sampling every 6 seconds or similar depending on device used) control with target values around 60-70%. Beginning with inspired oxygen of 50-60% and later increases based on above monitoring appears reasonable. Blood flow is usually set to 4 Llmin and eventually increased gradually. Thiagarajan et al. analyzed the ELSO Registry for ECPR and found an initial pump flow of 3 Llmin in both survivors and nonsurvivors and it remained constant after 24 hours. 39 Sweep flow initially matches that of the blood flow.

Complications during Implantation and Initial Launching Several serious complications may occur during percutanous cannulation. Unsuccesful cannulation itself occurs with usually fatal consequencies due to refractory CA. Surgical exploration ofthe groin, as mentioned, serves as an alternative; however, severe atherosclerotic femoral/iliac artery disease may preclude even a surgical approach. Alternative approaches usually for arerial cannulation should be imme­ diately tried iffemoral cannulation is unfeasible. The axillary approach with a 15 Fr cannula is probably best and if succesful, allows enough reperfusion into the ascending aorta but may need later conversion to a different site. Pro­ longed cannulation over 30 minutes usually means an overall time of CPR over 60 minutes, often associated with poor neurological out­ come or death. Vessel injury or rupture must be avoided by a careful technique during all stages of cannulation, espcially during wire insertion.

Extracorporeal Cardiopulmonary Resuscitation

The guiding principle is to be perfectly sure that the wire is properly inserted in the vessel lumen. Simultaneous distal cannulation of the same vessel (Figure 45-3) may happen and usually does not cause signficant problems. One sheath! wire is either removed when a large cannula is inserted into the vessel, or better left in situ for later safe removal. Despite proper arterial cannulation, arte­ rial wall dissection may occur, causing vessel narrowing or obstruction after removal of the cannula. This must be detected early and cor­ rected. Ifnot, it may cause severe limb ischemia. Venous cannulation is easier but improper position (not enough deep or too deep) may cause cannula dysfunction or inappropriate ECLS flow. Generally, ifthe ECLS flow proves adequate despite suboptimal cannula position, change of the cannula position is not recom­ mended. Cannula insertion of the same vessel (usually the vein) may prove fatal ifnot imme­ diately recognized and corrected. The telltale

Figure 45-4. Accidental double cannulation

of the femoral vein during ECPR. A 44-years

old male with primary pulmonary arterial

. hypertension with cardiac arrest cannulated

",0n ICU. The arterial cannula was erroneously inserted into the femoral vein, complication was immediately identified by a presence of recirculation (bright red blood in venous limb) and corrected by insertion ofa cannula into the femoral artery. Two cannulas in the femoral vein were left in situ, attached by V-connector and used for simultaneous drainage of inferior vena cava.

sign is highly oxygenated blood in the ve.!!2us cannula due to complete recirculation. A new cannula must be inserted into an appropriate vessel (Figure 45-4). Rhythm Analysis, Conversion of Ventricle Fibrillation (VF) and Further Investigation ECPR patients on ECLS may present a vari­ ety ofheart rhythms. In patients with refractory ventricular fibrillation (VF) as a cause of CA, ventricle arrhythmia usually persists. It remains unclear when another trial of conversion to sinus rhythm should be performed. However, if the anticipated cause of refractory VF is an acute coronary occlusion or chronic severe coronary artery disease, PCl on ECLS should preceed defibrillation (PH and potasium correc­ tion should also be performed). Other malignant rhythms or other causes ofrefractory CA should be identified as soon as possible. The causes of refractory CA differ from causes of usual CA and, unfortunately, may often be irreversible (aortic rupture, other severe bleeding, intracra­ nial hemorage). It is straightfoward to continue with coronary angiography following institution ofECLS. Ifthe cause is not established with pul­ monary angiography or aortography, and other causes including pericardial tamponade have been excluded by bedside echo cardiography, then CT scan of the brain and chest/abdominal exams should be performed. Frequent laboratory examination is ofpara­ mount importance in ECPR Initial examination should exclude severe electrolyte imbalances, assess organ functions, and allow thorough monitoring, mainly by means of lactate, blood gases, and hemoglobin levels. Not only at ECLS start, but also the trend in blood levels oflactate values and base excess40 and other variables (hemoglobin) may correlate with outcome in ECPR. Blood gas monitoring every 15 to 30 minutes after cannulation helps manage the dynamic changes that can occur.

507

Chapter 45

Postresuscitation Care Monitoring

Studies on ECPR in mCA

Monitoring during venoarterial ECMO differs from that of other ICU patients. Blood pressure must be monitored invasively, due to nonpulsatility. Pulsatility is defined as a pulse JJressure over 15 mmH~.41 Right radial artery for blood pressure monitoring and blood gases examination is mandatory. In any interpretation of monitored data, it must be kept in mind that the final result comes from a combination of spontaneous and extracorporeal circuits and is influenced by the ratio of those two flows. Therefore, cerebral and peripheral tissue NIRS may help monitor not only the brain and periph­ eral perfusion, but also overall hemodynamic status and further identifY deteriorations requir­ ing hemodynamic interventions or other inves­ tigations.42,43 Patients with non pulsatile flow or with severely impaired left ventricle function may develop pulmonary edema or left ventricle thrombosis which may require left ventricle venting or intraaortic balloon counterpulsation, (see Chapters 48 and 64).44 Target temperature and protective ventilator management on ECLS is also a part of routine care.

Multiple observational studies (Table 45-5) report encouraging results of ECPR in IHCA with neurologically favorable survival in up to 35% of cases. 45-49 In an analysis of 135 IHCA cases (50), average CPR duration lasted 56 min­ utes, 56% received subsequent intervention to treat the underlying etiology, weaning rate was 59% and 34% patients survived to discharge with 89% having acceptable neurologic status.50 Patients were resuscitated in the ICU (61%), cath lab (16%), and emergency room (19%). Importatnly, this study also demonstrated a declining chance of survival with prolonged duration of CPR. The probability of survival was approximately 50%, 30%, and 10% when CPR lasted 30, 60, or 90 minutes, respectively. Similar results of increased short-term and long-term survival benefit over conventional CPR in patients with in-hospital cardiac arrest of cardiac origin treated by ECPR have been confirmed in a propensity analysis. 51 Therefore, based on available studies, ECLS for in-hospital CA has become a standard ofcare in many tertiary institutions, despite the fact that no randomized study confirmed its effectivity.

Table 45-5. Characteristics and outcomes ofselected ECPR studies in IHCA. Study, Citation Country Chen et aI . (2008)46 Taiwan

Lin et a1. (20 lOt" Taiwan Shin et aI. (2011 )48 Taiwan Chou et al. (2014]10 Taiwan Zhao et aI. (2015)47 China Blumenstein et aI . (2016)'" Germany

508

Design

N

Prospective

59

Prospective

59

Retrospective

85

Retrospective

43

Retrospective

24

Retrospective

52

Age (yrs) Male(%) 18·75

Time to ECLS(min) 60

59 85% 60 62% 61 93% 59 79%

40

NeurologicaUy Favorable Survival 42% 30% 30% 18% (33% overall) 24%

42

28%

60

35%

36

33%

33

2 1%

72 54%

Extracorporeal Cardiopulmonary Resuscitation

Studies on ECPR

ORCA

in Japan, Taiwan, Italy. Currently,

Use ECPR in OHCA may appear even OHCA patients more attractive than in are generally younger, previously healthy, and cardiac origin. the cause of CA is usually Sudden often caused by a single reversible condition contrast to polymorbid in-hospital patients suffering IHCA. On the other hand, prein the care for OHCA victims hospital plays a major in selecting suitable patients in are for Trials assessing 45-6. were observa­ listed in tional, not randomized, and mainly

(NCTO 1511666 and NCTO 1605409). studies suffered a high risk ofconfounding bias due to the observational design.29 Nonetheless, patients included were usually under years of age, no-flow were below 5 minutes and low-flow periods were variable mainly close to 60 HI1JjU ....," to 140 PrognosticaHy favorable indicators included witnessed arrests, shockable rhythms, and re­ versible causes Treatments provided in addition to included PCI, targeted tem-

Table 45-6. Characteristics and outcomes of selected ECPR studies in OHCA. Study. Citation

Design

~~:~

N

Country

Nagao et at (2000)71 Japan Nagao et at (2010)12

Japan Le Guell et at (2011)16 France Megarbane et aI (2011)13 France Avalli et aI (2012)74 Italy

Haneya et aI (2012)75 Germany ~~?ul et aI (2013)76

Belgium Leick et aI (2013)54 Germany Maekawa et aI (2013io Japan

Tazarourte el aI (2013)17 France Kim ct aI (2014)56

Korea Mochizuki el aI (2014)19 Japan Sakamoto etal, (2014)'7 Japan I Stub el aI (2014)

• Australia ~ang et aI (2014)7&

Prospective Prospective Prospective Prospective Retrospective

54 (70) 57·62 171 (83_91)b 42 51 I (90) 46 47 (77) 46 18 (94) 36

I

Retrospective

26

Prospective

14

Retrospective

28

Prospective

53

Retrospective

27

Prospective

55

Retrospective

32

Prospective

260

Prospective

11

Retrospective · Republic of Korea ,,~~•. or pane lS in ~dy l1 'outcome

with

.,1.

li~

66 120 155 78 70

48 (58)

58"

54 39 (56) 53 (75)

12%

=

4% 2% 5% 15% 22% 29%

MR (83)

Neurologically F'llVorable Survival ,

61

(65) 48

49

15%

14()d

4%

69

14%

I :

I

I •

51 (78)

84

56.3

16%

I

12%

(90)

20"

27%

31

50.7 (75)

68

26%

320

56 (81)

44

9%

• Taimm

I Choi eta!. (2016)"

Time to ECLS(mln)

in non-ECLS patients

509

Chapter 45 management, and intraaortic baloon counterpulsation. Outcomes in studies varied substantially with neurologically favorable sur­ vival from 4% to 29%, mainly around 15_25%.16,54 Recent ''''".'''..... ies on ECPR in OHCA states overall survival in 833 patients in 20 studies of 22% including 13% with good neurological recovery.29 How­ ever, ECPR in OHCA remains challenging with many issues patient populations, variables asS'OClatflll a favorable neurological outcome, cost-benefit analysis of this resource strategy, etc. Different to minimize no-flow and low-flow states include prehospitalllRlulJ,,, to transport ECPR deployment55 under high-quality ongoing CPR (ie, mechani­ cal) with ECPR commencement after llV,'IJUal admission. IS randomized trials will help to defIne the future role of ECPR in OHCA.Efforts to increase survival in refractory OHCA should outweigh the risk neurologically impaired survivors.5&.58 Of note, donation is an in ECPR ,..,,"r.,h,,~p

_ _~~;;;~~~~~~~~~" a Studies on

in the Pediatric Population

In children, the survival rate after CA ranges from 9 to 47% for in-hospital events and 0%-29% events outside the hospital. survival rate for pediatric was reported to 33-51%, which was than for con­ ventional CPR (see Chapter 27).59-62

ECPR as a Stepwise Approach to Refractory CA ECPR itself does not assure favorable out­ come in refractory CA. Connecting a patient to ECLS must follow a protocolized approach CPR, therefore several centers have developed policies encompassing mechanical CPR, targeted temperature control, and early 510

reperfusion: study. IS

---

Weaning from ECLS after ECPR

Weaning follows a similar path of all veno­ arterial ECMO Chapter 51); close neurological monitoring and prognostication must occur. Patients after prolonged CPR may cardiac and favor­ able neurological outcome may the efforts to proceed with more advanced mechani­ cal circulatory support or heart transplantation. patient still Thus, early extubation of an dependent on ECLS may be very helpful in this 64,65 In ELSO ECMO duration in has been reported to range to 60-70 hours,39

Ethical Issues in ECPR ECPR is associated with potential harm to patients and/or their relatives. Although recov­ ery of the often occurs, a coma or "p,rPT"_ tive state could result. 57 It has earned the title of a in which the patient t'J'~U.V. oxygenation has been recently described. 18,19 technique has potential to hasten correction and prevent development of compartment '''lnrt..",rnt>

Chapter 47

limb complications in patients with ECMO

(Figure 47-6). Our practice is to place a distal perfusion cannula in all patients, and if a limb complication is suspected clinically or by a uni­ lateral drop in tissue oxygen saturation 0Z' we ensure the cannula is not kinked or thrombosed. Ifthe issue cannot be resolved, we then take the patient to the operating room for placement ofa chimney graft or an alternative inflow position such as the axillary artery. We use a perfusion cannula as a first-line treatment instead of the chimney graft because bleeding at the suture lines of the chimney graft can be problematic. ECMO flow should be high enough to allow "cardiac rest" and maintain adequate hemody­ namics and organ perfusion (Target MAP 65-70 mmHg; SVOz >55%). Total cardiac output is typically 60 ml/kg/min in adults. It is important to allow some blood flow to go through the right and left chambers to avoid stasis within the cardiac chambers. Ventricular ejections are

Figure 47-6. Lower extremity ischemia moni­ toring in peripheral EeMO using transcutane­ ous near-infrared spectroscopy. Sensors are placed on the calf of both lower extremities for continuous monitoring oftissue oxygenation to allow early detection of leg ischemia. 528

confirmed by pulsatility on the arterial wave­ fonn, or by observlng aortic valve opening on the echocardiogran~. If a mechanical valve is present, then main$ining opening and closing of the valve avoids lthrombus formation. Echo­ cardiogram should I be done within 6 hours of initiating ECMO, if not done during insertion to assess the left ventricular decompression. ECMO unloads the right ventricle but does not completely unlad the left ventricle, even though the LV prel2.0 ng/mL) clinical signs ofinfection and a sensitivity of 90% and speci­ this retrospective --;--"'-----:----:-­ 66 and other have li7 shown this to be useful in pediatrics. Candida, resistant Staphylococ­ cus aureus (MRSA) and llV'''''a\_'vc...... H::rmentmg gram bacilli (Pseudomonas, ;:)IenOlTO­ phomonas) infections are increasingly common, and appropriately broad spectrum antlmllCrCIOIa UHJLLV..,,,L

blood flow

whereas Vl!!!£2~~~~~~!& prev,ention checklists recommended for insertion~~~~~~;; uC;U'''''''''', however, use of antibiotic prophylaxis is not recommended.72 optimal care, vasodilatory septic lJvi>lnLv

(NIRS) and meticulous clinical examination with calf circumference measurements an early also of value by of hypolhyperperfusion that may intervention. 4o,41 attempts at distal compartment syndrome may occasionally require fasciotomy and commonly limb am­ putation. 42 Other with cannulation occurred 18% patients (n= 101) and include lacerations, arterial dissection, pseudoaneurysms, stenosis, and lymphoceles, all of which require surgical intervention. 3 !

VU"HU,lL­

lnl'rp",,,pri

COlnpHClltlC)flS

and

a Lower Extremity Ischemia In patients with lower extremity arterial can­ nulation, meta analytic studies demonstrate that 1 7% develop ischemia, 9-10% and ;;...;:;.:;...::...:=:.=:..::===...__ Qutation.2 ,3,29 is higher in patients with peripheral vascular Ul.:>',",r.:>", which may in 4% over the age of 40 particularly runong patients are diabetic, hypertensive, hyperlipidemic, chronic disease, or are active diruneter cannulas EeLS may occlude anterograde necessary flow native contributing to Iscillenrlla. Vascular ultrasound can the femoral during

536

Acute Neurologic Events

(36-37° C), and optimization of

Adult Cardiac ECLS Acute Complication and Comorbidity Management

cerebral oxygen delivery is prudent. Cerebral blood flow can be monitored with transcranial Doppler (TCD) sonographY,79,8o although this does not monitor for effects of differential hypoxia.81 Differential hypoxia occurs when oxygen desaturated blood ejected from the left ventricle mixes with oxygen rich retrograde blood flow from the ECLS circuit resulting in the potential to deliver poorly oxygenated blood to the coronary arteries and ascending arch of the aorta. Also termed Harlequin syndrome, this phenomenon was shown in one study to occur in 8% ofperipheral cannulations. 82Right radialar­ terial catheterization and use of pulse oximett,y on the right hand may be the best ways to ob­ tain a surrogate estimate of ECLS-oxygenated blood delivered via femoral artery cannula to the cerebral circulation (see Figure 48-1 ).83 The addition ofcerebral NIRS to monitor changes in cerebral perfusion may provide further guidance in select patients (Figure 48-1).84

M_f ·~ filstion:

- EtCO·t

- Lung protective veofiIaHon

• Wf>M

,.0,

RIght Upper Ljrnb: • A8G ooIlecfioo

• P\w.!o""mllY

centrffugal PUinp: • Koop ECMO orcuil as sJ\O(J

00 pos,.;!)le

· Optim:zq; flow to mflintaln ($SUO pcrlo$ioo

wh'Je avoicfclg noga(tIlC ptOS&JrQ events.

~es=~~kt\kxl · ,*,ume status

• Myoear(!;': reco40 kg! m2) adversely effects all major organ systems and in particular cardiovascular, respiratory, and metabolic function. 101 •J02 Management in obese patient is complicated by difficulty with imaging due to fluoroscopy and ultrasound power, table weight limitations on CT scanners, unpredictable lipophilic drug availability, with berno(1ynamlc TYI.r.nit,win.... and vascular access. 103-105 prevalence ofobesity hypoventilation syndrome and obstructive sleep apnea with associated pul­ monary hypertension, elevated blood pressures, and increased biventricular ventricular mass and cardiac output further challenges to the management of the obese patient on Initiation VA-ECMO may require lTrlr,urn pr(]ice(lun~S for cannulas to achieve adequate blood higher blood flow rates to adequately support basal oxygen consumption, and careful attention to wound care and pressure points for preven­ tion ofdecubitus ulcer formation and infection. In addition, mechanical ventilation pressures and intraabdominal pres­ sures be and can native

cardiac venous return and venous drain­ age flows in difficult to predict ways depending upon a patient's active or ventilation pattern. 106 With consideration for chronic hy­ pertensive and diabetic microvascular diatheses, the MAPs may need to be (MAP >75 to perfu­ sion pressure and organ function. Optimization "Y'Vaf'Tl delivery may be aided by use of cerebral and limb mRS. I07 In addition, moder­ ate glycemic control with a target of 140-180 mg!dL (7.7-10 mmolfL) and avoidance of hypoglycemia is recommended.108.109 insulin infusions may minimize large mic variability may increased mortalityllO while avoiding subcuta­ associated erratic insulin subcutaneous tissues. With attention to the care of morbidly obese critically ill patients, it may be possible to improve their outcomes over time. lll ,lI2 affects 10% of adults and is fre­ quently associated with other smoking related comorbidities such as airway ",."va"." and CAD. lOO.1I3,lJ4 COPD is associated with airflow limitation, chronic airway inflammation, cough, and Goals of manmechanically COPD patients include respiratory muscles, mamtain:ing bronchial hygiene, avoiding hyperventilation, minimizing the risk barotrauma, and avoid­ ing impairment ofECLS flows that result from dynamic Initially use of short acting bronchodilator and anticholinergic nebulizers and sedation is required to avoid air trapping and hyperventilation. use of stelro1(lS and antibiotics to considered for acute exacerbations manifested by increased secretions and worsening airflow obstruction. 115 clearance techniques range chest physical therapy to intrapulmonary tion therapy to fiberoptic bronchoscopy, 1l6(Print though there is no clear evidence to support routine use in COPD exacerbations except in 539

Chapter 48

pro­

acute complications and chronic comorbidi­

longed exhalation and reactive airways becomes

ties. An approach to monitoring and managing

challenging as patients emerge from sedation and during weaning from Recognizing and managing hyperinflation (intrinsic positive end expiratory via waveforms on mechanical ventilator and end tidal monitor prevents aSS:OCllate:d C()mlpm:anons barotrauma, as increased work of 117 and hemodynamic compromise. ,1l8 Unfortu­ natdy, inhaled beta agonist bronchodilators are frequently associated with which can detrimentally increase myocardial oxygen demand., especially in multimorbid patients with CAD. This becomes particularly important peripherally cannulated patients with tial hypoxia where coronary arterial blood may be inadequately saturated (as described ousJy). Usual monitoring for cardiac ischemia is performed with serial cardiac serum and antithrombin therapy the usual as well as minimizing myocardial demand. Beta blockers may be considered in this situation; however, their pulmonary side effects may limit their use in multimorbidity.119 Insertion an IABP was pre:viously conventional therapy augmenting myocardial diastolic perfusion dur­ cardiogenic shock. This, however, has fallen 120 While out use of an IABP was not to improve microcirculation ECLS in a recent small study, it can decrease LV end-diastolic pressure and pulmonary capillary pressure. 121 In and evaluation of ongoing myocardial postcardllot:orrlY cardiogenic shock, daily creatine kinase isoenzyme MB (CK-MB) relative index may valuable. 122,123

the complex interactions between the patients' acute illnesses, chronic diseases, and is necessary to short and long-term out­ comes 48-1). As multimorbidity becomes ofcommon "'''Ttl"...", more common, may providers and ing complications.

"'.....,,"'" with bronchiectasis. us

Summary

Adult cardiac ECLS can support patients with cardiopulmonary compromise; however, careful attention must be paid to managing 540

Adult Cardiac ECLS Acute Complication and Comorbidity Management

Table 48-1. Approach to Monitoring under Creative Commons Attribution '-'''JVU'''''-'.

VA-ECMO patient. (Permission to use granted

Monitor for Dysrhythmias such as ventricular fibriljation that lIlliy prevent ventricular

Rhythm

Treatment Carruoversion

I MAP J1UL~ 60 in the or of catecholamines, and should major metabolic bances. Third, pulmonary function should not impaired. IfPaOjFiOz .

•

ECMO cart must be placed near the patient bed on brake position. • ECMO system must be "",.,.",,,t1, nected to a secure plug red power outlet). All devices have a light indicating ifECMO is plugged in or on battery. • gas module must be securely connected gas to the gas sources (waH or module has to be safely connected to the oX'1{genat4:)r gas inlet port without any kink kinks and tenslOlOS on whole circuit (cannulae, and power hoses) must be checked. • Must check and ensure security of the cannula(e) and check placement with and reviewed with the aid of

MDT.

561

~"UVtt::f

50

A thorough look of all circuit components with a flashlight, looking for thrombin and/or clots. The more complex circuit "nlll""'''' pigtails, stop cocks), the higher the risk to develop and clots. 41 Check the security of all connectors and the presence of tie bands in tbe appropri­ ate places. 41 heater has to correctly connected to the heat the water has to full and the temperature set point adjusted to therapeutic goal. 41 Write down of ECMO sup­ port to aid guidance and decision making for optimal patient management. 41 The circulatory the Rotation Per Minute (RPM) the blood flow. The therapeutic is the blood flow. The ECMO specialist in charge of the pump has to control evolution blood flow in correlation with the RPM. 41 The flow rate must he high enough to provide adequate perfusion pressure and venous oxyhemoglobin saturation but low enough to preload to maintain left ventricular output. 41 The ventilatory the sweep gas flow should adjust regarding the and the Fi02 regarding the Pa02 41 The pressures: some are mea­ the venous pressure, the arterial pressure, and the but most ECMO teams add pigtails connected to pressure monitors. circulatory param­ the absolute pressure itself is not relevant but its evolution through time. 41 pump head must be placed and the ECMO specialist must check for disengagement. 41 The alarms: must be set your therapeutic Since the pump is nonocclusive, we recommend to maintain the blood flow above 2 Llmin to avoid any back specialist must know on which mode your ECMO is working.

In a mode, when an alarm is activated the circuit will working but when on Intervention mode, as soon as an alarm is set on, the pump stops working and an im­ mediate action must be set to resolve the problem. The choice of the depends of your human resources. If a nurse or a perfusionist is constantly at the patient bedside the intervention mode is possible, but if a nurse is taking care of more one ECMO patient and cannot intervene immediately when pump stops, the free will be safer.

41

0

562

41

The equipment needed in case of an emer­ gency: clamps, emergency hand emergency supplies at the bedside must be available; which include appropriate sized connectorslshears/tubing/rapid access line, fluid, tie-gun and Ttp,_"h'nc/,,'tp1"1 etc.

Vigilance in the circuit atten­ tion to detail by the ECMO specialist is of para­ mount importance. communication and escalation to the Duty ECMO consultant! Duty ECMO coordinator must be in a timely manner to ensure timely decision ......." ..,5.

Preventing Complications is potentially life-saving for the patient in severe cardiogenic shock refractory treatment, but it comes with to One points of ECMO management includes preventing and detecting complications. The ECMO role is an extended role and staff must be adequately/ex­ pertly trained to recognize the signs and compli­ cations of and mechanical ofECMO Specific and ongoing assessment is crucial to ensure clinical and skill is to the highest standards to minimize these risks Chapter 67). Attention to detail, vigilance ill and implementation of care in a

HUllUJJ",

Management ofAdults with ;anllO\las,':Ul/zr Disease on Extracorporeal

timely manner are paramount, along with effec­ tive communication and escalation to the Protocols team in the event of procedures must be in place to support aU team members in order to alleviate stress and improve practice in this highly charged envi­ ronment. between the ECMO coordinator, consultant, and specialist is unique. Together the team approach must verbalize and implement patient management goals, which include and assessment of daily patient! circuit parameters; for example, maintaining cardiovascular fluid mana)gelJtler:tt, manal$ellilertt, and guidelines. Commu­ nication must be concise, and consistent and shared with all members ofthe MDT. This will enable all team members to have a clear understanding of patient management. The nursing care of the patient must be individual, goal directed, and holistic.

Support

ECMO the anti-factor Xa can also be a better indicator ofthe heparin uw..ual"'....lJll.'1n"t> of adequate mean arterial pressure, (Cl), and (:::::30 .~.~.~~o/ during conditions inotropic support is ofhemodynamic improve­

ment in addition to decreases in the pulmonary

wedge pressure (PAWP) and the

venous (CVP) with with the

baseline values at the time of the

ofVA-ECMO. Therefore, the of the clinical signs of cardiac recovery and adequate assessments hemodynamic stability should the timing of weaning from VA-ECMO. For these reasons, VA-ECMO should stopped at the appropriate time based on both clinical condition of the patient and the indications for device implantation. With evidence of improved aortic pulsatility and contraction on echocardiography, the inotropes should be optimized and the flow reduced to 50% then 25% of the adequate cardiac output.

function are present, the reduction of the pump flow by approximately 50% can attempted. In conditions of lower circulatory support, cardiac output (CO) ;::;UJLH"-"'"lH to maintain end-organ hemodynamic stability, car­ diac function does not worsen, and neither LV dilatation nor mitral regurgitation occur, then 570

the circulatory support can be further decreased to 25% of the adequate COY Echocardiography in the Weaning Process Because contractility and opening (AV) are most important myocardial recovery, the use of echocardiogra­ phy with the transthoracic (TIE) and ...,".... v .. "Ul

oaulental to Although no standardized echocardiographic protocols from available, some echocardiographic parameters have recognized as good pre­ dictors ofsuccessful weaning. 3,7 The most used we:allltnil "tt••tp(]ru is on the reduction of the pump flow (at a minimum level of 1/ min for not more 30 minutes to reduce the risk ofthrombosis ofthe circuits) while clinical, hemodynamic, and assess­ ments are performed to evaluate the cardiac and the suitability of the discon­ from support. native LV recovers and improves in contractility, the reduction of the pump flow is followed by an in the LV ejection fraction (EF). Ac­ cording to the Frank-Starling law, as LV filling increases to the reduction of pump the improved LV contractility allows AV and of blood through outflow tract (LVOT) when the systolic pressure surpasses the aortic pressure. In ad­ to evaluations the LV EF and the LV echocardiographic mCllc .... , " '. . . "."

Factors Determining Neurologic Outcome

Neurologic injury on a variety of factors including hypoxia, acidosis, hypotension, and low cardiac output resulting in an undersupply of the brain play an important outcome. Also, neurologic diseases have a key impact on neurologic outcome after as preexisting pathologies to be ag­ gravated by extracorporeal support in addition to the necessity and influence of intcnsive care

medicine. It is deployment For example, ECLS cal cardiopulmonary resuscitation is associ­ ated with a higher risk for neurologic injury in comparison to an elective and controlled implantation in a patient with spontaneous cir­ culation, as mechanical chest compression and achieved compromised circulation per se are with injury.! The cannulation basically rl.i-I·o,.""i-. between venovenous (VV) and venoarterial (VA) as well. It is generally IS with a a VV vv• .LLLE, .... '''U~.ll protecting the central nervous terestingly, only minor differences are reported between support modalities (see l1tble according to ELSO Registry. For V.ru:t'.... JIV.the rate death in 2015 in the VV population comprising 1568 runs was 1.9%, and 2.9% 1769 VA-ECLS runs. This underlines the fact that neurologic during has multifactorial causes (U,",''''''' ing thromboembolic and altered coagulation cascade to the most prominent ones. In contrast to ELSO data, there is also evidence, according to the general believe and daily practice, that VV-ECMO is as­

,--,,,,,;u,e,

52

sociated with a lower risk for neurologic injury.2 VA support may be implanted using peripheral or cannulation techniques. Central can­ nulation can be differentiated in direct aortic arterial cannulation and cannulation of the sub­ clavian or or without the use These different cannulation techniques may have an impact on neurologic outcome, but it has never been investigated in a manner in ECMO Therefore, one can only assume that the peripheral cannulation strategy might be associated with a lower for injury in case of residual left ejection, since the retrograde aortic ECMO flow the brain supraaortic only partially. On the other cannulation the carotid which is mainly the neonatal field but also in adults, is with risk for neurologic injury.3 Along with the evolving in­ dications for ECLS, the support and monitoring technology has changed dramatically over the past 30 years (eg, centrifugal pumps, dual lumen cannulas, poly-methy l-pentene ",n",,,,on biocompatible and one can assume a benefit from the technological improvements.

Incidence ofNenrologic Injury The true incidence of neurologic compli­ cations is underestimated due to different fac­ tors including the difficulty to obtain reliable

imaging in critically ill patients during support in addition to the lack of participation of international Patients who sustain injury have an increased risk of death even without imaging. 4 Therefore, the number ofunrecognized can be to be Rastan et aL published intriguing autopsy findings in 78 ECMO patients or 50% ofthe ECMO population. He found in (79%) prior unrecognized causes of death including clinically u.!llliagnosed cerebral jnfilr.ction...in 7 patients (8% V The rate of undiagnosed and ru;t before postmortem examination evaluated thromboembolic events was 30%, there were major discrepancies between clinical and oo:,tmlort:em causes and is poorly ated due to the need for anticoagulation during ECLS support. Most common hemorrhagic injuries include intraventricular, intracerebral, subdural Unfortunately, is with a high mortality of approximately 80-90 % ac­ to the ELSO Infarction may be small or emboli or to vessel abnormalities ofthe The incidence ofimage documented and reported infarction is roughly similar to that of hemorrhage (see Table 52-2). ~~:-;:;

Table 52-1. Incidence of neurologic complications. Comparison ofthe years 2006 and 2015 data for adults).

576

Nel'lrO,(OfY,lC

.8% in comparison to 2.2% in the VA group, while cerebral infarction is slightly higher in the VA group 3.8% in comparison to 2.0% in the VV group. The ELSO ECMO runs and corresponding outcome rates for more than a quarter century the year starting with less than 20 runs a year, now reaching 1500-1800 international adult runs a year. The annually complication rates are the beginning. Notably, is a tion neurologic complication rates over years, which can marker therapy, manal;enllenlt, and also improvements. Another group of recorded neurologic complications are sel:i~Uf(~S on they occur f',.prn~~r,;;: a result of thromboembolic or hl",,,,£1'"'1(' LJ'-'LU.''-J

• ....,.,LLLUV. .L

and Pulmonary CO)'1IlJj'icati01'1S

retically gas exchange can be managed totally by extracorporeal machine, and some cen­ ters do follow practice. Acute injury leads to pulmonary fibrosis, particularly if the patient on high high oxygen VLunu,,,.>u for several has the ity to provide gas ...."'.~H'lHl':," to organism as as eliminating CO2 , but on the otber hand is also associated with complications. are a variety pulmonary complications all, an ECMO patient is no "',.,'""",.. ",1"1-""1'.0."1- than other patient requiring extended time of care medicine long-term mechanical ventilation . on top However, reqUlnng care medicine mechanical on inotropes can perceived as the sickest Pulmonary complications have a torial three me'cbamsm likely proposed. cations can be the £""'~""',fI,,n£",. and probably unrecogmzeo can easily because adequate oxygenation and pUlmonary perfusion on support. A pulmo­ nary complication after can evolve as "£">'1""'''"'''"(>#> of the cardiogenic shock ''''auw,!'. to congestion. 1be ",;'1,'£11£\_ shock leads to a multi organ with U1 ...C1 • .L1UV

rate 20-40% comparison to popUlation with a survival to discharge rate of40-60% regardless of support A""'UUlUY

"'"".f'"",..., and Incidence of Pulmonary Complications Patients on ECMO support are somehow unique care since theo­

Table 52-2. International summary of neurologic cmnpJllca,tio:as and corresjpolJldirlg survival Registry Data for Adults from 1 'c Complications

VA 7850 runs n

%

Survival

o

354

4.2

0

47 42

50 135

0.6 1.6

20 24

30

322

3.8

23

21

184

Cf=computer tomography

9

0

Chapter 52

impaired liver and kidney function with a cer­ tain negative impact on the pUlmonary function. acquired lung injury during extracorporeal support, like infection due to a compromised to fluid immune and congestion overload, can lead to pulmonary complications. Secondly, patients on ECMO usually reblood products due to bleeding as a consequence of cannulation difficulties or altered Daily with common in ECMO on pulmonary functionP Thirdly, VA support leads to pulmonary shunting and reduced transpulmonary blood 8 Pulmonary blood flow is inhomoge­ neously to pulmonary pathologies (atelectasis, carnification, fibrosis of etc.) and as a of LUvvUGUl cal This leads to underperfused pulmonary inducing pulmonary sions on top of the existing Koul et have shown in a animal model that total venoarterial bypass for longer periods of time, up to days, produces pulmonary damage with interstitial intraalveolar and parenchymal necrosis. 9 same authors have shown that maintaining pulmonary blood flow reduced rate parenchymal lesions. intllarr:umltory illleOlawrs in the animals undergoing VA support were seen when compared to VV sup­ port. It is noteworthy that all animal lungs were healthy_ However, these that prolonged pulmonary shunting on VA-ECMO support might have deleterious consequences on lung. In addition, there are some unique Table 52-3. Incidence of P"'''~'VLX"'J 2006 and 2015 (ELSO

complications for EeMO patients. One is the higher risk for pulmonary embolism caused by thrombotic material in and around the venous cannula. The risk for pUlmonary em­ bolism persists after weaning EeMO as mobile clots remain in the large prior-cannulated veins. Another rather unique feature of EeMO paon venoarterial support is left ventricular distension in case of very poor left ventricular function with no and pul­ monary corlgestl()fl It is rather impossible to describe the inci­ dence of the above mentioned ECMO related complications, since these complications are not registered in a manner. Even the ELSa cannot provide incidences of pulmonary infection on EeMO, pulmonary embolism, and so forth. provides incidences Hn'A/p.'lIpr

pneumothorax is 9.1 % in the VV population compared to 1.8 % in the VA population. Similar are found in of pulmonary (VV: the 6.3% vs. VA: 3.1%). The diseased can easily explain these differences. VV patients suffer from a diseased lung with a higher risk pneumothorax and hemorrhage. Looking at development of pulmonary complications one can note a clear reduction over the last decade (see Table 52-3).

IvVLllfl'1IvCl'ClVlJ''''.

VV 12lNA 141

Comparison of the years

VV 1568NA 1769

runs

Neurologic and Pulmonary Complications

Return to Work Expectations and Quality ofHfe Longitudinal outcomes after ECMO are important but mainly available for neonatal and pediatric patients and not adults. Functional neurologic outcome cannot be obtained from the ELSO Registry, most reports on functional outcome derive single centers over a short period. IO•I ! However, the ELSO Registry offers a hint about the functional status after ECMO due to the registered survival to discharge rate. The survival to discharge rate depends to a vast extent on the underlying The highest survival to rates are in pa­ tients with an acute lung failure following fulmi­ nant viral infection (eg, HI N 1) and VV support with survival rates of70-80%. There is evidence that this patient population has a near normal after ECMO 10,12,13 The quality of survival rates are for patIents ECMO cardiopulmonary resuscitation with survival rates of 25-35%. It is also important to point out that the survival to rate is generally lower in patients experiencing a complication in relation to '-''-,iH'V support. For example, patients on VV support developing a cerebral infarction regard­ less of diagnosis reach a survival to discharge rate of 30%. This did not significantly change over the years. However, it is difficult to extrapolate the return work rate after ECMO support based on the simple survival to discharge rate. A recently published review by Cochrane CoUabora­

records vH~"!LI''', tion found only 13 out of of which only 5 studies were included in the review including only 4 randomized controlled trials reaching only a patient cohort size of389 patlent:s. Two out of the randomized con­ trolled trials were conducted the Clinical heterogeneity across these studies prevented pooling data for a metaanalysis and on long-term outcome for adults is included trials was cally One of the trial which did not find a significant difference health related quality of life months after study randomization, nor did it show a clear survival benefit at 30 days and 6 month. 13 parenchymal on of fibrosis and high resolution CT minor pulmonary function abnormalities remain common and can be more than VV-ECMO. Furthermore, most patients ","vr'''''MPnc'''' a in quality of life due to pulmonary to Linden et alY Whether these apply also VA-ECMO patl,ents clear, but they can be highly assumed. our in­ This lack of evidence stitution at the University Medical Center KegerlSl)urg to evaluate the long-term outcome of our patient cohort comprising of more than 1000 patients. So far only of the VA-ECMO population our institution are available. Out of 465 VA-EeMO IJU.""Uo,,;) institution, including out of and !ll-llV"'" resuscitation cases, a survival rate to discharge on7% and a 2 year survival of45 % ofpatients rea,CnJlllg survival to after ECMO was IAnfi"_n"1"Tfl .,,",......., . .

Table 52-4. International summary of pulmonary complications and cor­ responding survival Registry Data for Adults from 1991-2015). VV (9102 runs)

Pulmonary Complications

VA (7850 runs)

I Survival % •

n

0/..

Survival %

n

0/0

I Relevant Pneumothorax

846

9.1

46

150

1,8

36

• Pulmonary Hemorrhage

588 , 6.3

39

,262

3.1

26

VV' '.vHvifenous; Vh,­

'IOHValLtbe:stst 2011 ;60(7):647­ 652. 24. Menut R, Larrieu N, Coni! JM, Georges Fourcade 0, Geeraerts T. Use of ECMO in a traumatic brain injured patient with severe hypoxemia. Ann Fr Anesth Reanim. 2013;32(10):701-703. 25. Muellenbach RM, M, Kunze et al. Prolonged heparin-free extracorporeal OX'lrgenatlon in multiple injured acute respiratory distress syndrome patients with J Trauma Acute Care 2012;72(5):1444-1447. 26. Ke Lin CY, Tsai YT, et aI. Increase the donor pool: transportation patient with 598

fatal head injury supported with extracor­

poreal membrane oxygenation. J Trauma. 2010;68(3):E87-88.

55 Tra][lSport of the Patient Supported with ECMO Bjorn Frenckner, MD, PhD, Wesley A. McKamie, RRT, CCP, Richard T.

From time to time need arises to move a patient supported with mem­ brane oxygenation (BCMO), either within a facility order to obtain a particular diagnostic or therapeutic intervention as cardiac catheterization or CT scan), or between fa­ cilities. of a patient on ECMO may be necessary in in which the facility does not provide ECMO services, or because the patient supported with ECMO requires other specialized not available at the such as to heart or transplantation. While transport of patients on ECMO has been reported since the earliest days of extracorpo­ support, relatively centers have experience inter-hospital ECMO transport. This hopes to provide informa­ tion on the history published outcomes ECMO transport, as well as practical informa­ fur~u'~V'~~L'U' tionon mechanically

Inter-Hospital Transport History and Published Outcomes Bartlett and colleagues in 1977 published the first report two pediatric patients trans­ ported between hospitals while supported with

MD

ECMO.! shortly after published successful use ofECMO by Hill colleagues. 2 The team at Wilford Hall U .S.AP. Medical Center published in 1991 the successful transport, using military aircraft, of 12 pediatric patients supported with ECMO over distances as great as 1400 miles. 3 From the late 1980s until recently, bulk ECMO ex­ t'\"""1PT'l'P in the U.S. had been reported by teams from the University of Michigan,4,5 Wilford Hall,3,6,7 Children's 8-10 and Columbia University Medical II From Europe substantial ECMO experience was reported by teams from Sweden,12,13 Ger­ manY,14,15 France/ Ii,17 and in Asia by Taiwan. 18 Reported survival rates of patients transported with support have essentially equaled whom ECMO was initiated house" and compared to international outcomes reported to the Extracorporeai Life Support 5,9,13 Although Organization the initial and atric patients, over the last two decades all groups have been successfully transported on in ECMO. In recent years, however, ECMO support ofadults with acute, refractory TP"t'\ITj~t(),'V failure has further increased follow­ ing the publication of the United and following TPn,(\rt" otsiUCi:leS~>ful ECMO support ofpatients with ARDS caused by pandemic HINl infiu­

599

Chapter 55

enza?(}'28 A number of these

involved

patients successfully transported to a regional

already on BCMO is referred to as a ~~;w

ECMO center

T!]psport, Obviously, centers with the ability to perform Primary Transports also can do Second­ ary Transports and we will therefore focus on Primary Transports

on ECMO.21.24,26,29.39 This

l1"..",.",cr

m

severe £1.",-'-"'-' with the introduction clinical use ingly compact, lightweight ECMO pumps and circuitry, which will be discussed below.

Indications for Inter-Hospital ECMO Transport The most common indication for transport­ ing a patient supported with ECMO is the need to move a patient from a center that not provide ECMO to an ECMO center. Conventional transport of these unstable criti­ cally ill patients may and have been described. 19•31 High frequency oscil­ latory ventilation (HFOV) and inhaled (iNO) may complicate a conventional Consequently, are often can­ nulated by the team at the hospital and this is referred to as Transport. n - Another common reason for inter-hospital transport of an ECMO patient is the need to move the already on to a cen­ ter with certain such as cat'di~lC or transplantation. Some centers have ability to initiate ECMO treatment in an emergency situation but not to continue for the whole treatment period. These patients may thus be to an established ECMO center. Occasionally, a patticular ECMO center may it necessary to transport a patient to another center if the number ex(:ee:>v:>:>lll'CUl of the patient, cannulation and ofECMO, preparing transport etc.), and 3) transport to the ECMO facility. In most instances transport starts with a phone call and a decision is made to launch the team patients who fulfill ECMO criteria. Once the decision is made, team members are contacted, equipment is packed, and transport vehicles are organized. The main priority to reach the referring hospital is to minimize time in order to secure the patient, while the dominating priority when transporting the patient to the ECMO center is patient safety. Time during the latter part is ofless importance. L . . . . ' ..... " J U

Table 55-1. Properties of ground ambulance, helicopter, and fixed wing aircraft (from ELSa guidelines32 ). (Courtesy ofCarl Chipman, R.N., Arkansas Children's Hospital)

Transport ofthe Patient Supported with ECMO

Most often the same vehicle travels to and from the referring hospital. However, on occa­ sion, such as when the desired vehicle remains unavailable, it may be quicker to reach the referring hospital with another vehicle (ground emergency vehicle, taxi plane, etc.) not capable of patient transportation. Local circumstances at different centers may also playa role. Several factors together dictate the choice of transport vehicle for inter-hospital ECMO transport, including distance, weather, team composition, and vehicle availability. Each mode of transport has unique advantages and disadvantages (Table 55-1). Transportation with fixed wing aircraft also requires ground transportation between hospital and airport at both ends. In hospitals equipped with helicopter pads the team loads the patient directly into the helicopter, obviating ground transporta­ tion. The ELSO guidelines recommend ground transportation for distances up to 400 km (250­ 300 miles), helicopter for distances up to 650 kIn (300-400 miles).32 Fixed wing aircraft can transport any distance. At Arkansas Children's Hospital, most ECMO transports have been accomplished us-

ing a helicopter, though fixed wing aircraft have been used for longer distances and occasionally ground ambulance has been used for transport within the state ofArkansas. Currently, Arkan­ sas Children's Hospital uses the Sikorsky S-76 aircraft for helicopter transport and a Beechcraft King Air for fixed wing transport. Aircraft used by other centers with ECMO transport experience may range from small jet planes like Cessna Citation llD to massive military transport aircraft. 3,7.12 In recent years, increased use ofcentrifugal pumps with smaller footprints has resulted in more compact transport ECMO circuits, allowing for use of smaller aircraft in some circumstances. One company currently markets a very compact, lightweight (10 kg), portable cardiopulmonary support system with self-contained membrane oxygenator, cen­ trifugal pump, and arterial and venous pressure monitoring capability. This system has been successfully used in air transport of ECMO patients in relatively small aircraft. 35 Any air­ craft chosen for inter-hospital ECMO transport must have doors that allow easy patient load­ ing and unloading (Figures 55-1 and 55-2) and mechanisms for quickly and easily locking the transport ECMO equipment to the floor. All ap­ propriate transport vehicles, ground or air, must

Figure 55-1. Arkansas Children's Hospital transport ECM0 sled is about to be loaded into the Beechcraft King Air Aircraft. Note that all equipment is mounted on the sled and that it is all in one piece.

Figure 55-2. This patient has just been unloaded from a Cessna Citation II aircraft. Note that the ECMO cart is separate from the stretcher and is standing on the ground. The ECMO physician is holding the tubing between the ECMO machine and the patient.

Transport Vehicle

601

Chapter 55

have an

source capable

standard

(110 V or 220 V), 60 cycle power to ECMO equipment as well as oxygen sup­ ply (other than transport oxygen cylinders) and must also pO!;Se1;S climate controL Personnel Different centers have very different com­ teams. position of their ECMO may depend on competencies in dif­ ferent professional groups, traditions, and other local circumstances. For example, in most centers perfusionists prime circuit, while in other centers is by a nurse (RN.) or a doctor with special education. The can be a (R.R.1), nurse, or intensivist. Most often a pediatric or cardiothoracic surgeon cannulates patient, although some centers reported percutaneous cannulation pediatric ECMO outside of the inter-hospital transport setting. least one case report de­ scribed percutaneous ECMO cannulation prior to ECMO transport.38 However, most published ofpre-transport percutaneous ECMO can­ nulation have been in adult patients?9-42 When configuring the transport responsibilities of each member should be clearly defined, including the capability to make HUUULIVU U'""vh)JlVU regarding candidacy for and mode 0fA, VV) of ECMO, cannulate the patient, the ECMO circuit, initiate ECMO treatment, sta­ bilize the patient, check cannula positions with ultrasound and/or radiograph, and to safely patient to ECMO facility. The team should also be prepared troubleshoot­ ing when unexpected difficulties arise. All responsibilities must be met by experienced will be no backup. For sec­ the patient already is on COIDPletelr1Ce for cannulation obviously does not have to be included in the team.

602

On

at the

institution, the

team must assess the patient and review labora­ data, radiographs, and any other pertinent clinical data. While in most it Oe(:onles at of the that the pais an ECMO occasionally patient deteriorates before the transport team eCtea

607

Chapter 55

The space is limited (Figure 55-5) and even minor patient or circuit procedures be difficult to accomplish.

Intra-Hospital Transport ofPatients Supported with ECMO Before leaving the topic of transporting patients supported with ECMO, it is important to address the less glamorous but still vital topic of intra-hospital transport. The need for ECMO patients to undergo diagnostic or therapeutic in­ terventions outside of the ICU arises with regu­ larity, and the ECMO team must be prepared. 5G-54 Such a transport should be viewed as an inter­ vention with potential risks and benefits, but can be perfOlIDed safely without complications. 54,55 The perceived high risk of moving an ECMO patients could potentially lead to reluctance to transport the patient for an intervention such as a computed tomography (CT) scan, and thus to a delay in diagnosis or therapy. In order to move a patient supported with ECMO safely within the hospital (eg, for CT scan, to the cardiac catheter­ ization suite, to the operating room), the team must have a pre-defined process that includes checklists of necessary equipment, personnel, and time required for such a move. Prodhan and colleagues reviewed the experience of the Arkansas Children's Hospital ECMO team over a 1O-year period in intra-hospital transport to evaluate the hypothesis that intra-hospital transport of ECMO patients is associated with clinically important diagnostic and therapeutic interventions. 55 During the time period studied, out of a total of 471 ECMO patients, 37 patients required 57 intra-hospital transports (37 trips for cardiac catheterization and 20 trips to CT scan). In the majority of patients transported for cardiac catheterization, a finding was identified that was not noted on echocardiogram and/or a management change (including cardiac surgery) occurred due to the information obtained during catheterization. CT scans of brain, thorax, and abdomen have also shown to have great c1ini­

608

cal value in many cases, frequently impacting

treatment. 53,55,56 One may conclude that intra-hospital transports ofECMO patients require extensive planning and logistics and a well-defined pro­ cess but often yield diagnostic information or therapeutic intervention crucial to patient care. Conclusion Transport of a patient supported with ECMO is technically feasible and reproduc­ ible by a number of centers for over two de­ cades. Successful ECMO transport requires careful planning and coordination of a skilled multidisciplinary team as well as clear com­ munication between the referring and receiving institutions. Recent advances in the design of equipment used for extracorporeallife support make ECMO transport more feasible than ever for an experienced ECMO team. ECMO is growing worldwide with an increasing demand for transports on ECMO. It has been suggested that networks of hospitals should be created around each ECMO center located in tertiary referral hospitals and that each such network should create mobile ECMO teams for retrieval of patients,57 which currently is the situation in Great Britain (see chapter 66).42

Figure 55-5. Patient on ECMO inside a Cessna Citation n aircraft. The environment is not only noisy but also cramped.

Transport o/the Patient Supported with ECMO

References

1. Bartlett RH, Gazzaniga AB, Fong SW, feries MR, Roohk HV, Haiduc N. Extracor­ poreal membrane oxygenator support for cardiopulmonary failure. Experience in 28 cases. 'The Journal thoracic and cardio­ vascular surgery. Mar 1977;73(3):375-386. 2. Hill ill, O'Brien TG, JJ, et al. Prolonged extracorporeai oxygenation for acute post-traumatic respiratory failure syndrome). Use Bramson lung. The New England journal ofmedicine. Mar 23 1972;286(12):629-634. 3. ComisbJD, Carter 1M, GerstmannDR, Null DM, Jr. Extracorporeal membrane oxygen­ ation as a means ofstabilizing and transporthigh neonates. ASArO transactions IAmerican Society for Artificial Internal Organs. Oct-Dec 1991;37(4):564-568. 4. DS, Pranikoff et al. .."''''''' of 100 patients on extracorporeal life Asaio J. NovDec 2002;48(6):612-619. 5. Bryner Cooley Copenhaver W, et al. Two decades' experience with interfacility on membrane oxy­ geriatlon. The Annals of thoracic ""...~"'... Oct 2014;98(4): 1363-1370. 6. Coppola CP, Tyree M, Larry DiGer­ onimo R. A 22-year experience in global transport extracorporeal membrane oxy­ geriation. Journal of pediatric surgery. Jan 2008;43(1 ):46-52; discussion 52. 7. Wilson BJ, Butler TJ, R. A 16-year neonatal/pediatric extracorporeal mem­ brane oxygenation transport experience. Pediatrics. Feb 2002;109(2):189-193. 8. Cabrera AG, Prodhan P, MA, et al. of children Interhospital ing extracorporeal membrane oxygenation support for cardiac dysfunction. Congenital heart May-Jun 1;6(3):202-208.

9. Clement KC, Fiser RT, Fiser WP, et al. LLlCH.HU'.'VH experience with lnr,F'rnnQ_ pital extracorporeal membrane oxygenation transport: A descriptive study. Pediatric critical care medicine : a journaJ of the Society of Critical Care Medicine and the World Federation ofPediatric Intensive and Critical JuI2010;11(4):509­ 513. 10. Heulitt MJ, Taylor BJ, et al. Inter-hospital transport of neonatal patients on extracorporeal membrane oxy­ genation: mobile-ECMO. Pediatrics. Apr 1995;95(4):562-566. 11. Biscotti M, Agerstrand C, Abrams D, et aL One Hundred Transports on Extracorporeal Support to an Extracorporeal Membrane Oxygenation The Annals ot1:holracLC surgery. Jul 20 15;100(1 ):34-39; discussion 39-40. 12. Reinhard J, et at Interhospital transportation of patients with severe acute on extra­ corporeal membrane oxygenation--national and international experience. Intensive care medicine. Oct 2001 ;27( 10): 1643-1648. Palmer 13. LM, '''''Lwn~'''. B. The Stockholm experience: Inter-hospital transports on extracorpo­ real membrane oxygenation. Critical care. 2015;19:278. 14. Rossaint R, Pappert D, Gerlach U/"""n"'''''''"V1 K, K. Extracor­ poreal membrane oxygenation for trans­ patients with severe port of ARDS. British journal Mar 1997;78(3):241-246. 15. Schopka Philipp A, Hilker M, et al. Clini­ cal course and long-tenn outcome following extracorporeal life support­ of patients facilitated interhospital with circulatory Resuscitation. Aug 2015;93:53-57. 16. Beurtheret S, Mordant P, Paoletti X, et al. Emergency circulatory support in refractory

609

LTtUiV',"'

55

cardiogenic shock patients in remote insti­ tutions: a pilot study (the cardiac-RESCUE program). European heart Jan 2013;34(2): 112-120. et Out­ 17. Roch A, Hraiech Masson come of acute respiratory distress syn­ drome patients treated with extracorporeal membrane oxygenation and brought to a center. care medicine. Jan 20 14;40( 1):74-83. 18. Huang YS, ation for adult shock Artificial organs. Jan 2006;30(1 ):24-28. 19. Peek Mugford M, Tiruvoipati R, et at and assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet. Oct 172009;374(9698): 1351-1 20. Turner DA, Rehder chael et Extracorporeal membrane oxygenation for severe refractory respirato­ ry failure secondary to 2009 HINl zaA. Respiratory care. Ju1201l ;56(7):941­ 946 . . Patroniti N, Zangrillo Pappalardo et al. The Italian ECMO network experi­ A(H1Nl) ence the 2009 pandemic: preparation for severe respira­ tory emergency outbreaks. Intensive care Hlv"""'UJlv. Sep 2011;37(9): 1447­ 22. Nair P, AR, J, et al. Extracor­ poreal membrane oxygenation for severe in pregnant and postpartum women during the 2009 HINt pandemic. Intensive care medicine. Apr 2011 ;37(4):648-654. 23. Cianchi Bonizzoli M, Pasquini A, et ECMO treatment of HI N 1­ Ventilatory induced severe respiratory failure: results of an Italian ECMO center. BMC pulmonary 2011; 11 :2. Holzgraefe B, M, Kalzen rad D, Palmer K, Frenckner B. Extracorpo­ 610

membrane ox~(genaTlon HINl 2009 respiratory Minerva anestesiologica. Dec 2010;76(12): 1043­

1 1. 25. Kumar A, Zarychanski Pinto R, et aL ill with 2009 influenza A(HINl) infection in Canada. Jama. Nov 42009;302(17): 1872-1879. Davies A, Jones Bailey M, et aL Ex­ tracorporeal Membrane 2009 Influenza A(HINl) Acute RespiraDistress Syndrome. Jama. Nov 4 2009;302(17): 1888-1895. 27. Zangrillo Biondi-Zoccai et Extracorporeal

genlat1Cm (ECMO) in " 10%

ture neutrophils

that define many of these parameters with age and can found in consensus statements ofvarious 100) • Has had aggressive fluid replacement and other pharmacological strategies described by the ACCM consensus statemenf • Continues to deteriorate with worsening hypotension, rising lactates, or rapidly progressive multiorgan dysfunction despite treatment The speed at which ECMO can be initi­ ated is institution-dependent and this must be borne in mind by clinicians seeking to try every possible less invasive strategy in children with rapidly progressive shock. Although several children have been successfully resuscitated from cardiac arrest caused by progressive sep­ tic shock and yet completely recovered,18 it is clearly more desirable to intervene before arrest occurs. The timing of this will depend on how quickly an institution can mobilize an emergency ECMO team. Parallels can be drawn to fulminant myocarditis, where the exact tim­ ing of mechanical support is based on clinical experience and institutional resources, rather than by prospectively studied, specific criteria. There have been increasing reports of ECMO for adult septic shock,29-32 with survival rates ranging from 15%-71 %. Poorer outcOl!les have been seen using peripheral venoarterial (VA) ECMO in patients with distributive shogk, in patients who have had cardiac arrest as a c.e­ s'iilt of septic shock, and in patients older than 6.Q..years.29-30 Better outcomes have been seen

in young adults who present with distributive shock then develop progressive left ventricular dysfunction and are placed on ECMO before cardiac arrest occurS. 32

Contraindications The standard contraindications to ECMO apply in septic patients, such as preexisting severe neurological dysfunction or incurable malignancy.33 An additional consideration is the septic oncology patient. Oncology patients have been regarded as poor ECMO candidates, but this view is anachronistic and outcomes are acceptable in some instances. One exception to this is allogeneic bone marrow transplant recipients, who have dismal outco~s. 34 Chil­ dren who have received stem cell transplanta­ tion and require ECMO have somewhat better outcomes but survival to hospital discharge is still only 10%.35 Neutropenic sepsis is not a contraindication to ECMO per se, but it is difficult to do because of the other morbidities that often coexist in this patient group - such as thrombocytopenia - and long-term survival may not be as good as for other indications. Nonethe­ less, successful outcomes have been reported. 36 The type of infecting organism should not be regarded as a major determinant ofthe appro­ priateness ofECMO although some organisms, most notably Bordetella pertussis and herpes simplex virus (HSV) in infants, are associated with poorer outcomes. 37.38 In the absence of other contraindications, the infecting microbe usually has minimal bearing on whether ECMO is offered or not, and is often not known at the time.

Cannulation Strategies Cannulation is one of the most important management issues in ECMO for sepsis and must be individually tailored to the patient's circulatory and respiratory status. An under­ standing of the pathophysiology of septic 615

shock coupled with adequate hemodynamic information is vital in planning an appropriate cannulation stnlte~!:V sepsis-induced isolated respiratory failure ECMO, VV is "'1"P·T"'''1''P£1 because it is associated with better outcomes. VV-ECMO avoids the complications VA-ECMO such as systemic embolization, arterial trauma, and ventricular afterload, while pUlmonary blood flow, pulsatile systemic flow, of blood in systemic ventricle and 43,44 VV-ECMO is also pre­ coronary ferred in those patients with ARDS that shock, when the patient is ready to be weaned off mechanical "nT"...... rt but not ready to cease pvtT!>r....1"'I'''·w.''!> I;;x\::nwllgt; because of ongoing respiratory In these consideration should to changing to VV cannulation if it is anticipated that lung recovery will require more than 1-2 days IfECMO is being considered primarily as circulatory support for refractory septic shock, then the patient's hemodynamic response to sepsis must first be established. Septic shock has three principle hemodynamic manifesta­ based on the most compromised part circulation: right heart failure, heart failure with poor oxygen or distributive shock with poor oxygen extraction.5 In advanced cases a mixture ofthese may occur, eg, adult patients who present with distributive but later develop ventricular failure, or children with a combination car­ diogenic vasoplegic shock 18,20,32 Right heart failure associated with tent pUlmonary hypertension ofthe newborn is most shock in neonates. Right failure from a combina­ tion of sepsis-induced ventricular dysfunction and high positive pressure ventilation can be seen in older patients. the neonatal period, septic children suffer from left ventricu­ lar failure with preserved vasomotor tone and 616

impaired oxygen delivery. The at which a child will alter their hemodynamic ,.",,,r... ,,n,u, left heart ('cold' shock) to distribu­ tive shock ('warm' shock) is highly variable and cannot be reliably predicted from child's However, by late adolescence and into adult­ hood, the near universal hemodynamic response to sepsis is distributive shock. This response a reduction in ventricular a reduction vasomotor tone, by a reduction in pvt....."r·T1C'.n at a mitochondrial a

"'VllUUU",L."'U

of clinical assessment, blood tests (eg, venous oximetry, lactate), and echocardiography, with or without measurement ofcardiac output, and IS done by an intensivist 5 Possible ECMO cannulation 0"''''''f;'''''' come apparent once the hemodynamic pattern ofshock has identified (Table 56-1). Those with right heart failure and concomitant respira­ tory can be supported with VV-ECMO is not particularly advanced, as the consequent reduction in intrathoracic pressure and optimization oxygenation and carbon dioxide clearance may be sufficient to improve myocardial performance and peripheral lation, especially in small children. Otherwise, peripheral VA-ECMO or central ECMO can be used. In heart failure, peripheral or high flow central ECMO is appropriate. Serial echocar­ diograms must be performed to monitor left heart distension. Ifthis is worsening, should be taken to it before left atrial distension and hypertension edema or pulmonary circuit flow may limit atrial distension; if un­ percutaneous atrial septostomy can be performed on or a left atrial vent cannula can ECMO, biatrial drainage. If the femoral is used in older chil­ dren or adults then some centers advocate the routine use ofan anterograde perfusion cannula

EeMO for Septic Shock

Table 56-1. EeMO

cannulation strategies in septic shock.

Hemodynamic Typical Pattern Patient Neonate Right heart failure

Cannulation Options VV

Advantages

-

Peripheral· VA (carotid)

-

Central· VA

-

Left heart failure

Young child

Peripheral VA (carotid or femoral)

-

Central VA

-

Distributive

Older child or adult

Central VA

-

Mixed shock (eg, cardio­ genic and vasoplegic) Concomitant refractory cardiac and respiratory failure

Any age

Central VA

-

Any age

Central VA

-

VAV

-

-

-

Prevents systemic embolization A voids differential cyanosis Fast May use 1 cannula Direct circulatory support Fast Allows the highest flow mtes Prevents differential 30 U/kglhr). In antithrom­ findings seen in adult 1'''''"''-'UlC':>. Inotropes can usually weaned com­ bin levels may be low, in which case there pletely or to minimal doses within a few hours may be a role administering intravenous of achieving goal-directed circuit flows. Vaso­ antithrom bin concentrate, aiming for 100-120% ('f\t,,,tT',,,t(W" may be necessary to maintain age of the value. In particularly appropriate mean arterial pressures but it is not cases of DIC and hPlTIArrh unusual to see hypertension ensue around this elements ofcoagulopathy which can particularly with high flows ECMO, in which case short-acting vasodilators then be targeted for treatment. haTh.,..

flow. Instead, circuit flows should

620

ECMOjor Other measures such as errectlVe

antibiotics and immediate treatment of any pharmacokinetics patIents """Pl""rIU extracorporeai inadequately "hu"",,'

increases in mortality,54,55 initial antibiotics should as as cover all likely pathogens, and be at maximum dose recommended by standard formularies, espe­ cially those with a wide therapeutic index such as ~-lactam antibiotics.56 role ofother extracorporeallife support modalities sepsis to remove inflammatory mediators or the response techniques, classified

(CRR]), plasmapheresis, plasma exchange, and hemoperfusion. 57 is frequently required in septic patients on ECMO to compensate LUUU"""'Uacute kidney injury provide adequate solute as as to prevent severe blood product, nu­ trient, and administration. Although some investigators have seen hemodynamic benefits in children receiving high-flux CRRT,5 CRRT should not been as standard manage­ septic patients on ECMO unless ""'r1P"T has or fluid over­ and plasmapheresis load. 58-59 have shown some promise in small trials but cannot be considered standard therapy and await proper evaluation in tive 48,60,61 Many of are undergoing phase II or III trials. One randomized, controlled trial intraabdominal sepsis and shock showed decreased mortality with use of polymyxin this was B hemoperfusion. 62 a secondary endpoint, statistical ':"15,llU"...,"'cuv,"" and has not been duplicated in more recent controlled trials.63

Most patients on ECMO for

Shock shock

recover quickly and do not require EeMO more than 3-4

Failure of

heart to re-

is usu­ a poor prognostic indicator. Occasionally, patients suffer from persistent ARDS, Ilt:l;t:::;~;l­ to VV-ECMO when the circu­ latory component illness "PC,,,i11Po'" scenario is seen particularly with disseminated Staphylococcus aureus and can challenging to deal with, as necrotizing staphylococcal pneumonia can cause substantial lung paren­ chymal destruction. After a trial prolonged VV-ECMO in some with this condition, the only other than withdrawal supmay be to perform lung transplantation directly from ECMO. However, scenario is uncommon and some patients can still successfully weaned. Outcomes

with rates of ap]Jro,xlIna1:ely This age group is unique in having sufficient data to comment on pathogen-specific outcomes. In one survey sent to 16 ICUs worldwide, 117 patients were identified, 107 were neonates. 39 Survival in patients with positive, or viral was 77%, 60%, 40% respectively, although the study was published nearly two decades ago and it is likely that outcomes are better now. One study of neonates on ECMO with herpes vi­ rus showed survival to discharge was infected only 25%.37 The survival rate in with~~~~~~~~~~~~ws

reveal that the neonatal period from bacterial or viral are and respec­ tively.21 The corresponding figures for 621

Chapter 56

patients are 61 % and 66%, of both adults children but this includes all causes ofARDS as that outn.-t1"t>r,pnt in patients.22,23 septic shock, historical experi­ .,.... "''''""." that the use ofECMO in children survival to hospital discharge 5,18 However, the use high flow, ECMO with modern circuitry intensive care is associated with survival rates approaching 75%,18,20 Hopefully, further assessment will support these more recent, im­ proved which are comparable in neonatal sepsis. to The increasing use of ECMO for adult septic shock has demonstrated that peripheral ECMO for distributive shock is most likely unhelpful that peripheral for severe sepsis-induced seen in younger adults, can save up to 70% of ..."T."",.,, 32 use of ECMO for adult distributive septic shock has been successfully reported in isolated cases but, thankfully, is never required. 64 IT'U'''PrJ

Conclusions for septic Our understanding of shock progressed considerably in the last 25 years, through being by many 1990s as a complete practitioners in the contraindication, to modern published case series demonstrating 75% survivaL ECMO is generally very in course of usually before antibiotics have taken effect. 20 effects it exerts on shock and U~H'~A'" of multiorgan failure are usually Ltll.HI ...".Uvl"'''-'''''r",_ real technology, combined with evolving ap­ proaches to both timing ofECMO initiation and management ofpatients receiving ECMO, there body suggest that role in .w'L-.'H'U has an care­ fully selected to lung in both. Cannulation at centers with and management strategies that facilitate an awake, extubated approach to maximize physi­ cal therapy are recommended when appropriate order to optimize posttransplant outcomes.

ECMOas

to

Transplantation

References

9. Javidfar J,

Thuita Nowicki U",iiro.."""n GB, Blackstone am,pumUUlOln be performed parespiratory '''''\Y'\nrt? J Thorac 39(3 ):765-773 Beaty CA, Kilic A, 2. Shah AS. Outcomes poral among high-risk patIents transplantation in the Transplant 2012;3] (1

venous extracorporeal membrane oxygen­ ation. Ann Thorac Surg 2011;91(6):1763­ 1768; discussion Hadem J, et aL 10. Fuehner T, Kuehn tracorporeal oxygenation in awake as bridge to lung trans­ plantation. Am J Respir Crit Care

D, et al. Use of

bicaval dual-lumen Catllleter for adult veno­ 1.

LLlVVUU,UL"'."'"

3.

J, Brodie D, Iribame Ex­ tracorporeal membrane as a to lung transplantation and recovery. Cardiovasc 20

K, et al. G, Riise Extracorporeal membrane oxygenation as a bridge to lung study. Eur J 2015;47(1):95-100; UlS\;US~aUIl 5. .LLV""J~'" CW, Kukreja J, DL, Diaz-Guzman

4.

as a bridge to pulmonary J Thorae Cardiovasc 2013;145(3):862­ 867; discussion 867-868. 6. M,Dauben HD,GamsE. ·Ant.... 1'"(H,I pumping routine open heart surgery improves clinical outcome. Artif Organs 1998;22(4):326-336. 7. Morgan IS, M, Sanger K, .lVJ.C14U.C"" PS. of centrifugal pru::atatnc cardiac 1998; 13(5):526-532. 8. Wang D, Zhou X, Sidor B, Lynch J, Zwischenberger JB. Wang-Zwische double lumen cannula-toward a percutaneous and ambulatory artificial lung. ASAIO J 2008;54(6):606-611.

H,l1,..Irrt KM, Abrams 11. Agerstrand D. Blood conserva­ Bacchetta MD, tion in membrane "'> membrane oxygenation. Semin Cardiovasc Surg 20 234. 29. Camboni D, Akay Sassalos P, et al. Use of venovenous extracorporeal membrane oxygenation and an atrial septostomy pulmonary and failure. Ann Thorac 2011;91(1):144-149. Strueber M, Hoeper MM, Fischer S, et al. Bridge to thoracic organ transplantation in patients with pulmonary arterial hyperten­ sion using a pumpless lung Am J 2009;9(4):853-857. 31. Pierre Keshavjee S. Lung transplanta­ tion: donor and recipient critical care as­ pects. Curr Opin erit 2005;11(4):339­ 344. J, Daven­ Zwischenberger membrane as a bridge to pulmonary transplantation. J Thorac 2013. Benza RL, Gomberg-Maitland M, Miller DP, et score calculator in patients with pulmonary arterial hypertension. Chest 2012;141(2):354-362. Delcroix M, R. Optimising the management of pulmonary arterial hyper­ tension patients: emergency treatments. European respiratory : an ViU',,","U journal ofthe European Respiratory Society 20 I0; 19( 117):204-211. 35. Biscotti M, Sonett 1, M. ECMO as to lung transplant. Thoracic sur­ gery clinics 2015;25(1):17-25. 36. Chiumello Coppola S, Froio Colombo A, Sorbo Extracorporeal support o

OVlilCTPlnn between 46_100%.44.49,50,68 In children who require ECMO after transplantation the survival typically reaches 50_60%,60,62,69 If a period, ECMO-treated """,(',.,1-",£1 with survival at I or 5 years.45 ,5l,52 The choice of mechanical assist device vs. ECMO support for severe PGD remains studied. survival between ... "t,Pt'ltc supported with RVAD or ECMO in isolated right-sided dysfunction but overall graft recovery was improved in the ECMO-treated successfully man­ patients. 70 PGD has aged with the Levitronix Centrimag, although a direct comparison with ECMO was not been made. 1l

Failure to Recover Donor Heart Should the not recover then a critical assessment of the anatomy of the donor heart with ultrasound, and catheter is required. A biopsy to assess degree of rejection is usually easily accomplished on ECMO. If rejection oc­ curs, we augment and review immunosuppres­ sion and assess the comorbidities. A discussion on the likelihood and feasibility successful medium term support to recovery or retrans­ plantation should then take place.n ,13

Late Graft FaUure Late failure can be supported with ECMO. usually occurs to acute cellular mediated rejection or chronic

658

allograft vasculopathy. When indicated, ECMO should considered soon after presentation to provide support during periods of augmenta­ tion of immunosuppression. Kittleson showed improved outcomes when ECMO is used as preemptive therapy as opposed to an element of cardiopulmonary resuscitation; survival to disfonner group, although was 79% in only 26% were alive at one year. 74,15 ECMO deployed to augmentation of immunosuppression may be effective and allow assessment of recovery, VAD, or retrans­ plantation. The latter is less likely to be successwithin one year ofthe initial transplant, and is a cofactor for early death in year 0Ids. 5 .14

Conclusion ECMO is still used the acute care of children with heart failure will save lives . Newer fonns ofcirculatory support are usually deployed to bridge to transplant because of longer waiting and have out­ comes. the relatively poor outcomes of bridge to with ECMO are due to ECMO or variables such as sedatives, depen­ and inability to rehabilitate is un""...,.., ECMO has a small role in treatment adult heart failure, probably restricted to acute resuscitation as a 'bridge to decision' or 'bridge to a bridge' .

EeLS in Heart Transplantation

References I. Colvin-Adams M, Smithy JM, Heubner BM, et al. OPTN/SRTR 20 12 annual data report: heart. Am J Transplant. 2014;14(Suppll): 1l3-138. 2. West LJ, Pollock-Barziv SM, Dipchand AI, et al. ABO-incompatible heart trans­ plantation in infants. N Engl J Med. 2001 ;344(11 ):793-800. 3. Boucek MM, Mashburn C, Dunn SM, et al. Pediatric heaIt transplantation after decla­ ration of cardiocirculatory death. N Engl J Med.2008;359(7):709-714. 4. Mancini D, Colombo PC. Left Ventricular Assist Devices: A Rapidly Evolving Alter­ native to Transplant. J Am Coil Cardiol. 2015;65(23):2542-2555. 5. Dipchand AI, Edwards LB, Kucheryavaya AY, et al. The registry of the International Society for Heart and Lung Transplanta­ tion: seventeenth official pediatric heart transplantation report--20 14; focus theme: retransplantation. J Heart Lung Transplant. 2014;33 :985-995. 6. Lund LH, Edwards LB, Kucheryavaya AY, et al. The Registry of the International Society for Heart and Lung Transplanta­ tion: Thirty-second Official Adult Heart Transplantation Report-20 15; Focus Theme: Early Graft Failure. J Heart Lung Trans­ plant. 2015; 34(10):1244-1254. 7. McCarthy PM, Smedira NO, Vargo RL, et al. One hundred patients with the Heart­ Mate left ventricular assist device: evolving concepts and technology. J Thorac Cardio­ vasc Surg. 1998; 115(4):904-912. 8. Pagani FD, Lynch W, Swaniker F, et al. Ex­ tracorporeal life support to left ventricular assist device bridge to heart transplant: A strategy to optimize survival and resource utilization. Circulation. 1999; 100(19 Suppl): II206-21 o. 9. Pagani FD, Aaronson KD, Dyke DB, Wright S, Swaniker F, Bartlett RH. Assess­

ment of an extracorporeal life support to LVAD bridge to heart transplant strategy. Ann Thorac Surg. 2000;70(6): 1977-1984; discussion 1984. 10. Lebreton G, Pozzi M, Mastroianni C, Leger P, Pavie A, Leprince P. Extracorporeallife support as a bridge to bridge: a strategy to optimize ventricular assist device results. Eur J Cardiothorac Surg. 2015;48(5):785­ 791. 11. Delius RE, Zwischen berger JB, Cilley R, et al. Prolonged extracorporeallife support of pediatric and adolescent cardiac transplant patients. Ann Thorac Surg. 1990;50(5):791­ 795. 12. del Nido PJ, Armitage JM, Fricker FJ, et al. Extracorporeal membrane oxygenation support as a bridge to pediatric heart trans­ plantation. Circulation. 1994;90(5 Pt 2): II66-69. l3. BarZiv SMP, McCrindle BW, West LJ, Edgell D. Outcomes of pediatric patients bridged to heart transplantation from extra­ corporeal membrane oxygenation support. ASAIO J. 2007;53(1):97-102. 14. Mah D, Singh TP, Thiagarajan RR, et al. Incidence and risk factors for mortality in infants awaiting heart transplantation in the USA. J Heart Lung Transplant. 2009;28(12): 1292-1298. 15. Kirk R, Naftel D, Hoffman TM, et al. Outcome of pediatric patients with dilated cardiomyopathy listed for transplant: a multi-institutional study. J Heart Lung Transplant. 2009; 28(12):1322-1328. 16. Brown KL, Wray J, Wood TL, McMahon AM, Burch M, Cairns J. Cost utility evalu­ ation of extracorporeal membrane oxygen­ ation as a bridge to transplant for children with end-stage heart failure due to dilated cardiomyopathy. J Heart Lung Transplant. 2009; 28(1):32-38. 17. Hetzer R, Loebe M, Potapov EV et al. Circulatory support with pneumatic paracorporeal ventricular assist device 659

Chapter 59

in infants and children. Ann Thorac

1998;66(4):1498-1505.

26. Almond

et al. Waiting list mortality among children

heart transplantation in the United Circulation 2009;119(5): 717-727. T, S, et at. 27. Cassidy J, A longer waiting game: bridging children to heart transplant with the Berlin EXCOR United Kingdom experience. J Heart Transplant. 2013;32(11):1101-1106. 28. Dipchand AI, Mahle WT, Tresler M, et at. Extracorporeal Oxygenation as to Pediatric Heart Transplantation on

tation Outcomes.

960-969.

oxygenation. Thorac Surg. 29. Duncan BW, Bohn DJ, AtzAM, French JW, 2009;87(6): 1894-1901. Laussen DL. Mechanical 21. Cassidy J, S, Kirk nttprf"" of to

support the treatment of children with acute fulminant myocarditis. J Thorac tion children. J Heart Lung Transplant.

2009;28(3):249-254.

Cardiovasc Surg. 2001;122(3):440-448. 22. Morales DL, Almond CS, Jaquiss RD, et 30. Rajagopal SK, Almond CS, Laussen Bridging children of all Rycus Wypij D, Thiagarajan diac the initial Extracorporeal oxygenation North American with the support of infants, children, and adults with acute myocarditis: a Heart EXCOR device. J Heart Transplant. 2011 ):1-8. review of the Extracorporeal Sup­ Cdt Care Med . . Almond CS, Buchholz H, Massicotte Pet al. port Organization 2010;38(2):382-387. Berlin Heart EXCOR Pediatric ventricular Investigational Device Ex­ 31. Teele SA, Allan CK, Laussen burger Gauvreau K, emption study design and rationale. Am Heart 1. 2011;162(3):425-435. Management and outcomes in pediatric 24. Almond CS, Singh TP, Gauvreau K et patients presenting with acute fulminant myocarditis. J Pediatr. 2011;158(4):638­ Extracorporeal membrane for 643. e1. to transplantation chil­ The dren in the United States: analysis of data from the Organ Procurement and Transplant and Life port 2011; 123(25):2975-2984. r2eJnation . Pediatr Crit 15(4):355-361. 25. CD Jaquiss RD, DN et al. Prospective trial of a pediatric 33. Wong.JK, Smith TN, Pitcher HT, Hirose ventricular N Engl J Med. Cavarocchi NC. Cerebral and Lower Limb Near Infrared Sn,~..J1·r.""'.r.",,, in Adults on 2012;367(6):

18. Goldman AP, Cassidy J, de Leval M et waiting to paediatric heart transplantation. Lancet 2003;362(9400): 1967-1970. 19. BlumeED,NafteIDC,Bastardi Duncan BW, Kirklin Webber SA. Outcomes Children Bridged to Transplantation With Ventricular Assist Devices A Multi-In­ stitutional Study. Circulation. 2006; 113(19): 23 19. Imamura M, Dossey AM, Prodhan P, et al. to cardiac chil­

660

listed

EeLS in Heart Transplantation

Extracorporeal Membrane Oxygenation. ArtifOrgans.2012;36(8):659-667. 34. Roussel A, AI-Attar N, KhaJiel F, et al. Ar­ terial vascular complications in peripheral extracorporeaJ membrane oxygenation sup­ port: a review oftechniques and outcomes. Future Cardiol. 2013;9(4):489-495. 35 . Hoeper MM, Tudorache I, Kuhn C, et al. Extracorporeal membrane oxygenation wa­ tershed. Circulation. 2014; 130( 10):864-865. 36. Cove ME. Disrupting differential hypoxia in peripheral veno-arterial extracorpo­ real membrane oxygenation. Crit Care. 2015;19:280. 37. Combes A, Leprince P, Luyt CE, et al. Outcomes and long-term quality-of-life of patients supported by extracorporeal mem­ brane oxygenation for refractory cardiogen­ ic shock. Crit Care Med. 2008;36(5):1404­ 141l. 38. Stub D, Bernard S, Pellegrino V, et al. Refractory cardiac arrest treated with mechanical CPR, hypothermia, ECMO and early reperfusion (the CHEER trial). Resuscitation. 2015;86:88-94. 39. Petroni T, Harrois A, Amour J, et al. Intra­ aortic balloon pump effects on macrocircu­ lation and microcirculation in cardiogenic shock patients supported by venoarterial extracorporeal membrane oxygenation. Crit Care Med. 2014;42(9):2075-2082. 40. CheungMMH, GoldmanAP, Shekerdemian LS, Brown KL, Cohen GA, Redington AN. Percutaneous left ventricular "vent" inser­ tion for left heart decompression during extracorporeal membrane oxygenation. Pe­ diatr Critic Care Med. 2003;4(4):447-449. 41. Aiyagari RM, Rocchini AP, Remenapp RT, Graziano IN. Decompression of the left atrium during extracorporeal membrane oxygenation using a transseptal catmula incorporated into the circuit. Crit Care Med. 2006;34(10):2603-2606. 42. De Rita F, Hasan A, Haynes S, et al. Out­ come of mechanical cardiac support in

children using more than one modality as a bridge to heart transplantation. Eur J Cardiothorac Surg. 2015;48(6):917-922. 43. Carmena MDGC, Bueno MG, Almenar L, et al. Primary graft failure after heart transplantation: characteristics in a con­ temporary cohort and performance of the RADIAL risk score. J Heart Lung Trans­ plant. 2013;32(12): 1187-1195. 44. D'Alessandro C, Aubert S, Golmard JL, et al. Extra-corporeal membrane oxygenation temporary support for early graft failure after cardiac transplantation. Eur J Cardio­ thorac Surg. 2010;37(2):343-349. 45. D'Alessandro C, Golmard JL, Barreda E, et al. Predictive risk factors for primary graft failure requiring temporary extra­ corporeal membrane oxygenation support after cardiac transplantation in adults. Eur J Cardiothorac Surg. 2011 ;40(4 ):962-969. 46. Ibrahim M, Hendry P, Masters R, et al. Management of acute severe perioperative failure ofcardiac allografts: a single-centre experience with a review of the literature. Can J Cardio\' 2007;23(5):363-367. 47. Kobashigawa J, ZuckermannA, Macdonald P, et al. Report from a consensus conference on primary graft dysfunction after cardiac transplantation. J Heart Lung Transplant. 2014;33(4):327-340. 48. Lim JH, HwangHY, Yeom SY, Cho ill, Lee HY, Kim KB. Percutaneous extracorporeal membrane oxygenation for graft dysfunc­ tion after heart transplantation. Korean J Thorac Cardiovasc Surg. 2014;47(2): 100­ 105. 49. Listijono DR, Watson A, Pye R, et al. Usefulness of extracorporeal membrane oxygenation for early cardiac allograft dysfunction. J Heart Lung Transplant. 2011 ;30(7):783-789. 50. Marasco SF, Vale M, Pellegrino V, et al. Extracorporeal membrane oxygenation in primary graft failure after heart transplan­

661

Chapter 59

tation. Thorac 2010;90(5):1 ­ 1546. 51. Seguchi 0, Fujita T, Murata Y, et aL Inci­ dence, etiology, and outcome of primary graft dysfunction in adult heart transplant recipients: a single-center experience in Japan. Heart Vessels. 2016;31 (4):555-562. 52. Segovia J, Cosio MD, JM, et al. RADIAL: a novel primary failure score in heart transplantation. J Heart Lung Transplant. 2011 ;30( 6):644-651. 53. Huang J, K, MendeloffEN, Spray TL, Canter CEo Risk factors for primary graft failure after pedi­ cardiac transplantation: importance of recipient and donor characteristics. J Heart Lung Transplant. 2004;23(6):716-722. 54. MB, DN, Ivy D, et Evidence of pulmonary vascular disease after heart transplantation for Fontan cir­ culation J Thorac Cardiovasc Surg. 2004; 128(5):693-702. 55. Hoskote A, Carter ventricular Burch M, Brown K. Acute failure after pediatric cardiac transplant: predictors and long-term outcome in current J Thorac era of transplantation Cardiovasc 2010;139(1): 146-153. 56. Mitchell MB, Campbell DN, Bielefeld MR, Doremus T. Utility of extracorporeal mem­ brane for early failure following heart transplantation in infancy. J Heart 2000; 19(9):834­ 839. 57. Kavarana MN, Sinha P, Naka Y, Oz MC, Edwards NM. Mechanical support the failing allograft: a center experience. J Transplant. 2003;22(5):542-547. 58. Marasco SF, Esmore J, et aL Early institution of mechanical support nt'''''''''' outcomes in primary cardiac allograft J Heart Lung Transplant. 2005;24(12):2037-2042.

662

59. Rodrigues W, Carr M, Ridout D et al. Total donor ischemic time: Relationship to early hemodynamics and intensive care morbid­ ity in pediatric cardiac transplant recipients. Pediatr Crit Care Med. 2011; 12(6):660-666. 60. Kaushal S, Matthews KL, X, et at. A multicenter study of primary failure after infant heart transplantation: Impact of extracorporeal membrane oxy­ genation on outcomes. Pediatr Transplant. 2014; 18(1):72-78. 61. Mettler B. Graft transplantation: Impact membrane oxygenation on outcomes. Pe­ diatr Transplant. 2014; 18( I): Su JA, Kelly RB, Grogan T, Elashoff D, Alejos JC. Extracorporeal membrane oxy­ pediatric orthotopic heart transplantation. Pediatr Transplant. 2015; 19(1 ):68-75. M, J, Crossland D, Hasan A. Elective extracorporeal mem­ i'S""c1(UIUll bridge to recovery in otherwise" unusable" donor hearts for children: Preliminary outcomes. J Heart Lung Transplant. 2013;32(8):839-840. 64. Wong Cavarocchi NC. :SD~~ctJroscor)v in Adult Patients Receiving Extracorporeal Membrane Oxygenation. Ann Thorac Surg. 2015;100(2): 766. et SM, Tribble Kaza

;)W...""";).lI.H extracorporeal mem­ oXYlgemluon (ECMO) ulP"'nina assistance for refractory cardiogenic shock. Intensive Care Med. 2011 11):1738­ 1745. 67. Simmonds J, Dominguez T, J, et and outcome of ex­ tracorporeal life support after pediatric transplantation. Ann Thorac Surg. 66-2172.

EeLS in Heart Transplantation

68. Lima EB, da Cunha CR, Barzilai VS, et al. Experience of ECMO in primary graft dysfunction after orthotopic hearttransplan­ tation. Arq Bras Cardiol. 2015; 105(3):285­ 291. 69. Tissot C, Buckvold S, Phelps CM, et al. Outcome of extracorporeal membrane oxygenation for early primary graft failure after pediatric heart transplantation. J Am Coli Cardiol. 2009;54(8):730-737. 70. Taghavi S, ZuckermannA, Ankersmit J, et al. Extracorporeal membrane oxygenation is superior to right ventricular assist de­ vice for acute right ventricular failure after heart transplantation. Ann Thorac Surg. 2004;78(5): 1644-1649. 71. Thomas HL, Dronavalli VB, Parameshwar J, Bonser RS, Banner NR:Steering Group of the UK Cardiothoracic Transplant Au­ dit.. Incidence and outcome of Levitronix CentriMag support as rescue therapy for early cardiac allograft failure: a United Kingdom national study. Eur J Cardiothorac Surg.2011;40(6):1348-1354. 72. Kanter KR, Vincent RN, Berg AM, Mahle WT, Forbess JM, Kirshbom PM. Cardiac retransplantation in children. Ann Thorac Surg.2004;78(2):644-649. 73. Mahle WT. Cardiac retransplantation in chil­ dren. Pediatr Transplant. 2008; 12(3):274­ 280. 74. Kittleson MM, Patel JK, Moriguchi JD, et al. Heart transplant recipients supported with extracorporeal membrane oxygenation: outcomes from a single-center experience. J Heart Lung Transplant. 2011;30(11): 1250­ 1256. 75. Perri G, Hasan A, Cassidy J, et al. Extra­ corporeal life support after paediatric heart transplantation. Eur J Cardiothorac Surg. 2012;42(4):696-701.

663

60 Immunodeficiency and EeLS Matteo Di Nardo, MD, Veronica Armijo-G 'Cia, MD, Thomas Staudinger, MD, Gerry Capatos, MD, Heidi Dalton, MD

Introduction Traditionally, Extracorporeal Life Sup~~rt (ECLS) has been reserved for patients suffe'rg acute cardiac or respiratory failure due to revrs­ ible diseases expected to have a good qua ity of life after recovery. Concerns whether EC S should be offered to immunocompromi ed patients have led to many thought-provok g discussions in the past. Furthermore, the I ck of specific criteria for ECLS use in the . u­ nocompromised renders patient selection m re challenging. This chapter provides current in­ sights and, where possible, new selection cnte­ ria for ECLS use in patients with solid and bloft cell tumors, patients receiving hematopoiefic stem cell transplantation (HSCT), and patie ts with autoimmune diseases or HIV.

ECLS Support in Pediatric and Adult Patients with Solid Organ and Blood can1er Survival for adult and pediatric patie ts with malignancy over the last 30 years has mark­ edly improved due to advancements in aggrfs­ sive chemotherapeutic regimens and supportive measures. During the last decade, a large bO~y of evidence suggests that adult leU surviv rs regain favorable quality of life, tolerate t e continuation of anticancer therapy, and ha e long-term survival defined by their underly" g

disease that may not differ from cancer patients who were never admitted to the lCUY Thus, a general reluctance to admit critically ill cancer patients to the lCU can no longer be justified.3

ECLS Utilization in Patients with Malignancy Previously, limited use ofECLS for patients with cancer was most likely influenced by cancer-related mortality as well as the ideol­ ogy ofECLS being a support modality offered only to acutely ill patients with no underlying chronic illness. Since 2009, reports describing ECLS use for patients (pediatric and adult) with malignancy have been published, im­ plying increased use in this population, most likely due to improved oncologic survival and more widespread ECLS use in complex patient populations.6,7 A 2008-20 12 analysis of the Ex­ tracorporeal Life Support Organization (ELSO) Registry documented that 178 pediatric patients with malignancy received ECLS, doubling uti­ lization compared to that reported previously.6 Furthermore, several pediatric and adult cases of successful use of ECLS as a bridge to thera­ peutic interventions or chemotherapy have been recently published. g,9 An analysis by Gow and colleagues of 72 adult cancer patients in the ELSO Registry from 1992-2008 reported an overall survival rate of 32%. Most patients had solid tumors and respi­

665

Chapter 60

ratory failure but for cardiac indicaf ns surprisingly yielded better outcomes. Comp ed to the general population, survival significantly lower. infections occ more commonly in cancer patients, bleeding events did not. 7 In another study 0 adult patients with hematological malign cy, VV-ECLS was used for ARDS due to p eu­ ed monia in 11 patients. Three due to septic cardiomyopathy, or ife tbreateOl!ng circulatory and respiratory re TTI""""'lP U,"'U'''''UH.:U bulky disease. F ve patients received rescue chemotherapy ng '-'''-'.LJV. Overall 50%. All ur­ vivors hematological remission and w re alive at followup 36 months. to A review of children from the istry from ] 994-2007 by Gow et 107 patients with oncologic patients were with 35% survi to discharge. lt More recently, an abstract Armijo-Garcia et a1. reported 178 tients with malignancy (also excluding HS patients) from 2008-2012, with 48% survi al to discharge despite similar complication ra es compared to those documented previously. Ty e of malignancy (hematologic or solid) did ot impact surviva1.6 Compared to ECLS patie ts without ra s patients from 2008-2012 had of gastrointestinal hemorrhage, leukopen a, hyperglycemia, myocardial stun, and rece' t of inotropes. However, multivariate eOTP 100-lSOxl 09, nd fibrinogen >2 gil. No studies report on brontho­ scopic transbronchial lung biopsy in pedia~c ECLS patients probably because of concclrns of bleeding. Air leaks rarely occur, prob Ibly due to low thoracopulmonary compliance, ow tidal volumes, and low ventilating pressure in patients supported with ECLS.92 I

Congenital Diaphragmatic Hernia Repair of congenital diaphragmatic hernia can safely be done on ECMO.93 The ideal tim­ ing of surgery is has been the focus of debate for many years (see Chapter 10). Most centers use antifibrinolytics for 24-72 hours to control postoperative bleeding. 7,93,94

Specific Surgical Procedures in Cardiac ECLS Sternotot,nylThoracotomy Blood clots can fill the pericardial or pleural space impairing venous return or pump flow, especially in ECLS patients with eentral can­ nulation. Clot debridement may prove neces­ sary in these patients. In neonates with shunt dependent circulation, clotting or obstruction of the systemic pulmonary shunt may lead to acute hypoxia or loss of systemic circulation, sometimes requiring MCS. Revision of the shunt can be safely performed on ECLS or af­ ter conversion to bypass, leading to improved outcomes. 95,96

Unloading the Left Ventricle (Left-vent) When the left ventricle (LV) functions poorly it can become over distended during ECLS resulting in acute pulmonary venous congestion and edema, or inadequate myocar­ dial recovery. Treatment is decompression of the LV. Elective LV decompression may reduce duration of ECLS97 and can be done surgically by cannulating the left atrium or a pulmonary vein, percutaneously by atrial stenting or bal­ loon atrial septostomy/6,98.IOO or via the trans­ diaphragmatic route. IOI In adults and children, echocardiographically guided percutaneous blade and balloon atrial septostomy can be perfonned safely atthe bedside. 102·104 Barbone et al. describe introducing a pigtail catheter in the femoral artery in 4 patients, advancing through 687

the aortic valve into the LV under allce.IOS

g id-

Conclusion

Ruprecht et aL have reviewed card'ac

decompression

ECLS, 106

lntracardiac Operations The removal of intracardiac clots, infect e endocarditis,107 or surgery for residual lesi ns usually requires transfer to CPB in opera g room. Two examples for using the ECLS cire it support: 1) a patient on VV-ECMO c n BT on vascu ar clamps are applied to the shunt and pulmon ry artery to control blood loss. A cell saver e be useful; 2) A on VA ECMO can ha e their coronary OPC equipment and techniques, (Table 61-2)

in the lexicon the surgeon, interventionist, or obstetrician can­ not be performed safely on ECLS preparation, we must merely await the patient circumstances which embolden pioneering protagonists.

Table 61-2. Techniqu for heart surgery on ECMO.

HEART SURGE e ECMO circuit is used to maintain oxygen livery and circulatory support whilst the surgery is p ormed. Use the ECLS circuit for DHCA

Convert to CPB

ted total

"\ Cool • • Give 3mgIKg Heparin and allow to circulate, check ACT >500 sec. •! ECLS circuit, connect CPB circuit to \AW.U\.U"", or additional cannulae if necessary. onCPB • \ Connect cnds ofECLS for '[. recirculate, turn the sweep off I few seconds if blood becomes aexoyg,enatea . •Check ACT, blood gas periodically. the surgery. It ~Oh~flt will not come offCPB then rccc.nne;ct same EeLS circuit ifit is clean. Ifit was a new one, Deal with in chapter 7. come offCPB their na~~IVfllP.(]

I

688

Procedures on EeLS

References 1.

2.

3.

4.

5.

6.

7.

8.

9.

Atkinson JB, Kitagawa H, Humphri s B. Major surgical intervention during e~tra­ corporeal membrane oxygenation. J ~ed Surg. Sep 1992;27(9):1197-1198. I Taghavi S, Jayarajan SN, Mangi , et al. Examining Noncardiac Surgical ro­ cedures in Patients on Extracorpo eal Membrane Oxygenation. ASArO J. Sep Oct 2015;61(5):520-525. Chestovich PJ, Kwon MR, Cryer HG Til­ lou A, Hiatt JR. Surgical procedures fOr pa­ tients receiving mechanical cardiac sup ort. Am Surgeon. Oct 2011;77(10):1314-1 17. Nagaraj HS, Mitchell KA, Fallat ME, 10ff DB, Cook LN. Surgical complications rnd procedures in neonates on extracorp0r.eal membrane oxygenation. J Ped Surg. f'ug 1992;27(8): 11 06-11 09; discussion I 09­ 1110. Downard CD, Betit P, Chang RW, Garza JJ, Arnold ill, Wilson 1M. Impact of CAR on hemorrhagic complication~ of ECMO: a ten-year review. J Ped Surg. tUg 2003;38(8): 1212-1216. Brunet F, Mira JP, Belghith M, et al. Effi cts ofaprotinin on hemorrhagic complicafons in ARDS patients during prolonged e a­ corporeal C02 removal. Intensive are Med. 1992; 18(6):364-367. van der Staak FH, de Haan AF, Geven ' Festen C. Surgical repair of congen tal diaphragmatic hernia during extraco real membrane oxygenation: hemorrh complications and the effect oftrane acid. J Ped Surg. Apr 1997;32(4):594- 99. Stone ME, Soong W, Krol M, Reich L. The anesthetic considerations in patiepts with ventricular assist devices present~g for non cardiac surgery: a review of ei t cases. Anesth Analg. JuI2002;95(1):42 49, table of contents. Factora FN, Bustamante S, SpiottaA, sian R. Intracranial hemorrhage surge

AMr­

patients on mechanical circulatory support: a case series. J Neurosurg Anesth. Jan 2011;23(1):30-34. 10. Kartha V, Gomez W, Wu B, Tremper K. Lap­ aroscopic cholecystectomy in a patient with an implantable left ventricular assist device. Brit J Anesth. May 2008;100(5):652-655. 11. Rubin S, Ali AN, Pages ON, Baehrel B. How to replace an extracorporeal life suppon without interruption of the cardio­ pulmonary assistance. Interact Cardiovasc Thorac Surg. Aug 2009;9(2):311-313. 12. TanakaD, Pitcher HT, Cavarocchi N, Hirose H. Migrated Avalon Veno-Venous Extracor­ poreal Membrane Oxygenation Cannula: How to Adjust Without Interruption ofFlow. J Cardiac Surg. Nov 2015;30(11 ):865-868. 13. Lidegran MK, Frenckner BP, Mosskin M, Nordell B, Palmer K, Linden VB. MRI of the brain and thorax during extracorporeal membrane oxygenation: preliminary re­ port from a pig model. ASArO J. Jan-Feb 2006;52( I): 104-1 09. 14. Goodwin SJ, Randle E, Iguchi A, Brown K, Hoskote A, Calder AD. Chest computed tomography in children undergoing extra­ corporeal membrane oxygenation: a 9-year single-centre experience. Ped Radiol. Jun 2014;44(6):750-760; quiz 747-759. 15. Hayes D, Jr., Tobias JD, Galantowicz M, Preston TJ, Tzemos KK, McConnell PI. Video fluoroscopy swallow study and nutritional support during ambulatory venovenous extracorporeal membrane oxy­ genation as a bridge to lung transplantation. World J Ped Congen Heart Surg. Jan I 2014;5(1):91-93. 16. Al-Ogaili Z, FouIner D, Passage J, et al. CT pulmonary angiography during veno­ arterial extracorporeal membrane oxygen­ ation in an adult. J Med Radia Oncol. Jun 2013 ;57(3):345-347. 17. KamraK, Russell I, Miller-Hance WC. Role of transesophageal echocardiography in the management of pediatric patients with 689

Chapter 61

congenital heart disease. Ped Anesth.

2011 ;21(5):479-493. 18. MichelenaHI, Suri RM, MaloufJ, Enriq ez­ Sarano M, Mankad Sv. Adult perioperat ve echocardiography: anatomy, mechanis s and effective communication. Prog ar­ diovasc Dis. Jul-Aug 2014;57(1):74-90 19. Haglund NA, Maltais S, Bick JS, et al. Hemodynamic transesophageal echo ar­ diography after left ventricular assist dev ce implantation. J Cardiothorac Vasc Ane h. Oct 2014;28(5): 1184-1190. 20. Dolch ME, Frey L, Buerkle MA, Weig T, Wassilowsky D, Irlbeck M. Transesop, a­ geal echocardiography-guided techniq e for extracorporeal membrane oxygenati I n dual-lumen catheter placement. ASAIO J. Jul-Aug 2011;57(4):341-343. 21. Marcus B, Atkinson JB, Wong PC, et 1. Successful use of transesophageal ec 0­ cardiography during extracorporeal mef­ brane oxygenation in infants after cardi c operations. JThorac Cardiovasc Surg. M y 1995; 109(5):846-848. 22. Johnston TA, Jaggers J, McGovern 'jJ, O'Laughlin MP. Bedside transsept~l balloon dilation atrial septostomy for ~e­ compression of the left heart during extk­ corporeal membrane oxygenation. Cath t Cardiovasc Intervent. Feb 1999;46(2): 19 ­ 199. 23. Cavarocchi NC, Pitcher HT, Yang Q, t al. Weaning of extracorporeal membr oxygenation using continuous hem ­ dynamic transesophageal echocardio ­ raphy. J Thorac Cardiovasc Surg. De 20l3; 146(6):1474-1479. 24. Daniel WG, Erbel R, Kasper W, et al. Safe of trans esophageal echocardiography. multicenter survey of I 0,419 examination . Circulation. Mar 1991;83(3):817-821. 25. O'Shea JP, Southern JF, D'Ambra MN, al. Effects of prolonged transesophage echocardiographic imaging and prob manipulation on the esophagus--an ech ­ 690

cardiographic-pathologic study. J Am ColI Cardiology. May 1991;17(6):1426-1429. 26. Karlson KH, Jr., Pickert CB, Schexnayder SM, Heulitt MJ. Flexible fiberoptic bron­ choscopy in children on extracorporeal membrane oxygenation. Ped Pulmonology. Oct 1993;16(4):215-218. 27. Kamat PP, Popler J, Davis J, et al. Use of flexible bronchoscopy in pediatric patients receiving extracorporeal membrane oxy­ genation (ECMO) support. Ped Pulmonol­ ogy. Nov 2011 ;46(11): 11 08-11l3. 28. Sharma NS, Peters T, Kulkarni T, et a1. Flex­ ible Bronchoscopy Is Safe and Effective in Adult Subjects Supported With Extracorpo­ real Membrane Oxygenation. Respir Care. Jan 26 2016. 29. Hodges AM, Gillham MJ, Lewis CA. Bed­ side placement of an endobronchial valve to aid invasive ventilation and weaning from extracorporeal membrane oxygen­ ation. Critical care and resuscitation : J Australasian Acad Crit Care Med. Sep 2015; 17(3):219-222. 30. Baqais KA, Mahoney M, Tobler K, Hui A, Noseworthy M. Pediatric sand aspiration managed using bronchoscopy and extracor­ poreal membrane oxygenation. Can Respira J. Sep-Oct 2015;22(5):261-262. 31. Panda BR, Alphonso N, Govindasamy M, Anderson B, Stocker C, Karl TR. Cardiac catheter procedures during extracorporeal life support: a risk -benefit analysis. World J Ped Congen Heart Surg. Jan 1 2014;5( 1):31­ 37. 32. Alsoufi B, Awan A, Manlhiot C, et al. Results of rapid-response extracorporea! cardiopulmonary resuscitation in children with refractory cardiac arrest following cardiac surgery. Eur J Cardiothorac Surg. Feb 2014;45(2):268-275. 33. Booth KL, Roth SJ, Perry SB, del Nido PJ, Wessel DL, Laussen PC. Cardiac catheter­ ization ofpatients supported by extracorpo­

Procedures on EeLS

real membrane oxygenation. J Amer 011 Cardiac Surg 6 2002;40(9): 1681-1686 34. Brown KL, Shekerdemian LS, Penn DJ. Transcatheter closure of a patent art rial duct in a patient on veno-arterial extr cor­ poreal membrane oxygenation. Inten ive Care Med. Apr 2002;28(4):501-503. 35. Silva IN, Erickson CC, Carter CD, 9t al. Management ofpediatric tachyarrh~fias on mechanical support. Circulation. ug 2014;7(4):658-663. 36. Alkhouli M, Narins CR, Lehoux J, Knfght PA, Waits B, Ling FS. Percutaneous IDe­ compression of the Left Ventricle infar­ diogenic Shock Patients on Venoart ial Extracorporeal Membrane Oxygen a ·on.

42.

43.

44.

~~~~al Cardiac Surg. Mar 2016;31 (3):t77­ 37. Kommineni M, Lang RM, Russo MJ, S ah AP. Percutaneous transcatheter c10sur of infarct related ventricular septal detl cts assisted with portable miniaturized e a­ corporeal membrane oxygenation: a series. Cardiovasc Revascularization Jul-Aug 2013;14(4):241-245. 38. Callahan R, Trucco SM, Wearden PD, eer­ man LB, Arora G, Kreutzer J. Outco es of pediatric patients undergoing car iac catheterization while on extracorpo eal membrane oxygenation. Ped Cardiol. 1ar 2015;36(3):625-632. 39. Scherrer V, Lasgi C, Hariri S, et al. Ra io­ frequency ablation under extracorpo~eal membrane oxygenation for atrial ta hy­ cardia in postpartum. J Cardiac Surg. ep 2012;27(5):647-649. 40. Cheruvu C, Walker B, Kuchar D, Subb ah RN. Successful ablation of incessant Y reentrant tachycardia in a patient on e a­ corporeal membrane oxygenation. H art Lung Circ. Jan 2014;23(1):eI2-15. 41. Rizkallah J, Shen S, Tischenko A, ziero! S, Freed DH, Khadem A. Successful abla on of idiopathic left ventricular tachycardi in an adult patient during extracorporeal m~m-

45.

46.

47.

l

48.

49.

brane oxygenation treatment. Can J CardioL Dec 2013;29(12):1741 eI717-1749. Thuys C, MacLaren G, d'Udekem Y, Eastaugh L. Vascular access for pediatric coronary angiography on extracorporeal membrane oxygenation. World J Ped Con­ gen Heart Surg. Jan 2015;6(1):126-129. Ucer E, Fredersdorf S, Jungbauer C, et al. A unique access for the ablation catheter to treat electrical storm in a patient with extracorporeallife support. Europace. Feb 2014; 16(2):299-302. Endemann DH, PhilippA, Hengstenberg C, et al. A simple method ofvascular access to perform emergency coronary angiography in patients with veno-arterial extJacorporeal membrane oxygenation. Intensive Care Med. Dec 2011;37(12):2046-2049. Brogan TV, Thiagarajan RR, Rycus PT, Bartlett RH, Bratton SL. Extracorpo­ real membrane oxygenation in adults with severe respiratory failure: a multi­ center database. Intensive Care Med. Dec 2009;35(12):2105-2114. Sarosiek K, Hirose H, Pitcher HT, Cavaroc­ chi NC. Adult extracorporeal membrane oxygenation and gastrointestinal bleeding from small bowel arteriovenous malforma­ tions: a novel treatment using spiral enter­ oscopy. J Thorac Cardiothorac Surg. May 2012; 143(5): 1221-1222. Zwischenberger JB, Cilley RE, Hirschi RB, Heiss KF, Conti VR, Bartlett RH. Life-threatening intrathoracic complica­ tions during treatment with extracorporeal membrane oxygenation. J Ped Surg. Jul 1988;23(7):599-604. Gross GW, Dougherty CH. Pleural hem­ orrhage in neonates on extracorporeal membrane oxygenation and after repair of congenital diaphragmatic hernia: imaging findings. AlR. Apr 1995;164(4):951-955. Jackson HT, Longshore S, Feldman J, Zirschky K, Gingalewski CA, GoUin G. Chest tube placement in children during 691

Chapter 61

extracorporeal membrane oxygena 'on

(BeMO). JPed Surg. Jan 2014;49(1):51 53; discussion 53-54.

50. Tashiro J, Perez EA, Lasko DS, Sola JE. Post-ECMO chest tube placement: A ro­ pensity score-matched survival analysi . J Ped Surg. May 2015;50(5):793-797. 51. Camporota L, Nicoletti E, Malafronte M, et al. International survey on the mao ge­ ment of mechanical ventilation dur ng ECMO in adults with severe resp ra­ tol)' failure. Minerva Anestesiologica. ov 2015;81(11):1170-1183 , 1177 p follow g 1183. 52. Braune S, Kienast S, Hadem J, et al. San ty of percutaneous dilatational tracheosto· y in patients on extracorporeal lung supp I rt. Intensive Care Med. Oct 20 13;39(10): 17t2­ 1799. 53. Hayes D, Jr., Galantowicz M, Preston J, Lloyd EA, Tobias .lD, McConnell PI. ~fa­ cheostomy in adolescent patients bridg d to lung transplantation with ambu a­ t0l)' venovenous extracorporeal membrJ,e oxygenation . Jap J Artif Organs : M r 2014; 17(1): 103-105. 54. Hayes D, Jr., Lloyd EA, Yates AR, c­ Connell PI, Galantowicz M, Preston J. Pediatric ambulatol)' ECMO. Lung. D c 2014;192(6): 1005. 55. Agar NJ, Berkowitz RG. Airway compli a­ tions ofpediatric extracorporeal membra e oxygenation. Ann Otol Rhinol Lal)'og I. Jun 2011 ;120(6):353-357. 56. Lysenko L, Zaleska-Dorobisz U, Blok et al. A successful cesarean section in a pregnant woman with A (HI N 1) influe a requiring ECMO support. Kardiochi To ­ kochirurgia Pol. Jun 2014;11(2):216-219 57. Panarello G, D ' Ancona G, Capitanio et al. Cesarean section during ECM support. Minerva Anestesiologica. Ju 2011 ;77( 6):654-657. 58. Park SH, Chin N, Choi MS, Choi JH, Chao YJ, Jung KT. Extracorporeal membran I

692

oxygenation saved a mother and her son from fulminant peripartum cardiomyopathy. J Obstet Gynecol Res. Jul20 I 4;40(7): 1940­ 1943. 59. Kim HY, Jeon HJ, Yun JH, Lee JH, Lee GG, Woo SC. Anesthetic experience using extracorporeal membrane oxygenation for cesarean section in the patient with peripar­ tum cardiomyopathy: a case report. Kor J Anesthe. May 2014;66(5):392-397. 60. Gevaert S, Van Belleghem Y, Bouchez S, et al. Acute and critically ill peripartum cardiomyopathy and 'bridge to' therapeutic options: a single center experience with intra-aortic balloon pump, extra corporeal membrane oxygenation and continuous­ flow left ventricular assist devices. Crit Care.2011;15(2):R93. 61. Landsman IS, Karsanac CJ. Case report: pediatric liver retransplantatioo on an extracorporeal membrane oxygenation­ dependent child. Anesth Analg. Nov 2010;111(5):1275-1278. 62. Howell CG, Hatley RM, Davis JB, Kanto WP. Successful gastrorrhaphy on ECMO. J Ped Surg. Dec 1988;23(12): 1161-1162. 63. Firstenberg MS, Abel E, Blais D, et al. The use of extracorporeal membrane oxy­ genation in severe necrotizing soft tissue infections comp licated by septic shock. Am Surgeon. Nov 2010;76(11):1287-1289. 64. Jacobs N, Hooft NM, Robinson BR, et al. The use of extracorporeal membrane oxygenation in blunt thoracic trauma: A study of the Extracorporeal Life Support Organization database. J Traum Acute Care Surg. Dec 2015;79(6):1049-1054. 65. Liston DE, Richards MJ. Venoarterial extracorporeal membrane oxygenation (VA ECMO) to facilitate combined pneu­ monectomy and tracheoesophageal fistula repair. J Cardiothorac Vasc Anesth. Aug 20] 4;28(4): 1021-1 023. 66. Redwan B, Ziegeler S, Freermann S, et al. Intraoperative veno-venous extracorporeal

Procedures on EeLS

lung support in thoracic surgery: a s· gle­ centre experience. Interact Cardoi asc Thorac Surg. Dec 2015;21(6):766-772 67. Ius F, Kuehn C, Tudorache I, et at. ung transplantation on cardiopulmonary sup­ port: venoarterial extracorporeal memb e oxygenation outperformed cardiopu nary bypass. J Thorac Cardiovasc Surg. 2012;144(6): 151 0-1516. 68. Zhou R, Liu B, Lin K, et at. ECMO sup­ port for right main bronchial disruptio in multiple trauma patient with brain inju --a case report and literature review. Perfu ion. JuI2015;30(5):403-406. 69. Ballouhey Q, Fesseau R, Benouaic V, Leobon B. Benefits of extracorpo eal membrane oxygenation for major bunt tracheobronchial trauma in the paedi tric age group. Eur J Cardiothorac Surg. pr 2013;43(4):864-865. 70. Chou NK, Chen YS, Ko WJ, et at. App~ica­ tion of extracorporeal membrane oxygen­ ation in adult burn patients. Artif 0lans. Aug 200 1;25(8):622-626. 71. Prodhan P, Imamura M, Garcia X, B es JW, Bhutta AT, Dyamenahalli U. Abd~mi­ nal compartment syndrome in newb rns and children supported on extracorpo eal membrane oxygenation. ASAIO J. Mar-~pr 2012;58(2): 143-147. 72. Rollins MD, Deamorim-Filho J, Scaife R, Hubbard A, Barnhart DC. Decompres ive ent laparotomy for abdominal compa syndrome in children on ECMO: e ect on support and survival. J Ped Surg. Jul 2013;48(7):1509-1513. 73. Beck R, Halberthal M, Zonis Z, Shos ani G, Hayari L, Bar-Joseph G. AbdomInal compartment syndrome in children. ed Crit Care Med. Jan 2001;2(1):51-56. 74. Lam MC, Yang PT, Skippen PW, Kiss on N, Skarsgard ED. Abdominal com art­ ment syndrome complicating paedia ic extracorporeallife support: diagnostic nd I

therapeutic challenges. Anaesth Intensive Care. Sep 2008;36(5):726-731. 75. Okhuysen-Cawley R, Prodhan P, Imamura M, DedmanAH, Anand KJ. Management of abdominal compartment syndrome during extracorporeal life support. Ped Crit Care Med. Mar 2007;8(2): 177-179. 76. Maj G, Calabro MG, Pieri M, Melisurgo G, Zangrillo A, Pappalardo F. Abdominal compartment syndrome during extracorpo­ real membrane oxygenation. J Cardiothora Vasc Anesth. Oct 2012;26(5):890-892. 77. Augustin P, Lasocki S, Dufour G, et at. Abdominal compartment syndrome due to extracorporeal membrane oxygen­ ation in adults. Ann Thorac Surg. Sep 2010;90(3):e40-41. 78. Yeo ill, Sung KH, Chung CY, et al. Acute compartment syndrome after extracorporeal membrane oxygenation. Jap J Ortho Sci. Mar 2015;20(2):444-448. 79. Brodt J, Gologorsky D, Waiter S, Pham SM, Gologorsky E. Orbital compartment syndrome following extracorporeal support. J Cardiac Surg. Sep 2013;28(5):522-524. 80. Avalli L, Maggioni E, Sangalli F, Favini G, Formica F, Fumagalli R. Percutaneous left­ heart decompression during extracorporeal membrane oxygenation: an alternative to surgical and transeptal venting in adult pa­ tients. ASAIO J. Jan-Feb 2011 ;57(1 ):38-40. 81. Friesenecker BE, Peer R, Rieder J, et at. Craniotomy during ECMO in a severely traumatized patient. Acta Neurochirurgica. Sep 2005;147(9):993-996; discllssion 996. 82. Krenzlin H, Rosenthal C, Wolf S, et at. Surgical treatment of intraparenchymal hemorrhage during mechanical circula­ tory support for heart-failure··-a single­ centre experience. Acta Neurochir. Sep 2014;156(9): 1729-1734. 83. Wilson TJ, Stetler WR, Jr., AI-Holou WN, Sullivan SE, Fletcher JJ. Management of intracranial hemorrhage in patients with

693

Chapter 61

left ventricular assist devices. J Neuros rg. May 2013;118(5): 1063-1068. 84. Niebler RA, Lew SM, Zangwill SD, e al. Incidence and outcome ofpediatric patj'ts e with intracranial hemorrhage while s p­ if ported on ventricular assist devices. Organs. Jan 2014;38(1):73-78. 85. Mayer RR, Hwang SW, Reddy GD, et al. Neurosurgical complications of left ven ic­ ular assist devices in children. J Neuros rg. Pediatrics. Nov 2012;10(5):370-375. 86. Marasco SF, Preovolos A, Lim K, Salam, n­ sen RF. Thoracotomy in adults while n ECMO is associated with uncontrolla Ie bleeding. Perfusion. Jan 2007;22(1):23-' 6. 87. Bressman M, Raad W, Levsky JM, stein S. Surgical therapy for complicatio s ofpneumonia on extracorporeal membr : e oxygenation can improve the ability 0 wean patients from support. Heart Lulg Vessel. 2015;7(4):330-331. 88. Bond SJ, Lee DJ, Stewart DL, Buchino J. Open lung biopsy in pediatric patients n extracorporeal membrane oxygenation. J Ped Surg. Oct 1996;31 (1 0): 1376-1378. 89. Jaklitsch MT, Linden BC, Braunlin E , Bolman RM, 3rd, Foker JE. Open-lu g biopsy guides therapy in children. Thorac Surg. Jun 2001 ;71(6): 1779-1785. 90. Inwald D, Brown K, Gensini F, Malone~, Goldman A. Open lung biopsy in neona 1 and paediatric patients referred for ex~­ corporeal membrane oxygenation (ECM ). Thorax. Apr 2004;59(4):328-333. 91. Ryan CA, Finer NN. Open lung biopsi s in neonates on ECMO: additional cas . Extracorporeal membrane oxygenation. J Ped Surg. Aug 1998;33(8):1327-1328. 92. Roze H, Thumerel M, Ouattara A, Jougon . Feasibility of bedside open lung biopsy' patients treated with extracorporeal me ­ brane oxygenation. Intensive Care Me Nov 2014;40(11): 1789-1790. 93. Keijzer R Wilschut DE, Houmes RJ, et . Congenital diaphragmatic hernia: to repa'

wm­

I.

694

on or offextracorporeal membrane oxygen­ ation? J Ped Surg. Apr 2012;47(4):631-636. 94. Dassinger MS, Copeland DR, Gossett J, et al. Early repair ofcongenital diaphragmatic hernia on extracorporeal membrane oxy­ genation. J Ped Surg. Apr 2010;45(4):693­ 697. 95. Miyashita T, Hayashi Y, Ohnishi Y, Inamori S, Kuro M. Anesthesia for an infant with hypoplastic left heart syndrome undergo­ ing reconstruction ofa systemic pulmonary shunt under extracorporeal membrane oxy­ genation. J Cardiothorac VascAnesth. Aug 1998; 12(4):497-498. 96. Allan CK, Thiagarajan RR, del Nido PJ, Roth SJ, Almodovar MC, Laussen PC. Indication for initiation of mechanical cir­ culatory support impacts survival of infants with shunted single-ventricle circulation supported with extracorporeal membrane oxygenation. J Thorac Cardiovasc Surg. Mar 2007;133(3):660-667. 97. Hacking DF, Best D, d 'Udekem Y, et al. Elective decompression of the left ven­ tricle in pediatric patients may reduce the duration of venoarterial extracorporeal membrane oxygenation. Artif Organs. Apr 2015;39(4):319-326. 98. Keenan JE, Schechter MA, Bonadonna DK, et al. Early Experience with a Novel Can­ nulation Strategy for Left Ventricular De­ compression during Non-postcardiotomy Venoarterial ECMO. ASAIO 1. Jan 5 2016. 99. Hong TH, Byun JH, Lee HM, et al. Initial Experience ofTransaortic Catheter Venting in Patients with Venoarterial Extracorporeal Membrane Oxygenation for Cardiogenic Shock. ASAIO J. Mar-Apr 2016;62(2): 117­ 122. 100. Eastaugh LJ, Thiagarajan RR, Darst JR, McElhinney DB, Lock JE, Marshall AC. Percutaneous left atrial decompression in patients supported with extracorporeal membrane oxygenation for cardiac disease. Ped Crit Care Med. Jan 2015;16(1):59-65.

Procedures on EeLS

101. Eudailey KW, Yi SY, Mongero LB, Wa­ gener G, Guarrera N, George 1. Trans dia­ phragmatic left ventricular venting d . g peripheral venous-arterial extracorpo eal membrane oxygenation. Perfusion. 2015 ;30(8):70 1-703. 102. O'Connor TA, Downing Gl, Ewing LL, Gowdamarajan R. Echocardiographi guided balloon atrial septostomy du ing extracorporeal membrane oxygena ion (ECMO). Ped Cardiol. lui 1993;14(3): 67­ 168. 103. Seib PM, Faulkner SC, Erickson CC, tal. Blade and balloon atrial septostomy for left heart decompression in patients with se ere ventricular dysfunction on extracorpo eal membrane oxygenation. Catheter C ~ iO­ vasc Interv. Feb 1999;46(2):179-186. 104. Dahdouh Z, Roule V, Lognone T, Sa­ batier R, Massetti M, Grollier G. At·al septostomy in cardiogenic shock relate to H1N1 infection. Acute Cardiac Care. ar

2013;15(1):7-9.

105. Barbone A, Malvindi PG, Ferrara P, Ta$li G. Left ventricle unloading by percutan ' us

pigtail during extracorporeal membr e

oxygenation. Interact Cardiovasc Thorac

Surg. Sep 2011;13(3):293-295.

106. Rupprecht L, Florchinger B, SChOPk~ S, et al. Cardiac decompression on extra ,or­ poreal life support: a review and dis us­ sion of the literature. ASAIO l. Nov­ ec 2013;59(6):547-553. 107. Noyes AM, Ramu B, Parker MW, derhill D, Gluck lA. Extracorpor Membrane Oxygenation as a Bridge to Surgery for Infective Endocarditis C plicated by Aorto-Atrial Fistula and diopulmonary Collapse. Texas Heart l. 2015;42(5):471-473.

tHY

695

62 Extracorporeal Elimination Meral Patel, MD, Rachel Sirignano,

Ryan Barbaro, MD, Matthew L. Paden, MD

Extracorporeal life support (ECLS) was traditionally thought of as providing isolated temporary cardiac and/or pulmonary support for However, as have more comfortable with this technology, the patients supported become more complex, often with multiple secondary organ that requite supportive therapy. The cardiac and pulmonary support that ECLS provides becomes the "platfonn" upon which multiple support therapies can be delivered. In vu"'I-' ....,,, we discuss use other extratherapies, such as continuous replacement (CRRT) and the use of apheresis procedures during VV",'VllJlV

Acute

(AKI) during EeLS

AKl commonly complicates critical illness and is a risk death in critically ill "1",,,n1',,, ofall ages and also for those ECLS. A complete review complicated topic is beyond the scope this chapter, but assessments of this topic been completed 1-3 for neonatal, pediatric, and adultICU Historically, the literature on AKl has ficult to assess, due to multiple differing defini­ ofthe condition. 4 Recently, standardized scores, such as the :'::";:-'""7""":-~;-'--;7----"'---:;;'-"'"

rum creatinine. 5-7 Uepel[lQlng mode, and definition used, occurs with reports of presence in 19-71 % of neo­ nates, 20-72% of pediatric, and to >70% of adult patients.8-12 In many of those same studies, an association is established between AKl and increased mortality. In addition to diagnosing AKl by urine and "'....",,-w.u.""'V meruiures, the of fluid overload as another manifestation tion that negatively organ function 13-15 In ECLS and mortality is well pal:IeIlts,FO has associated with ...... 1>'Ul........ oxygenation, increased duration of ECLS, and both AKl and FO are as­ mortality. 16-20 sociated with increased mortality, treatment of these comorbidities has been recommended in an attempt to improve ECLS outcomes. indications treatment of AKl and FO are not well established. In ll\Jl1-L>"L>L',' pallerus, consensus by ADQI on pharmacological and mechanical removal have published. 21 ,22 With patients, indications treatment are broadly divided into AKl, FO, electrolyte disturbances not araenable to medical therapy, and removal greatly by ECLS AU'",,"'>l"

697

Chapter 62

center. 23 The Extracorporeal Life Support Or­ ganization (ELSO) guidelines addressing AKl and FO state, goal of fluid management is to return the extracellular fluid volume to normal (dry reason is that weight) maintain it edema caused by critical illness or iatrogenic crystalloid fluid infusion causes lung and myo­ cardial failure, adding to the primary problem ... When the patient is hemodynamically stable (typically 12 hours) diuretics are instituted and continued until weight is diuretic response is not sutnclient negative fluid or if overt renal failure, continuous is added to the extracorporeal circuit to maintain fluid and electrolyte balance.''24

guidance criteria of renal therapy (~T) initiation, or optimal performance during While RRT can be intermittent or continuous, CRRT is the most common during and will the the following "nn,T.~.~n

23

Technical Aspects ofCRRT during ECLS Renal support therapies work by isolating blood on one side of a semipermeable mem­ brane, the properties of which permit solute and volume exchange through it. Seminars ;in Dialysis provides an entire with a review of CRRT technology, physiology, modes for a more look at this 26 In eral, multiple modes of providing CRRT however, the most techniquys ofdiffusion or convection rest on the Diffusion based therapies, such as continuous' venovenous hemodialysis (CVVHD), wOFk

698

by passing a solution of electrolytes on the nonblood side of the filter, in a counter CUf­ rent direction to the blood flow. This allows equilibration of solutes and water down concentration gradients in the plasma and the dialysate to occur. Diffusive is more at small solute removal (individual electrolytes, urea, etc.), compared to molecules. Alternately, convective therapies, such as venovenous ll\.",HV'lHI,latlUJJ (CVVH), utilize the transmembrane pressure to create a volume of ultraflltrate by pushing water through the membrane. Convective mass transfer occurs providing the ultrafiltrate with small solutes, which is determined by membrane nr(~1"\""t'1",,"· used allow clearance of molecules between 500-5000 Daltons.

reviewed. 27 a separate vascular access is not of the ECLS circuit and using a commercially available This standalone method sents the simplest approach in a patient who has a vascular prior to In this CRRT is similar to the patient who is not on with the exception that reductions or elimination of CRRT circuit anticoagulation is often possible due to the patient's ECLS antlcoagllJatlon. r.ruillIleJ;)L.!~llll!llilru1~d due to the increased anticoagulation. risk of bleeding due to ~~~"'u method to provide concomitant involves a ne:mo,nl­ ter in a shunt post-pump in the pump flow drives blood through UCIUU1.HLC:I. producing the ultrafiJtrate. The of ultrafiltrate produced is controlled by standard N pumps and is measured using a bedside urirneter. The ultrafiltrate can be discarded, user is seeking to provide only slow continuous ultrafiltration (SCUF), or a

Extracorporeal Elimination replacement fluid containing desired electro­ lytes can be delivered to continuous venovenous (CVVH). treated blood is to the ECLS pre:PUJnp. This method to provide amrall1Lagl~s of being and can by "I.IV"""""';'" However, it has multiple shunt that returns blood "'''''''''11"''''''' creates to the CRRT cuit, its efficiency, insubstantial magnitude. Thus pumplflow does not equal patient flow. Consequently, flow must be monitored distal to the shunt. Using technique can contribute to development of multiple electrolyte.anomalies, the smaller and is not

production as replacement fluids are not engineered to in this manner. are not pumps at flow restrictors, have a significant er­ this fashionJ.28 ror rate (~12.5%) when a laboratory setup ofin-line CRRT during ECLS, the difference between prescribed measured ultrafiltration rate was as high as 34 mLlhour (>800 mL/day).29 careful monitoring as well as both volume of ultrafiltrate replacement fluid delivered, is essential. This can be accomplished either volumetrically or by the use of accurate scales (+1- 1 but specialist U,,,..IIl;!,,!>,, which contributes to reduction ofthe use technique. a l l•. """"""

~~~~~~~~to prqvide

introducing a commercially available CRRT the circuit. connection ofthe into the depends on

including ECLS circuit design, pump, and CRRT software. In general, for head pump the CRRT device can integrated in prepump In this manner, CRRT device is usuaUy """'_41,.1""'''' the venous CRRT pressure monitoring software to function. ECLS Alternately, in a pump circuit, it is that the be connected VV",tVU.JUV of air entrainment due to the negative pressure in the prepump venous line. Additionally, the software on many CRRT devices will not allow the device to start with the magnitude pressures seen the venous line. In """"",,, .. '£'1 blood returns from the CRRT to postpump veThis configuration reduces recirculation through the CRRT device, and additionally allows oxygenator to act as a clot and air trap returns to the The advantages to a commercially available CRRT device during ECLS include improved balance accuracy to the inline builtin standard and flow monitoring, lon­ ger :filter life, and a engineered for purpose of providing However, no '-'Vl.1LUJ,'-'H,la.lyavailable device has been specifically designed and approved use dur­ ing ECLS. disadvantages ofthis approach include device costs and training

Outcomes ofCRRT during ECLS While theoretical ,-nl!>,,-r advantages of EeLS are above, few outcome compare these Santiago et al. children on ECLS who received concomitant CRRT (2 , 6 commercial CRRT 11 stand alone) and ..,,,,,"''''' 'oJ'

699

Chapter 62

contropl Additionally, of the 6 children with

a

device incorporated into

EeLS

increased (138.4 hour) 36.8 hours seen in nonCRRT 27.2 anticoagulation in addition to the et at evaluated 458 balance measurements berecelvmg during using a available device and 30 patients an system. 30 They dem­ onstrated significantly improved accuracy and a reduction in both ECLS and CRRT dura­ tion with the use a commercially available device. did not see changes in hospital or ICU length or mortality. largest set outcome data that exists for use of CRRT the ELSO n.v.ai>U l'"I';n1pnT"

and KDIGO AKI scoring complications are entered in once per run and no duration information is obtained. Using this definition, the presence of this complication age of respi­ ratory failure is associated with worse survival (Table 62-1) compared to the overall survival ofthese groups (74%,58%, exist for the cardiac population. Both kidney with injury and the use ofRRT are ...('•.".!t'~prl mortality. A subset

ELSO centers in the States and Canada formed Kidney Injury Membrane Oxygenation (KIDMO) research network to investigate relationship RRT use, and survival in J

resources, and physician ofmethod, an thrombosis 702

pertaining to procedures on ECLS currently Instead, the decision to proceed with apheresis should be based on evidence for the underlying of ECLS support. ease and £'Inrlpr~'

Contl'aindications Based on 2016 evidence based AFSA guidenil",,.,,,,,.,,, is or even harmful amyloidosis, amyotrophic ~v"..,,~,,"~. alloantibody coagulation

tor dermatomyositis/polymyositis, atypical with membrane cofactor protein mutations, refractory immune thrombocytope­ nia, inclusion body myositis, POEMS syndrome (polyneuropathy, organomegaly, endocrinopa­ thy, monoclonal gammopathy and changes syndrome), ABO incompatible donor transplant, schizophrenia, lupus nephritis, and Quinine or thrombotic microangiopathy.45 Other relative contraindications that need to be considered are related to the when used either as an for the procedure or the blood products that are being used during the procedure. Briefly, citrate's anticoagulation effects are due to the depletion of free, ionized extracellular calcium upon which many the coagulation are Uetlenoelllt. ever, cardiac function also be affected (especially in the neonatal population) low ionized calcium. Additionally, because ofthe anticoagulant oflarge amounts of IJle:OOllllgmay be in patients bleeding. both cases, the negative of citrate on calcium can be mitigated by infusion of additional calcium. As is common with complicated patients on ECLS, a thorough weighing of the risks and beIlellts ofthe both in the short and long time horizons, should guide of this therapy. nPITr"'. However its is burdened by the general complications of the device, so its use should be limited to with concomitant acute failure, when gas exchange has to be artificially maintained. Impella RP is a novel minimally invasive percutaneous RVAD, now available for tempo­ 725

---- ----------------

---~ -~,-~,'~~,

--~

Chapter 64

rary support of the RY. The recent RECOVER I3IGHT prospective trial demonstrate that Im­ pella RP is safe and reliable in patients with se­ vere RV failure refractory to medical treatment, with immediate and sustained hemodynamic benefit and favorable outcomes to 30 and 180 days.ls In our institution, we have adopted a step wise approach, using first a short-term percuta­ neous device (VA-ECMO or Impella RP). We proceed to surgical temporary RVAD only when recovery is not achieved in a few days and pr.o­ longed support is needed. The only exception might be the postcardiotomy setting.

Mobilization An early mobility program for patients receiving MCS devices has resulted in su.c­ cessful outcomes such as increased tolerance to activity and decreased mortality rate. In fa~t, complications of prolonged bed rest in the in­ tensive care unit are well known. 16 Prolonged bed rest increases risk of ventilator-associated pneumonia, need for prolonged ventilation, length of ICU and hospital stays, occurrence ofdelirium, and workload ofthe cardiovascular system. In particular, in patients assisted by tem­ porary MCS, avoiding deconditioning, muscle loss, and nutritional depletion is of paramount importance. In fact, all such conditions dramati­ cally affect future definitive therapeutic options (heart transplantation or LVAD implantation). Early progressive mobility goals for pa­ tients receiving MCS devices are no different than any other ICU patients and allow physiQal reconditioning. However, more limitations aJild safety considerations exist when mobilizing these individuals. 17 In fact, IABP, Impella, and peripheral VA.­ ECMO limit early mobilization because of the requirement for placement of cannulas within the femoral vessels. These devices require the patient to adhere to bed rest. However the reverse Trendelenburg position should be used 726

to prevent ventilator-associated pneumonia in ventilated patients or to allow for the ability to eat in nonventilated patients. Progressive mo­ bility with these devices encompasses turning from side to side, and passive/active range of motion, avoiding the extremity in which the device is located. When prolonged support is required, an ax­ illary placement approach for IABP or Impella should be considered. Surgical central implanta­ tion of the CentriMag ventricular assist device is another option. In fact, an early progressive mobility program permits such patients to be up in the chair a few days after implantation and tolerate up to 3D-minute ambulation sessions (Figure 64-5). Safety measures are required dur­ ing ambulation to not compromise the driveline to the devices.

Weaning Weaning of MCS occurs once noncardiac organ systems have recovered. In patients with irrecoverable end-organ and/or neurologic

Figure 64-5. Ambulatory patient during paracorporeal biventricular support with CentriMag.

Other Forms ofExtracorporeal Cardiovascular Support -1ABp, IMP ELLA, VAD

dysfunction withdraw of mechanical sUJilport

should be considered. 11 In the weaning prdcess, MCS support is decreased gradually (hours to days). If hemodynamics and other organ func­ tion remain adequate, the device is expla:nted or if hemodynamics and/or organ function deteriorate LVAD implantation or heart trans­ plantation need to be considered. For this reason the weaning process starts as soon as possible, ideally at the moment ofMCS implantation. In fact all patients who receive a temporary MCS should be assessed for advanced long-term heart failure therapies. When different devices are simultaneously required for MCS, weaning from the device which produces increased complications should be considered first. As a general rule, when ECMO is used, it should be the first device to be removed. Such strategy might reduce the duration ofECMO support, hasten its weaning, and eventually reduce its related complications.

727

Chapter 64

References

1. Cowger J, Shah P, Stulak J, et al. lNTER­

2.

3.

4.

5.

6.

7.

8.

728

MACS profiles and modifiers: Heterogene­ ity ofpatient classification and the impact (!)f modifiers on predicting patient outcome. J Heart Lung Transplant: the official publica­ tion of the International Society for Heart Transplantation. Nov 6 2015. Gilotra NA, Stevens GR. Temporary me­ chanical circulatory support: a review of the options, indications, and outcomes. Clin Med Insights. Cardiology. 2014;8(Suppl 1):75-85. Thiele H, Zeymer D, Neumann FJ, et al. Intraaortic balloon support for myocardial infarction with cardiogenic shock. New Engl J Med. Oct 42012;367(14): 1287-1296. Werdan K, Gielen S, Ebelt H, Hochman JS. Mechanical circulatory support in cardiogenic shock. Europ Heart J. Jan 2014;35(3): 156-167. Estep JD, Cordero-Reyes AM, Bhimaraj A, et al. Percutaneous placement of an intra­ aortic balloon pump in the left axillary/ subclavian position provides safe, ambula­ tory long-term support as bridge to heart transplantation. JACC. Heart failure. Ot1 perfusionists, physicians, 9 biomedical Identify the Employees and Eqnipment Keigls1ter€:,,_~.,., program that should meet criteria to apply for measures, decide which components of the ELSO Award in Life Support 26 ECMO should be controlled to deliver designation. the highest quality possible. Some ideas include Tracking conformity with EeLS work re­ from decision to cannulation, number of quirements and establishing corrective action a circuit is accessed, or number of blood helps preserve stability and standardization of the measured processes. As each deviation products. the norm is identified, it should be studied A new center should OellClllD1llrK ECMO centers incorporate practices so causation can be determined and control into guidelines and during prograDl de­ regained. Variation in processes can be detri­ to quality outcomes velopment. Joining ELSO and contributing data to the provides the opportunity for new mance. Simple audits measure or procedures. These may be ECMO to benchmark their outcomes and complications with other more experienced imbedded in the medical record for ease ofcollection and be reviewed daily, weekly, centers. The Registry data are compiled annually with reports sent to individual centers or monthly depending on the measure and the international volume at the center. measure­ data. With data from over 73,000 patients ments minimize costs and error. Improvement lected during its four decades, the ELSO work can focus on improving compliance with provides new centers with the opportunity established protocols and guidelines, then upon to compare their and identify op­ reducing variation. percent of variation portunities for improvement. Attending ELSO from protocols can be over time and conferences provides additional opportunities displayed on a run chart so key staff can develop to learn in the field and to network with other more experienced centers. The Continuously Improving and Sustaining a programs worldwide Program by having processes, procedures, and in place that promote While developing and an requires much eo­ excellence and exceptional ECMO care. The ECMO W>O'VvJlC:,,,,U

736

Implementing an EeLS Program

ergy, sustaining its success and ~U',U~"Hu.;nu may prove even more formidable. Creating the processes to is necessary. ungomg oversight is essential, particularly given tbe often sporadic nature ofECMO cases, especially the days program. The initial steering committee should be tran­ sitioned into an ongoing ECMO leadership council for medical and to meet routinely to discuss operational issues. Difficult cases can sap the resolve and spirit ofa ofstandardized rapid debriefing team. processes and routine case (morbidity and mor­ tality) to discuss care respectfully .,..f'r_ protected fashion is essential. 27 Training programs should built into the rou­ tine yearly calendars for staff and incoming spe­ cialty resident/fellow trainees who arrive each year. ELSO center membership helps sustain the program by permitting benchmark­ ing against global trends and similarly sized programs. Attendance at ELSO provides a source state of the art in care, sharing PYlr\pf1pr,{'p care and res:erurcn As a validation on importance. The Award of Excellence program allows a new center to to optimiza­ tion, and objective measures quality and effectiveness, beginning with Path to ExceBence recognition (see elso.org). The program application effectively serves as a to outline the development of the overall program and metrics, E>~'U~'E> the center's steps towards applying for AOE designation in the future. chapter has attempted to path of EeMO building roadmap ofthe journey. In the end, in ;=::..______~of TPV'PUf

737

Chapter 65

References 1. Guergueruain A-M, Ogino M, Dalton Shekerdernian L. Setup and rnamt{mallCe of life support nr,,,yrtt,I>"tn,p in terms of quality-adjusted life years if the therapy proves successful and leads to many productive years.

Fundamental Considerations Establishing uniformity in the quality and standard the delivered, reSUlting in comparable outcomes while cost help make ECMO an acceptable and ultimately 773

Chapter 69

more widely available therapy. This that all potential local barriers to success of EeMO, local infection or the rapid availability of blood products, be identi­ fied and addressed early. A successful BCMO program requires the commitment team, with and systematic (see Chapter 67) as well as with development evidence-based (or at least the sharing from larger centers) guidelines and protocols, and audit and quality analyses Chapter 68) in order to keep morbidity and mortality rates to a minimum. 2 evaluations involve as­ sessment costs at two potential interventions for any condition, the determination of the benefits associated with h"",,~n,o ...+ combination h"'r,"'nT" to both the The assessment ofboth cost and benefits a more consideration ofthe "value" ofan intervention; what additional benefit is provided for what additional cost. The to value ofa therapy can be described mUltiple those of the patients, the hospitals, medical personnel, or to industry as a whole. Economic evaluations health care professionals and ers to decide between competing interventions.

Measuring tbe Overall Cost ofECMO No single globally tool us to accurately model the true 'cost' ofECMO, and this variability relates to local healthcare models as well as staffing, equipment consumable and hospital costs of intensive care stay, laboratory resources and therapies including renal replacement therapy, drugs, etc. Notwithstanding, no two patients are ever to support and validate the the same. proposal that ECMO be used as a life mea­ sure, the long-tenu cost for those

774

vs. must be evaluated.

treated conventionally

Cost Effectiveness To judge plex, "'Vt'lPne,i"",

requires detailed analysis including Ah.p"t",,,,

measures of functional outcome, specifically

related to quality oflife, each patient. A num­

ber of tools have developed to help with

this evaluation. The quality-adjusted life years

(QALYs) tool is perhaps most ~~."L~"'"

applied for health interventions. It

assesses health-related quality life at a

point after an and bas

plied in adults and children, in multiple """.,'ll',,,,,

including ECMO and mechanical support. 3,4 In

UK, the National Health (and latterly, National Institute for Health lence [NICE]) has, for more than two decades, been using QALYs to measure the overall health benefits delivered by various treatment regimens. To apply such a tool broadly, it is important to a threshold below which the inter­ vention can be said to be cost effective in tenus of QALY gained, and above which the model would be as being too costly without .,ri''''''ll

Vr..'llJCUL'"

Medical Futility andPotentially Inappropriate Therapies

When deciding to or discon­ tinue ECLS prior to medical futility is often declared, Unfortunately, this term is neither readily definable nor from subjec­ tivity, For the past four medical futility has been extensively debated within the medi­ cal literature ,4 1-46 Futility are generally based and have as they to favor physicians family con­ cerns and do not standing surrogate. 45,47 guidelines recommend using the term futility only for the rare circumstances when an intervention can­ not achieve a desired which applying the term futility to ECLS dif­ ficult, as it can often provide physiological supAlternatively, the inappro­ describes "treatments at least some sought by the patient, but clinicians believe that competing ethical considerations 784

justify not providing terminology is intrinsically linked to """·l'''''r'>t..", ofimmedi­ ~~~~gand

~

depends on the

which can

members, clinicians, Further muddying the quality of life often change across a lifetime, and are frequently higher from a disabling illness than they would have been a priori. 18,4&-50 communication about llJ.L'''U~''''U of values, benefits can diminish the likelihood of ,",Villll'''' regarding futility arise, particularly prolonged arguments about costs to system and resource utilization are frequently voiced, but not typically helpful considerations. While arguments are on the ofECLS support they cannot with unique ClTlcurnstanc:es. Discontinuing EeLS Support Prior to Organ Recovery: Ethical Permissibility

When prior experience, or evidence demonstrate ECLS will not allow survival, stopping is America, or life-sustaining >11\.,,",11.'1.111....,,1 ventilation, dialysis, and nutrition or hydration is ethically and legally permissible even when these interventions are life prolonging, or highly eficiaPI,51-53 When incapacitated, the decision or advance directive this request pediatric patient, AVAJI~ 643

Intraventricular (IVH) 175

Invasive Mechanical Ventilation 639,713

IVH 260



J Vein 2]4

L Laparotomy 685

Latin-American ELSO 27

Left-Sided 652

Left Ventricular 480

Left Ventricular Afterload 39

Left Ventricular Distention 334

Left Ventricular Overload 538

Ischemia 525

85

Levetiracetam 803

Liberation And Decannulation 468

Limb Ischemia 335

802

'PVf'LU"'V"Y 795, 797

Livanova EOS ECMO 68

Livanova Revolution Centrifugal Pump 61

P 797,802,803

lfO(leVelopmental Outcomes 397

FoUowup 314

Neurological, Functional, And Sur­ vival Outcomes 327

Long-Term Outcomes 297, 472

Lower Ischemia 536

Low-Flow Phase of CPR 321

Luciano Gattinoni 9,21

292,687 224

Index

M Magnetic Levitation 55

Maintaining VV-ECLS 466

Major 593

Major Trauma 593

665

lVl'"'lUtlL Quadrox- iD 66

Massive Acute Pulmonary Embolism 486

Mechanical Chest Compression 501

Mechanical Circulatory Support 721

Meconium 123, 184

Mediastinal Masses 233

Medical Futility 784

Medos Deltastream 63

Medos Hilite LT 66

Medtronic

62

Membrane::staDUlzmg

Miller's Triangle 748

Mistake-Proofing 735

Mobilization 567

Monitoring 55

Morbid

Morphine 198

Motor Function 224

Motor Function 221

Multidisciplinary Teams 141

Multimorbidity 539

JySlunCtlOn Syndrome 703

LVlUlUurga.u Failure 479

Myocardial 570, 511

Myocardial Stun 368

Myocarditis 340. 651

N Native Venous Flow 41

Near Infrared Spectroscopy CNIRS) 505

Needs Assessment 131

Negative Pressures 58

Neonatal ECMO Trial off 212

Neonatal Encephalopathy 175

NeurodevelopmentalOutcomes ,211.395

Neurological Injury 362

Examinations 381

Neurologic Injury 323

Neuropsychological Development 222

Neurosurgery 686

Biocube 69

No-Flow 321

Norwood 340

Nosocomial Infections 380,473,554

Not 351

NovaJung 68

Nursing Care 201,283

Nutrition 273, 454,462

o

Organ Supply 650

Oseltamivir Pharmacokinetics 802

Outcomes 308

Outlet Pressure 60

Out Of Hospital Cardiac Arrest 325

Out Of Hospital Cardiac Arrest 518

Oxygenation Failure 240

Oxygen Content 32

. Debt 267

Oxygen Delivery 32,63,379

Oxygen Transfer 34

p _ 370

Palliative Care 781

Paracorporeal Pumps 650

Oxygenator 69

Patient Moves And Pressure Area Care 461

Patient Selection for ECPR 323

Pearl O'Rourke 5

PEEP 406

Percutaneous Access 248

Percutaneous Cardiopulmonary 501

Percutaneous Coronary Intervention 479

Perfusion Cannula 526

Perfusionist 742

Peripheral ECMO 525

venoarterial ECMO 561

Persistent Hypertension 125

Persistent Pulmonary Hypertension Of The Neo­ nate 184

Persistent Pulmonary Hypertension Of The New­ born (PPHN) 126, 169

Phannacokinetic 798,801,802,803

Phenobarbital 803

Physical 640

Plasma 109

Platelets 83. 85, 87

829

Index

Platelet Transfusion 109

Pueumocystis Jiroveci 669

Poisoning 627,631

Polymethyl Pentene 65

Polyvinylchloride 51

Postcardiotomy 481

Power Failure 59

Preeclampsia 589

Pre-ECMO Cardiac Arrest 359

'V'""''''''=5 Comorbidities 255 177

Preparation And Communication 680

Pressure Gradient 64

Pressure Ulcers 285

Primary Graft Dysfunction 655

rn""n,.,. 600

Prognosis 388

Prolonged Immobilization 382

Prone Positioning 203,278,292,450

799

Protamine 352

Prothrombin Complex Concentrates 114

Puerperium 583

ullIlonarv Contusion 593

Pulmonary Edema 553

Pulmonary Embolism 486

Pulmonary Hemorrhage 234

Pulmonary Hypertension 311

Pulsatility 52

Failure 59

Pump Related Complications 57

Q Quality-Adjusted Life Years 773

Quality Improvement 765

Quality Of Care 765

472

Quality Of Life

R Rated Flow 63

RBC 106

Recirculation 467

Recombinant Activated Factor VII 353

Recombinant Factor VIla 111

Reconstruction 2]4

Recovery 388, 571,657

Refractory Shock 479

Refractory Hypoxemia 416

Registry Addenda 811

830

Reports 811

Rehabilitation 234

Reimbursement 776

Renal Failure 261,272,700

Renal Insufficiency 260

Renal 551

Reperfusion Teclmique 506

Residual Cardiac Lesions 370

Residual Lesions 392

Resistance 248

Resource Utilization 782, 784

Failure 415

Rest 248

Rest Ventilatory Settings 449

Results and Outcomes Following ECMO for

ACHD 335

p'1T""lor,~rlP Flow 59

Retrograde Trial Off 213

Return Of Spontaneous Circulation (ROSC) 50]

507

348

Atrial Right Axillary/Subclavian Artery 350

Right Ventricular 486

Risk Prediction Models 440

Robert Bartlett 3

Robert E. Gross 18

Roller Pump 50,53, 188

Roller Pump Servo Regulation 54

RRT 260,261

RSV 240

s SAPS3 701

Secondary Transport 600

Sedation 193, 452

Sensorineural Loss 220 6]3,620,621, 703

619,621,622

Shock 613, Brukhonenko 1

Severe Brain Injury 595

Severe Hypercapnic Respiratory Failure 418

","P\IPI"1'''' Of Illness 810

Shortest Run Time 211

Short-Term Outcome 297

Single Care Giver Model 749

Ventricle 312, 341, 369

Ventricle and Failed Fontan Circulation

331

Index

Single Ventricle Patients 380

SkID Care 206, 285, 566

Skin Integrity 382

Smaller Centers 742

Smaller ECLS Center Volumes 261

SNHL 222

Social Justice 784

South and West Asian ELSO 28

Stability 569,570

Staff Support 463

Staged Palliation 358

Standardized Followup 224,225

Status Asthmaticus 232, 240, 418

Storage 106

Surfactant 128,292

Surrogate Decision Making 783

Survival Rate 395

Survival To Hospital Discharge 322

Sweep Gas 64

T

,.'

Tamponade Of Venous Drainage 351

Ted Kolobow 3, 18

Temperature Management 71

The Patient Who Cannot Be Weaned 469

Therapeutic Hypothermia 193

Thoracotomy Drains 684

Thrombin 83

1hrombocytopenia Associated Multi-Organ Fail­ ure 703

1hromboelastography 96

1hromboelastometry 96

1hrombosis 59

Total Artificial Heart 519

TPE 702

Trachea-Bronchial 594

Tracheostomy 451, 684

Training 749

Transcatheter Aortic Valve Implantation 488

Transcutaneous Continuous Near-Infrared Spec­ troscopy 527

Transfusion 105

Transfusion Thresholds 105

Transplantation 341,396

Transport 743

Transport Equipment 603

Transport Stretcher 605

Transport Teams 602

Transport Vehicle 601

Transpulmonary Flow 518

Trends in ECPR Utilization 322

Triage 741

Trialing 292

Trisomy 21, 176

Two-site VV-ECLS 431

Two Ventricle Physiology 380

u UK collaborative ECMO 774

Ultrasound 248

Unfractionated Heparin 93

Upper-Body 640

V

VA-ECLS 432

VA-ECMO 348

Vaginal Birth 588

Vancomycin 801, 804

Vascular Graft 348

Vasoactive Support 268

Vasopressor 451

Venoarterial ECMO 479,517

Venous Sinus Thrombosis 172

Ventilation Failure 242

Ventilator-Associated Lung Injury 713

Ventilator Management 275

Venting 350

Ventricular Assist Devices 479, 519, 649

Ventricular Decompression 368

Ventricular Failure 367

Ventricular Septal Defect 489

YILI 415,419

VITal Pneumonia 240

VV-ECLS 347,430

VV-ECLS Physiology 465

w Water Based Heater-Cooler Units 72

Weaning 211, 292,555

Weaning ECC02R 717

Weaning Predictors 389

Weaning Trial 389

Weaning VV-ECLS 468

Wound Care 286

831