Practical Old Age Psychopharmacology: a Multi-Professional Approach 9781498793988, 1498793983, 978-1-85775-958-7

Table of contents :
Content: Cover
Title Page
Copyright Page
Contents
Foreword
Preface
About the editors
List of contributors
1 Psychopharmacological history through the looking-glass of advancing years
2 Neuropathology
3 Neurochemical changes in old age
4 Practical issues in assessing and prescribing psychotropic drugs for older people
5 Old age psychopharmacology: the pharmacist's perspective
6 The role of the nurse in the assessment, diagnosis and treatment of older people with psychotropic drugs
7 The GP and 'medicines management' for older people with mental health problems in primary care 8 Management of geriatric mood disorders9 The treatment of psychosis in the elderly
10 Treatment of anxiety disorders
11 Sleep disturbance and its management in older patients Alison Diaper and Ian Hindmarch
12 Drug treatment in dementia
13 Pharmacogenetics and the treatment of Alzheimer's disease
14 The role of drug therapy in the management of behavioural and psychological symptoms of dementia
15 Treatment of delirium
Concluding remarks
Index

Citation preview

Practical Old Age Psychopharmacology : a multi-professional approach Edited by

Stephen Curran

Consultant in Old Age Psychiatry and Visiting Professor of Old Age Psychopharmacology Schoo/ of Human and Health Sciences University of Huddersfield Huddersfield

and

Roger Bullock Consultant Psychiatrist Kingshi/1 Research Centre Victoria Hospital Swindon

Foreword by

Alistair Burns

Boca Raton London New York

CRC Press is an imprint of the Taylor & Francis Group, an informa business

First published 2005 by Radcliffe Publishing Published 2018 by CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742

© 2005 Stephen Curran and Roger Bullock CRC Press is an imprint of Taylor & Francis Group, an Informa business No claim to original U.S. Government works ISBN-13: 978-1-85775-958-7 (pbk) This book contains information obtained from authentic and highly regarded sources. While all reasonable efforts have been made to publish reliable data and information, neither the author[s] nor the publisher can accept any legal responsibility or liability for any errors or omissions that may be made. The publishers wish to make clear that any views or opinions expressed in this book by individual editors, authors or contributors are personal to them and do not necessarily reflect the views/opinions of the publishers. The information or guidance contained in this book is intended for use by medical, scientific or health-care professionals and is provided strictly as a supplement to the medical or other professional's own judgement, their knowledge of the patient's medical history, relevant manufacturer's instructions and the appropriate best practice guidelines. Because of the rapid advances in medical science, any information or advice on dosages, procedures or diagnoses should be independently verified. The reader is strongly urged to consult the relevant national drug formulary and the drug companies' and device or material manufacturers' printed instructions, and their websites, before administering or utilizing any of the drugs, devices or materials mentioned in this book. This book does not indicate whether a particular treatment is appropriate or suitable for a particular individual. Ultimately it is the sole responsibility of the medical professional to make his or her own professional judgements, so as to advise and treat patients appropriately. The authors and publishers have also attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint. Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafrer invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. Typeset by Anne Joshua & Associates, Oxford

Contents Foreword Preface About the editors List of contributors 1 Psychopharm acological history through the looking-glass of advancing years David Healy

V

vi vii viii

1

2 N europathology Margaret M Esiri

11

3 N eurochem ical changes in old age Andrew W Procter

33

4 Practical issues in assessing and prescribing psychotropic drugs for older people Stephen Curran, Sharon Nightingale and John P Wattis

47

5 Old age psychopharm acology: the pharmacist's perspective Paul Hardy and Mary Crabb

73

The role of the nurse in the assessm ent, diagnosis and treatm ent of older people w ith psychotropic drugs Richard J Clibbens

85

1 The GP and 'm edicines m anagem ent' for older people w ith m ental h ealth problem s in primary care Owen P Dempsey

99

6

M anagem ent of geriatric m ood disorders Naila Jawaid and Robert C Baldwin

117

9 The treatm ent of psychosis in the elderly Roger Bullock

137

8

10 Treatment of anxiety disorders Srinivas Suribhatla and James Lindesay

165

11 Sleep disturbance and its m anagem ent in older patients Alison Diaper and Ian Hindmarch

177

12 Drug treatm ent in dem entia Roger Bullock

195

13 Pharm acogenetics and the treatm ent of Alzheim er's disease Clive Holmes and Michelle C McCulley

227

iii

iv

C o n te n ts

14 The role of drug therapy in the m anagem ent of behavioural and psychological sym ptom s of dem entia Roger Bullock

235

15 Treatment of delirium E Jane Byrne

253

Concluding remarks Stephen Curran and Roger Bullock

265

Index

267

Foreword This book is long overdue. The increased aw areness of drugs in psychiatry in general, and old age psychiatry in particular, m eans that a sum m ary of the key drugs and their m ain indications w ill be w ell received by everyone involved in the prescription and m onitoring of m edications. Psycopharm acology is a field w h ich develops quickly, and it is hard to keep up w ith every n e w initiative or ruling by a regulatory agency. N otw ithstanding that, this collection of contribu­ tions serves as a super addition to the literature, and provides an excellent sum m ary of the situation at present. The m ain areas covered include all those w hich an old age psychiatrist w ill encoun ter in practice. After introductory chapters on basic science issues, there is an admirable sum m ary of each of the m ain fields, both organic and functional illness, w h ich sum m arises inform ation available to date. As w ith m any projects, the trick is to get a niche in the market, and the editors and their authors are to be congratulated for putting together such a fine num ber of contributions. This book w ill serve as a baseline for everything else in this area of work - n ot just a baseline but a gold standard. It should be w elcom ed by all clinicians w orking in the area, and w ill serve as an excellent book for continuing professional developm ent. Professor Alistair Burns Professor of Old A ge Psychiatry University of M anchester April 2005

v

Preface In this book w e have brought together findings from recent research w ith a m ulti­ disciplinary perspective into the practical aspects of old age psychopharm acology. W e strongly believe in the im portance of bringing together the best in research and practice. W hen these are divorced, researchers m ay lose the vision of the real purpose of their work, and practitioners m ay not have access to the m ost recent advances in know ledge. W e hope that this book w ill help to bridge this gap. If w e ever develop a m ental illness in later life w e w ould like people to notice it early. W e w ou ld like them to respect our autonom y but steer us quickly in the direction of the best help available. W e w ou ld like to be seen by appropriately qualified people w h o could m ake a diagnosis and provide any useful m edical treatm ent and social, psychological, nursing and therapy support and advice to ourselves and our fam ilies. W e w ou ld like a truly personal approach to our needs. W e believe that this applies equally to drug treatm ents, but in our experience this is n ot alw ays the case. Psychotropic drugs have an im portant role to play in the m anagem ent of older people w ith m ental illness. W e hope that this book w ill help clinicians to use psychotropic drugs m ore appropriately so as to m axim ise their undoubted clinical benefits w h ile m inim ising som e of their u nw anted effects. Stephen Curran Roger Bullock April 2005

vi

About the editors S tep h en Curran works as a Consultant Old Age Psychiatrist in W akefield, and is a Visiting Professor in Old Age Psychopharm acology at the University of Hudders­ field. He graduated in psychology in 1983 and com pleted his undergraduate m edical training in 1986. He th en w orked as a Research Fellow and Lecturer in Old Age Psychiatry at the University of Leeds before m oving to W akefield in 1998. His research interests include old age psychopharm acology, particularly the use of qualitative m ethods to assess drug effects, and he leads the Ageing and M ental Health Research Group at the University. R oger B u llock com pleted his m edical and psychiatric training at Oxford and St B artholom ew 's Hospital, London. In 1993 he was appointed Consultant in Old Age Psychiatry in Sw indon, and tw o years later he founded the Kingshill Research Centre. The work of the centre is to study psychopharm acology and neuropsychology in the elderly, and to apply research from both fields to m ainstream clinical practice. Roger is n o w Clinical Lead of the Sw indon services and Director of the Research Centre, w h ich is n o w affiliated to the University of Bath, w here he is a Reader in Geriatric Psychopharm acology.

vii

List of contributors R obert C B a ld w in , Consultant in Old Age Psychiatry and Honorary Professor, M anchester M ental Health and Social Care Trust, M anchester Royal Infirmary, M anchester Roger B u llock, Consultant Psychiatrist, Kingshill Research Centre, Victoria Hospital, Sw indon, W iltshire E Jane Byrne, Senior Lecturer/Honorary Consultant Psychiatrist, University of M anchester, School of Psychiatry and Behavioural Sciences, Education and Research Centre, W ythen shaw e Hospital, M anchester Richard J C lib bens, Nurse Consultant, Older People's Services, Fieldhead Hospital, W akefield, W est Yorkshire M ary Crabb, Pharmacist, St Luke's Hospital, Huddersfield, W est Yorkshire S tep h en Curran, C onsultant in Old Age Psychiatry and Visiting Professor of Old Age Psychopharm acology, School of Hum an and Health Sciences, University of Huddersfield, Huddersfield, W est Yorkshire O w e n P D em p sey , General Practitioner and Lead Researcher, Lockwood Research Practice, Huddersfield, W est Yorkshire A liso n D iaper, Research Student, HPRU M edical Research Centre, School of Biom edical and M olecular Sciences, University of Surrey, Guildford, Surrey M argaret M Esiri, Professor of N europathology, The Radcliffe Infirmary, Oxford Paul Hardy, Yorkshire

Pharmacist,

Pinderfields

General

Hospital,

W akefield,

W est

D avid H ealy, Honorary C onsultant Psychiatrist/Director, North W ales Depart­ m en t of Psychological M edicine, University of W ales, College of M edicine, Hergest Unit, Gwynedd Hospital, Bangor, G wynedd Ian H indm arch, Professor of H um an Psychopharm acology and Head of HPRU M edical Research Centre, School of Biom edical and M olecular Sciences, U n i­ versity of Surrey, Guildford, Surrey C live H olm es, Senior Lecturer/Honorary Consultant in Old Age Psychiatry, M em ory A ssessm ent and Research Centre, M oorgreen Hospital, Southam pton N aila Jaw aid , Specialist Registrar in Old Age Psychiatry, M anchester M ental Health and Social Care NHS Trust, M anchester Royal Infirmary, M anchester Jam es L indesay, Professor of Psychiatry for the Elderly, University of Leicester, Leicester General Hospital, Leicester M ich elle C M cC ulley, Research Fellow, H um an Genetics Division, University of Southam pton, Southam pton

viii

List o f c o n trib u to rs

ix

Sharon N ig h tin g a le, Consultant in Old Age Psychiatry, Asket Croft C om m unity Unit for the Elderly, Leeds A n d rew W Procter, M anchester

Consultant Psychiatrist, M anchester Royal Infirmary,

S rinivas Suribhatla, Consultant in Old Age Psychiatry, B ennion Centre, Glenfield Hospital, Leicester John P W attis, C onsultant and Visiting Professor of Old Age Psychiatry, School of H um an and Health Sciences, University of Huddersfield, Huddersfield

Chapter I

Ps/chopharmacological history through the looking-glass of advancing years David Healy

Introduction In 1956, Roland K uhn described the effects of im ipram ine in a 56-year-old w om an, Paula J F, w h o had a severe delusional m elancholic disorder. Imipramine w as subsequently dem onstrated to be effective in a patient group suffering from vital depression w ith psychom otor retardation and often psychotic developm ents in older individuals. At the tim e, this condition w as regarded by m any as the distinct diagnostic entity of involutional m elancholia. The success of im ipram ine led to an alm ost im m ediate disappearance of this syndrom e. It m ight n ot be thought surprising that an effective psychotropic agent should lead to the disappearance of a syndrom e, on the basis that m ost people assum e this is w hat effective agents should do. W hat w as astonishing about this disappearance was that, far from im ipram ine being effective in the sense that penicillin w as for general paralysis of the insane (GPI), im ipram ine seem ed to be a treatm ent that validated the existence of the syndrom e and the n eed for its possible ongoing m aintenance treatm ent. Yet a few years later the diagnostic category of involutional m elancholia had effectively disappeared. This disappearance is at odds w ith the m ainstream of psychopharm acological history, in w h ich classification has b een driven largely by the response (or otherw ise) to psychotropic drugs rather than by the clinical criteria that drove classification in the pre-psychotropic era. The disappearance of involutional m elancholia gives a first hint that the history of psychotropic agents given to older individuals m ight be radically different to conventional histories of the psychotropic era. U nfortunately, there are few histories of w hat happens w h en psychotropic drugs are given to older individuals . 1' 2 The usual assum ption appears to have b een that w hat happens in the elderly w ill only reflect w hat is happening in a general adult group. H ow ever, there are good grounds to think that, in future, the books m ay reflect a quite different history, as this brief chapter seeks to illustrate.

The conventional history of the antidepressants The convention al origin of the antidepressant w as w h en Roland K uhn discovered the effects of im ipram ine in 1956. U npersuaded of his reports, Geigy sought confirm ation elsew here. At a tim e w h en com pounds could be synthesised in the

I

2

Practical old age p sy c h o p h arm ac o lo g y

laboratory and could enter m ainstream clinical use w ithin 3 m onths, im ipram ine took 2 years to market. A major reason for Geigy's reluctance appears to have stem m ed from the perception that depression was a relatively rare condition . 3 At alm ost the same tim e as K uhn reported on the effects of im ipram ine to Geigy, N athan Kline m ade the discovery that one of tw o recently synthesised anti-tubercular agents, iproniazid, had psychic energising properties. It later becam e k n ow n as on e of the first of the m on oam in e oxidase inhibitor (MAOI) antidepressants. Roche w ere not interested in such a com pound. Yet ev en before Kline or Kuhn, in 1952, M ax Lurie had discovered that isoniazid, another anti-tubercular agent, had antidepressant properties. This discovery w as replicated in Paris in the sam e year. H owever, n on e of the pharm aceutical com panies that Lurie approached w ere interested in the idea of an antidepressant. The very term w as n ew , having probably b een coined by Lurie. There had indeed b een yet another discovery of a drug w ith antidepressant properties before K uhn and Kline. In 1955, researchers from the Institute of Psychiatry dem onstrated in an article published in the Lancet that reserpine w as an antidepressant. H ow ever, Ciba, the makers of reserpine, w ere not interested in developing an antidepressant. It w as not until the discovery of am itriptyline and its m arketing by Merck that any pharm aceutical com pany sh ow ed an interest in developing an antidepres­ sant. Merck realised that if they w ere going to market this n ew class they w ould also have to sell the illness, so th ey purchased 50 000 copies of Frank Ayd's Recognising the Depressed Patient. 4 Despite this, sales of the antidepressants rem ained flat during the 1960s and 1970s. The antidepressant m arket on ly truly developed w h en the benzodiazepine tranquillisers ran into difficulties in the 1980s. Concerns about dependence on these drugs steered the m arketing developm ent of a n ew generation of drugs that acted on the serotonin system towards an antidepressant route. Thus the selective serotonin reuptake inhibitors (SSRIs) w ere developed in 1971 from a conjunction of clinical observations by Paul Kielholz and n eu ro­ scientific observations by Arvid Carlsson. Kielholz noted that not all of the tricyclic and MAOI antidepressants did the sam e things to people w h o w ere depressed, and he categorised 'antidepressants' according to w hether they w ere energy enhancing or w h ether th ey produced 'cognitive' changes. R eview ing K ielholz's classification in the late 1960s, Carlsson, one of the key pioneers of neuroscience, noted that the antidepressants w hich Kielholz stated w ere more energy enhancing acted on the catecholam ine system , w hereas those w ith cognitive effects acted on the serotonergic system . This led Carlsson to create the first SSRI, nam ely zim elidine, to see w hat the effects of such a drug w ou ld be. Zim elidine was launched in 1982, but w as later w ithdraw n because of toxicological problems. It w as succeeded by fluvoxam ine, fluoxetine, sertraline, parox­ etine and citalopram.

The unconventional history of the antidepressants Follow ing the em ergence of the SSRIs, the antidepressant market m ushroom ed. If antidepressants are effective, this should n ot have happened, and indeed this is not w hat happened in the case of involutional m elancholia. The m ost parsim o­

P sychopharm acological history

3

nious explanation of w h at happened w as that a process w as initiated w hereby cases of V alium w ere converted into cases of Prozac. This did not happen in Japan and m any other parts of the world w here the SSRIs did not em erge as antidepressants so early on. In these countries the antidepressant market remains a com paratively small one com pared w ith the market for tranquillisers. The developm ent of the SSRIs w as predicated on the notion that agents w hich act on catecholam ine system s w ere energy enhancing w hereas agents w hich act on the serotonin system appeared to do som ething else that was either anxiolytic or produced som e other cognitive effect. The SSRIs w ere therefore designed as drugs that w ere different from the tricyclic antidepressants, rather than 'cleanedup' versions of the sam e (w hich is h o w th ey had originally been m arketed). In fact the SSRIs have proved singularly ineffective for hospital depressions or depressions of the kind that Paula J F suffered from. In contrast, selective catecholam ine reuptake inhibitors such as reboxetine have proved very effective in hospital depressions, but apparently m uch less effective in com m unity depressions of the m ore anxious depressive type. This suggests that there m ay w ell be a distinctive com ponent to the kinds of depression cases w h o end up in hospital, particularly those w h o end up in hospital in their m iddle to later years. There rem ains a consensus of opinion that electroconvulsive therapy (ECT) is an effective treatm ent for the kinds of vital depression for w hich im ipram ine was first thou ght to be useful. ECT is n ot regarded as a treatm ent for those kinds of anxious depressions w h ich are found in the com m unity. Broadly speaking, the conditions thought to be responsive to ECT are the ones that in the 1980s w ere thought to be characterised by dexam ethasone non-suppression, a set of depres­ sions in w h ich there is thought to be substantial elevation of cortisol levels, possibly brought about by escape from circadian control. This is not found in the depression cases in the com m unity, and SSRIs in general appear to be less effective in dexam ethasone non-su p p ression . 5 Recently, how ever, m ifepristone, w h ich w as previously used as an abortifacient agent, has b een dem onstrated to have an effect on severe or psychotic depres­ sions . 6 It also appears to have effects on cortisol system s that are comparable to the effects of agents such as dexam ethasone or ECT. Should the clinical efficacy of m ifepristone be replicated, it w ill be an interesting test of pharm aceutical com pany perceptions of the depressive m arket to see w hether it, unlike rebox­ etine, is developed further. If this w ere to happen, there m ight be a n eed to reinvent m elancholia (if not involutional m elancholia) and redesignate the SSRIs as anxiolytics, a process that is arguably w ell under w ay. There are alternative fram eworks w ithin w hich to place such a developm ent. W hile involutional m elancholia m ay never have been the distinctive entity it was once thou gh t to be, this clinical classification arguably em bodied som e clinical insights of substance. The distinctive response of this syndrom e to som e agents and not to others points to different physiological underpinnings. One possibility is that a constitutional or tem peram ental com ponent shapes the differential response to selective agents seen in these vital depressions. Alternatively, a constitutional or tem peram ental com ponent m ay shape clinical presentations so that som e affective disorders in older years have features of anxiety, w hile others have features of a vital syndrom e. Older age m ay in fact be a key setting in

4

Practical old age p sy c h o p h arm ac o lo g y

w h ich som e of these issues that seem irresolvable in younger populations m ay be teased out.

The conventional history of the antipsychotics The conven tion al history of the antipsychotics has it that the discovery of chlorprom azine w as first interpreted as the discovery of a n e w form of sleep treatm ent. The subsequent synthesis of a num ber of non-sedative phenothiazine agents, such as proclorperazine, led to a recognition that these n ew agents m ight have neuroleptic rather than just sedative effects . 7 This created tw o schools of thought. One school based in Lyon argued for a classification of neuroleptic drugs ranging from the sedative to the incisive, and for a recognition that different agents w ould suit different clinical syndrom es and different constitutional types. This notion underpinned the developm ent of sedative neuroleptics such as thioridazine and levom eprom azine, in addition to perphenazine and other m ore incisive agents. In Paris, in contrast, D eniker and D elay argued that a neuroleptic effect w as the com m on therapeutic effect of all of these agents, and that other properties such as sedative properties w ere simply side-effects. The n otions of Delay and D eniker led to a research and drug d evelopm ent agenda that focused on optim ising the core neuroleptic effect. The battle b etw een Lyon and Paris was brief but bitter, w ith Paris trium phing. The developm ent in 1958 of haloperidol, a non-sedative agent, appeared to endorse the insights of the Paris school. Haloperidol becam e the ultim ate neuroleptic, and subsequent antipsychotics w ere developed on the basis of a set of laboratory tests aim ed at replicating its effects. The developm ent of haloperidol alm ost led to the abortion of clozapine. On a num ber of the em erging tests of neuroleptics the latter drug, w hich had first been synthesised in 1958, failed. This failure led to a relatively half-hearted develop ­ m ent programme, w hich w as abandoned once the propensity of clozapine to produce agranulocytosis w as recognised. H owever, a num ber of European clinicians, including Jules Angst, R aym ond Battegay and Hanns Hippius, objected and argued that clozapine w as indeed an effective antipsychotic ev en though it appeared not to have standard neuroleptic properties. The protest kept clozapine alive. W ith the discovery of the effect of the neuroleptic drugs on dopam ine systems, the develop m ent pathw ay focused on producing more selective and m ore potent D2-receptor antagonists. This culm inated in the benzam ide group of drugs, n am ely sulpiride, am isulpiride and rem oxipride, w hich arguably had som ew hat few er side-effects than other agents but w ere in general no m ore potent. W ith the re-em ergence of clozapine, argum ents w ere offered that the additional 5HT2 blockade w h ich this com pound offered m ight account for its apparently better effects than standard neuroleptics in schizophrenia.

The unconventional history of the antipsychotics In the m id-1960s, concerns em erged about the propensity of neuroleptics to produce tardive dyskinesia. This syndrom e, w h ich is som etim es present in untreated older patients w ith psychoses, appeared to occur w ith a m uch greater

P sychopharm acological history

5

frequency and severity in patients w h o w ere taking neuroleptics. At a tim e of rising antipsychiatric discontent w ith orthodox psychiatry, tardive dyskinesia provided a lightning rod to focus this discontent. In 1974, Sm ithKline and French settled their first m illion-dollar legal action w ith regard to tardive dyskinesia, and a subsequent generation of antipsychotic drug developm ent was abandoned. Apart from the benzam ides, w hich w ere not developed in the USA, drug develop m en t in this area w as only renew ed in the w ake of the re-em ergence of clozapine. Contrary to popular m ythologies, the re-em ergence of clozapine stem m ed n ot from any superior clinical efficacy, but rather from the fact that this w as a com pound that did n ot cause tardive dyskinesia. The agents that came in the w ake of clozapine, nam ely sertindole, olanzapine, quetiapine and risper­ idone, w ere also developed primarily as agents that w ou ld m inim ise the risk of tardive dyskinesia. Apart from a low er incidence of tardive dyskinesia, especially for quetiapine, n o n e of these n ew er agents are m ore effective than the older agents. Risperidone and olanzapine both produce dose-dependent extrapyramidal side-effects, and are in fact rather typical neuroleptics. The non-extrapyram idal spectrum of sideeffects of these agents in general suggests that th ey m ay in m any w ays be less safe than the older agents. O lanzapine and clozapine are associated w ith elevated lipid levels, the production of diabetes and w eight gain, all of w hich can be expected to elevate cardiovascular m ortality. The n e w agents differ in their propensity to block D 2 receptors and also in terms of their sedative properties. Am isulpiride and risperidone are m ost like co n v e n ­ tional neuroleptics, w h ile clozapine and quetiapine are m ore like sedative neuroleptics such as thioridazine or levom oprom azine. Recent studies w hich tested the ability of agents such as am isulpiride to augm ent the effects of clozapine in clozapine-refractory patients suggest that this cycle of drug d ev el­ opm ent has returned us to the position that faced clinicians from Lyon in 1958, nam ely that antipsychotic agents vary across a spectrum, and w hile there m ay be som e key elem ents in com m on, such as D2-receptor blockade, the differences b etw een agents are also significant. Classical neuroleptics m ay be effective in particular conditions, w hile others of a m ore sedative type m ay be m ore effective for other conditions. W hat is clear is that this n e w generation of agents has brought an increased interest in the use of antipsychotics in the elderly. The elderly population w ould seem to be the best population in w h ich to test out the com peting visions of Lyon and Paris, in that the onset of psychoses in older age throw s up a set of quite different syndrom es to those that are found in younger populations. For exam ple, w ill typical agents be m ore effective in system atised paranoid states, w ith atypical or sedative agents being m ore effective in disorganised states? There are in fact som e historical pointers that m ight guide clinical observations in this area. In addition to being effective for psychoses, the early results obtained w ith chlorprom azine suggested that this agent was effective in both m anic and delirious states. A lthough the neuroleptics in general are not seen as agents that cure psychoses, they can effectively cure delirious states. The contributions of neuroleptic pharm acology to the m anagem ent of delirious states w ere explored during the 1990s in an interesting set of clinical experim ents in Japan, w hich dem onstrated that m ianserin was an effective agent in the

6

Practical old age p sy c h o p h arm ac o lo g y

treatm ent of delirium . 8 The significance of this lies in the fact that m ianserin is an agent devoid of dopam ine receptor-blocking effects. Its efficacy in the treatm ent of delirious states appears to stem from its 5HT2-blocking properties. This has been confirm ed by subsequent dem onstrations of the efficacy of trazodone in delirious states. These findings suggest in general that agents w ith significant 5HT2-blocking properties, such as clozapine and quetiapine, m ay be particularly effective for disorganised or confusional psychoses in the elderly, w hereas the m ore selective paranoid psychoses m ay respond better to agents w ith potent D2-blocking properties, such as am isulpiride or risperidone.

The conventional history of cognitive enhancing agents The current role of ch olinom im etic drugs in the dem entias arose som ew hat by default. Early neuroscientific research indicated that antidepressants m ight act on both catecholam ine and serotonergic systems, w h ile the antipsychotics acted on dopam ine system s. This left one major brain system w ith ou t either a pathology or a set of drugs. The cholinergic system by default becam e the system that was linked to dem entia. This linkage w as helped by the fact that anticholinergic drugs can produce cognitive effects. This default option was arguably a m istake of historic proportions, in that anticholinergic agents produce confusion or delirium in high doses, rather than dysm nesias . 2 The am nestic effects of anticholinergic agents seem to be less clearcut than the am nestic effects of the benzodiazepines. For exam ple, m any older patients w ith chronic obstructive pulm onary disease have alw ays been m ain ­ tained on anticholinergic agents w ith ou t difficulty. There w ere accordingly few grounds for thinking that ch olinom im etic agents w ould be of significant utility in the dem entias. However, the cholinom im etic research paradigm led to a developm ent pro­ gram m e w hich successively w orked through a variety of cholinom im etic prin­ ciples, from replacem ent strategies w ith choline and lecithin to acetylcholinereleasing agents, such as piracetam and acetyl cholinesterase (ACE) inhibitors, to early ACE inhibitors such as physostigm ine and tacrine, and finally to a later generation of cholinesterase inhibitors such as galantam ine and donepezil. Som e recent cholinom im etic agents appear to have marginal efficacy across the dem entia spectrum. At present, this efficacy seem s to be best interpreted in terms of these drugs having significant effects in particular individuals or for particular syndrom es, such as dem entia w ith Lewy bodies (DLB). This m akes sense given that DLB appears to have a greater cholinergic deficit compared w ith other dem enting syndrom es. In parallel w ith these developm ents in the pharm acotherapy of the dem entias, there has b een an increasing aw areness of the possibilities of cognitive en h a n ce­ m ent in patients w ith conditions other than dem entia. This option opened up w ith early w ork by Krai in the late 1950s on senescent forgetfulness , 9 and has proceeded through to concepts such as age-associated m em ory im pairm ent.

P sychopharm acological history

7

The unconventional history of cognitive enhancing agents Latterly, w ith the em ergence of drugs that act on cholinergic system s, a greater appreciation of the potential role of such drugs has developed. This is perhaps best exem plified by exam ining one of the enduring m yths of psychopharm acology. W ith the em ergence of the m on oam ine theories of depression, antidepressants w ere supposed to act on either catecholam ine or serotonergic system s, according to different schools of thought. H ow ever, com m on to both schools of thought was the n otion that other actions, such as the anticholinergic effects of the tricyclics, could only contribute side-effects. Chief am ong these anticholinergic-induced side-effects was urinary retention. H ow ever, it is n o w clear that the urinary retention produced by tricyclic agents stem s from their catecholam ine reuptakeinhibiting properties. Anticholinergics do n ot cause urinary retention. In fact any clinical trials undertaken w ith anticholinergic agents suggest that these drugs, w h ich clearly cause a relatively rapid onset of euphoria w h en given in m odest doses, m ay in fact be antidepressant. M odulating the cholinergic system has the potential to produce a range of quality-of-life-enhancing effects, as the m arketing of ACE inhibitors as qu ality-of-life-enhancing antihypertensives helps to dem onstrate. Restricting considerations of the role of drugs that act on the cholinergic system to their possible effects in m em ory en h an cem en t or m em ory com prom ise is an unduly restrictive w ay of view ing their use. Further evidence for this thesis can be seen in the increasing use of drugs such as donepezil in trials for childhood disorders such as attention deficit hyper­ activity disorder (ADHD). C onversely, classic rem edies for ADHD, such as m ethylphenidate, have found increasing use in the behavioural m anagem ent of d em enting conditions in the elderly in recent years. A further recent develop m ent has b een the dem onstration that agents w hich act on glutam ate system s, and in particular on the N-m enthyl-D -aspartate (NMDA) receptor, m ay have effects in the dem enting disorders. Glutamate was in fact recognised as a neurotransm itter before dopam ine, and alm ost as early on as noradrenaline and seroton in . 10 H ow ever, despite its widespread availability in the brain, w here it is the com m onest neurotransm itter, a range of m isconceptions about the nature of neurotransm ission prevented an appreciation of its role, as did subsequent difficulties in generating agents that m odulate this system effectively. In recent years this picture has changed, and increasing num bers of agents are being developed, w ith the potential to help both as cognitive enhancers and as agents w h ich m ight prove useful in the m anagem ent of cerebral insults. The first agent of this type to com e on stream was m em antine, w hich has been w idely available on m any European markets from the early 1990s. There is in fact a body of evidence stem m ing back prior to the developm ent of specific NMDAactive agents, indicating that agents such as ketam ine or phencyclidine had prom nestic effects . 11 Future generations of such agents offer interesting prospects because these actions on the NMDA receptor m ay also offer the possibility of neuroprotective effects. The issues of neuroprotection and cognitive enhan cem en t potentially frame a n e w vision of psychopharm acotherapy. There is every chance that the answers to the problem s posed by the affective or schizophrenic psychoses w ill em erge from efforts to arrest neurodegeneration, rather than from the direct attack on these

8

Practical old age p sy c h o p h arm ac o lo g y

disorders that governs current drug developm ent. The issue of cognitive en h a n ce­ m ent in individuals w h o m ay not have frank disease opens up the prospects of smart drugs. At a tim e w h e n alm ost all previous grounds for discrim ination in our societies have been m ade illegal, w e still discriminate on the basis of intellectual advantage. H ow ever, smart drugs are likely to erode advantages, and therefore there m ay w ell be considerable debate surrounding the use of such com pounds in the future. Should th ey be restricted to 'medical' conditions such as ageassociated m em ory im pairm ent, or should they be m ore freely available? Som e of these agents m ay w ell be life prolonging , 12 - and w hat then? The increasing usage of 'drugs for the elderly' in other areas, such as donepezil for children, and the reciprocal use of m ethylphenidate in the elderly, suggests that the 'permafrost' that has encased established notions of psychotropic drug use in the elderly for so long is n o w 'm elting'. The n ew clinical realities w ill also in due course lead to a n e w set of histories.

Conclusions The im pact of treatm ent w ith psychotropic agents in old age psychiatry differs in m any significant respects from that found in other areas of psychiatry. Research­ ers interested in the history of psychotropic drugs w ill in future years probably learn m ore about the true place of these drugs in psychiatry from studying the impact of treatm ent on the clinical syndrom es found in old age psychiatry, and the classification of those syndrom es, than they w ill from studying the impact of treatm ent on syndrom es m anifest in any other age group.

Key points • The use of antidepressants, antipsychotics and cognitive enhancing agents in old age psychiatry dem onstrates interactions b etw een sy n ­ dromes and treatm ents that are not seen as readily in other areas of psychiatry. • The n otion that dem entia involves a cholinergic deficit is probably a historically grounded m istake.

References 1 Ban TA (1980) Psychopharmacology fo r the Aged. Karger, Basel. 2 Ban TA (1995) Bridging the gap b etw een psychogeriatric practice and research: a personal review . Neurol Psychiatry Brain Res. 3: 1 5 5 -6 0 . 3 Healy D (1997) The Antidepressant Era. Harvard U niversity Press, Cambridge, MA. 4 Ayd F (1961) Recognising the Depressed Patient. Grune & Stratton, N ew York. 5 Young EA, A ltem us M, Lopez JF et a l (2004) HPA axis activation in major depression and the response to flu oxetin e. Psychneuroendocrinology. 29: 1 1 9 8 -2 0 4 . 6 Belanoff JK, Flores BH, K alezhan M, Sund B and Schatzberg AF (2002) Rapid reversal of psychotic depression using m efipristone. J Clin Psychopharmacol. 21: 5 1 6 -2 1 . 7 Healy D (2002) The Creation o f Psychopharmacology. Harvard U niversity Press, Cambridge, MA.

P sychopharm acological history

9

8 U chiyam a M, Tanaka K, Isse K and Torn M (1996) Efficacy of m ianserin on sym ptom s of delirium in the aged. Prog Neuropsychopharmacol Biol Psychiatry. 20: 6 5 1 -6 . 9 Krai VA (1962) S enescent forgetfulness: b en ign and m alignant. J Can Med Assoc. 86: 2 5 7 -6 0 . 10 W atkins J (1998) Excitatory am ino acids: from basic science to therapeutic applications. In: D Healy (ed.) The Psychopharmacologists. Volume 2. A rnold, London. 11 Harborne GC, W atson FL, Healy D and Groves L (1996) The effects of sub-anaesthetic doses of ketam ine on m em ory, cognitive perform ance and subjective experience in healthy volunteers. J Psychopharmacol. 10: 1 3 4 -4 0 . 12 Knoll J (2000) The psychopharm acology of life and death. In: D Healy (ed.) The Psychopharmacologists. Volume 3. Arnold, London.

Chapter 2

Neuropathology Margaret M Esiri

Brain anatomy The extraordinary com plexity of the brain's structure can hardly be appreciated from its n aked-eye appearance. W eighing approxim ately 1.5 kg, w h en view ed from the outside it is made up of three m ain com ponents, nam ely the paired cerebral hemispheres w ith their deeply corrugated surfaces, below them the paired cerebellar hemispheres, and b etw een these the brainstem (see Figure 2.1a and b). M ore hints of com plexity of structure em erge as a result of cutting across the cerebral hem ispheres and brainstem (see Figure 2.2a, b, c and d), w h en it can be seen that there are distinct grey and w hite matter regions variously distributed around a central fluid-filled ventricular cavity. This cavity forms the paired lateral ventricles in the cerebral hem ispheres com m unicating w ith the single m idline third and fourth ventricles below . The grey m atter contains the nerve cells and their synaptic connections, and the w h ite matter contains the long axons that interconnect them , surrounded by their fatty m yelin sheaths. The alterations in naked -eye structure of the brain that occur in m ental illness in older people are relatively slight, and m ainly affect the regions that have been stippled in Figures 2.1 and 2. These are listed according to the num bers indicated in Figures 2.1 and 2 in B ox 2.1. It is therefore w ith these regions that w e are concerned, and they w ill be considered in m ore detail below . Not all of these regions are affected in all of the diseases to be considered here. Rather, different diseases are relatively selective in the regions that they damage, as indicated later. In this chapter the anatom y of the relevant brain regions is outlined, follow ed by a brief account of the changes in these regions that occur w ith ageing, and finally by sum m aries of the pathology of the various m ental illnesses that will be considered in this volu m e.

II

12

Practical old age p sy c h o p h arm ac o lo g y

(a)

(b)

F igure 2.1 (a) Lateral and (b) inferior v iew s of th e hu m an brain. Stippling indicates structures affected in m ental disorders in older people.

N e u ro p a th o lo g y

13

(a)

(b)

Figure 2.2 Slices across the m id-cerebrum at tw o adjacent levels, displaying (a) the cut slices and (b) a diagram matic representation, (c) M yelin -stain ed transverse section across the midbrain, (d) M yelin -stain ed transverse section across the m id-pons. The w h ite m atter appears black in the left-hand section, depicted in diagram matic form on the right. (For areas indicated by num bers, see B ox 2.1, w h ich indicates structures affected in m ental disorders in older people.)

14

(c)

(d)

Practical old age p sy c h o p h arm ac o lo g y

N e u ro p a th o lo g y

15

B ox 2.1 Sum m ary of brain structures relevant to m ental disorders in older people Grey-matter structures 1 Cerebral cortex - n ote especially particular subregions, stippled in Figure 2.1, association cortex, cingulate cortex (cing) 2 Parahippocampal gyrus 3 M amillary body 4 Amygdala 5 Basal nucleus of M eynert 6 Hippocam pus 7 H ypothalam us 8 Thalamus 9 Basal ganglia 13 Substantia nigra 14 Raphe nuclei 15 Locus ceruleus White-matter structures 10 Cerebral w h ite matter 11 Corpus callosum 12 Fornix

Grey-matter structures

Cerebral cortex The cerebral cortex consists of a layer of grey m atter of relatively uniform thickness (a few m illim etres) that lies beneath the m eninges and covers the w h o le surface of the corrugations, or gyri, of the cerebral hem ispheres, also extending into the gaps, or sulci, that separate the gyri. W ithin this layer are nerve cells (neurons) orientated perpendicular to the surface, arranged in six strata that differ slightly in their relative thickness and cellular content from one part of the cortex to another. Specialisation of function is found in different regions of the cortex, from reception of sensory inform ation to generation of im pulses respons­ ible for voluntary m ovem en t and execu tion of higher m ental functions including decision m aking, generation and reception of language, m em ory and calculation. T h e se h ig h e r m e n t a l fu n c t io n s a re la r g e ly ca rried o u t in w h a t is te r m e d association cortex, w h ich is found in all the m ain cerebral lobes, but particularly the frontal, tem poral and parietal lobes. A nother im portant part of the cortex for the execu tion of these functions, and for their integration w ith em otions, is the cingulate gyrus, w h ich lies in the m edial aspect of the cerebral hem isphere, just above the corpus callosum . In carrying out its functions the cerebral cortex does n ot act on its ow n, but in concert w ith im pulses conveyed to it from subcortical structures such as the thalam us, amygdala, raphe nuclei, locus ceruleus and nucleus basalis of M eynert (see below ). Activity in the cortex is dependent on the transm ission of im pulses across w ide netw orks of connections from one region of cortex to another, w h ich travel

16

Practical old age p sy c h o p h arm ac o lo g y

through the w h ite m atter w ith in each cerebral hem isphere and also from one hem isphere to the other through a w ide band of w hite-m atter nerve fibres called the corpus callosum. Additional im pulses from the cortex also reach m any subcortical structures, w ith w h ich each area is specifically and selectively co n ­ nected. These form com plex com m unication circuits that return to the cortex or are relayed to further subcortical groups of neurons, or nuclei. Electrical im pulses con veyed along nerve axons are even tu ally converted to chem ical signals that exert their influen ce across narrow synaptic junctions b etw een nerve cells, the chem ical released differing according to the nerve-cell type from w h ich the im pulse originates. Som e chem icals transmit an inhibitory signal to the postsynaptic neuron, and others are excitatory. Cortical neurons are of tw o m ain types, nam ely pyramidal (the majority) and non-pyram idal. Pyramidal neurons convey excitatory signals expressed m ainly by release of the neurotransm itter chem ical glutamate. N on-pyram idal neurons are in general inhibitory and release the chem ical transmitter gam m a-am inobutyric acid (GABA), in som e cases also accom panied by the release of a peptide such as substance P, vasoactive intestinal peptide (VIP) or corticotropin-releasing factor (CRF). To exert their influence on postsynaptic neurons, these neurons need to express specific receptors for the neurotransm itter or neuropeptides.

Hippocampus The hippocam pus is a folded layer of relatively sim plified cortex situated in the m edial part of the tem poral lobe (see Figure 2.2b). It is a three-layered structure containing, like the rest of the cerebral cortex, pyramidal and non-pyram idal neurons. The hippocam pus is closely connected w ith the adjacent entorhinal cortex and parahippocam pal gyrus and, via the fornix, w ith the septal area of the frontal lobe and m am illary body in the hypothalam us. It has a unique function in enabling n e w m em ories to be formed. It is dam aged relatively early in the course of Alzheim er's disease, in w hich im paired ability to form n e w m em ories is often a correspondingly early clinical feature.

Amygdala The amygdala is a closely interconnected cluster of neuronal nuclei lying in the anterior tem poral lobe just in front of the hippocam pus. It is closely connected w ith the hippocam pus, the olfactory sensory system , neighbouring regions of cerebral cortex and the cingulate cortex. It has a key role in em otional responses and feelings. It is dam aged in Alzheim er's disease and dem entia w ith Lewy bodies.

Basal nucleus of Meynert The basal nu cleu s of M eynert and som e related neurons lying m ore anteriorly, in the diagonal band of Broca, are the neurons that supply the cerebral cortex and hippocam pus w ith the neurotransm itter acetylcholine. These clusters of large neurons do not form a w ell-d efin ed grey-m atter structure, but are distributed in the substantia innom inata w hich lies just b elow the anterior com m issure and external segm ent of the globus pallidus, part of the basal ganglia. These clusters of n eurons assum ed im portance in our understanding of dem entia w h en it was discovered that acetylcholine deficiency is an early and conspicuous neurotrans­ m itter deficiency in the cerebral cortex in Alzheim er's disease (see Chapter 3).

N e u ro p a th o lo g y

17

Hypothalamus This is a key centre of autonom ic control in the brain. It lies in the wall of the third ventricle beneath the thalam us. It has m any different constituent nuclei that subserve different functions. One region, the m edian em inence, has co n ­ nections w ith the pituitary gland. A nother portion is the m am illary body, w h ich as w e have already seen has close connections w ith the hippocam pus. The m am illary bodies are dam aged in Korsakov's psychosis, m ost com m only seen as a com plication of alcoholism . A third com ponent is the suprachiasmatic nucleus, w h ich controls sleep rhythm s and develops pathology in Alzheim er's disease. Other nuclei control w ater and food intake and sexual function.

Thalamus The thalam us is a large mass of grey m atter lying in the walls of the third ventricle and extending as far laterally as the internal capsule, a large w hite-m atter tract that conveys axons to and from all parts of the cerebral cortex. There are m any different nuclei in the thalam us, som e of w h ich are concerned w ith relaying sensory inform ation from the periphery, w h ile others are concerned w ith m otor function and yet others w ith m em ory function.

Basal ganglia The basal ganglia form the other large subcortical group of nuclei, lying in front of the thalam us. The three major subdivisions of the basal ganglia are the caudate nucleus (w hich abuts the lateral w all of the frontal horn of the lateral ventricle), the putamen (w hich merges w ith the lateral aspect of the caudate nucleus anteriorly and is separated from it by the anterior limb of the internal capsule m ore posteriorly) and the globus pallidus (w hich lies m edial to the posterior portion of the putam en and lateral to the internal capsule). The basal ganglia are m ainly functionally concerned w ith control of m ovem ent. However, functional activity in this region is altered in obsessive-com pulsive disorder, and addictive behaviour is related to excess dopam ine stim ulation in the nucleus accum bens in the inferior part of the caudate nucleus. The caudate nucleus is relatively selectively dam aged in H untington's disease.

Raphe nuclei The raphe nuclei are a group of nuclei situated in the m edial part of the upper brainstem (midbrain) in the floor of the aqueduct, the narrow channel linking the third and fourth ventricles. Their neurons have axons projecting to all parts of the cerebrum, particularly the cortex and hippocam pus. The neurotransm itter used by raphe neurons is serotonin (5-HT). Damage to these nuclei occurs in Alzheim er's disease.

Locus ceruleus The locus ceruleus is a tightly localised colum n of pigm ented neurons lying in the upper and m id pons, just ben eath the floor of the upper part of the fourth ventricle. It is the source of the transmitter noradrenalin, w h ich it supplies through axonal connections to m ost parts of the cerebrum and cerebellar cortex. As w ith the substantia nigra, the pigm ent in its neurons is m elanin,

18

Practical old age p sy c h o p h arm ac o lo g y

w h ich accum ulates as a by-product of noradrenalin synthesis. The locus ceruleus is dam aged in Alzheim er's disease and to a lesser extent in Parkinson's disease.

W hite-m atter structures The w h ite m atter consists alm ost exclusively of axons, m any of w h ich are surrounded by m yelin sheaths, and the blood vessels needed to nourish them . R elatively short axons linking local regions of cortex run in the im m ediately subcortical w hite matter, w h ile longer connections linking m ore distant parts of cortex of the sam e or the contralateral hem isphere, or linking cerebral cortex w ith subcortical structures, run more deeply in the w h ite matter. A reduction in the density of nerve fibres in w hite m atter can arise as a result of directly inflicted damage, usually by ischaem ia, but it can also occur secondarily to loss of parent n erve cells in cortex or subcortical nuclei.

Changes in brain structure with normal ageing The pathology of specific diseases that affect the brain towards the end of life needs to be distinguished from changes that occur as an inevitable consequence of ageing. These are found ev en in the brains of older people k n ow n to have had w ell-preserved m ental function w h en they w ere alive . 1 These changes chiefly consist, at the m acroscopic level, of the following: • • • •

a slight but significant reduction in brain w eight slight enlargem ent of the lateral ventricles slight fibrous thickening of the leptom eninges that cover the brain a slight reduction in the volum e of the cerebral w hite matter.

The m ain m icroscopic changes that occur are as follows: • accum ulation of dense spherical structures called corpora am ylacea w hich develop m ainly in the processes of glial cells (the cells that nurture and support nerve cells) • a reduction in the size of neurons at som e sites, including the cerebral cortex • a reduction in the num ber of synapses in the cerebral cortex • a reduction in the num ber of neurons at som e sites, including the cerebral cortex and hippocam pus • the appearance in the association cortex of argyrophilic (silver-staining) plaques of the type also seen in Alzheim er's disease (see below ) • the appearance in the hippocam pus and adjacent cortex of small num bers of neurofibrillary tangles of the type also seen in Alzheim er's disease (see below ). The first tw o of these m icroscopic changes are not clearly correlated w ith degeneration of neurons. The reduction in the size of neurons is likely to be related to the reduction in the num ber of synapses, since the size of a nerve-cell body is related to the exten t of its processes and the num ber of synapses on those processes that it needs to support. The reduction in the num ber of neurons w ith age is of the order of 1 % per year after the age of 70 years for cerebral cortex 2 and for som e subregions of the hippocam pus . 3

N e u ro p a th o lo g y

19

The m ost likely reason for the changes seen in the brain w ith ageing is progressive damage to neuron constituents (m itochondria, DNA and cell m em ­ branes) caused by free radicals, coupled w ith progressive failure of the m ech an ­ isms norm ally intended to protect cell constituents from such dam age .4

Neuropathology of dementia Alzheimer's disease A bout 75% of cases of progressive dem entia in the elderly are due to Alzheim er's disease occurring either alone or w ith vascular or Lewy body disease. It is therefore not surprising that a major research effort has been under w ay for som e decades to enable a full understanding of the disease to be acquired. H ow ever, there are still m any aspects of this com plex disease that need to be clarified. Recently, anim al m odels based on genetic m utations that cause familial disease have helped to elucidate som e of the m ultiple factors that contribute to the disease . 5 In elderly sufferers from Alzheim er's disease the brain m ay sh ow only mild and non-specific changes to the naked eye, nam ely a m ild degree of cerebral cortical and hippocam pal atrophy, mild ventricular enlargem ent and loss of pigm ent from the locus ceruleus. In early-onset disease these changes tend to be m ore severe. H ow ever, in both age groups the distinctive pathology is only seen at the m icroscopic level. It w as m en tion ed earlier that ev en in norm al elderly people's brains, substantial num bers of argyrophilic plaques m ay be present in m icroscopic sections of association cortex. These plaques consist of abnormal fibrillary material deposited in clum ps w ith in the neuropil (the tissue that occupies the space b etw een cells in the cerebral cortex). H ow ever, detailed study of prospectively assessed elderly subjects has sh ow n that a particular subtype of these plaques, w h en present in large num bers, is associated w ith dem entia and occurs alongside the other hallm ark m icroscopic feature of Alzheim er's disease, nam ely neurofibrillary tangles. The subtype of plaque that is associated w ith dem entia is the neuritic plaque - one in w h ich there are abnorm al neuritic processes caught up in the structure (see Figure 2.3). Plaques that lack this feature are referred to as diffuse plaques (see Figure 2.4), and these are the ones that are frequently abundant in the brains of u ndem ented elderly people. It is though t likely that diffuse plaques are converted into neuritic ones over the course of tim e as additional pathological reactions take place in th e m . T h is v ie w is su p p o r te d b y t h e o b s e r v a tio n th a t in su b je c ts w it h D o w n 's

syndrom e, in w h o m the pathology of Alzheim er's disease invariably develops by late m iddle age, diffuse plaques are seen at an earlier age than neuritic o n es . 6 M any neuritic plaques have at their centre a dense core of am yloid surrounded by a cluster of microglial cells (glial cells that are of the m acrophage lineage). B eyond them lies a corona of neuritic and glial cell processes. W ithin the neuritic processes in such plaques can be found, at the ultrastructural level, abnormal m itochondria and abnorm al paired, helically w ou n d filam ents that are structu­ rally and chem ically identical to the abnormal filam ents that m ake up the inclusions in neuron cell bodies that are called neurofibrillary tangles (see Figure 2.5). It was tangles that A lzheim er particularly associated w ith dem entia,

20

Practical old age p sy c h o p h arm ac o lo g y

F igure 2.3 M icroscopic appearance of a neuritic plaque stained w ith silver. The black granules (indicated by arrowheads) represent neuritic processes. At the centre is an am yloid core.

F igure 2 .4 M icroscopic appearance of diffuse plaques stained w ith silver (indicated by arrows). There are n o neuritic processes, and n o am yloid core is present.

and more recent studies fully corroborate this association by dem onstrating that the severity of dem entia in an individual w ith Alzheim er's disease is closely correlated w ith the num bers of tangles that are present in the brain . 7,8 In contrast, the correlation of dem entia severity w ith the num ber of neuritic plaques is only m odest, and the correlation w ith the num ber of diffuse plaques is insignificant . 8

N e u ro p a th o lo g y

2 1

F igure 2.5 M icroscopic appearance of n eu ron s containing neurofibrillary tangles (indicated by arrows). Section stained w ith Galiyas silver stain.

Thus, in brief, Alzheim er's disease is characterised by a com plex pathology in w h ich there is deposition of plaques outside cells in regions of grey matter that are k n ow n to be im portant for cognitive function (association cortex and hippo­ cam pus), coupled w ith the appearance of tangles inside neurons in similar areas of the brain. At the start of the process, plaques are widespread but diffuse in type w hereas tangles are restricted to the hippocam pus and adjacent entorhinal cortex. This small am ount of tangle form ation does not initially interfere w ith m em ory and cognitive function, but as the process continues and m ore tangles are formed, impaired m em ory starts to m ake an appearance. Later, plaques becom e neuritic and tangles becom e m ore widespread, occurring m ainly in pyramidal neurons of the association cortex but also in subcortical nuclei that project to the cortex, particularly the nucleus basalis of M eynert, the raphe nuclei and the locus ceruleus. Tangles also develop in other n uclei in, for exam ple, the amygdala, the suprachiasm atic nucleu s and som e other hypothalam ic nuclei, and in som e parts of the thalam us. Plaques m ay also appear in subcortical nuclei such as the amygdala, caudate nucleus and putam en. It has b een noted that populations of neurons that are prone to develop tangles also lose a substantial proportion of their neurons, probably by an apoptotic-like process. Som e neurons that harbour tangles die, leaving an insoluble 'ghost' tangle marking their place, but other neurons seem to die w ith ou t developing tangles. Even tangle-bearing neurons that rem ain alive are unable to function norm ally because their cytoskeleton collapses. This is because the filam entous protein accum ulations that form tangles are com posed of a protein, tau, that is norm ally present in non-fibrillar form, helping to m aintain the cytoskeleton of m icrotubules and neurofilam ents in normal alignm ent. Filam entous tau is hyperphosphorylated by abnorm ally expressed enzym es, although exactly w h y this occurs and h o w it is related to the deposition of plaques, w hich probably occurs initially at synapses, is not fully understood. Plaques them selves are

22

Practical old age p sy c h o p h arm ac o lo g y

com posed of a different protein, /3-amyloid, w h ich is a polypeptide product of the enzym atic digestion of the am yloid precursor protein, a m em brane protein of u n k n o w n function. /3-amyloid, in the fibrillar form in w hich it is deposited in plaques, has been sh ow n to have toxic effects on neurons in culture, although it is by no m eans clear w h eth er tangle form ation in nerve cells in m ost cases of sporadic Alzheim er's disease is sim ply a con sequ en ce of the production of excess am ounts of /3-amyloid. One difficulty w ith this hypothesis is that, w h ile transgenic m ice that express disease-causing m utations in the gene coding for am yloid precursor protein develop large num bers of cortical plaques, they do not develop tangles . 9 H owever, the processes that are thought to give rise to Alzheim er's disease in hum ans operate over very long periods, of the order of m any years, w h ich m ay m ake a short-lived rodent an unsuitable anim al on w hich to m odel the disease. There are a num ber of risk factors for Alzheim er's disease other than the m ain on e of advancing age. Apart from the m inority of cases in w hich gene m utations in the am yloid precursor protein gene or in the genes for presenilin 1 and 2 are responsible for (usually) early-onset disease, these include the epsilon-4 p o ly ­ m orphism in the gene for apolipoprotein E, som e other less w ell-established gene polym orphism s, hypertension and exposure to head trauma. Exactly h o w such risk factors operate is as yet unclear. Discussion of them can be found in recent review s . 10,11 Som e of the sym ptom s of Alzheim er's disease, and indeed som e of the m ost troublesom e sym ptom s to deal w ith, are caused by changes in behaviour rather than in cognition. A lthough the pathological basis of such behavioural changes is m uch less w ell understood than that of cognitive decline, there has been som e research conducted on these aspects of the disease in the hope that increased understanding of the basis of these sym ptom s m ay lead to an im provem ent in their alleviation (see Chapter 13). In general, changes in behaviour in Alzheim er's disease relate to changes in subcortical nuclei or to attempts by the brain to overcom e the effects of subcortical pathology. Thus depression in the disease has b een correlated w ith loss of serotonin in the cortex secondary to loss of and tangle form ation in raphe neurons. A nxiety has been correlated w ith relatively good preservation of 5HT2a receptors in cortex , 12 aggression has b een linked to loss of locus ceruleus n eu ro n s 13 or relative preservation of substantia nigra n eurons , 14 and psychotic sym ptom s have b een linked to changes in m uscarinic acetylcholine receptors in the cortex . 15 W ith regard to dealing w ith the problem of Alzheim er's disease in ageing world populations, it is w orth noting that at present only treatm ent that am eliorates (albeit incom pletely) the cholinergic transm ission deficit, or that alleviates b eh av­ ioural changes, is available. Such treatm ent does not prevent the disease from progressing. M uch m ore effective w ou ld be a m eans of preventing or reversing the abnorm al fibrillar protein deposits that accum ulate inside and outside neurons, especially the intracellular tangles that are so closely related to defective cognition and neuron drop-out. A lthough there is som e experim ental evidence that plaques m ay be rem ovable , 16 there has been no sign as yet of a m echanism to rem ove tangles.

N e u ro p a th o lo g y

23

Vascular disease and dementia There has b een considerable confusion over several decades as to the role of cerebrovascular disease in dem entia, but som e clarification has em erged more recently. One of several difficulties that researchers have had to contend w ith is that cerebrovascular disease takes diverse forms, som e of w hich occur together, and som e of w hich are difficult to quantify w ith regard to their severity. One con ven ien t w ay to classify cerebrovascular disease is into tw o subdivisions as follows: •

lobar infarction - in volving cerebral cortex and contiguous w h ite matter, usually the result of obstruction to blood flow in a major artery or its branches, and classically giving rise to 'stroke' • subcortical infarction - in volving m ainly w hite matter in a relatively diffuse distribution and deep grey m atter in the form of lacunes. Subcortical infarction is com m only incom plete, producing attenuation rather than obliteration of tissue, particularly in w hite m atter (see Figure 2.6). It is associated w ith inadequate blood perfusion through arterioles and small arteries that have had their lum ens narrowed but not usually obstructed. The vessel walls are thickened by fibrosis. These changes are often consequent upon longstanding hypertension.

F igu re 2.6 L ow -pow er v ie w of a section of basal ganglia and w h ite m atter stained for m yelin. The 'lacy' appearance is due to w id en in g of perivascular spaces around small arteries and arterioles. M yelin, w h ich is stained black, is pale around som e of these vessels (see arrows), indicating loss of m yelin.

24

Practical old age p sy c h o p h arm ac o lo g y

Initial studies on the role of cerebrovascular disease in dem entia suggested that the latter was related to large volu m es of lobar infarction. H owever, recent studies have sh ow n that dem entia is m ore closely correlated w ith subcortical vascular disease . 17 This is the type of disease w hich, in advanced form, has long been recognised as Binswancjer's disease. A recently recognised familial disease w ith similar pathology is referred to as CADASIL (cerebral arteriopathy w ith dem entia and subacute ischaem ic encephalopathy), and tends to present w ith early-onset d em entia . 18 These diseases produce a type of dem entia that is characteristic of subcortical pathology, w ith poor concentration and slow ed responses. In general, it is unusual for vascular disease to be the sole cause of dem entia. Cerebrovascular disease and Alzheim er's disease can com bine to cause d em en ­ tia. It has b een recognised for m any years that cerebrovascular disease and Alzheim er's disease com m only occur together. There are several com m on risk factors for the tw o diseases, including possession of the ApoE e4 gene polym orph­ ism, hypertension, and elevated levels of cholesterol and hom ocysteine in the blood. One specific type of vascular disease, nam ely /3-amyloid deposition in the w alls of leptom eningeal and cortical vessels, does indeed occur in over 90% of cases of Alzheim er's disease, but it is unclear w hether this significantly com pro­ m ises blood supply to the brain. H owever, it can give rise to small and large lobar or subarachnoid haem orrhages. One form of dem entia occurs shortly after a stroke. A lthough initially it was though t that 'post-stroke' dem entia was likely to be due to vascular disease, it n o w seem s probable that som e cases are due to subclinical Alzheim er's disease w h ich has b een 'unm asked' by an additional insult to the brain in the form of a stroke. Pathological study of prospectively assessed elderly subjects has show n that those w h o w ere found at autopsy to have relatively mild degrees of A lzheim er-type pathology w ith neurofibrillary tangles confined to the entorhinal cortex and hippocam pus w ere not clinically dem ented in life unless they w ere also found to have additional cerebrovascular disease . 19 Subcortical lacunes also distinguished d em ented subjects am ong elderly nuns w ith abundant plaques in another study . 20 Thus Alzheim er's disease, particularly in its early stages, can sum m ate w ith cerebrovascular disease to cause dem entia. The im portant im plica­ tion is that prevention of cerebrovascular disease has the potential to reduce clinical expression of Alzheim er's disease in its com m on early stages.

Dementia with Lewy bodies The classical disease associated w ith the presence of Lewy bodies in the brain is Parkinson's disease, in w h ich a m ovem en t disorder characterised by bradykinesia, cogw heel rigidity and tremor is attributable to loss of pigm ented neurons from the substantia nigra and the presence of inclusions, term ed Lewy bodies, in som e surviving pigm ented neurons. Lewy bodies are rounded, frequently spherical structures com posed of granular and filam entous com ponents that occur in the cytoplasm of neurons (see Figure 2.7). Their principal chem ical constituent is a protein called a -syn u clein , w h ose norm al function is uncertain. A lthough they are m ost com m only found in the substantia nigra and locus ceruleus in Parkinson's disease, Lewy bodies also occur at several other sites, including the

N e u ro p a th o lo g y

25

F igure 2.7 M icroscopic appearance of Lew y bodies in a pigm ented n eu ron in the sub­ stantia nigra (h aem atoxylin and eosin stain). Inset sh ow s cortical Lewy body stained w ith an antibody to a -sy n u clein .

amygdala, nucleus basalis and hypothalam us, in that disease. A few m ay also be found in the cerebral cortex. D em entia is m ore com m on in people w ith Parkinson's disease than in healthy individuals of similar age. The pathology in som e cases of dem entia w ith Parkinson's disease com bines that of Alzheim er's disease w ith that of Parkinson's disease, but in m any cases A lzheim er-type pathology, if present, is on ly mild in extent and insufficient on its o w n to explain the dem entia. Such cases also have considerable num bers of Lewy bodies in cerebral cortical neurons, m ost com ­ m on ly in the parahippocam pal gyrus and cingulate gyrus. It is n o w recognised that a form of dem entia can occur in w h ich the m ain burden of Lewy body pathology falls on the cerebral cortex, rather than on subcortical nuclei such as the substantia nigra. These changes result in dem entia w ith ou t very m any, if any, Parkinsonian features, although at autopsy the substantia nigra does show som e neuron loss and Lewy body form ation. This condition is kn ow n as dementia with Lewy bodies.21 It has certain characteristic clinical features that m ake it recognisable, nam ely fluctuation in the severity of dem entia, visual hallucina­ tions and, often, mild parkinsonian features. It has been sh ow n that in dem entia w ith Lewy bodies the cortical cholinergic deficit is m ore severe than in A lzheim er's disease, probably because the nucleus basalis is dam aged by both pathologies . 22 Thus such patients m ay respond w ell to drugs that augm ent cholinergic function. They are also unduly sensitive to neuroleptic m edication, w hich can cause significant clinical deterioration. For these reasons it is im port­ ant to recognise dem entia w ith Lewy bodies clinically and to distinguish it from 'pure' Alzheim er's disease.

26

Practical old age p sy c h o p h arm ac o lo g y

Pick’s disease and other forms of frontal lobe dementia These diseases are m uch less com m on than Alzheim er's disease, and tend to present initially w ith altered behaviour, reflecting impaired frontal lobe function, or w ith language deficits rather than w ith m em ory problems. Som e cases present in m iddle age, but others present later. About 50% of the cases have a fam ily history of dem entia. In Pick's disease the naked-eye changes in the brain are usually more severe than in Alzheim er's disease, w ith marked atrophy of the poles of the frontal and/or tem poral lobes (see Figure 2.8), and marked expansion of the frontal and inferior horns of the lateral ventricles. The caudate nuclei m ay also be m oderately atrophied.

Figure 2.8 M acroscopic appearance of the brain from a case of Pick's disease. N ote the narrow ing of cortex and w id en in g of sulci (reflecting cortical atrophy) in th e frontal and tem poral lobes, w ith m assive com pensatory dilation of the lateral ventricles.

The microscopic features of Pick's disease are a dramatic loss of neurons from the atrophic cortical regions, reactive enlargem ent and an increase in the num ber of glial cells (astrocytes), and the presence in som e rem aining neurons of fibrillary rounded inclusions kn ow n as Pick bodies (see Figure 2.9). These bodies are com posed of the sam e protein, tau, that accum ulates as tangles in Alzheim er's disease, although it has a slightly different m olecular conform ation in Pick bodies to that in tangles. Som e neurons are greatly expanded to form 'ballooned' neurons by the intracytoplasm ic tau filam ents - these are called Pick cells. These inclusions are found principally in frontal and tem poral lobe cortex and hippocam pus. A few inherited cases of Pick's disease have been show n to be due to m utations in the tau gene on chrom osom e 17.

N europathology

27

F igure 2.9 M icroscopic appearance of Pick bodies in a frontal lobe section stained w ith an antibody to tau protein.

Som e cases w ith a clinical presentation similar to that of Pick's disease have m uch less severe frontal and tem poral lobe atrophy and lack the tau inclusions characteristic of that disease. Such cases frequently sh ow little in the w ay of specific m icroscopic pathology (mild neuron loss in the cortex of frontal and anterior tem poral lobes, m ild reactive astrocytosis and the presence of vacuoles in the second layer of the cortex). In these regions it m ay be possible to dem onstrate small distorted neuritic processes that react w ith the protein ubiquitin, a protein that binds to cytoplasm ic proteins destined for degeneration. (Ubiquitin anti­ bodies also bind to the proteins in Lewy bodies and neurofibrillary tangles.) This condition is frequently term ed non-specific frontal lobe dementia.

Dementia caused by alcohol misuse There are both reversible and irreversible effects of excess alcohol intake on brain structure. The reversible effect is atrophy of the frontal lobes, particularly the w hite matter. This is probably due to rem oval of water from the tissue. The irreversible effects are thought to be due in part to a direct toxic influence on the cerebellum , w h ich leads to patchy loss of Purkinje cells, particularly in the vermis (m idline region), and to loss of som e cerebral cortical nerve cells. The other irreversible pathology is due to vitam in Bx (thiam ine) deficiency, to w hich severe alcoholics are prone, possibly because the cerebral m etabolism of alcohol requires a high level of consum ption of the vitam in, and alcoholics tend to have a diet deficient in vitam ins. Vitamin B 2 deficiency causes damage to the capillaries in the regions of the brain close to the third ventricle, particularly the m am illary bodies and dorsom edial nucleus of the thalam us. This damage to the endothelial cells allow s leakage of red blood cells into the neuropil, w here they are phagocytosed by m acrophages, lysed and m etabolised to haem osiderin. N eurons

28

Practical old age p sy c h o p h arm ac o lo g y

are not the primary target of damage in this vitam in deficiency, but are damaged secondarily to the vascular damage, haem orrhage and iron deposition.

Other causes of dementia The three conditions of Alzheim er's disease, cerebrovascular disease and d em en ­ tia w ith Lewy bodies alone, or in com bination, account for the great majority of cases of dem entia in the elderly. H ow ever, there are m any other diseases that are occasionally encountered and w h ich should be borne in mind, particularly w h en there are atypical features in the clinical presentation. These include cerebral tum ours (gliomas, m etastases and m eningiom as), norm al-pressure hydro­ cephalus, HIV/AIDS, Creutzfeldt-Jakob disease, progressive supranuclear palsy, H untington's disease, leucodystrophies, and chronic subdural haem atom as. For accounts of the pathology of these conditions, the reader is referred to Graham and Lantos 23 and Esiri et al.24

Neuropathology of depression in older people The relatively effective treatm ent of depression w ith selective inhibitors of serotonin reuptake (see Chapter 8 ) im plicates the raphe nuclei and their projec­ tions in this condition. Until relatively recently there had been very little study of the neuropathology of depression, but there have n o w been neuroim aging and neuropathological studies of unipolar depression and of bipolar disorder in subjects of various ages, and of depression in the elderly, w hich are beginning to cast som e light on this subject. In familial unipolar depression and bipolar disorder, attention w as directed to the subgenual cortex, the m ost anterior and inferior part of the cingulate cortex, by positron em ission topography (PET) studies that found a reduced volu m e of cortex here . 25 Subsequent n europatho­ logical study of this area show ed a reduced num ber of glial cells, but no change in the n eu ron s . 26 Neuroim aging studies of depression in the elderly have d em on ­ strated changes in the frontal lobe w h ite m atter suggestive of vascular dam age , 27 and pathological studies have reported non-specific inflam m atory changes in adjacent dorsolateral prefrontal cortex consistent w ith reaction to vascular dam age . 28 Other neuropathological studies, in these instances directed at the brainstem nuclei sending serotonergic and noradrenergic projections to the cortex, the raphe n uclei and locus ceruleus, have reported reduced 29 or u n a l­ tered 30 num bers of neurons in these nuclei in depression. A lthough these initial studies of depression present a som ew hat bew ildering picture of disparate pathology in the cortex, cerebral w h ite m atter and brainstem, it could be the case that these reflect impaired delivery of serotonin to the cortex, w hich could conceivably influ en ce glial cell num bers (since glial cells have receptors for 5-HT on their surface), interruption of serotonergic and noradrenergic axons in w hite m atter by subcortical vascular disease, and a reduction in the num ber of raphe and locus ceruleus neurons in old age associated w ith A lzheim er-type pathology or age-related change. These possibilities w ill need to be investigated further in future studies.

N e u ro p a th o lo g y

29

Neuropathology of psychotic disorders Psychotic sym ptom s som etim es develop in dem ented individuals w h o m ay be suffering from either A lzheim er's disease or dem entia w ith Lewy bodies. The pathology of such sym ptom s has b een touched upon in the section on dem entia above. The neuropathology of schizophrenia is beyond the scope of this chapter, but those w h o w ish to read about it are referred to review s by Harrison 31 and Esiri and Crow . 32

K ey p o in ts • Brain structure and chem istry are subtly altered in m ental disorders of old age. M any w idespread areas of the brain can sh ow changes, w ith the region affected depending on the disease. • In Alzheim er's disease the hallmark changes are m icroscopic - principally the form ation of extracellular plaques com posed m ainly of /3-amyloid protein, and intraneuronal tangles com posed of tau protein. These d e­ posits are accom panied by neuron and synapse loss, particularly in the cerebral cortex and hippocam pus. • In vascular dem entia, subcortical sm all-vessel disease is m ore significant than major lobar infarction. Vascular disease on its ow n only rarely causes dem entia, but it com m on ly sum m ates w ith m ild A lzheim er-type pathology to do so. It m ay also cause depression in old age. • D em entia w ith Lewy bodies represents a form of dem entia in w hich pathology of the type seen in Parkinson's disease participates. It deserves clinical recognition because it responds relatively w ell to cholinesterase inhibitor drugs, but is adversely affected by neuroleptics. • A lthough Alzheim er's disease, cerebrovascular disease and dem entia w ith Lewy bodies are (in that order) the com m onest causes of dem entia in the elderly, there are m any other causes w hich should be considered, particularly in clinically atypical cases.

References 1 Esiri MM (1994) D em entia and norm al aging. In: FA Huppert, C Brayne and DW O 'Connor (eds) Dementia and N orm al Aging. Cambridge U niversity Press, Cambridge. 2 A nderson JM, Hubbard BM, Coghill GR and Slidders W (1983) The effect of advanced old age on the n eu ron e con ten t of th e cerebral cortex. Observations w ith an autom atic im age analyser point cou n tin g m ethod. J Neurol Sci. 58: 2 3 5 -4 6 . 3 W est MJ, C olem an PD, Flood DG and Troncoso JC (1994) D ifferences in the pattern of hippocam pal neuronal loss in norm al ageing and A lzheim er's disease. Lancet. 344: 7 6 9 -7 2 . 4 H alliwell B and Gutteridge JMC (1999) Free Radicals in Biology and Medicine. Oxford U niversity Press, Oxford. 5 Borchelt D (2003) Transgenic m ou se m odels of dem entia. In: M M Esiri, VM-Y Lee and J Trojanowski (eds) The Neuropathology o f Dementia (2e). Cambridge U niversity Press, Cambridge.

30

Practical old age p sy c h o p h arm ac o lo g y

6 Lemere CA, Blusztajn JK, Y am aguchi H, W isniew ski T, Saido TC and Selkoe DJ (1996) Sequence of deposition of h etero g en eo u s am yloid beta-peptides and Apo E in D ow n syndrom e: im plications for initial even ts in am yloid plaque form ation. Neurobiol Dis. 3: 16-32. 7 W ilcock GK and Esiri MM (1982) Plaques, tangles and dem entia. A quantitative study. J Neurol Sci. 56: 3 4 3 -5 6 . 8 Nagy Z, Esiri MM, Jobst KA et al. (1995) Relative roles of plaques and tangles in the dem entia of A lzheim er's disease: correlations using three sets of neuropathological criteria. Dementia. 6: 2 1 -3 1 . 9 Games D, Adams D, Alessandrini R et a l (1995) A lzheim er-type neuropathology in transgenic m ice overexpressing V717F beta-am yloid precursor protein. Nature. 373: 5 2 3 -7 . 10 Ritchie I< and L ovestone S (2002) The dem entias. Lancet. 360: 1 7 5 9 -6 6 . 11 Kawas CH and Katzman R (1999) Epidem iology of dem entia and A lzheim er disease. In: RD Terry, R Katzman, KL Bick and SS Sisodia (eds) Alzheim er Disease. Lippincott, W illiams and W ilkins, Philadelphia, PA. 12 Esiri MM (1996) The basis for behavioural disturbances in dem entia. J Neurol Neurosurg Psychiatry. 61: 1 2 7 -3 0 . 13 M atthew s KL, C hen CP, Esiri MM, K eene J, M inger SL and Francis PT (2002) Noradrenergic changes, aggressive behavior, and cognition in patients w ith dem entia. Biol Psychiatry. 51: 4 0 7 -1 6 . 14 Victoroff J, Mack WJ, Lyness SA and Chui HC (1995) M ulticenter clinicopathological correlation in dem entia. A m J Psychiatry. 152: 1 4 7 6 -8 4 . 15 Lai MK, Lai OF, K eene J et a l (2001) Psychosis of A lzheim er's disease is associated w ith elevated m uscarinic M 2 binding in the cortex. Neurology. 57: 8 0 5 -1 1 . 16 Bard F, C annon C, Barbour R et a l (2000) Peripherally adm inistered antibodies against am yloid beta-peptide enter the central nervous system and reduce pathology in a m ouse m odel of A lzheim er disease. N at Med. 6: 9 1 6 -1 9 . 17 Esiri MM, W ilcock GK and Morris JH (1997) N europathological assessm ent of the lesions of significance in vascular dem entia. J Neurol Neurosurg Psychiatry. 63: 7 4 9 -5 3 . 18 Morris JH and Kalim o H (2003) Vascular dem entia. In: MM Esiri, VM-Y Lee and J Trojanowski (eds) The Neuropathology o f Dementia (2e). Cambridge U niversity Press, Cambridge. 19 Esiri MM, Nagy Z, Sm ith MZ, Barnetson L and Sm ith AD (1999) Cerebrovascular disease and threshold for dem entia in the early stages of A lzheim er's disease (letter). Lancet. 354: 9 1 9 -2 0 . 20 Snow don DA, Greiner LH, M ortimer JA, Riley KP, Greiner PA and M arkesbery WR (1997) Brain infarction and the clinical expression of A lzheim er disease. The N un Study. JAM A. 277: 8 1 3 -1 7 . 21 M cKeith IG, Galasko D, Kosaka I< et a l (1996) C onsensus guidelines for the clinical and pathologic diagnosis of dem entia w ith Lewy bodies (DLB): report of the consortium on DLB international w orkshop. Neurology. 47: 1 1 1 3 -2 4 . 22 Perry EK, H aroutunian V, Davis KL et al. (1994) Neocortical cholinergic activities differentiate Lewy body dem entia from classical A lzheim er's disease. Neuroreport. 5: 7 4 7 -9 . 23 Graham DI and Lantos PL (eds) (2002) Greenfield's Neuropathology (7e). Arnold, London. 24 Esiri MM, Lee VM-Y and Trojanowski J (eds) (2004) The Neuropathology o f Dementia (2e). Cambridge U niversity Press, Cambridge. 25 D revets WC, Price JL, Sim pson JR Jr et a l (1997) Subgenual prefrontal cortex abnorm alities in m ood disorders. Nature. 386: 8 2 4 -7 . 26 Ongur D, Drevets WC and Price JL (1998) Glial reduction in the subgenual prefrontal cortex in m ood disorders. Proc N atl Acad Sci USA. 95: 1 3 2 9 0 -5 .

N e u ro p a th o lo g y

3 1

27 O'Brien J, A m es D, Chiu E, Schw eitzer I, D esm ond P and Tress B (1998) Severe deep w h ite m atter lesions and ou tcom e in elderly patients w ith major depressive disorder: follow -u p study. BM J. 317: 9 8 2 -4 . 28 Thomas AJ, O'Brien JT, Davis S et al. (2002) Ischem ic basis for deep w h ite m atter hyperintensities in major depression: a neuropathological study. Arch Gen Psychiatry. 59: 7 8 5 -9 2 . 29 C han-Palay V and Asan E (1989) A lterations in catecholam ine n eu ron s of the locus coeruleus in senile dem entia of the A lzheim er type and in Parkinson's disease w ith and w ith ou t dem entia and depression. J Comp Neurol. 287: 3 7 3 -9 2 . 30 Syed A, Chatfield M, M atthew s F, et al. (2005) D epression in the Elderly: pathological study of raphe and locus ceruleus. Neuropathol A ppl Neurobiol. In press. 31 Harrison PJ (1999) The n europathology of schizophrenia. A critical review of the data and their interpretation. Brain. 122: 5 9 3 -6 2 4 . 32 Esiri MM and Crow TJ (2002) Psychiatric disorders. In: DI Graham and PL Lantos (eds) Greenfield's Neuropathology (7e). Arnold, London.

Chapter 3

Neurochemical changes in old age Andrew W Procter

Introduction Effective drug treatm ents for depression and psychosis have been available since the 1950s. M uch is k n ow n about the pharm acological actions of such drugs, but the precise nature of the underlying neurochem ical pathology in these conditions rem ains u n kn ow n . C onversely, effective neurotransm itter-based therapies for the major neurodegenerative disease causing dem entia in old age, nam ely A lzheim er's disease (AD), have only becom e available in recent years, despite nearly three decades of study of this condition. The m odel for such treatm ent of a neurodegenerative condition has b een Parkinson's disease, the neurochem ical pathology of w h ich w as described in the 1960s. Understanding of m onoam ine and other neurotransm itter system s developed at around the same tim e and led to therapies for this condition. Thus, w ith this exception, it is broadly true to say that w e are currently in the position of having effective treatm ents for those conditions for w h ich w e have a poor understanding of the neurobiology, yet few treatm ents for those conditions about w hich w e k n ow m uch more. In part this m ay be true because the neurodegenerative disorders tend to be fatal after a relatively short period, and m any studies have therefore been m ade of post-mortem brain tissue. The investigation of the biology of functional psychiatric disorders has b een dependent on techniques w hich can be applied in life, w h en these conditions are relatively uncom plicated by other disorders and conditions. For m any years the investigation of post-mortem brain tissue has b een a standard against w h ich other studies have been compared, and these neurochem ical studies have provided inform ation about neurotransm itter fu n c­ tion and other m etabolic processes. H ow ever, such studies need to be inter­ preted carefully and conducted in such a w ay as to take account of the potential artefacts and epiphenom ena w h ich have often com plicated studies of AD. This disorder provides a m odel of h o w the effects of these factors m ay be taken into account. N onetheless, our present understanding of neurotransm itter systems, their receptors and second-m essenger system s in the hum an brain and their relevance to the psychiatric disorders of old age is largely inform ed by such post-mortem studies. It is these neurotransm itter system s w hich form the site of action of m ost of the drugs in use at present for psychiatric disorders, and their in v o lv e­ m ent in the pathology of neurodegenerative disorders w ill be the focus of this chapter.

33

34

Practical old age p sy c h o p h arm o co lo g y

The neurochemical study of post-mortem brain M any neurochem ical studies have b een undertaken of relatively small num bers of subjects using post-mortem tissue. Such studies w ere often conducted on patients previously resident in institutions, and thus are likely to have over­ represented subjects w ith prom inent behavioural sym ptom s. In AD, behavioural sym ptom s frequently determ ine w heth er patients com e to m edical attention and the care that they subsequently require. Yet conclusions have often been drawn about patients in general, even th ou gh the clinical characteristics of these subjects do not represent those of all patients w ith AD. Studies that have b een based upon m ore epidem iologically sound sam ples , 1 including subjects assessed during life for the range of clinical features of AD, are likely to be able to resolve these issues and dem onstrate w h ich biochem ical features are characteristic of AD, and w hich are those of subsets w ith particular behavioural syndrom es. Similarly, control subjects m ust definitely be free from the condition. A lthough it m ay be relatively easy to ensure that they are free from advanced disease, it m ay be m ore difficult to be confident that they are free from early or presym ptom atic stages of the disorder. If neurochem ical changes that occur early in the course of the disease and w hich m ay be of possible pathogenic significance are to be exam ined, it m ay be necessary to screen potential control subjects before death to confirm that th ey are gen u in ely asym ptom atic. Studies of m any aspects of the biochem istry of hum an brain have dem onstrated an apparent inherent variability com pared w ith the situation in experim ental anim als. W hile dem ographic factors such as age and sex, and other factors such as drug treatm ent m ust be taken into account, the m ode of the patient's death is a factor of particular note, indicated by m easures such as the duration of terminal com a . 2 As the precise details of this are rarely available to the biochem ist working w ith post-mortem samples, the pH value of brain tissue h om ogen ates 3,4 m ay be a valid index. Particularly in neurodegenerative conditions, tissue atrophy m ay confound interpretation of biochem ical data because it is norm al practice to report results per unit mass. This does n ot m ake allow ance for any reduction in the volum e of brain structure. Therefore shrinkage or loss of som e structures but not others m ay lead to reports of an increase in the markers in unaffected structures, such as increased 7 -am inobutyric acid (GABA) content of frontal cortex from A lzheim er biopsy tissu e . 5 A lthough conditions such as AD are progressive disorders, this is rarely acknow ledged in post-mortem studies, w h ich alm ost invariably exam ine patients dying at an advanced stage of disease. It has been possible to study samples rem oved earlier in the course of the disease, as neocortical tissue is occasion­ ally rem oved surgically for diagnostic purposes. These samples have b een compared w ith com prehensive control data obtained from neurosurgical pro­ cedures in w h ich the rem oval of neocortical tissue of norm al appearance was a necessary part of the surgical procedure . 6 In general, post-mortem studies reveal m ore exten sive and severe neurochem ical pathology than is found in biopsy material.

N e u ro ch em ical changes in old age

35

Neurotransmitter systems in human brain The cholinergic system

Anatomy and function of the neocortical cholinergic system The cerebral cortex receives tw o major distally projecting cholinergic pathw ays . 7 One of these originates in the basal forebrain, comprising the nuclei of the septum , diagonal band and M eynert. This term inates in all areas of the cortex, w ith particularly dense innervation of the hippocam pus as w ell as the amygdala. The other pathw ay originates in the brainstem, and w h ile it innervates the thalam us in particular, it also term inates in selected areas of the cortex, notably the frontal and occipital cortex. This cholinergic projection is thought to have a major role in processes such as attention and arousal. The basal forebrain cholinergic system is thought to have a role in regulating the activity of the entire cerebral cortex, and to m aintain the cortex in its operative m o d e . 7,8 The practical m anifestation of this is that acetylcholine (ACh) is thought to have a role in the control of selective attention , 9 by affecting discriminatory processes, the reception and evaluation of stim uli, and the m odification of cortical responsive­ ness. Cholinergic m echanism s therefore appear to govern m any brain functions associated w ith different cortical regions, such as perception, learning, cognition and judgem ent.

Neurochemistry ACh is synthesised in nerve term inals from its precursor, choline, w hich is transported to the brain via the bloodstream . Choline enters cholinergic nerve cells via a high-affinity uptake system w h ich is specific to these neurons. In the cytoplasm , choline reacts w ith acetyl-coen zym e A, a process that is catalysed by the enzym e choline acetyl transferase (ChAT). This enzym e is synthesised in the n erve-cell bodies and transported to the nerve terminals. The ACh thus synthesised is not uniform ly distributed in the neurons. About 2 0 % appears to be in solution in the cytoplasm, and the rem aining 80% is stored in a particulate com ponent of the nerve endings. M orphological studies have show n that there are A C h-containing vesicles in the presynaptic term inal, and it is these vesicles w h ich probably correspond to the particulate com ponent of the ACh. It is generally accepted that the vesicles represent an intracellular organelle in w hich neurotransm itter is stored prior to being released by neuronal stim ulation. R eleased ACh is not rem oved from the synaptic cleft by active reuptake system s, as is the case w ith m any other neurotransm itters, but is broken dow n by a m em brane-bound enzym e, acetylcholinesterase (AChE). This is an enzym e specific to ACh and can generally be distinguished from a plasma enzym e, butyrylcholinesterase, w h ich also m etabolises ACh but is not specific to synapses. The latter enzym e is often referred to as 'pseudocholinesterase'. Two distinct types of receptor, nicotinic and muscarinic, m ediate the postsynaptic actions of ACh. Nicotinic receptors occur at the neurom uscular junction and in the central nervous system (CNS). It is difficult to distinguish subtypes of nicotinic receptors using drugs, but m olecular biology studies have sh ow n distinct differ­ ences. All nicotinic receptors consist of five protein subunits arranged around a

36

Practical old age p sy c h o p h arm o co lo g y

trans-m em brane ion channel. It is the opening of this ion channel w hich m ediates the postsynaptic actions of the transmitter. The m uscarinic receptors m ediate slow er neuronal responses by activation of second-m essenger system s. The drug pirenzepine distinguishes tw o subclasses of m uscarinic receptors, those blocked by this drug being designated M! and all others being designated M 2. M olecular biology studies of these receptors have since confirm ed the existence of further subgroups, M 3 and M 5, w h ich sh ow similarities to M j, and M4, w hich show s similarities to M 2.

Cholinergic system in Alzheimer’s disease Early dem onstrations of substantial losses of ChAT from the brain tissue of patients w ith AD post mortem10~14 and ante mortem 15 have stim ulated m uch subsequent research on this neurotransm itter. This cholinergic deficit is one of the m ost prom inent features of A D . 6 N europathological studies have sh ow n that there is usually considerable loss of the neurons w hich give rise to the cortical cholinergic innervation, the neurons of the nucleus basalis of M eynert , 16-20 w hich is associated w ith loss of nucleolar volu m e of the cells . 20 B iochem ical m easures of cholinergic function have sh ow n consistent and extensive losses of those biochem ical activities w hich are associated w ith cholinergic term inals. In particular, ChAT activity seem s to be reduced post mortem in all areas of the cerebral cortex of patients w ith AD . 10-14,21 Early in the course of the disease, neurosurgical specim ens confirm this loss of activity and a reduced ability of the tissue to synthesise acetylch olin e . 15,22 It is generally accepted that there is probably little alteration in the subtype of postsynaptic receptor, at least until late in the disease, and the M 2 subtype is at m ost only m oderately affected , 23,24 and probably only in the frontal cortex .25 The in volvem en t of M 2 receptors m ay be involved in the pathogenesis of psychotic sym ptom s in A D . 25 H ow ever, loss of nicotinic receptors is a consistent finding , 26,27 and has led to the proposition that this m ay be an early and im portant aetiological e v e n t . 23 The m agnitude of the cholinergic dysfunction is correlated w ith the severity both of the cognitive im pairm ent assessed on global m easures of cognitive function , 28-31 and of the neuropathological changes, including senile plaque form ation 21,30,32 and loss of n eu ro n s .21 Considerable em phasis has b een placed on the significance of this cholinergic deficit, and it has b een suggested that the dem entia of AD is due primarily to th is . 33 Other conditions associated w ith cognitive im pairm ent are also associated w ith cholinergic pathology. These include Parkinson's disease, progressive supranuclear palsy and cerebrovascular dem entia, as w ell as head injury 34 and dem entia w ith Lewy bodies (DLB ) , 35 and they are associated w ith such pathology independently of the severity of Alzheim er pathology. However, AD is probably not solely a disorder of the cholinergic system . The neocortical cholinergic deficit probably only explains part of the cognitive decline, as has been suggested by neuropsychological studies of the effects of cholinergic antagonists . 36 The clinical syndrom e of AD consists of m ore than just cognitive im pairm ent, and frequently includes behavioural and psychiatric sym ptom s such as w andering, aggression and depression. A lthough the extent to w h ich these disorders of conduct and personality have been found in AD could be an overestim ate, especially in the presenium , these sym ptom s are alm ost certainly related to non-cholinergic transmitter pathologies. However,

N eu ro ch em ical changes in old age

37

som e non-cogn itive sym ptom s m ay have a cholinergic basis . 37 Visual hallucina­ tions are related to the severity of the cholinergic deficits in DLB , 38 a condition in w h ich the cholinergic deficits are generally m ore severe and visual hallucinations more com m on than in A D . 39

Monoamine neurotransmitters In addition to the cholinergic innervation, the cortex receives inputs from at least three other populations of subcortical neurons, each using a different transmitter. These are the catecholam ine neurotransm itters, noradrenaline and dopam ine, and the indoleam ine 5-hydroxytryptam ine (5-HT).

Anatomy and function of monoamine neurotransmitter systems W ithin the CNS the cell bodies of 5-HT neurons are located in the m idline of the upper brainstem , form ing distinct nuclei, the raphe nuclei. The inferior group of these project to the brainstem , cranial nerve nuclei and spinal cord. The superior group project rostrally and innervate the limbic and sensory areas of the forebrain. This group of nuclei show s a degree of topographic organisation of their projections such that the fibres of the dorsal raphe nuclei innervate the basal ganglia and cerebellum , w hereas the fibres of the m edian raphe nuclei innervate the hippocam pus and septum . As a result of its presence in these structures of the CNS, 5-HT plays a role in a great variety of behaviours, such as food intake, activity rhythm s, sexual behaviour and em otional states. The serotonergic system is also though t to play a significant role in learning and m em ory , 40,41 in particular by interacting w ith the cholinergic, glutamatergic, dopam inergic or GABA-ergic system s. The locus coeruleus (LC) is the major nucleus of origin of noradrenergic fibres in the m am m alian brain. Rostral and dorsal LC neurons innervate forebrain and cortical structures, w hilst the caudal and ventral neurons project to the cere­ bellum and spinal cord .42'43 It has b een proposed that this noradrenergic system , together w ith other sym pathetic system s, yields rapid adaptive responses to urgent stim uli , 44 although other proposed roles of LC-noradrenaline neurons include those in sleep, attention, m em ory and vigilance.

Neurochemistry of monoamine neurotransmitter systems The n eu rotra n sm itter 5-HT is sy n th esised from th e a m in o acid try p to p h a n by th e e n z y m e tryp top h an h y d ro x y la se. This e n z y m e has m a n y features in c o m m o n w ith tyro sin e h yd ro x y la se, w h ic h co n v erts ty ro sin e to L-dopa d uring th e sy n th esis of d o p a m in e an d n o rad ren alin e. The sy n th esised transm itters are fo u n d prim arily in storage v esicles. The u p tak e of tran sm itter in to th e se v esicles is an e n e r g y -d e p e n d e n t process w h ic h can b e disrupted by su ch drugs as m e th y le n e d io x y m e th a m p h e ta m in e (MDMA, 'ecstasy'). R elease of tran sm itter from th e v esicles an d n erv e en d in g s is regu lated by th e released tran sm itter via a n u m b er of au torecep tors, as w e ll as o th er n eu ro tra n sm itters b y w a y of h eterorecep tors. R eleased tran sm itters are in a ctiv a ted by p resy n a p tic active u ptake system s.

38

Practical old age p sy c h o p h arm o co lo g y

Monoamine neurotransmitter systems in AD Catecholamines

Studies of dopam ine neurotransm itter function in AD have not sh ow n any consistent pattern of in v o lv em en t . 6 Not surprisingly, therefore, there seem s to be little relationship b etw een m easures of dopam ine innervation and the clinical features of the syndrom e .45 The LC is k n ow n to be significantly dam aged in AD , 46-49 and a decreased noradrenaline content is found in the cerebral cortex . 50-54 Concentrations of the m ajor m etabolite are unaltered or ev en elevated , 6,52,53,55,56 probably reflecting increased turnover of NA. In the LC itself, alterations in neuronal m orphology, neurofibrillary tangle (NFT) form ation and loss of pigm ented noradrenaline neurons in topographically distinct regions have been observed .49,57 The clinical significance of changes in the noradrenaline system in AD is unclear. The extent of LC neuron loss has been reported to relate to disease duration 58 and severity of cognitive decline , 47,48 as has loss of neurotransm itter . 59 Studies have also found greater LC neuron loss in depressed 60-62 and aggressive 59 subjects w ith AD, although this m ay not be a consistent finding . 63

5-HT B iochem ical determ inations of neurons containing 5-HT in AD have m ostly relied on determ inations of the concentrations of 5-HT and its major m etabolite, 5hydroxy-indoleacetic acid (5-HIAA), in post-mortem samples. The content of these in m any areas of the neocortex of AD subjects m ay be reduced , 52,53,55,56 and neurofibrillary degeneration and neuronal loss in the raphe nucleus has b een reported . 64 H owever, this is by no m eans a consistent finding, and even in AD brains at autopsy half of the cortical areas m ay have no selective reduction of presynaptic 5-HT activity. As discussed above, this discrepancy b etw een studies m ay in part be explained by the inadvertent selection of hospitalised subjects w ith marked behavioural sym ptom s. In studies of epidem iologically representative samples of com m unitybased AD patients assessed in life, presynaptic cortical markers of 5-HT neurons w ere not uniform ly affected in AD. Thus although the density of 5-HT uptake sites in AD tem poral cortex was significantly reduced (61% of control), there was no significant alteration in the frontal cortex. Reduced 5-HT concentrations in the frontal cortex w ere also found to indicate those subjects w ith the m ost rapid cognitive decline. Patients judged to be aggressive during life show ed more severe loss of both 5-HT concentration and postsynaptic receptors in both retrospectively 64,65 and prospectively assessed 66 subjects. Disorders of 5-HT innervation have also b een im plicated in the aetiology of depression, anxiety and overactivity in AD . 66,67 Aggressive behaviour has been linked to m ore advanced disease, suggesting that loss of both 5-HT2 65 and 5-HT 1A67 receptors is a feature of more extensive pathology found towards the end stage of the disorder. H owever, brain im aging studies in healthy subjects also im plicate 5-HT1A in aggressive traits, 68 so the loss of cortical 5-HT]A receptors m ay w ell be particularly associated w ith aggressive tendencies in AD . 67

N e u ro ch em ical ch an g es in old age

39

Inhibitory amino acid neurotransmitters and neuropeptides The principal inhibitory neurotransm itters in the central nervous system are the am ino acids 7 -am inobutyric acid (GABA) and glycine. GABA appears to act as a transmitter in virtually all areas of the CNS, and glycine is largely confined to the caudal part of the brain and spinal cord . 69

GABA in AD W ithin the cortex there are large num bers of interneurons containing GABA, often co-localised w ith one or m ore of a variety of neuropeptides. The balance of evidence indicates that loss of these substances is not a fundam ental characteristic of AD. Post-mortem assessm ent of GABA-releasing neurons has been com plicated by artefacts and ep ip h en o m en a . 5 Thus no change in the activity of the enzym e responsible for GABA synthesis, glutam ate decarboxylase (GAD), w as found in a careful study in w h ich AD and control subjects w ere m atched for the nature of the term inal illness . 56 Normal GAD activity and GABA content have b een confirm ed w ith cortical biopsy tissu e . 6 A nother presynaptic m easure of GABA innervation, nam ely the GABA transporter GAT-1, was similarly unaltered in A D . 70 Perhaps not surprisingly, an attem pt to treat AD w ith a GABA agonist was unsuccessfu l . 71

Other amino acids and peptides in AD Both glycine and taurine are also thought to function as inhibitory neurotrans­ mitters, and their content is unchanged in AD, based on both post-mortem and biopsy tissu e . 5,72 The neuropeptides cholecystokinin, vasoactive intestinal p o ly ­ peptide and neuropeptide Y are unchanged in AD , 73,74 as is probably cortico­ tropin-releasing factor . 75 M any studies have indicated that som atostatin is reduced, but this was not confirm ed in biopsy sam ples . 6 Larger reductions in som atostatin and GABA content have b een reported in post-mortem studies that included only subjects displaying severe histopathology than in those w here no such selection criteria w ere em p loyed . 5 The loss of som atostatin correlates w ith the severity of cognitive im pairm ent . 76,77

Excitatory amino acid neurotransmitters Anatomy and function of excitatory amino acid neurotransmitters The am ino acid glutam ate has been kn ow n for som e tim e to be the major transmitter of excitatory signals of the long axonal projections of the CNS. In addition, a large proportion of sensory fibres contain glutam ate and aspartate, as is also the case for nerve cells linking different areas of the brain. H owever, because glutam ate has an im portant role in the m etabolism of all cells, and is also the precursor of the inhibitory transmitter GABA, the precise localisation of the transmitter glutam ate has b een difficult. N evertheless, a com bination of tech ­ niques, including histochem istry for glutam ate and its putative synthetic enzym e and retrograde transport of radiolabelled glucose (GLU), has provided strong evidence for a transmitter role for glutam ate in a num ber of pathw ays . 78 This has been convincingly dem onstrated for cortico-fugal fibres, the projections of

40

Practical old age p sy c h o p h arm o co lo g y

pyramidal neurons of layer V of the cerebral cortex. There is less definitive evidence available establishing a role for glutam ate as a transmitter of other cortical pyramidal neurons, in particular the cortico-cortical association and transcallosal fibres (typically found in cortical layers II and III). However, results obtained using a variety of techniques are consistent w ith glutam ate being the major neurotransm itter of these groups of neurons. There is m uch evidence that the initiation and m aintenance of epileptic seizures involve the release of glutam ate, and drugs w h ich block the NMDA receptor have anticonvulsant actions. Excitatory am ino acids are found in m ost sensory fibres and probably contribute to the transm ission of pain stim uli. It is w idely accepted that the p h en o m en on of long-term potentiation (LTP) is fundam entally involved in the form ation of m em ories. LTP is a long-lasting en h an cem en t of synaptic effectiveness w h ich follow s prolonged stim ulation of hippocam pal input pathw ays. Types of glutam ate receptor appear to be involved in this process. Prolonged activation of glutam ate receptors can cause neuronal damage due to excessive calcium entry, a p h en om en on k now n as excitotoxicity. This appears to have a major role in ischaem ic brain damage. Brief periods of ischaem ia can interrupt glutam ate reuptake by neurons and glia. This causes excessive stim ulation of postsynaptic receptors and calcium entry, w h ich can cause osm otic and m etabolic damage to the neuron.

Neurochemistry of excitatory amino acid neurotransmitters Glutamate has m any im portant functions in the body. It is a com ponent of proteins and is involved in a key step in cellular energy production. H owever, that of the glutam ate w h ich has a neurotransm itter role m ay be form ed by a different m etabolic pathw ay to that w hich is involved in m etabolism . Once glutam ate is released, there are high-affinity uptake sites in term inals and glia w h ich rem ove it from the synaptic cleft. S tudies w ith se lectiv e agon ists an d a n tag o n ists h a v e in d icated th at th ere are fou r m ajor classes of glu tam ate receptors. T h ese are th e N-m ethyl-D -aspartate (NMDA), k ain ate, a -a m in o -3 -h y d r o x y -5 -m e th y l-4 -is o x a z o le p r o p r io n a te (AMPA) an d m etab otrop ic classes. The AMPA receptors con sist of four p ro tein su b u n its that form a tra n s-m em b ra n e io n ch a n n e l.

The majority of AMPA receptors control sodium ion influx and thus cellular depolarisation, although som e also regulate calcium entry. Thus the AMPA receptors probably m ediate a fast depolarisation response to glutam ate release. M any kainate receptors appear to occur presynaptically, and probably therefore have a role in regulating glutam ate release. The m etabotropic receptors are least w ell understood, but act by w ay of a second-m essenger system coupled through G-proteins to potassium and calcium channels. The NMDA receptor is probably the best studied of the glutam ate receptors. A ctivation produces a slow , prolonged depolarisation, so these receptors do not m ediate the fast transm ission of excitation and the initiation of nerve im pulses. A ctivation of the receptor controls an ion channel that is perm eable to sodium and calcium . H owever, the entry of calcium alters intracellular pathw ays and this is probably the m eans by w h ich NM DA-receptor activation alters neuronal activity. The receptor is com plex, w ith recognition sites for several distinct groups of com pounds w h ich seem to have a regulatory role. In addition to the recognition site for glutam ate, there are also sites for glycine.

N eu ro ch em ical ch an g es in old age

4 1

Excitatory amino acids in AD The pyramidal cells of the tem poral and parietal lobes of the cerebral cortex use glutam ate as a neurotransm itter and are major and early sites of pathological changes in AD. Severity of cognitive im pairm ent is correlated w ith the pyramidal cell counts in layer III28 and the num ber of synapses in layer III. 79 Synaptosom al Na+ -dependent d- [ 3H] aspartate uptake (ASP uptake) has b een used as a marker of the relative density of glutam ate nerve terminals. ASP uptake m ay be reliably m easured in tissue from prom ptly performed autopsies, and in such specim ens w as reduced by 50% in the temporal cortex of AD patients , 80 indicating a loss of glutam ate synapses. The m ost likely interpretation is that this reflects loss of synapses of corticocortical fibres, since loss of the cells of origin of these fibres is the major pathology. Studies in hippocam pus have indicated that all the major input and output pathw ays (except the septohippocam pal ch oli­ nergic pathw ay) use glutam ate as transm itter . 78 There was cell loss and tangle form ation in the entorhinal cortex and in CA1 in AD patients at post-m ortem . Glutamatergic pathw ays are considered to be involved in m em ory , 81 and it follow s that this loss of presynaptic glutam atergic innervation is likely to co n ­ tribute to the m em ory dysfunction in AD. This m ay be com pounded by postsynaptic and other regulatory factors . 82 Cell death and the other pathological hallmarks of AD m ay be brought about through glutam ate-m ediated m echanism s. However, the distribution of pathology does support the idea that the disease spreads along cortical glutam ate pathw ays 83-85 w h ich m ay share the com m on property of synaptic plasticity . 86 The effects of reduced glutam atergic activity m ight include the prom otion of the form ation of senile plaques . 87 Hypoactivity of glutam atergic neurons is also im plicated in aberrant m echanism s of tau and thereby possible neurofibrillary tangle form ation . 6 82

K ey p o in ts • It has been generally assum ed that the neurochem ical study of post-mortem brain tissue reveals inform ation of relevance to the understanding of that condition in life. H ow ever, m any factors need to be considered for this to be a valid assum ption. It is probably true to say that the rigorous d em on ­ stration of reliability and validity w h ich is norm ally expected in m any other areas of psychiatry has rarely b een applied to biological m easures. • Studies controlling for confounding factors, and using epidem iologically sound sam ples of subjects, have indicated that early in the course of the disease abnorm alities of relatively few neurotransm itters are obvious. This is in contrast to the situation late in the disease, w h ich is usually exam ined in post-mortem tissue. • Thus the m ost reliable and consistent changes are those seen in the cholinergic innervation of the cortex and the cortical pyramidal neurons. H ow ever, by the tim e of death there is usually considerable in volvem en t of other neurons. The clinical effect of this appears to be to cause som e aspects of the cognitive dysfunction characteristic of AD. H owever, it is m ost likely that disorders of other (especially m onoam ine) neurotrans­ mitters cause m ost of the behavioural and other n on-cognitive sym ptom s.

42

Practical old age p sy c h o p h arm o co lo g y

References 1 Hope T, K eene J, Gedling K et al. (1997) B ehaviour changes in dem entia. 1. Point of entry data of a prospective study. Int J Geriatr Psychiatry. 12: 1 0 6 2 -7 3 . 2 Harrison PJ, Procter AW, Barton AJL et al. (1991) Terminal com a affects m essenger RNA detection in post mortem h u m an brain tissue. Mol Brain Res. 9: 16 1 -4 . 3 Kingsbury AE, Foster OJF, Nisbet AP et al. (1995) Tissue pH as an indicator of mRNA preservation in h u m an post mortem brain. Mol Brain Res. 28: 3 1 1 -1 8 . 4 Yates CM, B utterw orth J, T ennant MC et al. (1990) E nzym e activities in relation to pH and lactate in post mortem brain in A lzheim er-type and other dem entias. J Neurochem. 55: 1 6 2 4 -3 0 . 5 Lowe SL, Francis PT, Procter AW et al. (1988) G am m a-am inobutyric acid concentration in brain tissue at tw o stages of A lzheim er's disease. Brain. I l l : 7 8 5 -9 9 . 6 Procter AW (2002) N eurochem ical p athology of n eurodegenerative disorders in old age. In: R Jacoby and C O ppenheim er (eds) Psychiatry in the Elderly. Oxford U niversity Press, Oxford. 7 M esulam MM (1995) The cholinergic contribution to neu rom od u lation in the cerebral cortex. Neuroscience. 7: 2 9 7 -3 0 7 . 8 W enk GL (1997) The n u cleu s basalis m agnocellularis cholinergic system: on e hundred years of progress. Neurobiol Learn Mem. 67: 8 5 -9 5 . 9 Voytko ML (1996) C ognitive functions of the basal forebrain cholinergic system in m onkeys: m em ory or attention? Behav Brain Res. 75: 13 -2 5 . 10 B ow en DM, Smith CB, W hite P et al. (1976) N eurotransm itter-related enzym es and indices of hypoxia in senile dem entia and other abiotrophies. Brain. 99: 4 5 9 -9 6 . 11 Davies P (1979) N eurotransm itter-related en zym es in senile dem entia of the A lzheim er type. Brain Research. 171: 3 1 9 -2 7 . 12 D avies P and M aloney AJF (1976) Selective loss of central cholinergic n eu ron s in A lzheim er's disease. Lancet, ii: 1403. 13 Perry EK, Gibson PH, B lessed G et al. (1977) N eurotransm itter en zym e abnorm alities in senile dem entia. C holine acetyltransferase and glutam ic acid decarboxylase activities in necropsy brain tissue. J Neurol Sci. 34: 2 4 7 -6 5 . 14 Perry EK, Perry RH, B lessed G et al. (1977) N ecropsy evid en ce of central cholinergic deficits in senile dem entia. Lancet, i: 189. 15 B ow en DM, B enton JS, Spillane JA et al. (1982) C holine acetyltransferase activity and h istopathology of frontal neocortex from biopsies of d em ented patients. J Neurol Sci. 57: 191 -2 0 2 . 16 Nagai R, McGeer PL, Peng JH et al. (1983) C holine acetyltransferase im m u n oh istochem istry in brains of A lzheim er's disease patients and controls. Neurosci Lett. 36: 195-9. 17 W hitehouse PJ, Price DL, Struble RG et al. (1982) A lzheim er's disease and senile dem entia: loss of neurons in the basal forebrain. Science. 215: 1 2 3 7-9. 18 M ann DMA and Yates PO (1982) Is the loss of cerebral cortical ch olin e acetyltransfer­ ase activity in A lzheim er's disease due to degeneration of ascending cholinergic nerve cells. J Neurol Neurosurg Psychiatry. 45: 936. 19 Arendt T, Bigl V, T ennstedt A et al. (1985) N euronal loss in different parts of the nucleus basalis is related to neuritic plaque form ation in cortical target areas in A lzheim er's disease. Neuroscience. 14: 1-14. 20 M ann DM, Yates PO and M arcyniuk B (1984) Presenile A lzheim er's disease, senile dem entia of A lzheim er type and D ow n's syndrom e of m iddle age all form an agerelated contin u u m of pathological changes. Neuropathol A ppl Neurobiol. 10: 1 8 5 -2 0 7 . 21 M ountjoy CQ, Rossor MN, Iversen LL et al. (1984) Correlation of cortical cholinergic and GABA deficits w ith quantitative neuropathological findings in senile dem entia. Brain. 107: 5 0 7 -1 8 .

N e u ro ch em ical ch an g es in old age

43

22 Sims NR, B ow en DM, A llen SJ et al. (1983) Presynaptic cholinergic dysfunction in patients w ith dem entia. J Neurochem. 40: 5 0 3 -9 . 23 Perry EK (2000) The cholinergic system in A lzheim er's disease. In: J O'Brien, D A m es and A Burns (eds) Dementia. Arnold, London. 24 R oberson MR and Harrell LE (1997) C holinergic activity and am yloid precursor protein m etabolism . Brain Res Brain Res Rev. 25: 5 0 -6 9 . 25 Lai MK, Lai OF, K een e J et al. (2001) Psychosis of A lzheim er's disease is associated w ith elevated m uscarinic M 2 binding in the cortex. Neurology. 57: 8 0 5 -1 1 . 26 Aubert I, Araujo DM, Cecyre D et al. (1992) Comparative alterations of nicotinic and m uscarinic binding sites in A lzheim er's and Parkinson's diseases. J Neurochem. 58: 5 2 9 -4 1 . 27 Perry EK, Morris CM, Court JA et al. (1995) Alteration in nicotin e-b in d in g sites in Parkinson's disease, Lewy body dem entia and A lzheim er's disease: possible in d ex of early neuropathology. Neuroscience. 64: 3 8 5 -9 5 . 28 Neary D, Sn ow d en JS, M ann DM A et al. (1986) A lzheim er's disease: a correlative study. J Neurol Neurosurg Psychiatry. 49: 2 2 9 -3 7 . 29 Palmer AM, Francis PT, B o w en DM et al. (1987) C atecholam inergic n eu ron es assessed ante mortem in A lzheim er's disease. Brain Res. 414: 3 6 5 -7 5 . 30 Perry EK, T om linson BE, B lessed G et al. (1978) Correlation of cholinergic abnorm alities w ith senile plaques and m ental test scores in senile dem entia. BMJ. 2: 1 4 5 7 -9 . 31 M inger SL, H oner WG, Esiri M M et al. (2001) Synaptic pathology in prefrontal cortex is present on ly w ith severe dem entia in A lzheim er disease. J Neuropathol Exp Neurol. 60: 9 2 9 -3 6 . 32 Perry EK, Blessed G, T om linson BE et al. (1981) N eurochem ical activities in hum an tem poral lobe related to aging and A lzheim er-type changes. Neurobiol Aging. 2: 2 5 1 -6 . 33 Coyle JT, Price DL and DeLong MR (1983) A lzheim er's disease: a disorder of cortical cholinergic innervation. Science. 219: 1 1 8 4 -9 0 . 34 M urdoch I, Perry EK, Court JA et al. (1998) Cortical cholinergic dysfunction after hum an head injury. J Neurotrauma. 15: 2 9 5 -3 0 5 . 35 Sam uel W, Alford M, Hofstetter CR et al. (1997) D em entia w ith Lewy bodies versus pure A lzheim er disease: differences in cognition, n europathology, cholinergic dysfunction and synapse density. J Neuropathol Exp Neurol. 56: 4 9 9 -5 0 8 . 36 K opelm an MD and Corn TH (1988) C holinergic 'blockade' as a m odel for cholinergic depletion. Brain. I l l : 1 0 7 9 -1 1 0 . 37 M inger SL, Esiri MM, M cD onald B et al. (2000) Cholinergic deficits contribute to behavioral disturbance in patients w ith dem entia. Neurology. 55: 1 4 6 0 -7 . 38 Perry EK, H aroutunian V, Davis KL et al. (1994) Neocortical cholinergic activities differentiate Lewy body dem entia from classical A lzheim er's disease. Neuroreport. 5: 7 4 7 -9 . 39 Perry EK, M arshall E, Thom pson P et al. (1993) M onoam inergic activities in Lewy body dem entia: relation to h allucinosis and extrapyram idal features. J Neural Transm Park Dis Dement Sect. 6: 167-77. 40 B uhot MC, Martin S and Segu L (2000) Role of serotonin in m em ory im pairm ent. A n n Med. 32: 2 1 0 -2 1 . 41 M eltzer CC, Sm ith G, DeK osky ST et al. (1998) Serotonin in aging, late-life depression, and A lzheim er's disease: the em erging role of functional im aging. Neuropsychopharmacology. 18: 4 0 7 -3 0 . 42 W aterhouse BD, Lin CS, Burne RA et al. (1983) The distribution of neocortical projection neu ron es in the locus coeruleus. J Comp Neurol. 217: 4 1 8 -3 1 . 43 Loughlin SE, Foote SL and Fallon JH (1982) Locus coeruleus projections to cortex: topography, m orphology, and collateralisation. Brain Res Bull. 9: 1-16.

44

Practical old age p sy c h o p h arm o co lo g y

44 A ston-Jones G, Shipley MT and Grazanna R (1995) The locus coeruleus, A5 and A7 noradrenergic cell groups. In: G Paxinos (ed.) The Rat Nervous System. Academic Press, Sydney. 45 Bierer LM, Knott PJ, Schm eidler JM et al. (1993) Post-mortem exam in ation of dopam inergic parameters in A lzheim er's disease: relationship to non cogn itive sym ptom s. Psychiatr Res. 49: 2 1 1 -1 7 . 46 M ann DMA, Lincoln J, Yates PO et al. (1980) Changes in the m o n o am in e-con tain in g neurones of the h u m an CNS in senile dem entia. Br J Psychiatry. 136: 5 3 3 -4 1 . 47 Bondareff W, M ountjoy CQ and Roth M (1981) Selective loss of n eu ron es of origin of adrenergic projection to cerebral cortex (nucleus locus coeruleus) in senile dem entia. Lancet, i: 7 8 3 -4 . 48 Bondareff W, M ountjoy CQ and Roth M (1982) Loss of n eu ron s of origin of the adrenergic projection to cerebral cortex (nucleus locus coeruleus) in senile dem entia. Neurology. 32: 1 6 4 -8 . 4 9 Chan-Palay V and Asan E (1989) A lterations in catecholam inergic n eu ron es of the locus coeruleus in senile dem entia of the A lzheim er type and in Parkinson's disease and am yotropic lateral sclerosis. J Comp Neurol. 287: 3 7 3 -9 2 . 50 Francis PT, Palmer AM, Sims NR et al. (1985) N eurochem ical studies of early-onset A lzheim er's disease. Possible in flu en ce on treatm ent. NEJM. 313: 7 -1 1 . 51 A dolfsson R, Gottfries CG, Roos BE et al. (1979) Changes in brain catecholam ines in patients w ith dem entia of A lzheim er type. Br J Psychiatry. 135: 2 1 6 -2 3 . 52 Arai H, Kosaka K and Iizuka R (1984) Changes of biogenic am ines and their m etabolites in postm ortem brains from patients w ith A lzheim er-type dem entia. J Neurochem. 43: 3 8 8 -9 3 . 53 Gottfries CG, A dolfsson R, A quilonius SM et al. (1983) B iochem ical changes in dem entia disorders of A lzheim er type (AD/SDAT). Neurobiol Aging. 4: 2 6 1 -7 1 . 54 Palmer AM, W ilcock GK, Esiri MM et al. (1987) M onoam inergic in nervation of the frontal and tem poral lobes in A lzheim er's disease. Brain Res. 401: 2 3 1 -8 . 55 Cross AJ, Crow TJ, Joh n son JA et al. (1983) M on oam in e m etabolism in senile dem entia of A lzheim er type. J Neurol Sci. 60: 3 8 3 -9 2 . 56 R einikainen KJ, Paljarvi L, H uuskonen M et al. (1988) A post-mortem study of noradrenergic serotonergic and GABAergic n eu ron s in A lzheim er's disease. J Neurol Sci. 84: 1 0 1 -1 6 . 57 Iversen LL, Rossor MN, R eynolds GP et al. (1983) Loss of pigm ented dopam ine-/?hydroxylase positive cells from locus coeruleus in senile d em entia of A lzheim er's type. Neurosci Lett. 39: 9 5 -1 0 0 . 58 German DC, M anaye KF, W hite CL III et al. (1992) D isease-specific patterns of locus coeruleus cell loss. A n n Neurol. 32: 6 6 7 -7 6 . 59 M atthew s KL, Chen CP, Esiri M M et al. (2002) Noradrenergic changes, aggressive behavior, and cognition in patients w ith dem entia. Biol Psychiatry. 51: 4 0 7 -1 6 . 60 Zubenko GS and M oossy J (1988) Major depression in primary dem entia. Clinical and neuropathologic correlates. Arch Neurol. 45: 1 1 8 2 -6 . 61 Forstl H, Burns A, Luthert P et al. (1992) Clinical and neuropathological correlates of depression in A lzheim er's disease. Psychol Med. 22: 8 7 7 -8 4 . 62 Zweig RM, Ross CA, H edreen JC et al. (1988) The n eu rop ath ology of am inergic nuclei in A lzheim er's disease. A n n Neurol. 24: 2 3 3 -4 2 . 63 Hoogendijk WJ, Som m er IE, Pool CW et al. (1999) Lack of association b etw een depression and loss of n eu ron s in the locus coeruleus in A lzheim er disease. Arch Gen Psychiatry. 56: 4 5 -5 1 . 64 Palmer AM, Stratmann GC, Procter AW et al. (1988) Possible neurotransm itter basis of behavioral changes in A lzheim er's disease. A n n Neurol. 23: 6 1 6 -2 0 . 65 Procter AW, Francis PT, Stratmann GC et al. (1992) Serotonergic pathology is

N eu ro ch em ical changes in old age

66

67

68

69 70 71 72

73 74 75

76

77

78

79 80

81 82 83

84 85

45

not w idespread in A lzheim er patients w ith o u t prom inent aggressive sym ptom s. Neurochem Res. 17: 9 1 7 -2 2 . Chen CPL, Alder JT, B o w en DM et al. (1996) Presynaptic serotonergic markers in com m unity-acquired cases of A lzheim er's disease: correlations w ith depression and n euroleptic m edication. J Neurochem. 66: 1 5 9 2 -8 . Lai MK, Tsang SW, Francis PT et al. (2003) R educed serotonin 5-HT1A receptor binding in the tem poral cortex correlates w ith aggressive behavior in A lzheim er disease. Brain Res. 974: 8 2 -7 . Parsey RV, O quendo MA, Sim pson NR et al. (2002) Effects of sex, age, and aggressive traits in m an on brain serotonin 5-HT1A receptor-binding potential m easured by PET using [C -11]W A Y -100635. Brain Res. 954: 1 7 3 -8 2 . Farrant M (2001) A m ino acids: inhibitory. In: RA W ebster (ed.) Neurotransmitters, Drugs and Brain Function. John W iley and Sons, Chichester. Nagga K, B ogdanovic N and M arcusson J (1999) GABA transporters (GAT-1) in A lzheim er's disease. J Neural Transm. 106: 1 1 4 1 -9 . M ohr E, Bruno G, Foster N et al. (1986) GABA-agonist therapy for A lzheim er's disease. Clin Neuropharmacol. 9: 2 5 7 -6 3 . Lowe SL, B o w en DM, Francis PT et al. (1990) Ante-mortem cerebral am ino acid concentrations indicate selective degeneration of glutam ate-enriched n eu ron s in A lzheim er's disease. Neuroscience. 38: 5 7 1 -7 . Beal MF, e le v e n s RA, Chatta GK et al. (1988) G alanin-like im m unoreactivity is unchanged in A lzheim er's disease. J Neurochem. 51: 1 9 3 5 -4 1 . Gabriel SM, D avidson M, H aroutunian V et al. (1996) N europeptide deficits in schizophrenia vs. A lzheim er's disease cerebral cortex. Biol Psychiatry. 39: 8 2 -9 1 . D eSouza EB, W h iteh ou se PJ, Kuhar MJ et al. (1986) Reciprocal changes in cortico­ tropin-releasing factor (CRF)-like im m unoreactivity and CRF receptors in cerebral cortex of A lzheim er's disease. Nature. 319: 5 9 3 -5 . M inthon L, E dvinsson L and Gustafson L (1997) Som atostatin and neuropeptide Y in cerebrospinal fluid: correlations w ith severity of disease and clinical signs in A lzheim er's disease and frontotem poral dem entia. Dementia Geriatr Cogn Disord. 8: 2 3 2 -9 . Grouselle D, W insky-Som m erer R, David JP et al. (1998) Loss of som atostatin-like im m unoreactivity in the frontal cortex of A lzheim er patients carrying the apolipoprotein ep silon -4 allele. Neurosci Lett. 255: 2 1 -4 . Ottersen OP (1991) Excitatory am ino acid neurotransm itters: anatom ical system s. In: BS M eldrum (ed.) Excitatory A m ino Acid Antagonists. B lackw ell Scientific Publications, Oxford. D eKosky S and Scheff SW (1990) Synapse loss in frontal cortex biopsies in A lzheim er's disease: correlation w ith cognitive severity. A n n Neurol. 27: 4 5 7 -6 4 . Procter AW, Palmer AM, Francis PT et al. (1988) Evidence of glutam atergic den erva­ tion and possible abnorm al m etabolism in A lzheim er's disease. J Neurochem. 50: 7 9 0 -8 0 2 . H ym an BT, Van H oesen GW and Dam asio AR (1990) M em ory-related neural system s in A lzheim er's disease: an anatom ic study. Neurology. 40: 1 7 2 1 -3 0 . Francis PT (2003) Glutamatergic system s in A lzheim er's disease. Int J Geriatr Psychiatry. 18: S 1 5-21. Esiri MM, Pearson RC, Steele JE et al. (1990) A quantitative study of the neurofibrillary tangles and the ch olin e acetyltransferase activity in th e cerebral cortex and the am ygdala in A lzheim er's disease. J Neurol Neurosurg Psychiatry. 53: 1 6 1 -5 . Pearson RC and P ow ell TP (1989) The n eu roan atom y of A lzheim er's disease. Rev Neurosci. 2: 1 0 1 -2 3 . Pearson RCA, Esiri MM, Hiorns RW et al. (1985) A natom ical correlates of the distribution of the pathological changes in the neocortex in A lzheim er's disease. Proc N atl Acad Sci USA. 82: 4 5 3 1 ^ .

46

Practical old age p sy c h o p h arm o co lo g y

86 Arendt T, Bruckner MK, Gertz HJ et al. (1998) Cortical distribution of neurofibrillary tangles in A lzheim er's disease m atches the pattern of n eurons that retain their capacity of plastic rem odelling in the adult brain. Neuroscience. 83: 9 9 1 -1 0 0 2 . 87 B ow en DM, Francis PT, Chessell IP et al. (1994) N eurotransm ission - the link integrating A lzheim er's research. Trends Neurosci. 17: 1 4 9 -5 0 .

Chapter 4

Practical issues in assessing and prescribing psychotropic drugs for older people Stephen Curran, Sharon Nightingale and John P Wattis

Introduction The m anagem ent of older people w ith m ental illness invariably needs an integrated approach, w ith pharm acological, social and psychological inter­ ventions working together. H ow ever, this chapter w ill focus m ainly on pharm a­ cological treatm ents. Drug treatm ents have an im portant role to play alongside psychological and social interventions, but these broad approaches are not m utually exclusive. On the contrary, success w ith pharm acological interventions involves m any factors, only som e of w h ich are drug related. In general, the pharm acological evidence base is m uch better for younger patients. In particular, m any clinical trials exclude older people, and this m akes it difficult to extrapolate the findings to older people w ith m ental illness. Older people m ay also be excluded because it is m ore difficult to control for confounding variables such as concurrent diseases and the m edications needed to treat them . This m ust alw ays be borne in m ind w h en interpreting data from studies in younger people for use in older people. There is a n eed for more research on the use of psychotropic drugs in older people.

Pharmacological treatments In the sam e w ay that it has proved difficult to classify m ental disorders, so it has also proved difficult to classify psychotropic drugs, and a num ber of different classifications exist based on chem ical structures, m echanism s of action or the m ain effects on brain function. The classification is com plicated by the fact that m any drugs overlap and do not fit neatly into discrete categories. In general, the broad categories include antidepressants and m ood stabilisers, antipsychotics, anxiolytics and hypnotics and antidem entia drugs (these are discussed in Chapters 8 , 9, 10, 11 and 12). Sub-classification is usually on the basis of chem ical structures (e.g. tricyclic antidepressants), principal m echanism of action (e.g. cholinesterase inhibitors), pharm acokinetic properties (e.g. shortand long-acting benzodiazepines) and specific properties (e.g. sedation). H ow ­ ever, this classification, although useful, is very broad and has a num ber of lim itations.

47

48

Practical old age p sy c h o p h arm ac o lo g y

Assessment: general considerations Older people should be interview ed in a quiet, distraction-free environm ent, and every effort m ust be m ade to put th em at ease. In addition, they generally need m ore tim e than younger people and th ey should not be rushed. This is particu­ larly im portant w h en assessing 'liaison' referrals on m edical wards. Otherwise, any confusion w ill be com pounded and a falsely pessim istic im pression of function m ay result. Older patients m ay be very reluctant to discuss their personal feelings, so establishing a good rapport is essential. Som e patients, especially those w ith dem entia or com m unication difficulties, m ay not be able to give real consent. The doctor's ethical obligation is to act in the patient's best interest. This principle is particularly im portant w h en the patient's ability to exercise autonom y properly is dim inished. W here necessary, consideration should be given to using appropriate m ental health legislation. H ow ever, in England and W ales patients w ith dem entia w h o lack capacity can be treated or adm itted to hospital w ithou t the need to use m ental health legislation, provided that the healthcare professionals involved and the patient's family agree that this is the m ost appropriate course of action and that the patient is not actively refusing m edication or trying to leave hospital. A detailed assessm ent of the patient is im portant for a num ber of reasons. • First, it enables an accurate diagnosis to be made, and this has im plications for both the treatm ent and prognosis of the specific condition under consideration. • Secondly, it is the beginning of the process of developing a trusting relationship w ith the patient that w ill be crucial for providing a sound basis for treating the patient and m aintaining com pliance. • Thirdly, the assessm ent of the patient's past and current m edical problems will indicate w h ich drugs to avoid, and w ill highlight potential drug interactions. • Finally, a know ledge of the patient's past psychiatric history m ight point to the m ost appropriate treatm ents. A ssessm ent is often undertaken over a period of tim e. It usually involves several different disciplines w ith different insights into the needs of older people. All relevant assessm ents m ust be taken into account. The care plan should be m odified if necessary as n e w assessm ents are m ade. The self-discipline of good practice in record keeping and review can be im proved by the practice of regular peer-group audit, w h ich depends on careful record keeping. The use of standard­ ised assessm ents can facilitate audit. A ssessm ent of the patient is not a 'one-off' event. It should be repeated throughout treatm ent in order to evaluate progress and, if necessary, m odify treatm ent and care plans. This creates a 'feedback loop' w h ich should result in high-quality care m atched to the patient's current needs.

Psychiatric assessment History The psychiatric history starts w ith the presenting complaint (or com plaints), including h o w long it has b een present and h o w it developed. Quite often the patient lacks insight and believes that nothing is wrong. In these circum stances, careful probing is appropriate. Som etim es, w h e n it is difficult to obtain a clear

A ssessing an d prescribing p sy c h o tro p ic drugs fo r o ld e r p e o p le

49

history of the tim e course of an illness, the situation can be clarified by using 'tim e landmarks' such as the previous Christmas or som e im portant personal anniver­ sary. Often a proper history of the presenting com plaint can only be obtained by talking to a friend or relative before or after seeing the patient. In other cases inform ation m ay have to be pieced together from a variety of sources, such as hom e care staff, the social w orker and the GP. Usually it is best to follow the history of the presenting com plaint w ith an account of the personal history. M ost of us enjoy talking about ourselves, and it is quite easy to introduce the subject. A useful opening line is T ell m e a bit about yourself - w ere you born in this area?'. M em ory can be unobtrusively assessed w h ile going through the history by reference to im portant dates (e.g. the date of birth, call-up to the forces and date of marriage). The fam ily history and the history of past and current physical health can be w o v en into this brief account of the patient's lifetim e, and an assessm ent can be m ade of the patient's personality and characteristic w ays of dealing w ith stress. Old people, like you n g people, respond w ell to those w h o have a genuine interest in them . It is essential to ensure that the patient can see and hear the interview er. Courtesy is vital. Talking 'across' patients to other professionals or to relatives generates anxiety and resentm ent, as does lack of punctuality. It is also necessary to enquire about past psychiatric illness and any treatm ents.

Drug history It is im portant to establish the nam e, dose, frequency, duration, reason for starting, efficacy and tolerability of any psychotropic drug currently or previously prescribed. Inform ation should be sought about drug allergies, com pliance and any practical problem s (e.g. difficulty sw allow ing tablets). Som e drugs m ay cause or exacerbate psychiatric illness, and som e m ay cause side-effects that m ight exacerbate an underlying physical illness w h ich m ight th en cause further distress. Drugs m ay also interact w ith other drugs, so this needs to be considered and evaluated. M edication for physical illness should also be recorded. A ntihypertensives, digoxin and diuretics m ay be responsible for depressive sym ptom s, and all drugs w ith anti-cholinergic effects (including m any antidepres­ sants and antipsychotics) m ay produce confusion and constipation, am ong other side-effects. The list is constantly expanding, and the only safe advice is to assum e that any m edicine can potentially cause a w ide range of u nw anted effects. Postural h ypoten sion induced by tricyclic antidepressants or antipsychotics and other drugs m ay be m istaken for histrionic behaviour, and m ay be dism issed as part of the sym ptom s of an underlying depressive illness. B enzodiazepines often have a 'hangover' effect, and m ay accum ulate over m any days to produce confusion and daytim e drowsiness. M any drug interactions occur in older people, w h o are often on a num ber of different m edications. W hen an older patient presents w ith a n e w sym ptom , current m edication should alw ays be considered as a possible source of the sym ptom before further drugs are added. Ask about gastrointestinal sym ptom s, especially if selective serotonin reuptake inhibitors (SSRIs) are likely to be prescribed, as there is som e evidence that these can increase the risk of gastrointestinal bleeding, and the risk is similar to that associated w ith aspirin and non-steroidal anti-inflam m atory drugs (NSAIDs). These should be used w ith caution in older people (> 80 years) and in patients taking aspirin or an NSAID . 1 It is therefore im portant to be aware of current

50

Practical old age p sy c h o p h arm ac o lo g y

m edications (drugs for physical health, and psychotropic and over-the-counter drugs), previous drug sensitivities, response to treatm ent, side-effects and the potential for side-effects and drug interactions. In sum m ary, the content of the history w ill vary according to tim e and circum stance, but should generally include the following: • • • • • • •

the presenting com plaint and its history personal and family history (including illnesses and longevity) past illnesses and operations previous personality alcohol and tobacco consum ption current social circum stances and support drug history.

N\enta\ State Examination

Level of awareness At an early stage in the interview the patient's level of aw areness should be assessed. The patient m ay be drowsy as a result of lack of sleep or because of physical illness or m edication. A rapidly fluctuating level of aw areness is seen in acute confusional states, and a level of aw areness that fluctuates from day to day is on e of the clues to the diagnosis of chronic subdural haem atom a and/or dem entia w ith Lewy bodies. Impaired aw areness can lead to poor function on tests of cognition and m em ory, and if it is not recognised can lead to an underestim ation of the patient's true abilities. It is especially im portant to consider this if the patient has a recent-onset physical problem . The patient's ability to concentrate and pay attention is closely related to their level of awareness, but m ay be affected by m ore m undane factors. For exam ple, if the patient is in pain, it m ay be very hard for th em to understand the relevance of giving an account of their m ental state. Disturbance of m ood and abnormal perceptual experiences can also impair attention and concentration.

Behaviour (and general appearance) The patient's general appearance, behaviour, dress, personal hygiene and attitude to the interview er can be observed directly, and behaviour can also be deduced indirectly from the state of their house. Incontinence can often be sm elled (and occasionally felt). M obility can be checked by asking the patient to w alk a few steps. Especially if the patient lives alone, inconsistencies b etw een appearance and behaviour and the state of cleanliness and organisation of the h ousehold indicate either that there is a good social support netw ork or that the patient has deteriorated over a relatively short period of tim e. Various schedules enable the system atic assessm ent of behavioural 'problems'. These are best seen as a function of the interaction b etw een patient and environ m en t and not as intrinsic characteristics of the patient. The behavioural subscale of the Clifton A ssessm ent Procedure for the Elderly 2 and a shortened form of the Crighton Royal B ehavioural A ssessm ent Form 3 are useful exam ples. A scale has also been developed for the rating of behavioural sym ptom s by caregivers .4

A ssessing and prescribing p sy c h o tro p ic drugs fo r o ld e r p e o p le

5 1

B ehavioural and activities of daily living (ADL) scales enable num erical values to be attached to a person's needs, abilities and problem s in various important areas of behaviour. Such scales rem ind the assessor of im portant areas and enable discrepancies b etw een different areas of perform ance to be highlighted. P oten­ tially treatable problem s are m ore easily seen and dealt w ith. Standardised scales also enable a rough com parison to be m ade b etw een different patients and b etw een different points in tim e for the same patient, even w h en the assessm ent is m ade by a different person. Finally, som e scales give an overall rating of disability that can be used as a guide to the patient's future needs for care. There are m any such scales 5 and th ey are all imperfect, but they do at least enable a quick and system atic approach to the assessm ent of behaviour and abilities. The num erical values ascribed to such scales are, of course, arbitrary. For exam ple, a patient w h o is disturbed all night m ay not be m anageable at hom e, despite that factor only contributing a small am ount to the overall score. Scores m ust therefore be interpreted skilfully, taking into account the am ount of support available and the peculiar impact of certain behaviours.

Affect (mood) M ood in the technical sense used by psychiatrists is more than just h o w w e feel. It has b een described as 'a com plex background state of the organism', and it affects not only h o w w e feel but also h o w w e think and even the functioning of our m uscles and bow els. Older people are n ot always used to talking about their feelings, and it can som etim es be quite difficult to find the right words. This is particularly true for m en, w h o m ay not have the words to describe feelings (alexothym ia). Especially w here there are com m unication difficulties, one m ay have to resort to direct questioning (e.g. 'Do you feel happy?'). A lthough patients should alw ays be asked to give an account of their m ood, this account cannot alw ays be relied upon. Som e older patients w h o are quite depressed do not confess to a depressed m ood. This is often accom panied by som atisation - that is, the presentation of physical (hypochondriacal) com plaints. It m ay also signify a m ore or less deliberate 'cover-up' due to fear of hospital adm ission. A nhedonia (loss of the ability to take pleasure in life) is a useful indicator of depression. Psychom otor retardation (the slow ing of thought and action) can be so profound that the patient is unable to report their m ood or m ay even say 'I feel nothing', although their facial expression, tears, sighs, slow ed m ovem en t or agitation m ay reveal depression. Specific questions should be asked about guilt feelings, financial worries, and concerns about health. W here there is depressed m ood, careful enquiry should be m ade about suicidal feelings. This can be introduced in a non-threatening w ay by a question such as 'Have you ever felt that life was not w orth living?'. If the patient responds positively to this, further probes can be m ade about present ideas of self-harm. If psychom otor retardation is present, the answ er w ill take som e tim e to com e, and it is very easy to rush on to the n ext question before the patient has had tim e to respond to the previous one. Risk factors for suicide w hich should always be borne in m ind include the following: • m ale gender • depression • living alone

52

Practical old age p sy c h o p h arm ac o lo g y

• bereavem ent • long-standing physical illness or disability • alcohol abuse. Being aware of such risk factors com plem ents but does not replace individual questioning and judgem ent. One group of sym ptom s is often associated w ith severe 'biological' depression. This includes early-m orning w akening, low m ood w orse in the m orning (diurnal variation in m ood), and profound appetite loss and w eight loss. Som e self-rating scales have b een designed to avoid the confusion produced by the use of 'somatic' sym ptom s in other scales. They include the following: • the Geriatric Depression Scale (GDS ) 6 • short forms of the GDS 7 and • the brief assessm ent schedule depression cards (BASDEC ) 8 and related even briefer assessm ent scale for depression (EBAS-DEP ) 9 These scales can provide indications of possible m ood problem s and, together w ith other scales (especially the Hospital A nxiety and Depression Scale [HAD] ) , 10 can be useful in m easuring severity of depression, response to treatm ent and outcom e. H ow ever, the frequent som atisation of depression in old age m ay cause 'false negatives' on such scales. The opposite of depressed m ood is elated m ood, w hich is seen in m ania and hypom ania. In older patients, as in younger ones, decreased sleep, hyperactivity, excessive spending, flight of ideas, irritability and hypersexuality m ay be prom inent sym ptom s. A nxiety is also com m on in old age, som etim es in response to the stresses of ageing in our society. The patient m ay be so worried about falling that, in order to avoid anxiety, they restrict their life severely. Thus a patient w h o has had one or tw o falls m ay, instead of seeking m edical help, restrict him - or herself to a downstairs room in the hou se and never go out. As long as the patient continues to restrict their life, they experience little anxiety. W hereas in a you n g person such behaviour w ou ld alm ost certainly im m ediately lead to the patient being defined as 'sick' and a call for m edical attention, in older people this restriction is all too easily accepted as 'normal'. W hen assessing anxiety, attention should therefore be paid n ot only to h o w patients feel during the interview (w hich m ay in itself provoke anxiety!), but also to w hether they can engage in the tasks of daily living w ith out experiencing undue anxiety. A nxiety is an affect w ith physiological accom panim ents - a racing pulse, 'palpitations', 'butterflies in the stom ach', sw eating and diarrhoea. Patients not infrequently use the term 'dizziness' to describe not true vertigo, but a feeling of unreality associated w ith severe anxiety. Som etim es the physiological changes induced by over-breathing, such as tingling in the arms and ev en spasm of the m uscles of the hand and arm, m ay m ake matters w orse. The anxiety subscale of the HAD 10 provides an easy patient-rated m easure of the intensity of anxiety problems. Interestingly, anxiety and depression are often found together, and tend to im prove together w ith treatm ent. Panic attacks of rapidly m ounting anxiety, usually w ith physiological sym p­ tom s, m ay occur as part of a phobic disorder, in isolation or, perhaps m ost com m only in old age, in the context of a depressive disorder.

A ssessing an d prescribing p sy c h o tro p ic drugs fo r o ld e r p e o p le

53

Phobias occur w h en the patient is afraid of particular things or situations. Specific phobias (e.g. fear of spiders) are relatively uncom m on in old age, but generalised phobias (e.g. fear of going out or of social situations) are relatively com m on and can be crippling . 11

Thought (and talk) The form, speed and content of thought should all be assessed. Formal thought disorder occurs in schizophrenia and includes thought-blocking (w h en the patient's thoughts com e to an abrupt end), thought w ithdrawal (w h en thoughts are felt to be w ithdraw n from the patient's head), thought broadcast and thought insertion. For a fuller description of these p h enom ena, the reader is referred to a standard textbook of psychopathology . 12 Slow ing of the stream of thought (thought retardation) is found in m any depressive disorders. Slow thinking is also characteristic of som e of the organic brain syndrom es caused by m etabolic deficiencies. Thought is speeded up in m ania, often leading to 'flight of ideas', w here one thought is built upon another in a w ay that is founded upon tenuous associations. In dem entia, spontaneous thought is often dim inished, so-called 'poverty of thought'. The patient w ith an acute confusional state experiences difficulty in m aintaining a train of thought because of fluctuating levels of aw areness. In dem entias of m etabolic origin and in som e cases of m ulti-infarct dem entia, slow ing of th ought processes m ay be accom panied by difficulty in assem bling the necessary know ledge to solve particular problem s. The observer gains the im pression that the patient grasps that there is a problem but is frustrated in trying to cope w ith it. C ontent of thought is influenced by the patient's m ood. The depressed patient w ill often have gloom y thoughts, w ith ideas of poverty or physical illness. The anxious patient's thoughts m ay be preoccupied w ith h o w to avoid anxietyprovoking situations, and there m ay be unnecessary worries about all aspects of everyday living. The patient w h o feels persecuted m ay think of little else. Every noise or even t w ill be fitted into the persecutory framework. Talk generally reflects the patient's thought, unless suspicion leads to concealm ent. Speech is also influ en ced by various m otor functions. Slurred speech m ay be found in the patient w h o is drowsy or under the influence of drugs or alcohol. Som etim es it also results from specific neurological problems, such as a stroke. Patients w ith m ultiple sclerosis m ay produce so-called 'scanning' speech w here words are produced w ith ou t inflexion and w ith hesitation b etw een words. Patients w ith severe parkinsonism m ay have difficulty in form ing sounds at all (aphonia). Som e difficulty in finding words and putting speech together is found in m any patients w ith dem entia, particularly those w ith Alzheim er's disease. This is one form of dysphasia. A stream of apparent non sen se (so-called fluent dysphasia) m ay occur in dem entia but is also som etim es associated w ith a small stroke. The general behaviour of the patient, w h ich is n ot 'dem ented', and the sudden onset of the dysphasia provide im portant diagnostic clues. Occasionally, fluent dysphasia, especially w h e n it includes n e w words 'invented' by the patient (neologism s) m ay be m istaken for the so-called 'word salad' produced by som e schizophrenic patients. The sudden onset and absence of other signs of schizophrenia aid in diagnosis. An assessm ent by a speech therapist experienced in this area m ay also help not only in diagnosis but also in suggesting m anagem ent strategies that build upon the patient's residual (som etim es non-verbal) com m unication skills.

54

Practical old age psychopharmacology

Hallucinations These can be defined as perceptions w ith ou t external objects. Visual hallucina­ tions are usually found in patients w ith acute confusional states or dem entia, although occasionally th ey occur in patients w ith poor eyesight w ithout m easurable organic brain damage, especially if the patient is living alone in a relatively under-stim ulating environm ent. In dem entia they are m ore often found in dem entia w ith Lewy bodies, and their presence m ay correlate w ith neuropathological findings . 13 Auditory hallucinations (hearing 'voices' or so m e­ tim es m usic or sim ple sounds) occur in a variety of m ental illnesses, especially schizophrenia (w here th ey m ay consist of a voice repeating the patient's thoughts or of voices talking about the patient in the third person). They also occur in severe depressive illness and m ania, w h en they are often in keeping w ith the patient's m ood. Hallucinations of touch (tactile), sm ell (olfactory) and even taste (gustatory) also occur. H allucinations of being touched, especially w ith sexual connotations, occur in schizophrenia, and hallucinations of smell, especially of the patient believing him - or herself to sm ell 'rotten', in severe depression. Experiences like hallucinations som etim es occur in bereavem ent. These vary from 'hearing the footsteps' of the lost person to com plex p h en o m ­ ena occurring in more than one m odality. People explain these in different w ays. Som e see them as spiritual and com forting, w h ile others m ay be afraid th ey are 'going mad'.

Delusions A delusion is a false unshakeable belief that is out of keeping w ith the patient's cultural background. D elusions occur in fragmentary forms in organic m ental states, but w ell-d evelop ed delusions are usually found only in schizophrenia and in severe affective disorders, w h en ideas of poverty, guilt or illness m ay develop into absolute convictions. Ideas of persecution are also som etim es found in patients w ith depression of m oderate severity, and these too can develop into full-blow n delusions. D elusions of grandeur (e.g. that the patient has extra­ ordinary pow ers of perception or is fabulously rich) are found in m anic states. In paranoid schizophrenia, the delusional content is often very com plicated and m ay involve persecutory activities by w h o le groups of people. These delusions m ay be su p p o r te d b y h a llu c in a to r y e x p e r ie n c e s .

Obsessions and compulsions Obsessions occur w h e n the patient feels com pelled to repeat the sam e thought over and over again. They can be distinguished from schizophrenic phenom ena such as th ought insertion by the fact that obsessional patients recognise the thoughts as their o w n and try to resist them . Som etim es such thoughts m ay result in com pulsive actions (e.g. returning m any tim es to check that the door has b een locked). A lthough they are characteristically a part of obsessive-com pulsive disorder, obsessional sym ptom s also occur in depressed patients, and repetitive (not strictly com pulsive) behaviour can also be a result of m em ory loss (e.g. w h en a patient repeatedly checks that the door is locked because they have forgotten that they have already done so).

A ssessing and prescribing p sy c h o tro p ic drugs fo r o ld e r p e o p le

55

Illusions Illusions occur w h en a patient m isinterprets a real perception. Som e som atic (hypochondriacal) worries can be based on this. For exam ple, m any older people have various aches and pains, but som etim es patients m ay becom e over-concerned by these and m ay begin to worry that they indicate som e physical illness. Such m isinterpretations of internal perceptions are not usually described as illusions, although the term w ould be quite appropriate. Acute confusional states also produce illusions w h e n the patient, seeing the doctor approaching them , m isinterprets this as som eon e com ing to do them harm and strikes out.

Cognition: orientation, memory, concentration and attention These are considered together because they are so interdependent. Orientation w ith regard to tim e, place and person should be recorded in a system atic w ay. The degree of detail w ou ld depend upon the tim e available and the purpose of the exam ination. Orientation for tim e can easily be divided into gross orientation, such as the year or approxim ate tim e of day (m orning, afternoon, evening or night), and finer orientation, such as the m onth, the day of the w eek and the hour of the day. Orientation w ith regard to person depends upon the familiarity of the person w h o is chosen as a point of reference. Orientation w ith regard to place also depends upon familiarity. A useful brief scale w hich includes som e item s of orientation as w ell as som e item s of m em ory is the Abbreviated M ental Test (AMT) score, developed by H odkinson 14 from a longer scale w hich has previously b een correlated w ith the degree of brain pathology in dem ented patients . 15 A slightly longer related scale has been developed for com m unity u se , 16 although m any people 'adapt' the AMT (e.g. by asking for the patient's hom e address rather than the hospital nam e). Like m ost short scales, the AMT is not alw ays w ell used . 17 It is also frequently overused, leading to false positives. Orientation is to a large extent dependent upon m em ory, although it should n ever be forgotten that the patient m ay not k n ow the nam e of the hospital they are in, sim ply because they have n ever been told. M em ory for rem ote events can be assessed w h e n taking the patient's history. This can be done by checking against k n ow n facts such as date of birth, checking w ith family members, and also by the internal consistency of the history. The ability to encode n e w material can be assessed by the capacity to rem em ber a short address or to rem em ber the interview er's nam e w h e n tested a few m inutes later. M any patients w ith dem entia w ill have great difficulty in encoding and storing n ew m em ories. Som etim es, especially in the m etabolic dem entias, one can form the im pression that the patient is encoding and storing n e w material but that they are having great difficulty in retrieving the m em ory w h en asked to do so. This has b een described as 'forgetfulness'. Apraxia (dem onstrated by the inability to copy sim ple drawings) and nom inal aphasia (the inability to rem em ber the nam es of com m on objects) can also be sim ply tested. A popular and relatively brief assessm ent of organic m ental state is the M ini-M ental State E xam ination (MMSE ) 18 w hich exam ines m em ory and a variety of other functions. H ow ever, the MMSE has been criticised because it is sensitive to educational attainm ent, its 'parallel' forms are not truly parallel, and it is not sufficiently sensitive to change. Longer scales such as the Cambridge cognitive exam ination (CAMCOG ) 19 are probably too lengthy for routine clinical

56

Practical old age p sy c h o p h arm ac o lo g y

use. A variety of m easures are used for evaluating outcom es in drug trials, 20 but these are generally too lengthy for routine use. A prom ising developm ent is the construction of scales designed to elicit changes in the patient's m em ory from relatives and carers, such as the inform ant questionnaire on cognitive decline in the elderly (IQCODE ) . 21 A nother useful brief test w h ich assesses a variety of cognitive functions is the 'clock test', w here the assessor presents the patient w ith a circle and asks them to fill in the num bers as on a clock face and set the tim e to (for exam ple) ten to tw o. This test is a useful screening tool for a broad range of cortical fu n ction . 22 In m ore severe dem entia, w here other tests are disabled by 'floor' effects, the Hierarchical D em entia Scale (HDS ) 23 m ay be useful. M ore detailed descriptions of organic m ental state exam ination can be found in Lishman's Organic Psychiatry. 24

Insight and judgement In severe psychiatric illness, insight (in the technical psychiatric sense) is often lost. Depressed patients m ay be unable to accept that they w ill get better, despite rem em bering previous episodes that have im proved w ith treatm ent. M anic and paraphrenic patients m ay act on their delusions, w ith disastrous consequences. Patients w ith severe dem entia often do not fully realise their plight, w hich is perhaps fortunate. Patients w ith milder dem entia m ay have som e insight, especially in the m etabolic and vascular types of dem entia w here m ood is, not surprisingly, also often depressed. Judgem ent is related to insight. This can be a particularly difficult question w ith a m oderately dem ented patient w h o is living alone, or living w ith relatives but left alone for a substantial part of the day. Patients m ay leave gas taps on and be dangerous to them selves and others, but at the sam e tim e m aintain that they are looking after them selves perfectly w ell and do not need any help, m uch less residential or nursing hom e care. They m ay have a m istaken im age of the care that th ey are refusing. C om m on sense, professional ju dgem ent and m ulti-professional, inter-disciplinary w orking are needed to make decisions in these cases.

Physical assessment Specialists in the psychiatry of old age need to work closely w ith their medical colleagues, particularly GPs and physicians for older people, because of the overlap and interactions b etw een physical and psychiatric illn ess . 25 No psychiatric exam ination, particularly in older people, is com plete w ithout a physical exam ination. Even in the patient's hom e a selective exam ination m ay be carried out, although it m ay be m ore appropriate to bring the patient to the clinic or the surgery for a m ore thorough exam ination. For a fuller account of assessm ent from the point of v iew of geriatric m edicine, the reader is referred to Assessing Elderly Patients.26 A joint psychiatric-geriatric clinic can facilitate the m anagem ent of difficult cases. H ow ever, simple things can m ake a big difference to patients, such as im proving vision and hearing, foot care and dentures, as w ell as suitable w alking and other aids.

Sensory impairment D im inished sensory input, one of the techniques used in 'brainwashing', is often inflicted on old people by our slow ness in recognising and correcting defects of

A ssessing an d prescribing p sy c h o tro p ic drugs fo r o ld e r p e o p le

57

sight and hearing. Sensory deprivation m ay be instrum ental in producing paranoid states and in precipitating or w orsening confusion. Poor hearing is also associated w ith depression. A n estim ate of visual and auditory acuity is part of the exam ina­ tion of every older person. W ax in the ears is an easily rem edied cause of poor hearing. Other forms of deafness m ay require a hearing-aid. A good deal of patience m ay be need ed to learn to use an aid properly, especially if poor hearing has been present for som e tim e. Look out for flat or dirty battery contacts in hearing-aids. For assessm ent purposes, m ore pow erful portable amplifiers are useful. Even the in expen sive amplifiers linked to sim ple headphones that are advertised in popular m agazines can be surprisingly effective. Visual defects range from those that are easily corrected by spectacles and other aids to those like cataract and glaucom a that require m ore com plicated surgical or m edical intervention.

Laboratory investigations Further investigations m ay be planned in the light of findings from the history, m ental state and physical exam inations. M any doctors w ou ld confine them selves to haem oglobin, full blood count and film, urea and electrolytes, liver function tests and thyroid function tests, and som e w ou ld routinely add serum vitam in B 12 and folate and a serological test for syphilis. A lthough the necessity for the last is disputed, clinics that perform such tests still report unexpected positive findings. Other tests, such as chest X-ray, skull X-ray, electroencephalogram , com puterised axial tom ography (CAT), nuclear m agnetic resonance (NMR) im aging and other forms of brain scan are at present only justified by specific indications. Hopes that CAT scans m ight provide an easy and definitive diagnosis of 'senile dem entia' by dem onstrating brain atrophy have not been realised due to w ide overlaps in the picture b etw een norm al, functionally ill and dem ented patients, although occasionally a scan w ill reveal an unsuspected tum our or other problem. N ew er scanning techniques such as NMR 27 or positron em ission tom ography (PET) m ay even tu ally be able to help in definitive diagnosis before death. In one fascinating study involving post-m ortem verification, even SPECT scanning only had similar accuracy to rigorously applied diagnostic criteria (NINCDS-ADRDA), and a m edial tem poral lobe CT scan w as not m uch better than the use of DSMIII-R criteria . 28 The Royal College of Psychiatrists Consensus Statement gives a reasonable account of w hat m ight be expected of the evaluation of dem entia by a secondary service in the UK . 29

Additional investigations A num ber of additional investigations m ight need to be performed prior to starting treatm ent w ith a specific drug. The list b elow is not exhaustive, and the reader is referred to the latest British National Formulary, the Sum m ary of Product Characteristics (SPC) produced for each drug by the drug m anufacturer, and The Maudsley Prescribing Guidelines.30

Antipsychotics A full blood count (FBC) should be done before initiating treatm ent w ith clozapine. A n FBC should th en be m easured every w eek for 18 w eeks, th en every 2 w eeks for 1 year, and thereafter every m onth.

58

Practical old age p sy c h o p h arm ac o lo g y

Sudden unexplained death has b een k n ow n for som e tim e to be associated w ith antipsychotics, but the frequency has not b een clearly defined. Som e antipsy­ chotics are associated w ith QTc-interval prolongation, w hich increases the risk of sudden death from the potentially fatal ventricular arrhythmia k now n as torsades de pointes, and thioridazine has been linked m ost frequently w ith th is . 31 Drugs w ith a m oderate effect on QTc interval include chlorprom azine and quetiapine, and those w ith a high risk include thioridazine and sertindole. The risk is increased in patients w ith cardiovascular disease (e.g. long QT syndrom e, bradycardia and ischaem ic heart disease), use of high doses, m etabolic disorders (hypokalaem ia, hypocalcaem ia, hypom agnesaem ia) and other factors such as stress/shock, fem ale gender and increasing age. Patients at risk should have an ECG at baseline and thereafter every 6 m onths. Sw itching to a different drug is required if the QTc interval is > 4 4 0 ms in m en and > 4 7 0 ms in w om en . Antipsychotics can also cause hyperglycaem ia, so blood sugar levels should be m onitored, especially w ith olanzapine.

Antidepressants and mood stabilisers M ost antidepressants, but especially SSRIs and venlafaxine, can low er the serum sodium concentration, so this is w orth m easuring before and after initiating treatm ent. Sodium levels of 1.25 mmol/1 or low er require urgent advice and treatm ent from a physician. Lofepramine can cause abnorm al liver function tests (LFTs) and not u ncom m on ly frank jaundice, m aking LFT m easurem ent advisable. M irtazepine and m ianserin (not often used now ) can suppress the w hite cell count, so this should be m easured before treatm ent and regularly thereafter. Patients w h o are being prescribed lithium should have a baseline ECG, thyroid function tests (TFTs) and urine and electrolytes (U&Es). The lithium level should be m easured w eekly until levels are in the therapeutic range and stable, and thereafter should be m easured every 3 to 6 m onths. U&Es should then be m easured every 3 m onths and TFTs every 6 m onths. Carbamazepine requires a baseline FBC to be repeated every 2 w eeks for 2 m onths and thereafter every 3 to 6 m onths. Sodium valproate requires a baseline FBC, LFTs and U&Es to be repeated every 6 m o n th s . 30

Antidementia drugs The cholinesterase inhibitors m ay cause bradycardia, and m any clinics routinely undertake ECGs before com m encing treatm ent. Patients w ill norm ally be screened for a range of problem s to exclude the treatable causes of dem entia, and this w ould usually include LFTs. The latter are m ore im portant in patients taking donepezil or galantam ine, as these drugs are m etabolised by the liver . 32 A recent study w ith rivastigm ine that specifically looked at cardiovascular effects found no effect on heart rate, QRS and QT interval compared w ith placebo . 33 Before initiating treatm ent w ith an antidem entia drug, the diagnosis should be established. The laboratory tests recom m ended by the National Institute of H ealth 34 in the USA for a patient w ith suspected dem entia include the following: • com plete blood count • electrolyte and m etabolic screen • thyroid function • vitam in B 12 and folate levels

A ssessing an d prescribing p sy c h o tro p ic drugs fo r o ld e r p e o p le

• • • • •

59

syphilis serology urine analysis chest radiograph ECG head CT scan.

Hom e assessment This is usually an integral part of the assessm ent, and m ost older patients referred for psychiatric assessm ent should preferably be seen initially in their ow n hom es. This is also an ideal opportunity to undertake a social assessm ent, including a review of social netw orks, relationships and coping strategies used by fam ily and friends. H om e assessm ents have a num ber of advantages, including the follow ing. • The patient is seen in the situation w ith w hich he or she is familiar. • The confusion, disorientation and distress w h ich m ay be caused by a trip to hospital or consulting room s are avoided. • The environm ent can be assessed as w ell as the patient. • The patient's function in his or her o w n environm ent and the level of social support can be assessed. • Neighbours and relatives are often readily available to give a history of the illness and its impact on them . Set against this are the disadvantages from the clinician's perspective, including tim e spent travelling, difficulties parking and problem s performing physical exam inations and tests in the patient's h om e. H owever, the hom e assessm ent allow s a range of im portant areas to be assessed quickly w hich w ould not be apparent if the patient w as seen in a clinic setting. Som e exam ples are sum m arised in Box 4.1.

Box 4.1 The h om e assessm ent: w hat can be assessed? (m odified from Curran and W attis35) • Local environment Location of shops and busy roads. • General state of the house: General state of repair and cleanliness. • Cooker and fire: Is the cooker being used? Is there any evidence of fire risk? • Fridge: Is the fridge w ell stocked w ith fresh food? Is any of the food out of date or rotten? • Heating, lighting and ventilation: Is the house cold? Is there adequate lighting? • Fire risks: Cookers and fires are obvious exam ples. Is there any evidence of cigarette burns (e.g. in the bed area)? • Toilet and bathing facilities: Are these being used appropriately? • Accident hazards: Can the patient use the stairs safely? Are the floors cluttered? • Sleeping arrangements: Is the patient getting enou gh sleep? W here is the patient sleeping? If the patient is sleeping downstairs, are the arrangem ents adequate? • Alcohol: Is there any evidence of alcohol m isuse (e.g. lost or em pty bottles)?

60

Practical old age p sy c h o p h arm ac o lo g y

•

Vulnerability: Is the patient vulnerable (e.g. leaving doors unlocked, leaving large am ounts of m o n ey in exposed places)? • Medication: Are there any problem s w ith the patient's m edication? Can the m edication be found? Is it the same as the prescribed m edication? Is the patient taking the m edication?

The patient's family and neighbours often have a key role to play in assessm ent and continuing m anagem ent. Therefore a good relationship w ith them is very im portant. At the first interview , the patient and family w ill express m any anxieties. Som e of these m ay be founded upon their ow n ideas about the purpose of the assessm ent. Time is w ell spent listening to the problem s as the patient and relatives see them . A still popular m isconception is that the doctor, nurse or social worker has com e to 'put aw ay' the patient in the local institution. The elderly patient's idea of w hat institutional care involves m ay also be quite different from that of the assessor. Older people som etim es find it difficult to conceive that an adm ission to hospital or a residential h om e could be anything other than perm anent. W e need to take tim e to listen to these fears and to explain w h y w e are visiting, and the scope and lim itations of any help w e can offer. A nxiety m ay impair the patient's and relatives' ability to grasp and rem em ber w hat is being said. It m ay be necessary to repeat the same inform ation several tim es and to ask questions to clarify w hether explanations have really b een understood. A lthough an assessm ent m ay be com m onplace to us, for the patient and their relatives it is often taking place at a crisis point in their life. An em pathetic m anner, acknow ledging the patient's and relatives' concerns, w ill help them to realise that their worries have b een taken seriously. This w ill help to form a good relationship, w hich is the basis for further treatm ent. The quality of relationships can be assessed both indirectly from the past history of the patient and the family, and m ore directly during a joint interview . It is w orth trying to assess w hat each m em ber of the fam ily is aim ing for and h o w open fam ily m em bers are in their com m unication w ith each other. Som etim es there is a pathological attachm ent b etw een fam ily m em bers, resulting in a m aladaptive pattern of caregiving. The grow ing physical and psychological dependence of old people w ith progressive illnesses such as Parkinson's disease and Alzheim er's dem entia can put extraordinary stress on fam ily relationships that w ill reveal previously care­ fully disguised 'fault lines'. Because services are som etim es only m ade available w h en a crisis has occurred and m em bers of the family are at the end of their tether, w e do see relatives w h o m ay be labelled as 'rejecting'. This happens less frequently than it used to, as the develop m ent of m ore effective services has led to earlier intervention before a crisis has occurred. Even w h en a crisis has occurred, it m ay be possible to m anage it in such a w ay that the relatives realise that they can continue to cope if they have the help of appropriate services. In addition, carers n o w have a statutory right to have their ow n care needs assessed by the social services.

A ssessing an d prescribing p sy c h o tro p ic drugs fo r o ld e r p e o p le

6 1

M aking allies o f the family An im portant factor here is a prom pt response, usually in the form of a h om e visit. This is the first step in im pressing upon the fam ily the fact that help is available. Carers are relieved to find som eone w h o has tim e and is w illing to listen to the problem s they are facing and provide practical help. This can cause family m em bers to re-evaluate their attitudes and avoid prem ature decisions to put an elderly relative into a care hom e. The understanding and assessm ent of a situation by fam ily m em bers m ay be quite different from the professional view point, and m ust be 'heard' and respected by the team that is planning help. The appropriate use of short-term care hom e or hospital adm ission, day care, fam ily care and h om e care services to relieve perceived strains can enable the fam ily to cope. A lthough the m edical m em bers of the team should include social and support needs in their initial assessm ent, w here these are complicated, further expert assessm ent by another team m em ber is justified, w h eth er by a social worker, an occupational therapist or a com m unity psychiatric nurse.

The need for 24-hour care W hen an older person lives alone and suffers from m oderate or severe dem entia, it m ay be im possible to provide adequate supervision w ith ou t adm itting them to a long-stay care facility. This should not be view ed as a 'failure' to keep the person 'in the com m unity', but as the appropriate use of one of a range of options for providing care. M anagem ent depends upon psychiatric and m edical diagnosis as w ell as the fam ily and social situation, and the skill of the psychogeriatric team lies in understanding the various com ponents of the situation and h o w they interact in order to produce the best possible m anagem ent plan.

Problem formulation and management: ‘care planning’ Each patient needs a 'tailor-m ade' care plan of medical, psychological, nursing and social m anagem ent. This w ill often need to be agreed on a m ulti-disciplinary basis after assessm ents have been m ade by several different m em bers of the team . Som e team s insist that the initial assessm ent is m ade by a senior m em ber of the m edical staff, as physical and severe psychiatric illness usually needs to take precedence in any m anagem ent plan. A nother m em ber of the team m ay attend either w ith the doctor or later. The alternative pattern of using specially trained and supervised nursing staff, for exam ple, m ay becom e increasingly com m on in the face of rising dem and and lim ited resources. In any case, it is essential that the initial assessm ent includes an overall v iew of m edical, psychological and support needs and social factors, and that the fam ily is given som e idea of the likely m anagem ent and the help available at that first assessm ent. This includes at least an outline know ledge of local h om e care, day care, voluntary groups, financial allow ances and other relevant sources of help. The care plan w ill usually evolve and develop as the patient's condition and circum stances change. The care planning procedures 36 provided an official fram ework for this process in the UK. They introduced a m ore carefully m onitored procedure to replace the inform al approach that previously prevailed. The advantages of a m ore formal

62

Practical old age p sy c h o p h arm ac o lo g y

procedure are that all stakeholders are m ore likely to have a 'say' in producing the plans, and that there is less likelihood of poor com m unication leading to failure. The disadvantages are that the n e w procedures can be tim e-consum ing and im pose a bureaucratic burden on overw orked clinicians. This in turn can lead to a slower and less flexible team response w here the central purpose of caring for the patient is obscured by the necessity of filling in forms. Ideally a synthesis of the personal clinical approach and m ethodical docum entation and process can be achieved, but not w ith ou t supporting clerical resources and inform ation technology. Case m anagem ent is a related philosophy w hich is used by social services departm ents in the UK and elsew here. This approach potentially allow s a m ore flexible deploym ent of resources, although there is also a danger that it w ill sim ply becom e a crude m echanism to control costs and transfer blam e. A review of research into care m anagem ent suggests that approaches in w hich the care m anager is a clinician actively involved w ith the patient work better than those in w hich the care m anager is a 'broker' w h o is n ot directly involved w ith the patient . 37 A medical, psychosocial and nursing 'diagnosis' is only the beginning of patient m anagem ent. Because of the m ultiplicity of problems faced by patients, a haphazard approach to m anagem ent is tim e-w asting, inefficient and potentially dangerous. For exam ple, arranging long-term hom e care support for people w h o are suffering from undiagnosed and untreated depressive episodes m ay 'keep th em going' in the com m unity, but it does not permit them to have the best possible quality of life, and it m ay lead to a treatm ent-resistant depressive illness or even to suicide. Arranging nursing hom e care for people w h ose confusion is due to an undiagnosed m edical condition denies them proper treatm ent, reduces their independence, w astes resources and m ay also have fatal consequences. On the other hand, a narrowly m edical approach to people's problem s will also reduce their prospects of independent living. Team work is therefore essential, and it is m ost im portant that case m anagem ent and care planning procedures do not detract from the tradition of m ulti­ disciplinary team w ork that has been built up in m any psychogeriatric services. The traditional psychiatric form ulation of a patient's problem s has always included not only the psychiatric diagnosis but also all relevant m edical, p sycho­ logical and social factors. Care plans achieve the same thing by listing the patient's needs, planning interventions to m eet those needs, and agreeing responsibility b etw een professionals and agencies for different com ponents of the plan. A lthough these plans are ideally agreed at a m eeting of the patient, caregiver(s), h ealth service and social services staff, in practice w e have found this too tim econsum ing to be accom m odated w ith in the available resources. W e n o w reserve full m eetings for difficult cases, and w e use an abbreviated (although still firmly m ulti-disciplinary) procedure for the majority of cases. C om m unity team m em ­ bers review patients regularly in a m ulti-disciplinary forum, but only arrange formal m eetings as and w h en circum stances dem and.

Starting drug treatments Lader and H errington 38 have suggested that 'unless there is intense distress there is no need to proceed w ith haste'. There are a num ber of im portant reasons for delaying treatm ent.

A ssessing and prescribing p sy c h o tro p ic drugs fo r o ld e r p e o p le

63

• The additional tim e allow s for m ore detailed assessm ent, w hich w ill m ake the eventual treatm ent m ore inform ed. • Psychotropic drugs can be harmful, and older people are often very sensitive to their side-effects, so th ey should not be given unnecessarily. • Som e patients im prove after the initial assessm ent/adm ission, rendering psychotropic drugs unnecessary. For exam ple, in general practice patients w ith m inor disorders often im prove after sim ple discussion, and drugs m ay reinforce the patient's v iew that they have a 'serious illness ' . 39 Adm ission to hospital m ay be associated w ith a significant im provem ent in sym ptom s. H ow ever, this is n ot alw ays sustained. • Other reasons include the increased risk of drug interactions and the financial im plications. W here possible, older people should be on as few drugs as possible. H ow ever, if a decision is m ade to prescribe a particular drug, then choosing the m ost appro­ priate m edication can be a com plex process. Response to previous treatm ents is not alw ays a good guide , 38 but this needs to be considered. In general, drugs should be introduced gradually (unless there is a single dose) and stopped gradually. Som etim es small, apparently 'subtherapeutic' doses can be useful for people w h o are intolerant of psychoactive drugs (e.g. the patient can be given half or a quarter of the norm al dose). H ow ever, som e patients w ill require standard doses of drugs, and antidepressants in particular should not generally be abandoned w ith ou t a trial of an adequate dose for an adequate period of tim e. There is also variation in response depending on the illness and comorbid clinical factors such as personality factors . 40 Age itself is an im portant determ inant of response to treatm ent, due to reduced drug clearance, increased side-effects, reductions in receptor density and increased risk of interactions w ith other drugs. Physical illness can also impair the response to treatm ent through a variety of m echanism s. Reduced diet can also have an impact through changes in the proportion of body fat, reduced plasma proteins, reductions in the levels of am ino acids and vitam ins, w ith consequences for enzym e function , 41 and changes in receptor density. In general, it is better to k now a few drugs in detail so that on e becom es familiar w ith their doses, side-effects and interactions w ith other drugs. Confusion is also reduced if the generic nam e is used w h en ever possible. Com plex drug regimes should be avoided, and in general it is not usually necessary to use tw o or m ore drugs from the sam e class of drugs (e.g. tw o or more antipsychotic drugs). Drugs should be com m enced cautiously. Similarly, drugs should be stopped cautiously a n d o v e r a f e w d a y s o r w e e k s , p a r tic u la r ly if th e p a tie n t is o n h ig h d o se s a n d /o r

has been taking the drug for extended periods of tim e. This w ill depend in part on the individual patient, the drug in question, the dose and the duration of treatm ent. There is a range of factors that have to be considered before prescribing drugs in older people. One of the possibilities that should alw ays be considered is the advantage of prescribing n o th in g . 38 C onsent is another im portant issue. In order to consent to treatm ent, the patient should have a broad understanding of the treatm ent, including the risks and benefits, the consequences of not taking the treatm ent, and the alternatives currently available (w ith their risks and benefits). Patients w h o lack capacity to give consent should norm ally be treated under the M ental Health Act 1983, but

64

Practical old age p sy c h o p h arm ac o lo g y

patients w ith dem entia w h o lack capacity are frequently treated w ithout this legal fram ework. This has generated considerable m edical and legal debate .42 The current legal position in England and W ales is that patients w ith dem entia w h o lack capacity can be treated or adm itted to hospital w ithout the need to use m ental health legislation, provided that the healthcare professionals in ­ volved and the patient's fam ily agree that this is the m ost appropriate course of action, and that the patient is not actively refusing m edication or trying to leave hospital. Thus som e of the factors that n eed to be considered before com m encing treatm ent include the diagnosis, the likelihood of safe com pliance, and the response to previous treatm ents, including tolerability and patient preferences. The efficacy and tolerability of the drug for a particular patient need to be considered. D oes the drug have a licence for the indication for w hich it is being used? Is the patient aware of this? Are there any age-specific considerations? W hat are the likely side-effects based on the patient's physical health? Are there likely to be any drug interactions? Has the patient been given adequate inform ation? Som e of these questions w ill n o w be explored in slightly m ore detail.

Age-related changes Older people are m ore sensitive to the side-effects of drugs and experience these m ore frequently than younger patients . 43 Pharm acodynam ic and pharm aco­ kinetic changes w ith increasing age both play a part in explaining increased drug sensitivity .44 Pharm acodynam ic changes include increased receptor sen si­ tivity and a reduction in total receptor num bers w ith increasing age. Changes in drug distribution, m etabolism and excretion are the m ain pharm acokinetic factors involved. A shift in body com position w ith a relative reduction in total body w ater com pared w ith fat alters the distribution of lipid-soluble drugs in particular. As a result, the half-life of m any lipid-soluble psychotropic drugs is prolonged. Hepatic m etabolism of psychotropic drugs is reduced because of decreased hepatic blood flow and slow ed enzym e m etabolism . The glom erular filtration rate declines by 50% b etw een young adulthood and the age of 70 years , 44 w h ich is of particular im portance in explaining the increased risk of lithium toxicity in older patients. In sum m ary, som e of the im portant pharm a­ cokinetic and dynam ic changes that occur w ith increasing age include the following: • • • • • • • • •

increased risk of side-effects and drug interactions increased receptor sensitivity decreased receptor num bers prolonged half-life of lipid-soluble drugs increased blood levels of w ater-soluble drugs decreased plasma album in levels decreased gastric acid production and em ptying reduced hepatic m etabolism decreased glom erular filtration rate . 45

A ssessing an d prescribing p sy c h o tro p ic drugs fo r o ld e r p e o p le

65

Large variations in plasm a concentrations of tricyclic antidepressants occur after similar doses, and this variation is greater in older people, particularly w ith am itriptyline, nortriptyline and im ipram ine. Reduced clearance and increased half-lives are also observed. The clearance of trazodone is also reduced in older people, and there is an increased half-life. Age-related increases in plasma concentrations have b een reported w ith citalopram, paroxetine, fluoxetine and sertraline. Lithium clearance is reduced in older people, so smaller doses are n eed ed to achieve the sam e therapeutic benefit. Benzodiazepines tend to have an increased half-life in older people, especially diazepam. There has b een very little research w ith regard to antipsychotics, but in general smaller doses are needed to achieve a therapeutic outcom e.

Side-effects o f psychotropic drugs in older people M ental illness is com m on in older people, especially the very old, and p sycho­ tropic drugs are com m on ly prescribed. Older people are m ore likely to have a range of physical illnesses, and psychotropic drugs m ay exacerbate these. These conditions include cardiovascular (e.g. arrhythmias and hypertension), endocrine (e.g. diabetes), gastrointestinal (e.g. constipation, m alnutrition), urological (e.g. prostatic hypertrophy), central nervous system (e.g. dem entia, seizure disorders) and ophthalm ic (e.g. glaucom a) conditions .46 These disorders can cause particular risks for older patients treated w ith psychotropic drugs, but an aw areness of them can lead to m ore focused prescribing. The side-effects of psychotropic drugs can be categorised according to their effects on neurotransm itter system s. These include anticholinergic (e.g. reduced salivation, sw eating and gastric acid production, increased intraocular pressure, blurred vision, urinary retention, tachycardia, im potence and delirium ), antidopam inergic (extrapyramidal side-effects, galactorrhoea, gynaecom astia and pigm entation), antihistam inic (sedation, hypotension, w eight gain) and anti-a!-adrenergic (tachycardia, arrhythmia, angina, insom nia and tremor) effects .46 K now ledge of the m echanism of action of particular p sycho­ tropic drugs w ill enable the prescriber to predict the likely side-effects in older people w ith specific physical illnesses, and thus avoid them . In addition, a know ledge of drug interactions should reduce side-effects through m ore focused prescribing.

Drug interactions Older people w ill often be taking a w ide range of psychotropic drugs as w ell as drugs for physical health problems. Hurwitz 47 reported a higher level of drug interactions in inpatients, and studies in outpatients have reported similar findings . 48 There is therefore an increased risk of drug interactions. Older people are m ore sensitive to these and the consequences can be serious, and are a com m on cause of poor efficacy w ith psychotropic drugs. Som e of the com m on and/or im portant drug interactions are sum m arised in Tables 4.1 and 4.2.

66

Practical old age p sy c h o p h arm ac o lo g y

Table 4.1 C om m on adverse drug reactions and interactions: antidepressants and m ood stabilisers (m odified from Curran and W attis45) Drug

Comments

Antidepressants

• Tricyclic antidepressants (TCAs): Increased drow siness w ith sedative drugs; en h an ced anticholinergic effects w ith anticholinergic drugs; increased levels w ith haloperidol; reduced levels w ith high-fibre diet; reduced m etabolism w ith cim etidine; increased m etabolism of im ipram ine, d oxepin and am itriptyline by carbamazepine; raised p h en y to in levels w ith im ipram ine • Citalopram: Increased risk of hyponatraem ia and serotonin syndrom e w ith buspirone • Fluoxetine: Carbamazepine levels increased; cardiac arrhythm ias w ith cisapride; increased risk of serotonin syndrom e w ith lithium ; increased p h en ytoin levels; increased risk of cardiotoxicity w ith terfenadine • Paroxetine: R educed bioavailability w ith cim etidine and p h en ytoin • Sertraline: Possible cardiac arrhythm ias w ith cisapride; raised p h en y to in levels • Mirtazepine: Increased sedation w ith benzodiazepines • Moclobemide: Increased half-life w ith cim etidine • Reboxetine: Increased risks of hypokalaem ia w ith diuretics; avoid using w ith erythrom ycin and flecainide • Venlafaxine: Increased diazepam clearance; reduced clearance w ith cim etidine. Serious adverse reaction w ith selegiline

M ood stabilisers

• L ith iu m : These are exten sive. R educed excretion caused by ACE inhibitors and NSAIDs and increased excretion caused by sodium bicarbonate, acetazolam ide, loop diuretics, thiazides and th eop h yllin e. Lithium toxicity reported w ith SSRIs. Increased risk of n eu rotoxicity w ith carbam azepine and p h en ytoin , diltiazem and verapam il and m ethyl-dopa. Increased risk of EPS w ith clozapine, haloperidol and phen oth iazin es. E nhanced effect of m uscle relaxants • Sodium valproate: E nhanced effect of aspirin. D ecreased m etabolism caused by erythrom ycin and cim etidine. Increased effect of warfarin. Increased risk of neutropenia w h en com bined w ith olanzapine • Carbamazepine: These are exten sive, and on ly a brief sum m ary is provided here. Increased level of carbam azepine w ith erythrom ycin and isoniazid. Reduced anticoagulant effect w ith warfarin. Increased plasm a concentrations of flu oxetin e and flu voxam in e. M etabolism of m ianserin and TCAs is accelerated. Avoid w ith MAOIs. Plasma concentrations of m irtazepine and paroxetine are decreased. D ecreased m etabolism w ith cim etidine. M etabolism of clozapine, haloperidol, olanzapine, quetiapine and risperidone is increased

A ssessing and prescribing p sy c h o tro p ic drugs fo r o ld e r p e o p le

67

Table 4.2 C om m on adverse drug reactions and interactions: benzodiazepines, antipsychotics and antidem entia drugs (m odified from Curran and W attis45) Drug

Comments

Benzodiazepines

• Increased drow siness w ith sedative drugs (e.g. alcohol, antihistam ines, TCAs, p h enothiazines) • A bsorption reduced by antacids and anticholinergic drugs • Clearance of diazepam reduced by propranolol • C im etidine inhibits the m etabolism of long-acting benzodiazepines • Levels of diazepam increased by flu oxetin e • P otentiation of p h en ytoin by diazepam and chlordiazepoxide • Clearance increased by sm oking (nicotine) • A ntagonism of levodopa by diazepam , nitrazepam and chlordiazepoxide

A ntidem entia drugs

• General: A ntagonism by and of anticholinergic m edication; exacerbation of su ccin ylch olin e-typ e m uscle relaxation during anaesthesia. E nhance the effect of su xam eth on iu m and antagonise the effect of non-depolarising m uscle relaxants • Galantam ine: Plasma levels increased by paroxetine; clearance reduced by paroxetine and flu oxetin e • M em antine: Risk of CNS toxicity increased w ith ketam ine and dextrom ethorphan. E nhances the effect of antim uscarinics and possibly decreases the effect of antipsychotics. A void w ith dopam inergics because of th e increased risk of CNS toxicity

Antipsychotics

• General: Increased drow siness w ith sedative drugs; en h an ced anticholinergic effects w ith anticholinergic drugs • Chlorpromazine: E nhanced h yp oten sive effect w ith hypotensives; reduced absorption w ith antacids; reduced levels w ith cim etidine; reduced m etabolism of sodium valproate • Haloperidol: A bsorption reduced by antacids; levels decreased by rifampicin • Risperidone: Clearance increased by carbam azepine • Olanzapine: Clearance increased by carbam azepine • Quetiapine: Levels reduced w ith carbamazepine; raised levels w ith erythrom ycin; slightly raised lithium levels; levels reduced by p h en ytoin

Information for patients Psychotropic drugs are pow erful agents and they can som etim es have serious consequences. Before starting treatm ent, patients should be given adequate inform ation and tim e to m ake an inform ed choice. U nfortunately, because of lim ited tim e as w ell as clinical pressures this is often a neglected part of the process. The Drug and Therapeutics Bulletin 49 has recom m ended that patients should be given the follow ing inform ation before starting treatment:

68

Practical old age p sy c h o p h arm ac o lo g y

• the nam e of the drug • the aim of the treatm ent (relief of sym ptom s, cure, prevention of relapse or prophylaxis) • h o w the patient w ill k n ow if the drug is or is not working • w h en and h o w to take the m edication • w hat to do if a dose is m issed • h o w long to take the drug • side-effects • effects on perform ance (e.g. driving ability) • interactions w ith other drugs. There are few w h o w ou ld disagree w ith these recom m endations, but unfortu­ nately this is a very tim e-consum ing process and it can som etim es be difficult to determ ine h o w m uch choice and inform ation should be given. Som e years ago a patient w ith bipolar affective disorder stormed out of SC's clinic in a very angry m ood. She later contacted SC to explain that the reason she had left w as because she had b een given a choice - 'You are the doctor, I expect you to decide'. She w as not w ell en ou gh to m ake a decision. A nother exam ple is that of a carer w h o com plained that her husband had had a heart attack from his chlorprom azine because the list of side-effects for the drug had increased his anxiety so m uch that this had precipitated the heart attack. This can be a difficult balance to achieve. Patient inform ation leaflets such as those available from the Royal College of Psychiatrists (w w w .rcpsych.ac.uk) and the Alzheim er's Society (w w w .alzh eim er.org.u k) have b een specifically w ritten for patients and carers, and they strike a balance b etw een providing sufficient inform ation to enable the patient to m ake an inform ed choice, and not scaring the patient and preventing them from benefiting from w hat can be life-saving treatm ents.

Assessing response to treatment Evaluation of the response to treatm ent is intricately linked w ith assessm ent and initiation of treatm ent. R esponse should be evaluated fairly frequently after com m encing treatm ent, but there are n o universally accepted guidelines, and practice depends in part upon clinical judgem ent (see below ). It is usual for contact to be m aintained w ith the patient for as long as the patient remains on psychoactive drugs, although again there are no guidelines. In any even t there should be clear discussion w ith the GP and the team if formal contact com es to an end. It is also im portant to determ ine the aim of treatm ent (e.g. relief of acute sym ptom s, prevention of relapse or prophylaxis).

Stopping drugs Drugs also n eed to be stopped carefully and in a planned m anner. M any drugs, if stopped quickly, especially if th ey have b een given at high doses for long period of tim e, can cause w ithdraw al sym ptom s. In general, drugs w ith longer half-lives cause few er problems. For exam ple, the w ithdraw al sym ptom s frequently associated w ith SSRI w ithdraw al include dizziness, lethargy, nausea, vivid dreams and irritability. These w ere significantly m ore com m on in patients

A ssessing an d prescribing p sy c h o tro p ic drugs fo r o ld e r p e o p le

69

stopping paroxetine (17.2% ) than in those stopping fluoxetine (1.5% ) - flu o x ­ e tin e h a s a m u c h lo n g e r h a lf - lif e .50

Yellow Card system In the UK the Yellow Card schem e was introduced in 1964 after the thalidom ide tragedy, to allow for the spontaneous reporting of suspected adverse drug reactions. Since th en m ore than 400 000 reports have been subm itted to the Com m ittee on Safety of M edicines on a voluntary basis by doctors, dentists, pharmacists, coroners and nurses, and by pharm aceutical com panies under statutory obligations. Forms are available in issues of the British National Formulary. An electronic version becam e available in 2002 and is easy to use. M ore inform ation is available at the M edicines and Healthcare Products Regulatory A gency w ebsite (h ttp://m edicines.m hra.govu k ).

Measuring outcome The em phasis on accountability and, in the UK, the em phasis on clinical governance adds a n e w im perative to the n eed to m easure ou tcom es . 51 Ideally, outcom es should be assessed from a variety of perspectives (e.g. those of the patient, carer, provider and com m issioner), and m easures should be found w hich w ill reflect a broad consensus on 'good' and 'bad' outcom es b etw een different view points. A distinction should also be m ade betw een overall m easures of outcom e such as the HoNOS 6 5 + , individualised m easures focused on each patient, and specific m easures designed for use in particular diagnostic groups. The HoNOS 6 5 + belongs to the HoNOS fam ily (Health of the Nation Outcom e Scales), the usefulness of w h ich as tools in routine practice has been questioned . 52 In one small study of outcom es on an acute psychiatric ward for old people, the HAD com pleted by the patient em erged as a good overall m easure for the outcom e of depression, and correlated w ell w ith a variety of 'view points ' . 53 M easures such as this, w hich can 'summarise' outcom es in specific disorders and settings, ideally need to be com bined w ith the less sensitive but m ore global measures. Building such m easures into care pathways w ould enable routine audit of perform ance. This could produce feedback to the clinical team and enable com parisons of perform ance both b etw een team s and w ithin team s over time. As w ell as increasing accountability, this kind of m easure should encourage reflective practice and im prove clinical quality. The snag is that researching and applying such m easures requires tim e and m oney. The tim e needed can be reduced by increasing efficiency through good inform ation technology, but this also requires investm ent.

K ey p o in ts • Psychotropic drugs have an important role to play in the treatm ent of m ental disorders in older people. • A detailed assessm ent before starting treatm ent is essential in order to achieve the best therapeutic outcom e w ith m inim al side-effects and good com pliance.

70

Practical old age p sy c h o p h arm ac o lo g y

• Patients should be given adequate inform ation and tim e to m ake an inform ed choice before starting treatm ent. • Once initiated, treatm ents should be properly m onitored for clinical benefit and tolerability and, w h en appropriate, cautiously discontinued. • M ore research is n eeded that specifically focuses on the use of psych o­ tropic drugs in older people.

References 1 Drug and Therapeutics B ulletin (2004) Do SSRIs cause gastrointestinal bleeding? Drug Ther Bull. 42: 17 -1 8 . 2 Pattie A and Gilleard C (1979) M anual o f the Clifton Assessment Procedure fo r the Elderly (CAPE). Hodder and Stoughton Educational, Sevenoaks. 3 Cole MG (1989) Inter-rater reliability of the Crichton Geriatric B ehaviour Rating Scale. Age Ageing. 18: 5 7 -6 0 . 4 Rabins PV (1994) The validity of a caregiver-rated brief B ehaviour Sym ptom Rating Scale (BSRS) for use in the cognitively im paired. In t J Geriatr Psychiatry. 9: 2 0 5 -1 0 . 5 Burns A, Lawlor B and Craig S (1999) Assessment Scales in Old Age Psychiatry. Martin Dunitz, London. 6 Yesavage J, Brink T, Rose T et al. (1983) D evelop m en t and validation of a geriatric depression screening scale: a prelim inary report. J Psychiatr Res. 17, 3 7 -4 9 . 7 Yesavage J (1986) Geriatric D epression Scale (GDS): recent evid en ce and d evelop m en t of a shorter version. Clin Gerontol. 9: 1 6 5 -7 3 . 8 Adshead F, Day CD and Pitt B (1992) BASDEC: a n ovel screening instrum ent for depression in elderly m edical inpatients. BMJ. 305: 397. 9 A llen N, A m es D, A shby D, B en n etts K, Tuckwell V and W est C (1994) A brief sensitive screening instrum ent for depression in late life. Age Ageing. 23: 2 1 3 -1 8 . 10 Zigmond A and Snaith P (1983) The Hospital A n xiety and D epression (HAD) Scale. Acta Psychiatr Scand. 67: 3 6 1 -7 0 . 11 Lindesay J (1991) Phobic disorders in the elderly. Br J Psychiatry. 159: 5 3 1 -4 1 . 12 Sims AC (1988) Symptoms in the Mind: an introduction to descriptive psychopathology (2e). WB Saunders, London. 13 Forstl H, Burns A, Levy R and Cairns N (1994) N europathological correlates of psychotic p h en om en a in confirm ed A lzheim er's disease. Br J Psychiatry. 165: 5 3 -9 . 14 H odkinson HM (1972) Evaluation of a m ental test score for assessm ent of m ental im pairm ent in the elderly. Age Ageing. 1, 2 3 3 -8 . 15 Blessed G, Tom linson BE and Roth M (1968) The association b etw een quantitative m easures of dem entia and senile change in the grey m atter of elderly people. Br J Psychiatry. 144: 7 9 7 -8 1 1 . 16 Kay DW, Black SE, B lessed G, Jachuck SJ and Sahgal A (1972) The prevalence of dem entia in a general practice sample: upward revision of reported rate after follow -u p and reassessm ent. Int J Geriatr Psychiatry. 5: 1 7 9 -8 6 . 17 Holm es J and Gilbody S (1996) Differences in use of abbreviated m ental test score by geriatricians and psychiatrists. BMJ. 313: 465. 18 Folstein MF, Folstein SE and M cHugh PR (1975) 'M ini-M ental State'. A practical m ethod for grading the cognitive state of patients for the clinician. J Psychiatr Res. 12: 189-98. 19 G reifenhagen A, Kurz A, W isem an M, Haupt M and Zimmer R (1994) C ognitive assessm ent in A lzheim er's disease: w hat does the CAMCOG assess? Int J Geriatr Psychiatry. 9: 7 4 3 -5 0 .

A ssessing an d prescribing p sy c h o tro p ic drugs fo r o ld e r p e o p le

7 1

20 Curran S and W attis J (1997) M easuring th e effects of antidem entia drugs in patients w ith A lzheim er's disease. H um Psychopharmacol. 12: 3 4 7 -5 9 . 21 C hristensen H and Jorm AJ (1992) The effect of premorbid intelligen ce on the M iniM ental State and IQCODE. In t J Geriatr Psychiatry. 7: 1 5 9 -6 0 . 22 Shulm an KI, Gold DP, C ohen CA and Zucchero CA (1993) Clock-drawing and dem entia in the com m unity: a longitudinal study. Int J Geriatr Psychiatry. 8: 4 8 7 -9 6 . 23 Ronnberg L and Ericsson K (1994) Reliability and validity of the hierarchic dem entia scale. Int Psychogeriatr. 6: 8 7 -9 4 . 24 Lishman W A (1987) Organic Psychiatry: the psychological consequences o f cerebral disorder. B lackw ell Scientific Publications, Oxford. 25 W attis JP and Curran S (2003) Physical com plaints and psychiatric disorders. Geriatr Med. 23: 3 3 -9 . 26 Philp I and Ebrahim S (1994) Assessing Elderly Patients. Farrand Press, N ottingham . 27 O'Brien JT (1995) Is hippocam pal atrophy on m agnetic resonance im aging a marker for A lzheim er's disease? Int J Geriatr Psychiatry. 10: 4 3 1 -5 . 28 Jobst KA, B arnetson LP and Shepstone BJ (1998) Accurate prediction of histologically confirm ed A lzheim er's disease and the differential diagnosis of dem entia: the use of NINCDS-ADRDA and DSM-III-R criteria, SPECT, X-ray CT and Apo E4 in m edial tem poral lobe dem entias. Int Psychogeriatr. 10: 2 7 1 -3 0 2 . 29 Royal College of Psychiatrists (1995) Consensus Statement on the Assessment and Investi­ gation o f an Elderly Person w ith Suspected Cognitive Im pairm ent by a Specialist Old Age Psychiatry Service. Royal College of Psychiatrists, London. 30 Taylor D, Paton C and Kerwin R (2003) The M auds ley Prescribing Guidelines (7e). Martin Dunitz, London. 31 Reilly JG, Thomas HL and Ferrier IN (2002) R ecent studies on ECG changes, antipsychotic use and sudden death in psychiatric patients. Psychiatr Bull. 26: 1 1 0 -1 2 . 32 Inglis F (2002) The tolerability and safety of cholinesterase inhibitors in th e treatm ent of dem entia. Int J Clin Pract. 1 2 7 (S u p p l. 1): 4 5 -6 3 . 33 M organroth J, Graham S, Hartman R and A nand R (2002) Electrocardiographic effects of rivastigm ine. J Clin Pharmacol 42: 5 5 8 -6 8 . 34 Arnold SE and Kumar A (1993) Reversible dem entias. Med Clin North A m . 77: 2 1 5 -3 0 . 35 Curran S and W attis JP (2001) Practical Psychiatry o f Old Age (3e). Radcliffe M edical Press, Oxford. 36 D epartm ent of Health (1990) Caring for People: the Care Programme Approach for people w ith a m ental illness referred to the specialist psychiatric services. Joint Health/Social Services Circular. Health and Social Services Development. D epartm ent of Health Publications Unit, London. 37 Burns T (1997) Case m anagem ent, care m an agem en t and care program m ing. Br J Psychiatry. 170: 3 9 3 -5 . 38 Lader M and Herrington R (1990) Biological Treatments in Psychiatry. Oxford M edical Publications, Oxford. 39 Thomas KB (1978) The consultation and the therapeutic illusion. BMJ. 1: 1 3 2 7 -8 . 40 Young MA, Keller MB, Lavori PW et al. (1987) Lack of stability of the RDC en d ogen ou s subtype in con secu tive episodes of major depression. J Affect Disord. 12: 1 3 9 -4 3 . 41 W illiam s RT (1978) N utrients in drug detoxification reactions. In: JN Hathcock and J Coon (eds) N utrition and Drug Interactions. A cadem ic Press, N ew York. 42 Livingston G, Hollins S, Katona C et al. (1998) Treatm ent of patients w h o lack capacity. Psychiatr B u ll 22: 4 0 2 -4 . 43 Ray WA, Fought RL and Decker MD (1992) Psychoactive drugs and risk of injurious m otor v eh icle crashes in elderly drivers. A m J Epidemiol. 136: 8 7 3 -8 3 . 4 4 Lader M (1994) N europharm acology and pharm acokinetics of psychotropic drugs in old age. In: J Copeland, M A bou-Saleh and D Blazer (eds) Principles and Practice o f Geriatric Psychiatry. John W iley and Sons, Chichester.

72

Practical old age p sy c h o p h arm ac o lo g y

45 Curran S and W attis JP (2004) Psychopharm acology in th e elderly. In: DJ King (ed.) Seminars in Clinical Psychopharmacology (2e). Gaskell, London. 46 Sm ith D (1998) Side-effects of psychotropic drugs. In: D W heatley and D Sm ith (eds) Psychopharmacology o f Cognitive and Psychiatric Disorders in the Elderly. C hapm an and Hall M edical, London. 47 Hurwitz N (1969) Predisposing factors in adverse reactions to drugs. BMJ. 1: 5 3 6 -4 0 . 48 Learoyd BM (1972) Psychotropic drugs and th e elderly patient. M ed J Aust. 1: 1 1 3 1 -3 . 49 Drug and Therapeutics B ulletin (1981) W hat should w e tell patients about their m edicines? Drug Ther Bull. 19: 7 3 -4 . 50 C oupland NJ, Bell CJ and Potokar JP (1996) Serotonin reuptake inhibitor w ithdraw al. J Clin Psychopharmacol. 16: 3 5 6 -6 2 . 51 C harlw ood P, M ason A, Goldacre M, Cleary R and Eilkinson E (1999) Health Outcome Indicators: severe m ental illness. Report o f a working group to the Department o f Health. N ational Centre for Health O utcom es D evelop m en t, Oxford. 52 Stein GS (1999) U sefulness of th e Health of the N ation O utcom e Scales. Br J Psychiatry. 174: 3 7 5 -7 . 53 Wattis JP, Butler A, M artin C and Sum ner T (1994) O utcom e of adm ission to an acute psychiatric facility for older people: a pluralistic evaluation. Int J Geriatr Psychiatry. 9: 8 3 5 -4 0 .

Chapter 5

Old age psychopharmacology: the pharmacist’s perspective Paul Hardy and Mary Crabb

Introduction This chapter w ill discuss the role of the pharmacist in the care of the older m entally ill person. In the first part it w ill describe specific m edicines w hich m ay require special care due to the physiological changes that occur w ith increasing age, and it w ill then go on to discuss m edicines m anagem ent using the exam ple of a patient journey.

Pharmacokinetic changes in old age As people get older their use of m edicines tends to increase. Four in five people aged over 75 years take at least one prescribed m edicine, w ith 36% taking four or m ore m edicines . 1 These changes, together w ith polypharm acy (taking m ultiple m edicines for m ultiple m edical conditions), m ean that the elderly are m ore prone to adverse drug reactions (ADRs). Adverse drug interactions are an important cause of morbidity, and ev en mortality, in older people. There is a threefold greater incidence of drug-induced morbidity in patients over the age of 60 years com pared w ith th ose under 30 years. ADRs have b een reported to be solely or partly the cause of 10-16% of adm issions of elderly people to acute geriatric w ards . 2

Pharmacokinetic changes Pharm acokinetics can be defined as 'how the body handles a drug', w hereas pharm acodynam ics is described as 'w hat a drug does to a patient'. Pharm acoki­ netic parameters of drug handling (absorption, distribution, m etabolism and elim ination) m ay be significantly altered in the elderly (see Table 5.1).

73

74

Practical old age p sy c h o p h arm ac o lo g y

Table 5.1 Pharm acokinetic changes w ith increasing age Absorption

Increased gastric pH Decreased intestinal blood flow Decreased surface area for absorption D ecreased gastrointestinal m otility Decreased gastric em ptying Decreased gastric acid secretion

Distribution

Increased total fat con ten t D ecreased total body w ater Decreased lean body mass D ecreased serum album in

M etabolism

D ecreased liver mass Decreased liver blood flow Decreased liver en zym e activity

Elim ination

Decreased glom erular filtration D ecreased renal tubular filtration

Absorption Difficulties w ith sw allow ing are relatively com m on am ong elderly patients w ith psychiatric disorders, especially those that are associated w ith neurodegenerative or cerebrovascular conditions. Factors such as an increase in gastric pH and a reduction in absorptive surface area and m esenteric blood flow m ay contribute to a reduced rate of drug absorption. This slow ing can be com pounded by drugs that cause anticholinergic effects, the adm inistration of antacids and the ingestion of food . 3

Distribution The distribution of drugs m ay be significantly altered because of age-related changes in body fat, total body water, lean body mass and plasma album in levels. Total body fat increases b etw een the ages of 18 and 35 years from 18% to 36% in m ales and from 33% to 48% in fem ales .4 This increase in body fat results in an increase in the volu m e of distribution of lipid-soluble drugs such as diazepam, and can result in a prolonged elim ination half-life leading to accum ulation. A decrease in lean body mass reduces the volum e of distribution of w ater-soluble drugs such as digoxin and lithium that bind to m uscle in older adults. Therefore smaller doses m ay be required in elderly people. Factors w hich m ay affect binding and that have special applicability in older people include protein concentration, disease states, co-adm inistration of other drugs, and nutritional status . 4 Follow ing absorption into the bloodstream , m ost m edication binds to plasma proteins, album in and cel-acid glycoprotein. Acidic drugs tend to bind to plasma album in, w hereas basic drugs bind to a 1-acid glycoprotein. Only an unbound drug is able to cross the blood-brain barrier and interact w ith the intended target in the brain. In the elderly, plasma album in levels fall resulting in higher levels of protein-bound drugs (e.g. diazepam, phen ytoin and warfarin), w hich m ay have an increased free concentration and therefore an enhanced effect. Plasma a; 1-acid glycoprotein levels m ay rem ain unchanged or m ay rise slightly w ith ageing.

O ld age p sychopharm acology: th e p h arm acist’s p e rsp e c tiv e

75

Total body water m ay decrease by 15%, reducing the volum e of distribution of w ater-soluble drugs such as lithium and leading to higher plasma concentrations. C oncom itant adm inistration of diuretics m ay exacerbate this problem.

Metabolism Hepatic m etabolism is decreased in the elderly due to a loss of liver mass and a decrease in liver blood flow . Other disease states such as heart failure w ill further decrease liver blood flow . First-pass m etabolism of drugs w ith a higher extraction ratio, such as barbitu­ rates and propranolol, are blood flow dependent and are impaired, resulting in increased drug levels. Liver m etabolism of drugs by phase 1 reactions (e.g. oxidation) decreases w ith age, and this leads to age-related reductions in clearance for drugs such as diazepam, chlordiazepoxide and nortriptyline. Phase 2 reactions (e.g. conjugation) sh o w few or n o age-related changes . 2 Note that in the elderly, norm al liver function tests (LFTs) do not im ply norm al m etabolism of drugs. The cytochrom e P450 group of isoenzym es is responsible for the m etabo­ lism of m ost com m only used drugs. The CYP1, CYP2 and CYP3 fam ilies are responsible for the majority of drug m etabolism in h u m an s . 5 Antidepressants such as selective serotonin reuptake inhibitors, m irtazepine and venlafaxine are all primarily elim inated by oxidative m etabolism via the cytochrom e P450 e n zy m es . 6 A ntipsychotic drugs are also m etabolised by this pathw ay, and caution is advised as these drugs m ay have reduced clearance in the elderly.

Elimination This is the m ost im portant age-related change, and disease states such as diabetes and heart failure can w orsen renal function, as can an acute illness w h ich leads to dehydration, such as a chest in fection . 7 Renal function deteriorates w ith increas­ ing age. The glom erular filtration rate decreases by 1 % per year after the age of 40 years. A decrease in renal excretion of cleared m edicines m ay lead to an increase in serum concentration, resulting in accum ulation and toxicity. Drugs that are rem oved by renal excretion include allopurinol, atenolol, digoxin and spirono­ lactone. Lithium is dependent on renal function for elim ination. Serum levels of lithium above the therapeutic range m ay result in nephrogenic diabetes insipidus. In older adults, toxicity m ay occur at levels that w ou ld be considered therapeutic in younger adults. Signs of toxicity include increased tremor, confusion and ataxia. Creatinine clearance can be estim ated from serum creatinine levels using the Cockcroft-Gault eq u ation :8 . . F x (140 — age) x body w eight (kg) Creatinine clearance = ---------------------- — ------------------& — serum creatinine concentration (/im o l/ 1)

w here F = 1 .0 3 for fem ales and 1.23 for m ales. In obese patients, ideal body w eigh t should be used, as total body w eight w ill overestim ate renal function. Renal im pairm ent is often m issed in the elderly, especially as a decrease in m uscle mass m ay prevent an increase in plasma creatinine levels. One w ay to grade renal im pairm ent is sum m arised in Table 5.2.

76

Practical old age psychopham nacology

Table 5.2 Glomerular filtration rate and degree of renal im pairm ent Degree o f renal im pairm ent

Glomerular filtration rate (m l/m in)

Mild

2 0 -5 0

M oderate

1 0 -2 0

Severe

< 10

Pharmacodynamic changes Pharm acodynam ic changes m ay be defined as 'w hat a drug does to the patient'. The m ost im portant changes seen in older people are sum m arised in Table 5.3. Table 5.3 Pharm acodynam ic changes observed in older people Increased sensitivity to som e drugs D ecreased drug receptor population D ecreased cholinergic transm ission D ecreased therm oregulatory m echanism s Decreased orthostatic circulatory responses Increased postural instability

Pharm acodynam ic changes w ith age have not b een studied to the sam e extent as pharm acokinetic changes, but it is evident that som e system s in the body becom e increasingly sensitive to drugs that act on them . There is an increased sensitivity to central nervous system drugs, such as benzodiazepines. These m ay cause sedation, confusion, disorientation, m em ory im pairm ent and delirium. Reduced cholinergic transm ission in the brain w ith increasing age m ay explain w h y the elderly are m ore prone to drug-induced cognitive disorder. Drugs w hich m ay cause cognitive im pairm ent include anticholinergics (benzhexol, orphenadrine and procyclidine), benzodiazepines, opioids and tricyclic antidepressants. The selective serotonin reuptake inhibitors (SSRIs) and reversible inhibitors of m on oam ine oxidase A have not b een sh ow n to have negative effects on cognition. The elderly are m ore susceptible to anticholinergic side-effects, and m any m edication classes, including tricyclic antidepressants, antipsychotics, antihista­ m ines and anti-arrhythm ics, have anticholinergic properties . 9 A nticholinergic side-effects include dry m outh, blurred vision, constipation, confusion and urinary retention. Therm oregulatory m echanism s are often impaired in the elderly, and h yp o­ therm ia m ay occur. C om m only im plicated drugs include phenothiazines, tricyclic antidepressants, benzodiazepines, opioids and alcohol. Older people are m ore prone to drug-induced h ypotension because of an impaired reflex tachycardia

O ld age p sychopharm acology: th e p h arm acist’s p e rsp e c tiv e

77

response. M edicines responsible for orthostatic effects include tricyclic antide­ pressants, phenothiazines and levodopa. It is appropriate to assum e that there is an impaired baroreflex function in elderly p eople , 10 and to m onitor these m edicines carefully. Other effects of psychotropic m edicines in the older person include agitation, m ood and perceptual disturbances, headache, sweating, sexual dysfunction, gastrointestinal disturbances (nausea, anorexia, changes in w eight or bow el habits) and hyponatraem ia . 3

Medication and falls in the older person About 30% of older people over 65 years of age fall at least once a year, often resulting in fractures . 11 The aetiology of falls is com plex, and m edicines are seldom the sole cause, but th ey do m ake a major contribution to the problem. M edication m ay cause falls by various m echanism s, including the following: • • • • • • • •

increased sedation reduced balance reduced reaction tim e postural hypotension drug-induced parkinsonism confusion hypoglycaem ia postural instability - this increases w ith age and drugs that increase postural sw ay m ay precipitate falls.

Tinetti 12 has sh ow n the strongest link to an increased risk of falling to be w ith selective serotonin reuptake inhibitors, tricyclic antidepressants, neuroleptic agents, benzodiazepines, anticonvulsants and class IA anti-arrhythm ic m edica­ tions. O ne-third of prescriptions given to older patients adm itted to hospital could be stopped w ith ou t detrim ent to the patient, either because the drugs are unnecessary or because they are absolutely contraindicated. Despite this, m any drugs are still prescribed unnecessarily . 13

Pharmaceutical care of older people with mental illness Pharm aceutical care has b een defined as 'the responsible provision of drug therapy for the purpose of achieving definite outcom es w hich im prove the patient's quality of life . ' 14 Philosophically, it m ay be seen as the latest stage in a process of m oving the focus of pharmacists' activity from a supply-oriented role to a patient-oriented one. Typically a series of pharm aceutical care needs w ill be identified for a patient (see Table 5.4) to guide the pharm acist in devising that patient's care plan.

78

Practical old age p sy c h o p h arm ac o lo g y

Table 5.4 Pharm aceutical care planning: exam ples of som e of the key issues (adapted w ith perm ission from Shaw et al.15) A dherence problem s A dm inistration problem s Side-effects/adverse drug reactions M edication review Therapeutic drug m onitoring (TDM) Special m onitoring requirem ents Drug interactions Polypharm acy Other m edications (non-prescribed rem edies, illicit drug use)

In the area of old age psychiatry there are relatively few exam ples of pharm a­ ceutical care initiatives directed towards the elderly m entally ill w hich have been sh ow n to have positive outcom es for patients. Exam ples of program mes designed for the elderly 25,28-9 or for the m entally ill, w hich can give pointers to the types of service w h ich m ight be developed in the future, are included in this book. To dem onstrate the range of activity in this area, w e shall follow a hypothetical patient journey through secondary and primary care and highlight som e of the program m es w h ich have been developed at each stage on the journey. Note that the programmes discussed are largely pilot schem es and not available in all areas.

Hospital admission W hen a patient is adm itted to hospital, it can often be their drug therapy w hich has precipitated the adm ission. It has b een estim ated that 10-20% of hospital adm issions are caused by adverse drug reactions. M annesse et al.16 exam ined adm issions of patients over 70 years of age to general m edical wards. A severe adverse drug reaction w as present in 24 of cases, and 12% of adm issions w ere classified as probably caused by an adverse drug reaction. Badcott et a l .} 7 in a 10-day audit of adm issions via Accident and Emergency, identified 90 patients (45% ) in w h o m drug-related problem s w ere a likely causative factor w ith regard to their adm ission. F ollow ing adm ission to hospital, the patient is prescribed m edication for inpatient use w h ich is usually intended to replicate the drug regime taken by the patient im m ediately prior to adm ission. For a variety of reasons this m ay not alw ays be accurate. D rew ett 18 studied adm issions to a general hospital in order to quantify these discrepancies on their m edication charts. In total, 34% of patients w ere found to have errors. As w ell as m is-transcriptions of dose and frequency, patients had m edicines om itted in m any instances. Pharm acist-conducted drug histories on adm ission have b een used to avoid som e of these problems. The pharmacist's greater familiarity w ith the range of

O ld age p sychopharm acology: th e p h arm acist’s p e rsp e c tiv e

79

products available com m ercially can help to im prove the accuracy of the m edication history. H owever, the scope for pharm acy intervention in this area is lim ited. It is not clear w h eth er the inform ation gleaned is m ore helpful than careful checks w ith practice records or w ith m edication records in com m unity pharm acies. Also, patients adm itted to psychiatric settings m ay be unable to provide the pharmacist w ith an adequate m edication history for various reasons, and the introduction of another health professional to the patient at this early stage of treatm ent can be counter-productive. M any pharm acies n o w im plem ent strategies of reusing patients' ow n drugs, or the 'one-stop' dispensing system , as advocated in the National Service Framework for Older P eople . 19 Reuse of patients' ow n drugs has b een associated w ith a reduction in m edication errors .20 O ne-stop dispensing reduces the risk of tran­ scription errors w h e n inpatient m edication is rew ritten for discharge, and provides the opportunity for patients to retain responsibility for their m edication before discharge. H ow ever, this approach m ay not suit patients in m ental health environm ents. Generally, inpatient stay for psychiatric conditions tends to be of a longer duration, w h ich requires the resupply of m edication that onestop dispensing is designed to avoid. In addition, small supplies of m edication are still required for short periods of h om e leave.

The inpatient stay Pharmacist in volvem en t w ith prescription m onitoring at ward level, and attend­ ance at m ulti-disciplinary ward rounds, is n o w an established practice in m any UK hospitals. In addition, m any pharmacists have developed other services to help to m eet the needs of their patients. Longitudinal drug histories are one exam ple. Paton 21 review ed the drug history request service provided at their hospital (an acute adult m ental health unit). A longitudinal history for a long-stay patient could take an experienced pharmacist up to a full day to prepare, and on the basis of its findings w ou ld conclude by providing som e suggestions for future pharm acological m anagem ent to be considered for the patient. A questionnaire w as circulated am ong users (m edical and nursing staff) to assess w h eth er this service offered good value. R espondents indicated an overall high level of satisfaction w ith the service. Of ten patients studied for 6 m onths follow ing the drug history, nine had changes to their treatm ent plan as a result of the recom m endations m ade, and the majority of these led to im provem ents in the patient's m ental state. The use of com puterised pharm acy records has assisted this process at Ram pton Hospital . 22 The TPS' database presents a patient 'tim eline', w ith m edication record, blood histories and other patient data and events co-ordinated on the tim eline. The relationship b etw een onset of drug therapy and progression of disease has becom e easier to observe, and has added a layer of objective observational data to the assessm ent of w hether or not psychiatric drugs are working. Increasingly as evidence-based and protocol-driven practice takes hold in patient m anagem ent, m ore objective criteria for directing pharmacist activity w ill be needed. Beers et a l 23 have attem pted to devise a list of m edication deem ed inappropriate for use in the elderly. Using an expert panel, a review of the literature resulted in a list of 33 m edications or categories of m edications co n ­

80

Practical old age p sy c h o p h arm ac o lo g y

sidered less suitable or unsuitable for use in elderly patients. The list was later revised 24 to keep abreast of developm ents in the field of pharmacotherapy, and is n o w an im portant assessm ent tool in studies of prescribing practice. Similarly, Osborne et al.25 attem pted to devise an indicator for appropriate neuroleptic prescribing in nursing hom es in the UK. D eveloping an indicator based on US consensus criteria, the authors exam ined neuroleptic prescribing in 934 residents in 22 nursing hom es in the South Thames region. In total, 45% of prescriptions for antipsychotics in this cohort w ere initiated prior to adm ission to the nursing hom e. Only 17.8% of prescriptions fulfilled the criteria for appropriate prescribing.

Hospital discharge: the secondary/primary care interface The risk of readm ission during the first 6 m onths after discharge is very high for psychiatric patients .26 Therefore, as our patient m oves towards hospital discharge, continuity of care across the interface becom es a major concern. A num ber of studies have looked at im proving com m unication b etw een primary and secondary care pharmacists by m eans of pharmacy discharge inform ation. In a study of general m edical patients , 27 discharge prescription details were passed on to com m unity pharmacists in an active group, and not in a com parison group. The num ber of unintentional discrepancies in subsequent com m unitybased supply of m edication was low er for the intervention group than for the com parison group, and the num ber of discrepancies judged to have a definite adverse effect was also low er. The authors calculated that giving such inform a­ tion to 19 patients w ould prevent one patient from being exposed to one unintentional discrepancy w ith a definite adverse effect. In another study on pharm acy discharge planning in adult patients from a large psychiatric hospital , 15 97 patients w ere random ly assigned to either an inter­ vention group or a control group. Intervention took the form of pharm aceutical needs assessm ent, provision of m edication inform ation and sending discharge planning inform ation to the patient's com m unity pharmacy. Patient know ledge did not differ b etw een the tw o groups. M edication problems post-discharge (supply, adm inistration and storage/handling issues, and therapeutic problems) w ere greater for the control group than for the intervention group. There was a trend in the control group for low er readm ission rates in the 3 m onths follow ing discharge, but this did not reach the level of clinical significance. However, the overall benefits of discharge pharm aceutical care planning m ay be limited. Nazareth et al.28 studied the effectiveness of a pharmacy discharge plan in 362 older m edical patients. The primary outcom e m easure of readmission rates at 3 and 6 m onths did not differ b etw een the intervention and control groups. Secondary outcom es of the num ber of deaths, hospital outpatient or general practice appointm ents, and the num ber of days in hospital during the follow -up period also did not differ. This finding contrasts w ith the positive assessm ent of discharge planning seen in the other studies. It is theoretically possible that m ental health patients w ould gain greater benefit from this type of intervention than m edical patients, precisely because of the high risk for readm ission during this period, but this cannot be confirm ed on the basis of the present data.

O ld age p sy c hopharm acology: th e p h arm acist’s p e rsp e c tiv e

8 1

Community-based interventions As our patient becom es reintegrated in their dom estic environm ent, ongoing pharm aceutical care can con tin ue to offer health benefits. Specialisation in m ental health pharm acy has b een alm ost exclusively a secondary care p h en o m ­ en o n in the past. C om m unity pharmacists w h o have supported patients w ith m ental health problem s in the com m unity have had no formal structures to support or recognise their contribution. H ow ever, the study program mes available from the Centre for Postgraduate Pharmacy Education and the United Kingdom Psychiatric Pharmacy Group are beginning to im prove this situation. Bernsten et al.29 studied com m unity pharm acy-based care for elderly patients w ith a variety of m edical and/or psychiatric problem s in seven European countries. A harm onised, structured pharm aceutical care programme was pro­ vided by com m unity pharmacists w h o received training. A general decline in health for the w h o le study cohort w as seen over the course of the 18-m onth study. H ow ever, significant im provem ents w ere seen in the intervention group com pared w ith the control group. Patients in the intervention group reported better control of their m edical conditions as a result of the survey. The authors noted that the chief beneficial outcom es of the study seem ed to be in the hum anistic aspects, such as satisfaction w ith treatm ent and sign and sym ptom control, and (perhaps surprisingly) less im pact was seen on drug therapy, drug know ledge and adherence. One of the m ost w ell-regarded com m unity-based schem es of recent years has b een the w ork conducted in South Derbyshire by Harris. 30 In this project, 24 com m unity pharmacists w ere recruited to work alongside other health profes­ sionals in providing advice, support and inform ation, visiting patients in their o w n hom es, and rationalising treatm ent regim es and supporting patient adher­ ence. The pharmacist provides a pharm aceutical care plan to the patient or carer. This schem e has proved popular w ith other health workers and w ith carers. Follow ing the success of this project, Harris w as com m issioned by the M ental Health Foundation to investigate the level of provision of com m unity-based pharm aceutical care for the elderly m entally ill throughout the UK. This work has recently b een updated . 31 R espondents reported that the m ain perceived m edication difficulties for this patient group included obtaining inform ation about m edicines, understanding the instructions for taking m edicines, rem oving m edicines from the container, and taking m edicines as prescribed. Only a third of respondents considered that routine support for elderly m entally ill patients was available. Provision of support for com m unity team s or hom e-based services was inconsistent, and the level of continuity of care b etw een hospital and the com m unity w as low . Funding problem s w ere com m only seen as a barrier to provision of services.

82

Practical old age psychopharmacology

Key points • This chapter has discussed the im pact on patient care made by pharm a­ cists at all stages of the patient journey. The m odel schem es that have been highlighted dem onstrate the untapped potential of pharmacists to make significant contributions to m edicines m anagem ent. • In tim e it is hoped that provision of these value-added services w ill be able to make im provem ents to the lot of greater num bers of elderly patients w ith m ental health problems.

References 1 Erens B and Primatesta P (eds) (1999) The Health Survey fo r England in 1998: cardiovascular disease. The Stationery Office, London. 2 D enham MJ and Barnett NL (1998) Drug therapy and the older person - role of the pharmacist. Drug Safety. 4: 2 4 3 -5 0 . 3 Zubenko GS and Sunderland T (2000) Geriatric psychopharm acology: w h y does age matter? Harvard Rev Psychiatry. 7: 3 1 1 -3 3 . 4 Young LY and Koda-Kim ble MA (eds) (1995) Applied Therapeutics: the clinical use o f drugs (6e). Lippincott W illiams & W ilkins, Philadelphia, PA. 5 Adams P (1998) Drug interactions that matter: m ech an ism and m anagem ent. Pharm J. 261: 6 1 8 -2 1 . 6 D evane CL and Pollock BG (1999) Pharm acokinetic considerations of antidepressant use in the elderly. J Clin Psychiatry. 60 (Suppl. 20): 3 8 -4 4 . 7 Phillips PA, Johnston Cl and Gray L (1993) Disturbed fluid and electrolyte h om oeostasis follow in g dehydration in elderly people. Age Aging. 22: S 2 6 -3 3 . 8 Kuczynska J and Evans H (2000) Prescribing drugs for patients w ith renal failure. Prescriber. 11(5): 1 1 5 -2 3 . 9 Gray SL, Lai KV and Larson EB (1999) D rug-induced cognition disorders in the elderly: incidence, p reven tion and m anagem ent. Drug Safety. 2: 1 0 1 -2 2 . 10 M ets TF (1995) D rug-induced orthostatic h y p oten sion in older patients. Drugs Aging. 6: 2 1 9 -2 8 . 11 A non (2000) M anaging falls in older people. Drug Ther Bull. 38: 6 8 -7 2 . 12 Tinetti M (2003) Preventing falls in elderly persons. NEJM. 348: 4 2 -9 . 13 Royal C ollege of Physicians (1997) Medication fo r Older People (2e). Royal College of Physicians, London. 14 Hepler CD and Strand LM (1990) Opportunities and responsibilities in pharm aceutical care. A m J Hosp Pharm. 47: 5 3 3 -4 3 . 15 Shaw H, M ackie CA and Sharkie I (2000) Evaluation of effect of pharm acy discharge planning on m edication problem s experienced by discharged acute adm ission m ental health patients. Int J Pharm Pract. 8: 1 4 4 -5 3 . 16 M annesse CK, D erxx FHM, De Ridder MAJ et al. (2000) C ontribution of adverse drug reactions to hospital adm ission of older patients. Age Aging. 29: 3 5 -9 . 17 Badcott S, Fernandes R, Hargreaves D and Freij R (1999) Target drug-related adm issions. Pharm Pract. 9: 3 5 8 -6 1 . 18 D rew ett NM (1998) Stop regular m edicine errors. Pharm Pract. 8: 1 9 3 -6 . 19 D epartm ent of Health (2001) N ational Service Framework fo r Older People. The Stationery Office, London. 20 N icholls M (2000) M edicines: peace in a POD. Health Serv J. I l l : 35. 21 Paton C (2000) Are w e clinically effective? Pharm Pract. 8: 1 9 8 -2 0 0 .

O ld age p sychopharm acology: th e p h arm acist's p e rsp e c tiv e

83

22 A non. (2002) In n ovative database gives pharmacists passport into clinical team . Med Manag. 1: 7 -8 . 23 Beers M, Ouslander JG, R ollingher I et a l (1991) Explicit criteria for determ ining inappropriate m edication use in nursing h o m e residents. Arch Intern Med. 151: 1 8 2 5 -3 2 . 24 Beers M (1997) Explicit criteria for determ ining inappropriate m edication use by the elderly: an update. Arch Intern Med. 157: 1 5 3 1 -6 . 25 Osborne CA, Hooper R, Li KC et a l (2002) A n indicator of appropriate neuroleptic prescribing in nursing hom es. Age Ageing. 31: 4 3 5 -9 . 26 W alker SA and Eagles JM (2002) Discharging psychiatric patients from hospital. Psychiatr Bull. 26: 2 4 1 -2 . 27 D uggan C, Feldm an R, H ough J and Bates I (1998) R educing adverse prescribing discrepancies follow in g hospital discharge. Int J Pharm Pract. 6: 7 7 -8 2 . 28 Nazareth I, B urton A, Shulm an S, Sm ith P, Haines A and Timberall H (2001) A pharm acy discharge plan for hospitalised elderly patients - a random ised controlled trial. Age Aging. 30: 3 3 -4 0 . 29 B ernsten C, Bjorkman I, Caramona M et al. (2001) Im proving th e w ell-b ein g of elderly patients via com m u n ity pharm acy-based provision of pharm aceutical care. Drugs Aging. 18: 6 3 -7 7 . 30 Harris D (1999) H elping elderly people w ith m ental health problem s. Pharm J. 263: 2 3 -4 . 31 M ilne A and Lingard J (2001) Medicines and Good Health in Later Life. The M ental Health Foundation Updates. Volume 3, Issue 8 ; h ttp ://w w w .m en ta lh ea lth .o rg .u k /h tm l/co n ten t/ U pdatev03i08.pdf

Chapter 6

The role of the nurse in the assessment, diagnosis and treatment of older people with psychotropic drugs Richard J Gibbens

Introduction For m any older people w ith m ental health needs, nurses are the healthcare professionals w ith w h o m they are likely to have m ost frequent contact. The nurse m ay also be the m ain link and inform ation source for the person's fam ily and carers. W here the Care Programme Approach or social care m anagem ent packages are in place for the older person, this m ay be a form alised arrangement, w here the nurse acts as the com m unication hub for all agencies and professionals involved in the person's package of care. W ith regard to psychotropic m edication, the nurse's role in w orking w ith older people requires a special aw areness and know ledge of the potential effects of ageing on h o w psychotropic drugs are m etabolised and excreted, and the nature and prevalence of potential adverse effects of these drugs. Psychotropic m edications are those intended to have an effect on m ental functioning, such as antidepressants, anxiolytics, hypnotics, antipsychotics and antidem entia drugs. These drugs are frequently prescribed for older people experiencing m ood disorder, sleep disorder, delerium , dem entia, and thought or behavioural disturbance. The potential benefits of prescribing any drug m ust be w eighed against the possible risks. The nurse m ust be skilled in identifying w h e n a prescription is not the best first option for treatm ent or intervention. He or she m ust be familiar w ith the range of m ethods available for the adm inistration of drug treatm ents and the appropriate use of m easured delivery system s, especially in com m unity settings, such as 'blister packs' and other m easured 'daily' system s designed to enable the o ld e r p e r s o n to ta k e th e ir m e d ic in e s in d e p e n d e n t ly a n d sa fe ly .

The nurse m ust alw ays consider issues of concordance. Will the person take the drug as prescribed? Do th ey w ant to take this drug? Do th ey understand w hat it is for and h o w it should be taken? W hat w ill they do if they experience side-effects? The nurse w orking w ith older people w ith m ental health problem s and their fam ilies and carers has a vital role in providing appropriate inform a­ tion and education to enable an effective partnership approach w ith the older person. From initial assessm ent and diagnosis to treatm ent w ith and m onitoring of psychotropic m edication, nurses have an essential role to play in ensuring

85

86

Practical old age p sy c h o p h arm ac o lo g y

that best prescribing practice is enabled in true partnership w ith the older person.

Assessment All nurses caring for older people m ust be alert to the im portance of screening for and assessm ent of m ental health needs, w hatever the setting. Early detection of m ental health needs in older people enables appropriate intervention to m inim ise further distress or deterioration . 1 W here drug treatm ents for m ental health problem s are being considered, com prehensive nursing assessm ent will provide key inform ation to enable accurate diagnosis and appropriate prescribing practice that is tailored to the patient's individual needs. A ssessm ent inform ation gathered by the nurse m ay form part of a global health screening assessm ent such as the 'EASY-Care' tool , 2 w hich contains specific brief questions to identify possible depression or cognitive im pairm ent. Nurses in contact w ith older people in the com m unity or acute general hospital en viron ­ m ent w ill often have sustained contact w ith the person, w hich enables the identification of possible m ental health needs through their privileged know ledge of the individual. Nurses in these environm ents can utilise screening tools 3 and thoughtful day-to-day practice w ith older people, to ensure that conditions such as depression are not m issed in cases w here there is a primary focus on physical health or social care interventions. This inform ation m ay inform m ulti-disciplin­ ary discussion and decision m aking on appropriate interventions and treatm ent, or prompt referral for a m ore detailed specialist m ental health assessm ent. Nurses w orking w ith in m ental health roles m ay com plete a m ore detailed specialist m ental health assessm ent, including the com pletion of screening instrum ents such as the Geriatric Depression Scale 4 or the M ini-M ental State E xam ination . 5 Com pletion of a com prehensive m ental health assessm ent w ith the older person w ill enable the nurse to identify any problems or needs that require intervention. The nurse m ay use this inform ation to form a treatm ent plan w ith the person for relevant psychological or psychosocial interventions, or to inform an accurate diagnosis to enable appropriate drug treatm ent prescribing by the GP or psychiatrist. Effective com prehensive assessm ent w ill ensure that the nurse provides key inform ation to enable appropriate and effective prescribing. A know ledge of the person's previous and current m edication, physical health and social history and situation w ill enable the nurse to provide the prescriber w ith key inform ation to enable the best treatm ent decision. The nurse's know ledge of the person's previous experience of m edication and preferences for any particular in terven ­ tions and therapies w ill all inform individually tailored prescribing decisions w hich increase the likelihood of concordance w ith the person receiving the prescription. Nurses are often w ell placed to obtain key assessm ent inform ation to enable the m ulti-disciplinary team to w eigh up the potential benefits and risks of drug treatm ent for the older person w ith m ental health problems. Other interventions such as facilitating increased social contact and activity in depression, for exam ple, or the use of cognitive behavioural therapy (see Box 6.1), m ay be m ore appropriate first-line interventions to be im plem ented and evaluated prior

T h e role o f th e n u rse in th e tr e a tm e n t o f o ld e r p e o p le w ith p sy c h o tro p ic drugs

87

to considering the suitability of drug treatm ents. The nurse has a responsibility to ensure that other non-pharm acological interventions are appropriately co n ­ sidered either as an alternative to or in conjunction w ith any pharm acological treatm ents. Clear com m unication w ith and the provision of tim ely inform ation to the older person at all stages of assessm ent and diagnosis w ill ensure a partnership approach w hereby individual preferences are recognised and considered.

Box 6.1 Conditions that can be helped by cognitive behavioural therapy • • • • • • • • • •

Phobias Certain anxiety disorders, including panic attacks and panic disorder D epression Eating disorders O bsessive-com pulsive disorder Anger Post-traumatic stress disorder Sexual and relationship problem s Drug or alcohol abuse Som e sleep problems

Adapted from Prodigy . 6

Nurses working w ith older people in specialist roles m ay have a key function in ensuring the appropriate prescribing and adm inistration of psychotropic m edica­ tion. For exam ple, specialist liaison nurses m ay be requested to provide inform ation and advice on appropriate prescribing for older people's m ental health in general hospital wards or in the care h om e sector. W here the older person is identified as dem onstrating difficult or 'challenging' behaviour, the liaison nurse can work w ith the care or treatm ent team to ensure that appropriate assessm ent inform ation is gathered, alternatives to drug treatm ent are considered and, if appropriate, that prescribing follow s national or local good practice guidelines or protocols . 7,8

Diagnosis Increasingly, nurses working w ith older people w ith m ental health problems are formally establishing a nursing diagnosis for conditions such as depression or anxiety, from w h ich a treatm ent plan is identified w h ich m ay include social interventions (such as reducing isolation) and psychological interventions (such as cognitive behavioural therapy ) . 9 M ental health nurses m ay utilise the diag­ nostic criteria outlined in the DSM -IV 10 or IC D -10 11 classification system s to assist them in reaching an effective nursing diagnosis and to ensure a compatible approach w ith medical colleagues w h o m ay also be involved in the person's care, particularly w here prescribing of a psychotropic m edication is appropriate. Nurses w ill often be w ell placed to offer advice or interventions directly to the older person before a psychotropic m edication is prescribed. For exam ple, if the older person is experiencing sleep disturbance, m anagem ent strategies should be the first intervention before consideration of the prescribing of a hypnotic m edication (see Box 6.2):

88

Practical old age p sy c h o p h arm ac o lo g y

Hypnotics and anxiolytics should be avoided w here possible and reserved for short courses to alleviate conditions after causal factors have b een established . 12

Box 6.2 General tips to help w ith sleep • Body rhythm - get up at the sam e tim e each day and never sleep during the day. • Routines - are useful before bedtim e (e.g. a warm drink, hot bath and reading). • Bedroom - should be free of noise and not used for w atching television, eating or work. • Stim ulants - such as alcohol, caffeine and sm oking should be avoided in the evening. • Exercise - regularly earlier in the day, but not w ith in a few hours of bedtim e. • Eat - only a light snack before bedtim e and not a large m eal. • R elaxation - m ay be im proved by a relaxation tape. Adapted from Prodigy . 12

Locally agreed protocols for the detection and treatm ent of specific conditions such as depression increasingly provide a fram ework for the diagnosis of and interventions for com m on m ental health needs such as dem entia and depression. These protocols indicate diagnostic steps w ith appropriate intervention strategies, including pharm acological intervention w here appropriate. Such agreed proto­ cols provide evidence-based practice guidelines w h ich the nurse can utilise both to inform their ow n individual care and treatm ent of the older person w ith m ental health needs and to guide appropriate interventions in discussion w ith the w ider m ulti-disciplinary team . 13 Increasingly in the care of older people w ith m ental health needs, nurses are com pleting initial com prehensive assessm ents and formal psychom etric tests w hich directly inform m edical colleagues w ith regard to the provision of appropriate prescribing decisions. This partnership approach w ith m edical col­ leagues reduces unnecessary repetition of diagnostic assessm ent and enhances nurse-led, com m unity-focused care for older people w ith m ental health needs.

T reatment W hen adm inistering any psychotropic drug the nurse m ust ensure that they adhere to the Nursing and M idwifery Council (NMC) guidelines for the adm in­ istration of m ed icin es 14 (see Box 6.3). It is good practice to enable the self­ adm inistration of m edicines by the patient or carers w here appropriate, particu­ larly for hospital patients approaching discharge . 14,15 H owever, effective policies and procedures m ust be in place to ensure safe practice, and for som e older people w ith m ental health problem s the potential risks and benefits of this approach m ust be carefully considered. W here a concordance aid such as a m onitored-dose container or d aily/w eek ly dosing aid is proposed, this should ideally be dispensed,

T h e role o f th e n u rse in th e tr e a tm e n t o f o ld e r p e o p le w ith p sy c h o tro p ic drugs

89

labelled and sealed by a pharmacist. The nurse w ill n eed to ensure both the ongoing appropriateness of any such aid and that the patient or their carer has sufficient inform ation and understanding to use this equipm ent safely . 14

B ox 6.3 Principles for the adm inistration of m edicines W hen exercising your professional accountability in the best interests of your patients you must: • k n o w the therapeutic uses of the m edicine to be administered, its norm al dosage, side-effects, precautions and contraindications • be certain of the identity of the patient to w h om the m edicine is to be adm inistered • be aware of the patient's care plan • check that the prescription, or the label on m edicine dispensed by a pharmacist, is clearly w ritten and unam biguous • have considered the dosage, m ethod of adm inistration, route and tim ing of the adm inistration in the context of the condition of the patient and coexisting therapies.

During the adm inistration of m edicines to older people, or m onitoring of treatm ent concordance, the nurse w ill n eed to evaluate w h eth er any prescribed drug is prescribed in the m ost appropriate form for the individual. Alternatives to tablets, such as liquids or solubles, m ay be available and m ore comfortable or suitable for som e older people w ith m ental health needs. Nurses have an essential role in effective m edication m anagem ent, and m ay find it helpful to review local trust policies and work w ith local pharmacists, clinical governance com m ittees and prescribing com m ittees to achieve a system ­ atic approach to ensuring that appropriate prescribing and m onitoring of psych o­ tropic m edication is a reality for patients in their care. Nurses are required to m aintain their professional com petence in relation to the adm inistration of m edicines, and w ill n eed to ensure that th ey participate in effective continuing professional developm ent linked to their role, job description and identified com petencies. Formally identifying and m eeting ongoing com petency and train­ ing needs through annual appraisal and training plans is an essential part of ensuring effective m edications m anagem ent. An increasing range of electronic resources are available to guide and support nursing practice, such as the electronic BNF 15 and Prodigy 6 internet sites. W hen adm inistering drug treatm ents for older people w ith m ental health problems, nurses m ust take account of the physiology of ageing and the potential consequences of this for those in their care. T rounce 16 has identified that older people m ay be m ore at risk of altered distribution, m etabolism and excretion of drugs. An older person w ith reduced blood album in levels m ay experience greater pharm acological effects than a younger person at the sam e drug dose, due to reduced protein binding w ith the drug. The enzym es w hich break dow n m any drugs in order to m etabolise them m ay be less active w ith increased age, and the blood supply to the liver m ay also be reduced. This can potentially lead to an over-accum ulation of the drug and sym ptom s of overdose. Renal function in an

90

Practical old age p sy c h o p h arm ac o lo g y

80-year-old m ay be half that of som eon e aged 40 years, and the resultant reduced excretion of drugs m ay cause potential accum ulation. Som e body organs and system s m ay also be m ore sensitive to drug treatm ent in older patients, w ith a higher incidence of disturbance of blood pressure, for exam ple, in an older population. Adverse drug reactions m ay be tw o to three tim es m ore com m on in older people, due to a range of potential factors including the num ber of drugs taken, the presence of severe illness and reduced elim ination of drugs from the b o d y . 1617

Managem ent o f behavioural disturbance In 2004, im portant concerns w ere identified regarding the use of specific atypical antipsychotics for behavioural and psychological sym ptom s in d em entia . 8 This report produced expert guidance to the effect that w h en behavioural and psychi­ atric sym ptom s are identified in people w ith dem entia, assessm ent of the person's needs m ust consider the following: A ny changes in environm ent, relationships or physical health? To w h o m is the sym ptom a problem and w hy? Do family carers and care staff need additional training to im prove therapeutic interactions ? 8 In considering the use of drug treatm ent for behavioural and psychiatric sym p­ tom s, the Royal College of General Practitioners recom m ends that psychosocial, behavioural and environm ental interventions are tried first, w ith additional consideration of the appropriate use of arom atherapy. The Nursing and M id­ wifery Council gu id elin es 14 suggest that nurses w h o practise com plem entary and alternative therapies m ust have successfully undertaken training and be com ­ petent in this area. It is essential to have considered the appropriateness of any such therapy to both the condition of the patient and any coexisting treatm ents. It is also essential that the patient is aware of the therapy and gives their inform ed consent. Nurses are often in key roles of responsibility w ith regard to planning and delivering the care and treatm ent of people w h o are experiencing dem entia. Nurses m ust ensure that appropriate assessm ent and education have taken place w ith the person w ith dem entia and their family, carers and care staff in cases w here there is distressing behaviour or psychiatric sym ptom s. Nurses w ill often be able to lead the w ay in providing non-pharm acological interventions, working creatively w ith the person w ith dem entia to identify individual solutions that are acceptable to that person and their carers . 18 W ithin busy and stressful roles w ith dem anding workloads, m edication can often be view ed as a 'quick fix' for behaviour that presents challenges or difficulties. A m ore individualised approach, focused on the individual w ith dem entia and their unique needs, w ill often be more tim e-consum ing, but is ultim ately often m ore rewarding as the nurse works creatively to identify personcentred solutions. M any nurses in this situation m ay be required to adopt a clear stance and dem onstrate leadership both in trying alternative individual approaches, and in resisting any inappropriate pressure for a drug to be prescribed before other options have b een im plem ented and evaluated. This approach w ill require clear com m unication w ith all m em bers of the m ulti-disciplinary or

T h e role o f t h e n u rse in th e tr e a tm e n t o f o ld e r p e o p le w ith p sy c h o tro p ic drugs

9 1

m ulti-agency team , as w ell as crucially w ith the person him - or herself and his or her fam ily. This nursing role can be a challenging one, particularly if carers or care staff sh ow resistance to exploring the im pact of their ow n behaviour and response to the person w ith dem entia. H ow ever, persistence in finding individual solutions w ill bring benefits as those w h o are being introduced to this more creative w ay of responding to difficult behaviour recognise their successes and becom e m ore familiar and com fortable w ith this approach.

Use of shared care guidelines Nurses w orking w ith older people w ith m ental health problem s have played a key role in the developm en t and im plem entation of shared care protocols for conditions such as dem entia and depression, w ith target dates for the im p lem en ­ tation of these protocols previously identified w ithin the National Service Framework for Older P eople 1 (see Box 6.4).

B ox 6.4 An exam ple of a shared care protocol for antidem entia drug treatm ents In South W est Yorkshire M ental Health NHS Trust, local shared care protocols have b een im plem ented for the prescribing of antidem entia drugs. M em ory service nurses respond to initial referrals, com pleting h om e-based assessm ents and liaising w ith psychiatrist colleagues in diag­ nosis and the m onitoring and titration of prescribed antidem entia m edica­ tion, prior to the GP taking over prescribing responsibility. The m em ory services rem ain involved in m onitoring the person w ith dem entia at agreed intervals and inform ing the GP of their progress and the suitability of continuation of treatm ent. This role has provided the m em ory service nurses w ith the opportunity to develop im proved inform ation and support for people w ith dem entia and their carers, both pre- and post-diagnosis.

A com parison study of tw o locality m em ory services for people w ith dem entia has highlighted the value that both the person w ith dem entia and their carers place on the opportunity to have consistent access to the sam e nurse, w h o visits them at hom e at a con ven ien t tim e and provides pre-diagnostic support and inform ation, diagnostic and m onitoring assessm ents and follow -up m onitoring and support . 19 This approach, w hich has been developed in response to the availability of antidem entia drugs, dem onstrates that in addition to the availability of the drug itself, it is the com m unication, inform ation and support that this brings w ith it from the m em ory service nurse that is of significant value to the person w ith dem entia and their fam ily. A valid concern m ay be w hether people w ith dem entia w h o are not identified as suitable for antidem entia drug treatm ent receive equal access to these benefits from local health and social care providers. Seeking the consent of the older person is an essential com ponent of all aspects of the provision of nursing care and treatm ent. W hen prescribing or adm inister­ ing psychotropic m edication, as w ith any intervention, for a person's consent to treatm ent to be valid they m ust be:

92

Practical old age p sy c h o p h arm ac o lo g y

• capable of taking that particular decision • acting voluntarily • provided w ith en ou gh inform ation to enable them to m ake the decision . 20 It is the right of any older person to refuse m edicines. If prescribed drugs are refused, the nurse should record the reason for refusal and identify an opportun­ ity to review this w ith the prescriber and other relevant m em bers of the m u lti­ disciplinary team . W hen som eone refuses offered prescribed m edication (e.g. in a hospital ward), every effort should be m ade to identify the reasons for this refusal. D oes the person have any safety concerns about the drug? Has it m ade them feel u nw ell? Have the purpose and anticipated benefits of the drug been clearly explained? Do th ey require any w ritten inform ation about the drug or an opportunity to discuss the treatm ent w ith their doctor? If the tablets are difficult for the person to sw allow , or have an unpleasant taste, are there any alternatives available? It w ill often be useful to discuss this w ith a pharmacist in order to obtain advice and inform ation on alternatives. Any nurse w h o identifies the potential need for the covert adm inistration of m edicines to older people w ith m ental health problem s w ill need to follow the guidelines set out in the UKCC Position Statement on the Covert Administration of Medicines, 21 endorsed by the Nursing and M idwifery Council. H owever, this practice is best avoided if at all possible, and should only be considered in exceptional circum stances. It is the right of an individual to refuse any treatm ent, and every adult is assum ed to have the capacity to consent to or refuse a treatm ent until the contrary is proven. This includes people w ith severe dem entia, as no older person should be assum ed to lack capacity in any situation unless this has been form ally assessed and identified. A ssessm ents of capacity to m ake a particular decision w ill only relate to that decision at that particular time, and m ust not be generalised to any other situation or intervention. W hen adults lack capacity and this has been formally determ ined, it is possible to law fully provide treatm ent and care, but this m ust be in the person's 'best interests'. A ny determ ination of the person's best interests m ust include inform a­ tion from those w h o k n o w the person w ell, and also identification of previously expressed w ishes and preferences. In addition to the person's 'medical' or health need, attention m ust be given to the im pact of any treatm ent on their p sycho­ logical, social and spiritual w ell-being. No one can give consent on behalf of another adult. The healthcare professional responsible for the person's care is legally accountable for deciding w h eth er the treatm ent is in the person's best interests, but ideally decisions w ill reflect agreem ent b etw een professionals and those 'close to the older person ' . 20

Efficacy and safety The National Service Fram ework for Older P eop le 1 identifies the need for regular m edication review s for older people to m inim ise the risk and harm of p o ly ­ pharm acy. Once an older person has b een prescribed a psychotropic drug, nurses are often best placed to m onitor the effects of the drug, in terms of its benefit and any possible experienced adverse effects. This requires a w orking know ledge of the drug, its anticipated treatm ent benefits and possible side-effects. The nurse can consider this inform ation in the context of the individual patient's lifestyle

T h e role o f th e n u rse in th e tr e a tm e n t o f o ld e r p e o p le w ith p sy c h o tro p ic drugs

93

and health history, to ensure that appropriate dose titration, m edication review , consideration of alternatives or discontinuation of treatm ent all occur w h en appropriate. Polypharm acy in older people has b een identified as a significant risk factor for falls . 1 M edication review and m onitoring of possible side-effects of prescribed drugs, such as h ypotension, can play an im portant role in falls prevention. The prescribing of antipsychotic m edications for patients in nursing and residential hom es has b een identified as a particularly significant concern : 17 Such m edicines used to treat behavioural com plications of dem entia m ay hasten cognitive declin e . 17 Nurses have a clear role in ensuring that any behavioural disturbance is fully and appropriately assessed, and that the prescribing of m edication is view ed as a last resort once other interventions have b een appropriately im plem ented. The provision of tim ely, appropriate and understandable inform ation on drug treatm ents, anticipated benefits and possible side-effects is likely to im prove the patient's concordance and provide him or her w ith an opportunity to raise any concerns prior to treatm ent. If the patient has a com m unicative, trusting relation­ ship w ith the nurse, a m ore open discussion of prescribed m edication is likely to occur, m inim ising the risk of prescriptions that are not dispensed or unused stockpiles of tablets in the patient's m edicine chest.

Reporting adverse events The Y ellow Card schem e is available to enable nurses to report any suspected adverse drug reactions. It is available in a paper version at the back of the Nurse Prescribers Formulary (NPF)22 and the British National Formulary (BNF).15 The C om m ittee on Safety of M edicines (CSM) and the M edicines Control A gency (MCA ) 23 recom m end that nurses use the electronic version of the Yellow Card, w hich is available at: w w w .m ca.gov.u k /yellow card . The CSM/MCA identify an adverse drug reaction as follows: an unw an ted or harm ful reaction experienced follow ing the adm inis­ tration of a drug or com bination of drugs under norm al conditions of use and suspected to be related to the m ed icin e . 23 All adverse reactions, including non-serious ones, should be reported for black triangle m edicines (w hich indicates intensive m onitoring of that product by the CSM/MCA). For established m edicines or vaccines, all serious suspected adverse reactions should be reported, including those that are: fatal - life-threatening - disabling - incapacitating - congenital abnorm ality - involve hospitalisation - and/or m edically significant .23

Nurse prescribing Nurses have a key role in the assessm ent, diagnosis and treatm ent of older people w ith psychotropic drugs. Historically, nurses have supported m edical prescribing practice, providing tim ely accurate inform ation to enable appropriate prescribing decisions by m edical staff. Registered m ental nurses w orking w ith older people in

94

Practical old age p sy c h o p h arm ac o lo g y

com m unity or inpatient settings have for m any years prom pted tim ely appro­ priate prescribing and advised n o n -m en tal-h ealth prescribers (e.g. in primary care settings) on suitable and appropriate drug treatm ents in the field of older people's m ental health. M ore recently, national procedures and validated training pro­ gram m es have b een established to enable nurses to achieve a registrable qualifi­ cation in independent, extended independent and supplem entary prescribing, w here auton om ous appropriate prescribing is undertaken directly by m ental health nurses, rather than 'through the back door' by advising m edical colleagues. The role of experienced m ental health nurses in advising doctors on appropriate prescribing for m ental health needs has b een recognised for m any years . 24 In this context the developm ent of extended roles for nurses, enabling them to supple­ m ent the m edical workforce in m ental health prescribing, is a natural d evelop ­ m ent. H ow ever, concern has b een expressed that in undertaking formal prescribing roles, nurses m ay becom e less focused on care and concentrate m ore on m edical treatm ent, at a tim e w h en a m uch greater em phasis on psychosocial approaches, talking therapies and health prom otion is required in older people's m ental health nursing. De-facto prescribing by com m unity m ental health nurses has b een identified , 25 w here nurses have described influencing and advising doctors on prescribing decisions. User view s on m ental health nurse prescribing have been identified as includ­ ing concerns about the appropriateness of prescribing by nurses rather than doctors , 26 and currently any psychotropic drugs prescribed by a nurse as a supplem entary prescriber require that the agreem ent of the patient to this form of prescribing is docum ented. M ental health nurses have been described as having a m ixed level of existing know ledge of pharm acology, and concern has been raised about potential underestim ation of the level of know ledge and experience needed to prescribe psychotropic m ed ication .27,28 This m ay be addressed by the developm ent of a nurse prescribing training curriculum to include additional content on psych o­ tropic pharm acology for nurses w orking in the field of m ental h ealth . 29 Since 2003 the opportunity has b een available for nurses to undertake training to becom e extended independent and supplementary prescribers. At the tim e of writing, psychotropic m edications for older people's m ental health problem s are not included w ithin the list of approved m edicines that registered nurse prescribers can prescribe independently. H ow ever, qualification as a supplem entary pre­ scriber does enable the prescription of psychotropic m edications by the nurse, w ithin strict specific param eters . 30 The supplem entary prescribing role is described as follows: A voluntary prescribing partnership b etw een the independent pre­ scriber (a doctor or dentist) and a supplem entary prescriber, to im plem ent an agreed patient-specific Clinical M anagem ent Plan . 30 The independent prescriber (IP) is responsible for diagnosis and the parameters of the clinical m anagem ent plan. The supplem entary prescriber (SP) has discretion w ith regard to the dosage, frequency and product, w ith in the limits of the clinical m anagem ent plan. Once a clinical m anagem ent plan has b een com pleted, this m ust be review ed by the IP at least annually if not m ore frequently. The IP and SP m ust share access to a com m on patient record. A greem ent to the clinical

T h e role o f th e n u rse in th e tr e a tm e n t o f o ld e r p e o p le w ith p sy c h o tro p ic drugs

95

m anagem ent plan m ust be recorded by the independent prescriber and supple­ m entary prescriber before prescribing can b egin . 30 Clearly the supplem entary prescriber should not prescribe any m edicine that th ey do n ot feel com petent to prescribe. The independent prescriber can choose n ot to im plem ent a clinical m anagem ent plan w here there is concern for any reason that supplem entary prescribing m ay not be appropriate. The independent prescriber can also resum e full direct responsibility for prescribing at any tim e if required. An exam ple of supplem entary nurse prescribing for antidem entia drug treatm ents is given in Box 6.5.

B ox 6.5 Supplem entary prescribing of antidem entia m edication in a m em ory service C om m unity assessm ent

m em ory

clinic

nurse

com pletes

com prehensive

hom e

i

Patient attends m em ory clinic for com pletion of baseline psychom etric tests and diagnostic interview w ith m edical and nursing team . Inform ation is provided on diagnosis and available drug treatm ents i

Patient is in agreem ent w ith supplem entary prescribing by the nurse i

Clinical m anagem ent plan for nam ed patient is signed by (independent prescriber) and nurse (supplem entary prescriber)

doctor

I Nurse provides ongoing m onthly prescriptions, including dose titration, in accordance w ith NICE guidance 31 and locally approved shared care gu id e­ lines for antidem entia m edication i

At 4 to 6 m onths from initial prescription, GP takes over prescribing in accordance w ith the shared care guidelines

There is no legal restriction on the clinical conditions that the supplem entary prescriber m ay treat. It is anticipated that supplem entary prescribing m ay be m ost useful in the treatm ent of som e long-term conditions. All prescription-only m edicines can be prescribed except controlled drugs . 30 A scoping study conducted by the National Prescribing Centre for the Depart­ m ent of Health identified that clinical m anagem ent plans need to be relatively sim ple and quick to com plete in order to be w orthw hile, and should not include a lot of inform ation that is already available w ith in the shared record. The clinical m anagem ent plan should m ake reference to appropriate reputable guidelines or agreed protocols for the treatm ent of specific conditions . 32 Registered nurses are personally accountable for their prescribing practice, and m ust work to the sam e standard or com petence that applies to all other prescribers . 30 Nurse prescribers are also accountable to the Nursing and M idwifery Council and m ust act in accordance w ith its code of professional conduct. All nurse prescribers should have professional indem nity. The National Prescribing

96

Practical old age psychopharmacology

Centre has produced a framework document to enable nurses to m aintain ongoing competency in their prescribing.33

Key points • Nurses have contact with older people in many different settings, and have significant opportunities to enhance the appropriate detection, assessment and treatm ent of mental health problems for these people. • Knowledge of psychotropic medications for older people, including w hen and how these should be prescribed, m onitored and evaluated, together with a knowledge of alternative or complementary non-pharmacological interventions, is essential for all nurses involved in the care of older people.

References 1 D epartm ent of Health (2001) National Service Framework for Older People. D epartm ent of Health, London. 2 Sheffield Institute for Studies on Ageing (1999) EASY-Care Elderly Assessment System. Sheffield Institute for Studies on Ageing, University of Sheffield, Sheffield. 3 Royal College of N ursing (1997) Guidelines for Assessing Mental Health Needs in Old Age. Royal College of Nursing, London. 4 V an-M arw ijk H, W allace P, de-Boc G et a l (1995) E valuation of th e feasibility, reliability and diagnostic value of shortened versions of th e Geriatric Depression Scale. Acta Psychiatr Scand. 67: 361-70. 5 Folstein M, Folstein S and M cHugh P (1975) 'M ini-M ental State': a practical m eth o d for grading th e cognitive state of patients for th e clinician. J Psychiatr Records. 12: 189-98. 6 Prodigy (2004) What is Cognitive Behaviour Therapy (CBT)? w w w .prodigy.nhs.uk 7 Royal Pharm aceutical Society of Great Britain (2003) The Administration and Control of Medicines in Care Homes and Children's Services. Royal Pharm aceutical Society of Great Britain, London. 8 Royal College of G eneral Practitioners (2004) Guidance for the Management of Behavioural and Psychiatric Symptoms of Dementia and the Treatment of Psychosis in People with a History of Stroke/TIA. Royal College of General Practitioners, London. 9 Woods R (1999) Psychological Problems of Ageing: assessment, treatment and care. Jo h n Wiley and Sons, Chichester. 10 A m erican Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental Disorders (4e). A m erican Psychiatric Association, W ashington, DC. 11 W orld H ealth O rganization (1992) The ICD-10 Classification of Mental and Behavioural Disorders: clinical descriptions and diagnostic guidelines. W orld H ealth Organization, Geneva. 12 Prodigy (2003) Guidance: hypnotic or anxiolytic dependence; w w w .prodigy.nhs.uk 13 A lzheim er's Society (accessed 2004) Dementia: diagnosis and management in primary care. An evidence-based educational package about dementia for primary care teams in CD-ROM format; w w w .alzheim er's.org.uk 14 Nursing and M idwifery Council (2002) Guidelines for the Administration of Medicines. Nursing and M idwifery Council, London. 15 The Pharm aceutical Press (2003) Nurse Prescribes' Formulary 2003-2005. The Pharm aceutical Press, London.

T h e role o f t h e n u rse in th e tr e a tm e n t o f o ld e r p e o p le w ith p sy c h o tro p ic drugs

97

16 Trounce J (2000) Clinical Pharmacology fo r Nurses (16e). C hurchill-Livingstone, Edinburgh. 17 D epartm ent of Health (2001) Medicines and Older People: implementing medicines-related aspects o f the NSF fo r Older People. D epartm ent of Health, London. 18 Clibbens RJ and Lewis D (2004) The role of the nurse in the assessm ent, diagnosis and m anagem ent of patients w ith dem entia. In: S Curran and J W attis (eds) Practical M anagement o f Dementia: a multi-professional approach. Radcliffe Publishing Ltd, Oxford. 19 Timlin A, Gibson G, Curran S and W atttis JP (2005) Memory Matters: A report exploring issues around the delivery o f anti-dementia medication. U niversity of Huddersfield, Hudders­ field. 20 D epartm ent of Health (2001) Seeking Consent: working w ith older people. D epartm ent of Health, London. 21 Nursing and M idw ifery Council (2002) UKCC Position Statement on the Covert A dministration o f Medicines. Nursing and M idw ifery Council, London. 22 The Pharmacutical Press (2003) The British N ational Formulary. The Pharm aceutical Press, London; w w w .b n f.org/ 23 C om m ittee on Safety of M edicines and the M edicines Control A gency (2002) Extension o f the Yellow Card Scheme to Nurse Reporters. M edicines Control A gency, London. 24 G ournay K and Barker P (2002) Prescribing: the great debate. Nurs. Standard. 17: 2 2 -3 . 25 R am charan P, H em m ingw ay S and Flow ers K (2001) A client-centred base for nurse prescribing. M ent Health Nurs. 21: 6 -1 1 . 26 Harrison A (2003) M ental h ealth service users' view s of nurse prescribing. Nurse Prescrib. 1: 7 8 -8 5 . 27 M cCann TV and Baker H (2002) C om m unity m ental h ealth nurses and authority to prescribe m edications: the w ay forward? J Psychiatr M ent Health Nurs. 9: 1 7 5 -8 2 . 28 Davis J and H em m ingw ay S (2003) Supplem entary prescribing in m en tal h ealth nursing. N urs Times. 99: 2 8 -3 0 . 29 D epartm ent of Health (2004) Nurse Prescribing Training and Preparation: extended form ulary nurse prescribing and supplementary prescribing. D epartm ent of Health, London. 30 D epartm ent of Health (2004) Supplementary Prescribing: key principles. D epartm ent of Health, London. 31 D epartm ent of Health (2001) Guidance on the Use o f Donepezil Rivastigmine and Galantamine for the Treatment o f Alzheim er's Disease. NICE T echnology Appraisal guidance No. 19, D epartm ent of Health, London. 32 D epartm ent of Health (2004) Scoping Study o f Supplementary Prescribing. D epartm ent of Health, London. 33 N ational Prescribing Centre (2001) M aintaining Competency in Prescribing: an outline fram ew ork to help nurse prescribers. N ational Prescribing Centre, London.

Chapter 7

The GP and ‘medicines management’ for older people with mental health problems in primary care Owen P Dempsey

Introduction The general practitioner (GP) in the UK has a pivotal gate-keeping role w ith u nique opportunities for the early detection of m em ory or behavioural problems and m ental illness in older people. H ow ever, the early recognition of problems such as dem entia and depression can be very difficult in the hurly-burly of everyday practice, and unfortunately GPs do not always have the tim e, the m otivation or the skills required . 1 In older people in particular, depression and dem entia can be very difficult to recognise, but despite the large and increasing prevalence of these problems, less than 40% of GPs have received any specialist training in gerontology, and even less w ill have received training in old age psychiatry . 1,2 A report from the Audit C om m ission has recom m ended that m ental health experts provide m ore training for primary care health w orkers . 1 As w ell as the early recognition of m ental health problems, GPs also have clinical and legal responsibility for the vast majority of prescribing for older people w h o have already b een established on psychotropic m edication. This is a huge and im portant problem . Four out of five people over 75 years of age take at least one prescribed m edicine, w ith 36% taking four or m ore m edicines, and alm ost half of the NHS drugs bill is spent on m edicines for older p eop le . 3,4 Care hom es are a special case. Residents receive up to four tim es as m any prescription item s as older people living in their o w n hom es, and up to 53% of care hom e residents receive at least on e inappropriate prescription 5,6 (see Case study 7.2). U nder­ standably, GPs struggle to cope. Zermansky found that 72% of repeat pre­ scriptions sam pled in 50 practices had not been review ed in the past 15 m on th s . 7 In order to address this issue, the National Service Framework for Older People (w w w .n elh .n h s.u k /n sf/o ld er_ p eo p le/d efa u lt.h tm ) has em phasised the need for GPs to perform regular face-to-face clinical m edication review s and to consider flexible w ays of m anaging this issue, such as collaborative working w ith pharmacists or nurse specialists/practitioners .8 For those w h o are interested, there is an excellent pack produced by the U niversity of Leeds on behalf of the Task Force on M edicines M anage­ m ent Partnership. This is available online to support m edication review s (w w w .m edicines-partnersh ip.org/m ed ication-review /toolkit).

99

10 0

Practical old age p sy c h o p h arm o co lo g y

Depression Special considerations and epidemiology D epression is com m on in older people, w ith perhaps 10-15% of the over-65s experiencing significant depressive sym ptom s. It is often 'masked'. The patient m ay have predom inantly anxiety sym ptom s m asking an underlying depression, or they m ay have a m ultitude of physical problem s causing anxiety about potential illness, and m ay feel that physical com plaints are m ore 'acceptable' than em otional ones. Depression is under-recognised, is associated w ith tw ice the suicide rate for younger people, and is strongly associated w ith physical illness . 9 1 0 It is associated w ith a poorer prognosis than physical illness, even allow ing for illness severity . 10 M ilder depressive sym ptom s m ay be recognised by GPs but are often not treated, perhaps because of uncertainty about the effectiveness of treatm ent.

Diagnosis GPs should be alert for the early w arning signs and prepared to ask specifically about anhedonia (lack of pleasure in life), poor appetite, w eight loss and suicidal ideas (see B oxes 7.1 and 7.2). If in doubt, the Geriatric Depression Scale (GDS) or its shorter version (G D S-15) can be used. It contains 30 questions w ith a yes/n o response, and concentrates m ainly on cognitive rather than physical sym ptom s . 11 A score of > 11 suggests depression . 12

Box 7.1 Early w arning sym ptom s of depression 13 • • • • • •

Sleep disturbance Fatigue Failure to care for oneself W ithdrawal from social life U nexplainable som atic sym ptom s H opelessness about a coexisting physical disorder

B ox 7.2 People at increased risk of depression and/or suicide 10 • • • • • • • • •

Living in an institution Older w h ite m en Living alone M ultiple physical illness Sleep disturbance Recent bereavem ent Previous suicide attem pt Other loss (incom e, children m oving away) A lcohol or drug abuse

T h e G P and ‘m ed icin es m a n a g e m e n t’ fo r o ld e r p e o p le w ith m en tal health p ro b le m s

10 1

Assessment Possible m edical causes or drug-related causes should be evaluated. The list of drugs associated w ith depression is long, and the m edications are com m on and n ot alw ays easily reduced or stopped (see Box 7.3). The GP should consider reduction of the drugs listed in B ox 7.3 if depression is diagnosed . 14

B ox 7.3 C om m on drugs associated w ith depression • B enzodiazepines and buspirone • A nticonvulsants

• Anti-parkinsonian drugs (e.g. anticholinergics and L-dopa) • Cardiovascular system drugs (e.g. beta-blockers, enalapril, m ethyl-dopa) • Gastrointestinal drugs (e.g rantidine/cim etidine) • N on-steroidal anti-inflam m atory drugs • A m inophylline • Steroids

The patient should be exam ined w ith particular reference to signs of conditions associated w ith depression, including appearance (unkem pt, parkinsonian facies, oedem a of m yxoedem a, pallor of anaem ia), blood pressure (may affect choice of treatm ent), pulse (m yxoedem a), w eigh t (useful baseline), fundi (raised intracra­ nial pressure - brain tum ours) and central nervous system , particularly rigidity, tremor, cog-w h eelin g and ataxia (Parkinson's disease or m ultiple sclerosis). Consider, as a m inim um , blood tests for the following: • thyroid function tests - hypothyroidism • calcium /phosphate - hypercalcaem ia (m alignancy) • full blood count, ferritin, vitam in B 12/folate - iron, vitam in B 12/folate deficiency (anaem ia) • random plasma glucose - diabetes • electrolytes and creatinine - renal failure. Consider form ally testing for cognitive im pairm ent w ith the M ini-M ental State E xam ination (MMSE) (see section on dem entia) to check for early dem entia and ask (both the patient and the carer if possible) about the four dom ains of functioning that are com m only affected in dem entia, nam ely: •

u s e o f tra n sp o r t

• m anaging a budget • using a teleph on e • m anaging m edication.

Treatment The GP and the primary healthcare team can provide im portant 'general support' and encouragem ent, including attention to practical issues, access to social services, b ereavem ent counselling, and a sharing w ith the patient of the problem, its nature, possible causes, effective treatm ents, the likely length of the illness

10 2

Practical old age p sy c h o p h arm o co lo g y

and its possible consequences. There are tw o useful leaflets that the GP can provide: • Depression in the Elderly (w w w .rcp sych.ac.uk/public.help/depeld /depeld.htm l) • Bereavement (w w w .rcpsych .ac.u k/p ub lic/help/b ereav/bereavem .htm l).

Drugs The first-line treatm ent for older people should be a selective serotonin reuptake inhibitor (SSRI). The N ational Service Fram ework on M ental H ealth 15 (w w w .d o h .g o v .u k /p u b /d o cs/d o h /m h m a in .p d f) recom m ends that tricyclic anti­ depressants (TCAs) should not be prescribed for depression in patients over 70 years of age. Even th ou gh older people are particularly susceptible to the adverse effects of the older TCAs, analysis of prescribing show s that elderly patients are m ore likely to be prescribed an older TCA and less likely to be prescribed an SSRI than younger patients . 16 In a prescribing analysis study using appropriate doses for primary care, only 43% of those over 65 years of age received an adequate dose in cases w here a TCA was prescribed . 16 The response takes longer in older people, and m ay take up to 12 w eeks. Treatment should be continued for 6 to 12 m onths, and lifelong treatm ent m ay be indicated in recurrent or severe depression. GPs need to be aware of potential side-effects and interactions. R eduction in or cessation of treatm ent should be gradual. The side-effects include anticholinergic (urinary retention, constipation and glaucom a), antihistam inic (sedation) and a xadrenergic-blocking (postural hypotension) effects (see Table 7.1). Table 7.1 Side-effects of com m on antidepressants 17

TCAs A m itriptyline D othiepin Lofepramine SSRIs Paroxetine Citalopram Atypical Trazodone

a i- adrenergic blocking

Anticholinergic

Antihistam inic

+++ +

++ +

++++ ++ +

0 /+

0

0

0

+++

+

Referral to secondary care This should be considered in cases w here there is a high risk of self-harm (e.g. serious suicidal thoughts, previous suicide attem pts), self-neglect, failure to respond to treatm ent after 1 2 w eeks on a therapeutic dosage, or significant carer stress (see Case study 7.1).

T h e G P and ‘m edicines m a n a g e m e n t’ fo r o ld e r p e o p le w ith m ental health p ro b lem s

10 3

Case study 7.1 Mr A lived w ith his 'young-for-her-age', active and busy 67-year-old wife w h o continued to work as a seam stress/dressmaker from hom e. At the age of 72 years he presented at the surgery w ith low back and left leg pain and a left-sided foot drop and absent left-sided ankle jerk. His wife was w ith him and she did m ost of the talking for him . He had always had an introspective personality, had had no hobbies or interests since retirement, and for m onths had spent all day on the sofa, and was reluctant to m ix w ith others. He presented a very flat expressionless affect, denied feeling depressed but kept referring back to the awful pain in his leg that w as preventing him from sleeping. His w ife said that he m ade 'a big thing' of keeping the blankets off his leg at night w ith a hom e-m ade cradle. He was anxious about the possibility of spinal cancer. He was given opiate analgesics (tramadol). His w ife did everything for him, and seem ed exasperated. He was referred to the orthopaedic surgeons. Investigations of the spine revealed a prolapsed disc, but surgery w as decided against because of the chronicity of the sym ptom s. Blood tests, including bone screen and plasma viscosity, w ere all normal. The surgeons suggested am itriptyline for relief of the neuralgia. This (and the strong opiate analgesia) led to a hom e visit because of acute urinary retention and Mr A was adm itted and eventually had a transurethral resection of the prostate. He rem ained torm ented and obsessed by the pain in his foot, was apathetic and becam e forgetful. An MMSE revealed a score of 13/20. He was started on fluoxetine, w ith little noticeable benefit. He was referred to the com m unity psychiatric nurse for the elderly, and m uch to his discomfort (and his w ife's guilt) started on day care 3 days a w eek to provide respite for his w ife. During this tim e his w ife becam e ill, w ith faints and w eakness of no apparent cause, and w as adm itted acutely from her hom e. All her in vesti­ gations w ere negative and her o w n depressive symptom s, heavily overlaid w ith anxiety, eventually becam e apparent. She responded w ell to a com bination of counselling and paroxetine. Mr A's m ood has proved resistant to treatm ent w ith a range of antidepressants, and he remains on citalopram and tramadol. His deteriorat­ ing cognitive im pairm ent has made attempts to reduce his anxiety about cancer by attempts to link his fears to his m ood very difficult. He asks repetitive questions of his w ife and becom es extrem ely anxious if he can't see her. He attends surgery regularly w ith his wife, w h o is adapting to her circumstances.

Dementia Special considerations and epidemiology Based on the Morbidity Statistics in General Practice, 18 it is possible to estim ate the incidence, prevalence and workload for patients w ith dem entia. The figures in Table 7.2 are based on an assum ed list size of 2000 patients per GP principal.

104

Practical old age p sy c h o p h arm o co lo g y

Table 7.2 Incidence, prevalence and w orkload for patients w ith d em en tia 18 Classification category

Early- and lateonset organic psychotic conditions

Incidence

Prevalence

Workload

N ew patients per general practitioner per year

Patients consulting per general practitioner per year

C ondition-related consultations per general practitioner per year

1.6

3.6

7.4

A bout 5% of people aged 65 years or over have som e form of dem entia, w ith the prevalence rising to about 20% by 80 years, and one in three by 90 years . 19 In care hom es, 70-80% of residents w ill have functional im pairm ent due to dem entia. There are guidelines for the primary care m anagem ent of dem entia from the North of England Evidence-based G uidelines D evelopm ent Group .20 These are helpful if used in conjunction w ith local guidelines (e.g for referring to a m em ory m onitoring nursing service for assessm ent for suitability for an acetylcholinesterase inhibitor).

Diagnosis There is evid ence that GPs lack the skills and know ledge to diagnose and m anage dem entia effectively, and increased training has b een recom m en ded . 1 W ith the em erging n e w treatm ents recom m ended for use w ith in the NHS for mild to m oderate dem entia 21 (w w w .n ice.o rg .u k ), there is a renew ed em phasis on early diagnosis. H ow ever, early diagnosis is difficult and the GP needs to be alert for early signs (see Table 7.3). The diagnosis should be suspected if the patient has difficulty w ith one or m ore of the follow ing four dom ains of daily activity, and questions about these should be directed to the carer as w ell as the patient: • • • •

m anaging m anaging m anaging m anaging

m edication use of the telep h on e a budget public transport.

Note that problem s w ith these dom ains of functioning m ay appear before obvious m em ory loss, especially in patients w ith vascular dem entia.

T h e G P and ‘m edicines m a n a g e m e n t’ fo r o ld e r p e o p le w ith m en tal health p ro b le m s

105

Table 7.3 Features of dem entia at different points in its path (adapted from Iliffe and D ren n an 19) Early changes

Later changes

E m otional changes

Shallow ness of m ood Lack of em otion al responsiveness and consideration of others

Irritability and hostility Aggression

C ognitive changes

M em ory deficits Perseveration and repetitive speech Getting lost

R eceptive and expressive dysphasia Thought processes fragm ented Psychotic features in 30-40% (paranoid delusions) Auditory and visual hallucinations

B ehavioural changes

Social w ithdraw al Self-neglect D isorientation

W andering E vening and nocturnal restlessness prom inent Turning night into day A ggression and v iolen ce

Physical changes

Usually later in the disease process

W eight loss Incon tin en ce Rigidity Seizures (late)

Once the diagnosis is suspected it is recom m ended that general practitioners or a m em ber of their team use a cognitive function test, such as the M ini-M ental State E xam ination (M M SE ) . 11 This gives a score out of 30 and takes about ten m inutes to com plete, so is not easy to incorporate into an ordinary consultation. It w ould probably be best to set up a separate consultation specifically to perform this test, as it is easy to perform it inaccurately. It is also probably best to becom e familiar w ith on e test. In Huddersfield the guidelines for referral to the m em ory m onitoring service (for consideration for treatm ent w ith an acetylcholinesterase inhibitor) include the criterion of an MMSE score of > 1 2 .

Assessment A physical exam ination is required, w ith the sam e level of detail as for depression (see above). The patient should be exam ined w ith particular reference to signs of conditions associated w ith m em ory problem s or depression, including appearance (unkem pt, parkinsonian facies, oedem a of m yxoedem a, pallor of anaem ia), blood pressure (m ay affect choice of treatm ent), pulse (m yxoedem a), w eight (useful baseline), fundi (raised intracranial pressure - brain tum ours) and the central nervous system , particularly rigidity, tremor, cog-w h eelin g and ataxia (Parkinson's disease or m ultiple sclerosis). For vascular dem entia, w h ich is characterised by a step-w ise deterioration in cognitive function, it is w orth assessing for additional risk factors for athero­ sclerosis, such as blood pressure, carotid bruits, body mass index (BMI), lipids (if under 75 years of age) and sm oking status. H ow ever, there is increased

106

Practical old age p sy c h o p h arm o co lo g y

recognition of the overlap b etw een Alzheim er's disease (AD) and vascular dem entia, and vascular risk factors also increase the risk of AD. Consider using the Geriatric D epression Scale (GDS) to exclude depression, w h ich is m ore com m on w ith dem entia and can also cause cognitive im pairm ent . 11 Other problem s m ay contribute to cognitive im pairm ent (e.g. infections, drug toxicity, alcohol, and pain and its underlying causes) (see Case study 7.2). Blood tests are as described for depression above, nam ely: • thyroid function tests - hypothyroidism • calcium /phosphate - hypercalcaem ia • full blood count, ferritin, vitam in B 12/folate deficiency • random plasm a glucose • electrolytes and creatinine.

iron, vitam in B 12/folate

There are a num ber of rarer and potentially reversible causes of dem entia, such as norm al-pressure hydrocephalus (classical triad of dem entia, ataxia and in con tinence). The GP should also take tim e to discuss w ith the patient and their carer the diagnosis of early cognitive im pairm ent. There is evidence that people value inform ation, although GPs often shy aw ay from a full discussion . 22 Discussion of h o w m em ory im pairm ent and dem entia affect people (especially in terms of em otional or behavioural changes), w hat causes these conditions, w hat can be done to help (both the illness and in terms of support), and the possible tim e course and consequences can be achieved over a num ber of consultations. This process can also be supported by referral to the old age psychiatry service, w hich can co-ordinate further assessm ents, consider the treatm ent options and provide support for the carer.

Case stu d y 7.2 Mr P w as a n ew patient at the practice, adm itted to a local residential hom e for older people. He w as 87 years old, and had b een bereaved 3 years previously. He had had dem entia for about 3 years, and w as no longer able to cope on his ow n. He had on e daughter w h o visited him regularly. The GPs at his n ew practice w ere alerted to his presence w h en a repeat prescription request cam e through w h ich included aspirin, ramipril, m etform in, frusem ide 80 mg per day, oxpentifylline, atorvastatin, am itrip­ tyline 50 mg at night, and tem azepam 2 0 m g at night. The GP visited the patient, w h o se old notes had not yet com e through from the previous practice. He w as pleasantly confused, sitting m ost of the tim e in a chair dozing, w ell-m an n ered w h en w oken, and obese. He had m arked peripheral oedem a but n o other abnorm alities on exam ination. His blood sugars w ere around 10 mmol/1. His blood pressure w as 130/80 mmHg, his pulse rate w as regular, and he w as a little unsteady on his feet. He was not causing the staff any problem s. He w as disorientated w ith regard to season, place and tim e, and appeared cheerful. The GP decided to reduce the tem azepam and the am itriptyline by half, and to stop the atorvastatin (it has unproven benefit at this age and can cause gastrointestinal upset) and

T h e G P and ‘m edicines m a n a g e m e n t’ fo r o ld e r p e o p le w ith m en tal health p ro b le m s

10 7

oxpentifylline (lim ited value), to check Mr P's electrolytes, full blood count and thyroid function tests, and to await the records, arranging to review the patient in 2 w eeks tim e. The notes revealed that he had b een on am itripty­ line for several years for dysaesthesiae affecting his legs and thought to be due to am iodarone that had b een discontinued a few years previously. There w as no history of depression, and he was thought to have congestive cardiac failure. The GP visited again, and the electrolytes, full blood count and thyroid function tests w ere norm al. In him self Mr P w as m uch the same. A lthough h e had peripheral oedem a, his lungs w ere clear and there was no breathlessness or tachycardia. He rem ained unsteady. The GP stopped the tem azepam and the am itriptyline com pletely, and reduced the frusem ide to 40 m g per day, and arranged a follow -up visit in a m onth's tim e. The hom e w as attem pting a w eight-reducing diet, and was also m onitoring sugars. Two w eeks later Mr P's daughter rang the GP to ask w h y the tablets had b een stopped. She said her father w as anxious as he w as used to taking his sleeping tablets, and he had not been sleeping w ell. The GP explained the rationale for the changes, and prom ised to review Mr P soon. He rang the hom e, w h o said that Mr P was not sleeping w ell at night, tending to wander, and becam e anxious at bedtim e w ith ou t his sleeping tablet, but w as m uch m ore alert during the day, less confused, and taking part in som e of the activities to a lim ited exten t during the day. He w as also less unsteady in his gait. The GP planned to review the patient w ith a view to further reduction of m edication if clinically possible. The daughter agreed that the hom e could offer her father a paracetam ol tablet at night. A w eek later the h om e rang the GP to report that Mr P was unsettled at night, getting up and getting dressed, and he was upset about not having his sleeping tablets. The GP agreed to restarting the tem azepam , and arranged to review Mr P in 1 m onth's tim e.

Issues raised by the above case study include the follow ing. • This patient's m edication had clearly not b een adequately review ed for som e tim e. • Should the psychotropic m edication have been w ithdraw n m ore slowly? • Should the GP have discussed the changes in m edication w ith the patient and the daughter before initiating them ? • Is it ethical to offer a paracetam ol tablet at night to allay such a patient's anxieties about m issing his sleeping tablets? • Should the patient rem ain on tem azepam indefinitely?

Special considerations in dementia These include the following: • use of the n ew er antidem entia drug • m anagem ent of behavioural and psychiatric problems.

108

Practical old age p sy c h o p h arm o co lo g y

Use of the newer antidementia drugs: the acetylcholinesterase inhibitors The National Institute for Clinical E xcellence (NICE) has recom m ended that these drugs should be m ade available on the NHS, but only according to strict guide­ lin es . 21 H ow ever, one major study on effectiveness has dem onstrated that approxim ately seven (95% Cl, 4 -1 6 ) people w ould have to be treated w ith donepezil for on e person to benefit in terms of an outcom e based on a clinical im pression of im pairm ent and a carer rating . 23 • The patient m ust have m ild to m oderate Alzheim er's disease, w ith an MMSE score of > 1 2 . • The drug m ust be prescribed by a specialist. • The patient m ust be carefully m onitored after com m encing treatm ent by the specialist services. • If there is no im provem ent after 3 m onths (e.g. on the MMSE score or the activities of daily living score), the drug should be discontinued. • The GP should only take over prescribing w ith an agreed shared care protocol w ith clear end points.

Use of antipsychotics This is a particularly thorny issue and deserves special attention. Up to 60% of residents in care hom es are on psychotropic m edication, and up to 30% are on antipsychotics. The drugs prescribed have the potential to cause serious side-effects, and there is good evidence that this level of prescribing is excessive and causing harm, and that it can be reduced w ith ou t causing significant problem s . 24-29 In our ow n survey of 200 residents in 20 nursing hom es (all non-'Elderly M entally Infirm' care hom es) and residential hom es across Huddersfield, 56% of residents w ere taking a regular psychotropic m edication (60% in nursing hom es and 51% in residential hom es), and of these 31% w ere taking antipsychotics, 2 1 % w ere taking benzodiazepines, and 2 2 % w ere taking antidepressants (13% TCAs vs. 63% SSRIs) . 30 There w as significant cognitive im pairm ent in approxi­ m ately 80% of residents in both nursing and residential hom es, as m easured by capacity to provide inform ed consent. The antipsychotic drug bill is also rising. B etw een 1999 and 2000 there was a 70% increase in the use of atypical antipsychotic drugs for people aged 60 years or over in the UK. Som e of this increase was due to the discontinuation of thioridazine and the introduction of m ore expensive atypical antipsychotics, but nevertheless overall prescribing of all antipsychotic drugs rose by 6 %. In the year 2 0 0 0 -0 1 , antipsychotic prescribing costs rose by £39 m illion (32% ), com pared w ith statins w h ich rose by £103 m illion (3 3 % ) . 8 In the UK there are guidelines available such as Interventions in the Management of Behavioural and Psychological Aspects of Dementia issued by the Scottish Inter­ collegiate G uidelines N etw ork (SIGN) (w w w .sign .ac.u k ) , 31 and the North of England Group guidelines on the primary care m anagem ent of d em entia . 20

W hen to prescribe and when not to It is very com m on for GPs to be presented w ith behavioural problem s of patients, particularly in care hom es. Som etim es this is m anifested as agitation of som e form

T h e G P and ‘m ed icin es m a n a g e m e n t’ fo r o ld e r p e o p le w ith m en tal health p ro b le m s

109

(see below ), and som etim es it is a poor sleep pattern. It is also com m on for GPs to have several older residents on repeat prescriptions for antipsychotic m edication (up to 30% of residents in care h om es). There is often pressure for the GP to prescribe (see Case studies 7.3 and 7.4).

Case stu d y 7.3 In the office at the practice the GP w as given a prescription to sign for Mrs Z, a resident in a local nursing hom e. It w as for zopiclone, lansoprazole, co-codam ol lactulose and ferrous sulphate, n on e of w h ich w as on the patient's list of repeat prescriptions. The GP had a look at Mrs Z's electronic record. She had last b een seen 13 m onths previously for a chest infection, w h en she had b een prescribed som e antibiotics. Her paper notes w ere found. They w ere very thin indeed, w ith a few scribbled entries from m any years previously. The GP rang the m atron at the h om e. It transpired that Mrs Z, a thin, norm ally quiet 74-year-old, had had a fall one Saturday about a m on th ago and fractured her fem ur. She had been discharged to the hom e 2 w eeks ago, on zopiclone from the hospital. She was glad that the GP had telephoned, because Mrs Z seem ed confused at tim es, she had lost w eight, she w as not eating, she w as shouting out at people w h o w eren't there, and had b een trying to hit out at other residents and staff w h o w alked near her. The m atron had b een thinking of asking for a visit. In the office at the practice the GP found a discharge 'flimsy' that had not yet b een filed, w ith illegible writing on it relating to the recent adm ission. The GP visited, and Mrs Z's m edications had all b een started during her recent hospital adm ission. She w as frail, dozing and difficult to rouse. The scar from the internal fixation w as w ell healed, but m ovem en t at the hip joint provoked grim acing and an angry aggressive attem pt to push the GP aw ay. Her abdom en w as distended and tender w ith a suspicion of an epigastric mass, and rectal exam ination revealed a rectum loaded w ith hard faeces. Her chest was clear. She w ou ld not respond to any com m ands and did not say anything. The m atron said that Mrs Z had becom e increasingly tetchy and w ithdraw n prior to her adm ission, w ith a tendency to w ander at night. The GP contacted the hospital. Her haem oglobin level had b een 9.5 postoperatively and she had b een started on iron and lactulose. She had been disturbed at night and had also b een prescribed zopiclone. Her post­ operative electrolytes and sugar levels w ere norm al. The GP arranged for a repeat full blood count, plasma viscosity, ferritin, vitam in B 12 and folate, random plasma glucose, electrolytes, calcium and thyroid function tests. The nurse perform ed an enem a w ith good results. The co-codam ol was changed to paracetam ol, and the zopiclone, lansoprazole and ferrous sulphate w ere stopped. Her haem oglobin level w as 10.2, plasma viscosity w as 1.9 (< 1 .7 2 norm al range), ferritin w as 2 2 (1 5 -1 5 0 ), vitam in B 12 was 185 (2 0 0 -1 1 0 0 ), urea was elevated consistent w ith dehydration, and glucose, calcium and thyroid function tests w ere norm al. She rem ained agitated and restless w ith w hat seem ed to be visual hallucinations, and was striking out at staff w h ile being dressed, fed or bathed. The GP discussed w ith the relatives their expectations and concerns. They just w anted their

I 10

Practical old age p sy c h o p h arm o co lo g y

m other to be com fortable, and they appreciated that she was seriously ill. The GP explained that he though t that Mrs Z probably had an underlying m alignancy, possibly of the stom ach, that she w as suffering from dem entia w h ich w as causing distressing hallucinations (probably accelerated by her recent operation), that she w as too frail to m ove (and this w ould probably cause m ore distress than benefit), and that the aim was to reduce her distress. She w as started on a small dose of risperidone, and her bow els w ere m onitored. Fluids w ere encouraged, together w ith a light diet. She becam e less agitated, started to eat and started to m obilise. She developed som e rigidity and tremor, but no akathisia or tardive dyskinesia. Over the next 3 m onths she rem ained com fortable and less agitated, but she slow ly continued to lose w eight.

Case study 7.4 Mrs N w as discharged from hospital follow ing a hip operation after a fall, back to her nursing h om e. She w as 90 years old, physically quite strong and on no m edication apart from paracetam ol. She had severe cognitive im pairm ent, and the GP was asked to visit because of w orsening agitation, w ith verbal aggression and physical violence towards other residents and staff, w h o she accused of taking her things. Exam ination and blood tests w ere norm al. The agitation and violen ce w ere m ost troublesom e w h en Mrs N w as being dressed and undressed. She had som e d ay-n igh t reversal, w ith a tendency to w ander into other residents' rooms. After discussion w ith relatives she w as started on haloperidol, w hich did calm her agitation, but during the n ext 4 w eeks she becam e increasingly im m obile and restless at rest, persistently tapping her feet and m oving in her seat. On exam ination she was found to have becom e quite rigid w ith a parkinsonian tremor and she exhibited akathisia. She w as started on procyclidine, an anticholinergic, but this led to w orsening confusion and a dry m outh. The local psych o­ geriatrician w as asked for advice. The haloperidol and procyclidine w ere stopped, and he explained that anticholinergics should always be avoided in patients w ith dem entia, and that the new er antipsychotics had few er side-effects, and Mrs N w as started on a small dose of risperidone. He suggested that she should be given m ore tim e w h en dressing and undres­ sing, and allow ed to do as m u ch as she could by herself. The akathisia and parkinsonian sym ptom s abated, she rem ained calmer, and after 2 m onths the risperidone w as reduced and stopped w ith no ill effects.

The introduction to this chapter included a discussion of the evidence for excessive prescribing of antipsychotics to older people (i.e. w h e n there is no valid indication for their use), especially in care hom es. Faced w ith agitated behaviours (agitation can cover a range of issues, including hallucinations, delusions, sleeplessness, excitem ent, hostility, belligerence or em otional instability) it is im portant to exclude underlying, often reversible causes. Iliffe

T h e G P and ‘m edicines m a n a g e m e n t’ fo r o ld e r p e o p le w ith m ental health p ro b le m s

I I I

and D rennan have described a useful m nem onic, nam ely PAID (Physical, A ctivity related, Intrinsic to dem entia, D epression or delusion) for underlying causes (see Table 7.4). Table 7.4 The causes of behavioural problem s (after Iliffe and D rennan19) Physical

Pain (e.g. from osteoarthritis) causing aggression or w andering, or infection (e.g urinary tract infection); also consider reten tion of urine, and constipation

A ctivity related

For exam ple, w h e n a carer is helping w ith an intim ate task such as dressing or bathing; also in response to a lack of activity or com pany

Intrinsic to dem entia

W andering, repetitive stroking of another person, uncontrollable laughter or crying (som etim es seen in vascular dem entia)

D epression or delusions

H allucinations or d elusions w ith paranoia; treat the psychosis itself but be careful w ith m edication (see b elow )

W hen presented w ith a behavioural problem the GP should: • • • • •

exclude other causes as described above obtain a clear description identify contributory factors identify others in volved and w h at they w ere doing consider com ing back to review the problem after a diary has b een kept by the hom e for a short tim e period.

Not all agitation requires an antipsychotic drug, and som e behavioural problem s m ay be am eliorated by the use of skilful approaches that avoid confrontation. These include the following: • calm ness • approaching the patient in their line of vision • doing things at the patient's pace • using non-verbal cues (e.g. sm iling) • providing reassurance. A small dose of an antipsychotic m ay be indicated for som e of the 30-40% of patients w ith dem entia w h o develop psychosis in terms of paranoid delusions or hallucinations associated w ith aggression and violence. Before m edication is considered the GP should establish that these behaviours are actually causing a significant problem , as they do not alw ays do so. System atic review s have suggested that the antipsychotic drugs can have a m odest beneficial effect on 'agitation' in dem entia. The effect is on the behaviours, and the studies did not m easure the effects on the patient's overall w ell-being or any adverse effects. The trials also sh ow a large placebo effect. The antipsychotic drugs include the traditional dopam ine antagonists thiorida­ zine (n ow w ithdraw n), haloperidol and chlorprom azine, as w ell as the n ew er atypical drugs such as olanzapine, sulpiride and risperidone, w h ich have an additional therapeutic serotonin-blocking action. The n ew er drugs cause few er extrapyramidal (parkinsonian) side-effects, although they m ay still cause

I 12

Practical old age p sy c h o p h arm o co lo g y

confusion, sedation and postural hypotension, and there is little evidence for their effectiveness in dem entia even th ough they are w idely and increasingly being prescribed to the older population (see introduction to this chapter). The drugs take up to 1 w eek to have an effect on psychotic behaviours and, given the num ber n eed ed to treat to benefit (NNT) of 5 (i.e. five patients w ould n eed to be prescribed the drug for one to benefit from it), for the majority there w ill be no beneficial effect . 32 This m eans that consideration should be given to stopping the drug after a 4 to 6 -w eek trial. There are no easy-to-use, reliable and objective m easures of behavioural problems, although a simple diary kept by care staff m ight be useful.

Guidelines for the use and monitoring o f antipsychotic drugs33 • The atypical antipsychotics have a better side-effect profile. • Start w ith the sm allest dose prescribable. • Make increases slow ly, w ith careful m onitoring for em ergent side-effects. • You m ay n eed to w ait several w eeks for the full effects to occur. • Only con tinu e if there is dem onstrable evidence of efficacy (this is not easy to obtain; you could consider asking the hom e to m onitor it w ith one of the behavioural scales, such as the 'Nursing Hom e B ehavior Problem Scale'34). • Care should be taken to identify Lewy body dem entia (see below ). • A void the use of anticholinergics. • R eview continued need for treatm ent every 3 m onths. • Encourage trials of drug-free periods to test w hether treatm ent is still needed. • Careful m onitoring of em ergent side-effects should take place at least fortnightly (this requires the patient to be seen and exam ined). Such side-effects include: - confusion - a w orsening of cognitive im pairm ent over tim e - falls - tardive dyskinesia (orofacial or buccolingual m ovem ents, or athetoid m ovem en ts of limbs - due to a rebound dopam ine hypersensitivity; 15% prevalence; m ay w orsen if the drug is w ithdraw n too quickly; no effective treatm ents) - parkinsonian features (tremor and rigidity) - akathisia (restlessness).

Sleep problems A bout a quarter of patients w ith dem entia develop sleep problems. These can range from interrupted sleep to a com plete n igh t-d ay reversal. The GP needs to take the follow ing steps: 1 Establish w hether there really is a problem. • Does the care hom e have unrealistic expectations, putting the patient to bed at 9 pm only for him or her to w ake 6 hours later at 3am, w ell rested and fully awake? • Is the resident him - or herself adversely affected by the change in sleeping pattern? If not, th en the h om e m ay be able to work round the problem, being up w ith the resident, keeping them occupied and then trying them

T h e G P and ‘m edicines m a n a g e m e n t’ fo r o ld e r p e o p le w ith m en tal health p ro b le m s

2

I 13

back to bed in half an hour, rather than trying to keep them in bed throughout the night. H ow ever, for lon e carers this w ould be difficult. Check for any factors that m ay disturb sleep, including the following: • pain • urinary frequency (nocturnal dose of diuretic) • disturbance by others • psychotic sym ptom s • depressed m ood.

If there is a persistent problem w ith sleep w ith ou t psychotic sym ptom s, a short course of a short-acting hypnotic can be tried.

Lewy body dementia Patients w ith Lewy body dem entia are excessively sensitive to antipsychotics. These drugs should therefore be avoided and, if necessary, specialist help sought. The features suggestive of Lewy body dem entia are as follows: • • • •

interm ittent clouding of consciousness visual hallucinations parkinsonian features falls.

Schizophrenia and other psychoses The antipsychotics have an im portant role to play in the m anagem ent of schizophrenia and other forms of psychosis not related to dem entia, and usually in these situations a specialist has initiated the drug. Once the GP takes over prescribing responsibility, th ey m ust be careful not to stop or reduce the m edication injudiciously.

K ey p o in ts • • • •

•

Organise a system for the regular face-to-face m edication review of older patients in care hom es. Be aware of and seek out the early w arning signs of dem entia and depression (think Transport, M edication, Phone, Budget). Avoid using tricyclic antidepressants in older people. A void using antipsychotics for older people. They are rarely indicated and rarely work, and th ey cause side-effects. If they are used, consider a short course only. Seek out the underlying causes of agitation before trying m edication (think Physical, A ctivity, Intrinsic to dem entia, D epression).

References 1 Audit C om m ission (2000) Forget Me Not: m ental health services for older people. Audit Com m ission Publications, London.

I 14

Practical old age p sy c h o p h arm o co lo g y

2 Millard P et al. (1999) Nursing Home Placements o f Older People in England and Wales: a National A udit 1995-1998. D epartm ent of Geriatric M edicine, St George's Hospital M edical School, London. 3 D epartm ent of Health (2000) Prescriptions Dispensed in the Community. Statistics fo r 1989 to 1999: England, Bulletin 2000/02. D epartm ent of Health, London. 4 UK Data Archive (1998) Health Survey fo r England: findings, w w w .d ata-arch ive.ac.u k 5 W alley T and Scott A (1995) Prescribing in th e elderly. Postgrad Med J. 71: 4 6 6 -7 1 . 6 Lunn J, Chan K and D on o h u g h e J (2003) A study of the appropriateness of prescribing in nursing hom es. Int J Pharm Pract. 5: 6 -1 0 . 7 Zermansky A (1996) W ho controls repeats? Br J Gen Pract. 46: 6 4 3 -7 . 8 D epartm ent of Health (2001) Medicines and Older People. Supplem ent to the NSF fo r Older People. D epartm ent of Health, London. 9 Iliffe S and D rennan V (2000) D epression, an xiety and dem entia. In: Primary Care for Older People. Oxford U niversity Press, Oxford. 10 Katona C (1989) The ep id em iology and natural history of depression in old age. In: K Ghose (ed.) Antidepressants fo r Elderly People. C hapm an and Hall, London. 11 Burns A, Lawlor B and Craig S (1999) Assessment Scales in Old Age Psychiatry. Martin D unitz, London. 12 Yesavage J, Brink T and Rose T (2003) D evelop m en t and validation of the Geriatric D epression Screening Scale: a prelim inary report. J Psychiatr Res. 17: 3 7 -4 9 . 13 W orld Health Organization (1992) The ICD-10 Classification o f M ental and Behavioural Disorders. World Health Organization, G eneva. 14 D hondt T, Derksen P, Hooijer C et al. (2003) D epressogenic m edication as an aetiological factor in major depression: an analysis in a population of depressed elderly people. In t J Geriatr Psychiatry 14: 8 7 5 -8 1 . 15 Departm ent of Health (1999) A N ational Service Framework for M ental Health. D epartm ent of Health, London. 16 D on ogh u e J, Katona C and Tyle A (1998) The treatm ent of depression: antidepressant prescribing for elderly patients in primary care. Pharm J. 260: 5 0 0 -2 . 17 Katona C and L ivingstone G (2004) Drug Treatment in Old Age Psychiatry. Martin Dunitz, London. 18 M cCormick A, Flem ing D and Charlton J (1995) Morbidity Statistics in General Practice: fourth national study 1991-1992. HMSO, London. 19 Iliffe S and D rennan V (2001) Primary Care and Dementia. Jessica K ingsley Publishers, London. 20 Eccles M, Clarke J, L ivingstone M, Freem antle N and M ason J (1998) North of England E vidence-based G uidelines D evelop m en t Project: guidelines for the primary care m anagem en t of dem entia. BMJ. 317: 8 0 2 -8 . 21 N ational Institute for Clinical E xcellence (2003) Guidance on the Use o f Donepezil Rivastigmine and Galantamine fo r the Treatment o f Alzheim er's Disease. N ational Institute for Clinical E xcellence, London. 22 M aguire C, Kirby M, Coen R et al. (2003) Family m em bers' attitudes towards telling the patient w ith A lzheim er's disease their diagnosis. BMJ. 313: 5 2 9 -3 0 . 23 Rogers S, Farlow M, D oody R et al. (2003) A 2 4 -w e e k double-blind, placebo-controlled trial of donepezil in patients w ith A lzheim er's disease. Neurology. 35: 1 3 6 -4 5 . 24 W axm an H, Klein M and Carner E (1985) Drug m isuse in nursing hom es: an institutional addiction? Hosp Commun Psychiatry. 36: 8 8 6 -7 . 25 McGrath AM and Jackson GA (1996) Survey of neuroleptic prescribing in residents of nursing h om es in Glasgow. BMJ. 312: 6 1 1 -1 2 . 26 Ruths S, Straand J and Nygaard HA (2001) Psychotropic drug use in nursing h om es diagnostic indications and variations b etw een institutions. E ur J Clin Pharmacol. 57: 52 3 -8 .

T h e G P and ‘m edicines m a n a g e m e n t' fo r o ld e r p e o p le w ith m ental health p ro b le m s

I 15

27 Furniss L, Burns A and Craig S (2000) Effects of a pharm acist's m edication review in nursing hom es. R andom ised controlled trial. Br J Psychiatry. 176: 5 6 3 -7 . 28 Thapa PB, M eador KG, Gideon P, Fought RL and Ray W A (1994) Effects of a n ti­ psychotic w ithdraw al in elderly nursing h o m e residents. J A m Geriatr Soc. 44: 2 8 0 -6 . 29 Hopker S (2003) Drug Treatments and Dementia. Jessica K ingsley Publishers, London. 30 D em psey OP, M oore H and M ulligan E (2003) A survey o f psychotropic prescribing and behavioural problems in residential and nursing homes. U npublished paper. 31 Scottish Intercollegiate G uidelines N etw ork (1998) Interventions in the M anagement o f Behavioural and Psychological Aaspects o f Dementia. Scottish Intercollegiate G uidelines N etwork, Edinburgh. 32 Schneider LS, Pollock V and Lyness S (1994) A m eta-analysis of controlled trials of neuroleptic treatm ent in dem entia. J A m Geriatr Soc. 38: 5 5 3 -6 3 . 33 Howard R, Ballard C, O'Brien J et al. (2001) G uidelines for the m an agem en t of agitation in dem entia. In t J Geriatr Psychiatry. 16: 7 1 4 -1 7 . 34 Ray WA, Taylor JA, Lichtenstein MJ and M eador KG (1992) The N ursing Hom e B ehavior Problem Scale. J Gerontol. 24: M 9 -1 6 .

Chapter 8

Management of geriatric mood disorders Naila Jawaid and Robert C Baldwin

Unipolar depression Introduction Despite its prevalence and seriousness, depressive disorder in older people rem ains under-treated . 1,2 K now ledge of the efficacy of antidepressant treatm ent in the elderly has b een restricted by the exclusion from m ost antidepressant drug trials of the very old, and those w h o are frail and/or w ith cognitive im pairm ent, w h ich are the majority of those seen in the routine practice of old age psychiatry. The optim al treatm ent of depression in later life is crucial, and requires appreciation of several age-related factors such as comorbidity, polypharm acy, altered drug kinetics , 3 variable treatm ent response and increased predisposition to side-effects .4,5 This chapter review s the develop m ent and use of antidepressants in older depressed patients. The relevant literature on neurochem istry, kinetics, dosing, efficacy and differential side-effect profiles of various classes of drugs w ill be review ed, w ith special em phasis on geriatric data.

Subtypes of depression and treatment modalities There is insufficient space to discuss the nosology of m ood disorders in older people. H ow ever, there is som e relationship b etw een different types of depressive disorder and treatm ent options, sum m arised in Table 8 . 1 . 6 In dysthym ia there are the sam e sym ptom s as for major depression, but few er of them and w ith a duration of at least 2 years. D ysthym ia in older people is often associated w ith poor health. 'Minor' depression is still an experim ental category, but it seem s to be especially com m on in older people and is certainly not a trivial condition. It is defined in the same w ay as dysthym ia, but w ith a m uch shorter duration and often w ith additional sym ptom s such as anergia, poor m otivation and poor cognition. The only evidence that pharm acological treatm ent works com es from a study in w hich sym ptom s w ere present for at least 4 w eeks and w ere associated w ith functional declin e . 7

I 17

118

Practical old age p sy c h o p h arm ac o lo g y

T able 8.1 Sum m ary of evidence: treatm ent m odality and type of depression 6 Type o f depression

Treatment modality

Psychotic depression

C om bined antidepressant/antipsychotic or electroconvulsive therapy (ECT)

Severe (non-psychotic) depression

C om bined antidepressant and psychological therapy (e.g. cognitive behavioural therapy, problem -solving, interpersonal therapy or brief p sychodynam ic psychotherapy)

M oderate depressive episode

A ntidepressant or psychological therapy (as above)

D ysthym ia

Consider an antidepressant

R ecent onset sub-threshold (m inor) depression (1^1 w eeks)

W atchful w aiting

Persistent sub-threshold (m inor) depression w ith functional im pairm ent ( > 4 w eeks)

Consider an antidepressant

Assessment This includes a detailed history from both patient and inform ant covering aspects such as recent change in m ood state, physical illness, drug and alcohol use, adverse life events and available support system s. A m ental state exam ination w ith em phasis on suicidal ideation and plans, delusional p h en om en a and cognitive im pairm ent (w herever possible using a standardised test such as the M ini-M ental State Exam ination, MMSE8) is an essential com ponent of the assessm ent process along w ith relevant physical exam ination and laboratory investigations. An issue of particular im portance is the assessm ent of suicidal risk. A ny deliberate self-harm in an older person should be taken very seriously, as depressive disorder often underlies it.

Measuring depression in older people Screening for depressive disorder in older people can be achieved w ith the Geriatric Depression Scale (GDS ) , 9 w h ich has b een validated and w idely used in elderly subjects. It is quick to adm inister and exists in several versions (ranging from 4 to 30 item s). It has b een translated into a num ber of languages and further inform ation can be obtained from its public-dom ain w ebsite (h ttp://stanford.edu/~yesavage/G D S.htm l). The severity of a diagnosed case of depression can be assessed w ith the M ontgom ery-A sberg Depression Rating Scale (MADRS ) 10 or the Ham ilton Rating Scale for Depression (HRSD ) . 11 Both have b een validated in elderly depressed subjects , 12 although som e clinicians find that the MADRS has greater face validity in patients w ith com orbid m edical conditions. Both are useful in assessing response to treatm ent.

Principles o f drug treatment The first challenge is to break d ow n the barriers w hich prevent treatm ent from being instituted in the first place. All treatm ent options available to depression

M a n ag em en t o f geriatric m o o d d iso rd e rs

I 19

earlier in life rem ain effective in old age and, as w ith younger people, therapy in m ore than one m odality m ay be m ore effective and appropriate than one m odality a lo n e . 13,14 Education about depressive disorder and its treatm ents is vital in building a trusting and therapeutic relationship w ith the patient. It m ay cover the following: • • •

counselling the patient that depression is a treatable disorder explaining that antidepressants are non-addictive discussing the im portance of com pliance (taking tablets regularly, not m issing tablets, no sudden discontinuation) • em phasising to the patient that th ey m ust not stop treatm ent w h en recovery has occurred • explaining that acute treatm ent in older people m ay typically last for 6 - 8 w eeks at the m inim um .

Pharmacotherapy There are n o w m any antidepressants available, so it is im perative to operate w ith in som e general guidelines. The choice of antidepressant should be guided by factors such as type and severity of depression, tolerability, side-effects and contraindications to the drug. Antidepressants available in the UK can be classified as sh ow n in Box 8.1.

B ox 8.1 Classification of antidepressants Tricyclics Secondary am ines (nortriptyline, desipram ine) Tertiary am ines (im ipram ine, am itriptyline, dosulepin, clom ipram ine) N ew er tricyclics (lofepram ine) Monoamine oxidase inhibitors Irreversible (phenelzine) Reversible (m oclobem ide) Selective serotonin reuptake inhibitors (SSRIs) (fluvoxam ine, fluoxetine, paroxetine, sertraline, citalopram, escitalopram) Serotonin/noradrenaline reuptake inhibitors (SNRIs) (venlafaxine, duloxetine) Noradrenaline reuptake inhibitors (NARIs) (reboxetine) Noradrenaline and specific serotonin enhancers (NASSs) (mirtazepine) Atypical antidepressants (trazodone, m inaserin) Others

(L-tryptophan) (licensed use only)

12 0

Practical old age p sy c h o p h arm ac o lo g y

Table 8.2 lists antidepressants according to their m ain m ode of action, side-effects, starting doses and average daily dosages. Altered pharm acokinetics (see below ) in older people m ean that for the tricyclics (but not for SSRIs and m any of the new er drugs) there is considerable inter-individual variation in h o w these drugs are handled, and therefore predicting the final dose can be difficult. Furthermore, a dose-titration period is necessary, w h ich w ill len gth en treatm ent trials compared w ith new er antidepressants. Table 8.2 Side-effect profiles and dosages of the m ain antidepressants (Baldwin et al.6) Starting dosage (mg)

Average daily dose (mg)

Drug

M ain mode o f action

A n ti­ cholinergic

A n ti­ histaminic

otladrenergic blocking

Am itriptyline

N E++

++++

++++

++++

2 5 -5 0

7 5 -1 0 0

+++

++

+++

25

7 5 -1 0 0

+++

++

++

1 0 -3 0

7 5 -1 0 0

+++

++

++

5 0 -7 5

7 5 -1 5 0

5HT+ Im ipram ine

N E++ 5HT+

Nortriptyline

NE++

D othiepin (D osulepin)

N E++

M ianserin

a2

0 /+

+++

0 /+

30

3 0 -9 0

Lofepramine

N E++

+

+

+

7 0 -1 4 0

7 0 -2 1 0

5HT+

5HT+

5HT+ 100

Trazodone

5HT2

0

+++

+

F luvoxam ine

5HT

0 /+

0 /+

0

2 5 -1 0 0

300 1 0 0 -2 0 0

Sertraline

5HT

0 /+

0

0

2 5 -5 0

5 0 -1 0 0

F luoxetine

5HT

0 /+

0

0

10

20

P aroxetine

5HT

0 /+

0

0

1 0 -2 0

2 0 -3 0

Citalopram

5HT

0 /+

0

0

1 0 -2 0

2 0 -3 0

Escitalopram

5HT

0

0

0

5 -1 0

0 -2 0

P henelzine

M AO-I (nonre versible)

0 /+

0

++

M oclobem ide

RIMA

0 /+

0

0

V enlafaxine XR

N E+

0 /+

0

0 /+

M irtazapine

«2

0

++

0

15

3 0 -4 5 (divided dosages)

300

3 0 0 -4 0 0

2 5 -7 5

7 5 -2 0 0

5H T ++ 7 .5 -1 5

1 5 -3 0

5HT2

5HT, serotonin reuptake inhibitor; NE, norepinephrine (noradrenaline) reuptake inhibitor; DA, dopamine reuptake inhibition; RIMA, reversible inhibitor of monoamine oxidase ihibitor a 2 antagonism at the presynaptic a 2 receptor. a 2 + /+ + indicates magnitude of effect from small (+) to marked (++++).

M a n ag em en t o f geriatric m o o d d iso rd e rs

12 1

Tricyclic antidepressants The majority of tricyclic antidepressants have three rings in the m olecule, w ith tw o b en zen e rings fused to a central seven-m em bered carboxylic or heterocyclic ring. In attem pts to avoid som e of the side-effects, com pounds w ith differing ring structures, ranging from m onocyclic to tetracyclic, w ere synthesised. All of these com pounds are inhibitors of noradrenaline uptake, w ith a variable potency of inhibition of 5-hydroxytryptam ine (5-HT) uptake as w ell. Tricyclics are som e­ tim es subdivided into secondary and tertiary am ines (see Box 8.1). Secondary am ines are predom inantly noradrenaline reuptake inhibitors and have a m ore benign side-effect profile. Pharmacokinetics and pharmacodynamics

Tricyclic antidepressants are by and large w ell absorbed orally. They are lipophilic and w id ely distributed in the body, being strongly bound to plasma proteins and tissue constituents. They are m etabolised in the liver to a variety of m etabolites, som e of w hich are active antidepressant com pounds. The major route of elim ina­ tion is via renal excretion. M inor am ounts enter the bile and are th en excreted in the faeces. The low er rates of both hepatic m etabolism and renal clearance in this age group need to be taken into account w h en dosing, otherw ise considerably higher than expected steady-state plasma levels m ay develop in older patients. For this reason, som e clinicians have recom m ended plasma level m onitoring for tricyclic antidepressants in elderly patients. The new er tricyclic, lofepramine, has m inim al anticholinergic properties and cardiac toxicity in older patients com pared w ith older drugs. Inter-individual differences usually account for the majority of the variance in kinetic parameters, rather than the effect of ageing perse.15 Side-effects

(see Table 8.2)

Drugs w ith tertiary am ine structure tend to produce m ore antagonism of ax-adrenergic receptors w ith subsequent hypotension, blockage of histam ine receptors (leading to sedation) and blockage of cholinergic receptors (causing dry m outh, blurred vision, urinary retention, dizziness, tachycardia, m em ory im pairm ent and, at high and toxic doses, delirium ). The tendency to produce orthostatic h ypotension is a serious side-effect in the elderly that leads to an increased risk of falls and limb fracture.

Monoamine oxidase inhibitors (MAOIs) Traditional MAOIs bind non -selectively, either irreversibly or alm ost so, to both type A and type B m on oam in e oxidase enxym es to prevent m onoam ine destruction. Because th ey lack selectivity, traditional MAOIs can lead to serious side-effects w h e n food rich in tyram ine or other am ines is ingested. In addition, these drugs sh o w strong interactions w ith other drugs that seriously lim it their use com pared w ith other antidepressants. They m ay have a place in the treatm ent of resistant depression , 16 but they are hardly ever used as first-line drugs now adays and w ill therefore not be discussed further here. Reversible MAOIs, represented by m oclobem ide, have considerably few er diet­ ary constraints and problem atic drug interactions. M oclobem ide achieves this by reversibly inhibiting largely m on oam in e type A enzym e. The MAOIs are all rapidly absorbed. The reversible ones have shorter half-lives and require m ultiple daily

12 2

Practical old age p sy c h o p h arm ac o lo g y

doses. A period of 2 w eeks m ust elapse w h en irreversible MAOIs are stopped before starting other drugs that m ay interact w ith the MAOIs and w ith SSRIs.

Selective serotonin reuptake inhibitors (SSRIs) SSRIs are th ird -g en era tio n an tid ep ressan ts w ith a sp ecific effect o n sero to n in reup take, m ak in g th e m less lik ely to cau se th e sid e-effects en c o u n te r e d w ith th e u se of tricyclics. The m a in class m em b ers are rep resen ted by flu v o x a m in e, flu o x e tin e , p a ro x etin e, sertraline, citalopram an d escitalop ram . The latter is th e active s-en a n tio m er of Rs-citalopram. The R -enantiom er is essen tia lly p h a rm a co ­ lo gically in active. E scitalopram has b e e n d em o n stra ted to sh o w greater efficacy an d to p rovid e a m ore rapid resp o n se th an cita lo p ra m .17 Pharmacokinetics and pharmacodynamics

SSRIs are rapidly absorbed from the gut and m etabolised to several m etabolites (often w ith antidepressant activity), each w ith a different drug half-life. All SSRIs are extensively protein-bound and have a large volu m e of distribution. For exam ple, the plasma half-lives of fluvoxam ine and fluoxetine are relatively unaffected by age, w hereas the steady-state plasm a concentrations of paroxetine, sertraline and citalopram are higher in the elderly and the elim ination half-life is longer than for the sam e dose in adult patients . 18 Paroxetine also has a notable affinity for m uscarinic cholinergic receptors, thus m aking it the m ost sedative of the SSRIs. Adm inistration of all SSRIs can augm ent TCA activity, as w ell as the activity of other drugs m etabolised by the hepatic cytochrom e P450 system . SSRI interactions w ith the P450 system appear to be dose dependent and are im portant in elderly people, w h o are m uch m ore likely to be on concurrent m edication. Side-effects

The m ost com m only reported side-effects of SSRIs include nausea, agitation, anxiety, headache, sleep disturbance and tremor. Less frequently there are changes in appetite and w eight, dry m outh, sw eating, w eight loss and sexual dysfunction, especially anorgasmia. The SSRIs have a major advantage in that they lack the cardiotoxicity and major cholinergic and adrenergic side-effects of tricyclics. W ith­ drawal reactions w ith som e SSRIs can be severe and prolonged . 19-21 They are usually characterised by rebound anxiety, dysphoria, insom nia and parasthesiae, but are relatively short-lived. M ost problem s arise w ith SSRIs that have short half-lives. T able 8.3 Cytochrom e P450 isozym es inhibited by SSRI antidepressants (in vitro) 3A4

D ru g

1A 2

2C 9

2C 19

2D 6

Fluoxetine

+

++

+

+++

++

Sertraline

+

+

+

+

+

Paroxetine

+

+

+

+++

+

Citalopram

+ ++

+++

+

++

Escitalopram Fluvoxam ine

+++

For more details see Greenblatt et al.22 and von Moltke et a l ,23

M a n ag em en t o f geriatric m o o d d iso rd e rs

12 3

SSRIs such as paroxetine and flu oxetin e are inhibitors of P450 isoenzym es (CYP 2D 6), w h ich can considerably influ ence the m etabolism of tricyclic anti­ depressants and other m edication (see Table 8.3).

Other antidepressants Venlafaxine selectively inhibits the uptake of serotonin and noradrenaline, and in contrast to tricyclics show s no affinity for other neuroreceptors. No dosage adjustm ents are recom m ended for older people. Compared w ith SSRIs, venlafax­ ine affects the cytochrom e P450 enzym e system relatively little. The m ost com m on side-effects are nausea, headache, dry m outh, dizziness, constipation and hypertension, although postural hyp otension can be a problem in older people. To m inim ise the risk of discontinuation sym ptom s, the dose of venaflaxine should be gradually tapered over a period of a few w eeks. The possibility of serotonergic syndrom e should be taken into account w h en venlafaxine is co ­ adm inistered w ith other serotonergic drugs . 24 SSRIs and venlafaxine are also associated w ith hyponatraem ia in elderly inpatients . 25 D uloxetine is a n ew er SNRI w hich m ay have a safer cardiovascular profile than venlaxafine but data on older patients are awaited. Mianserin and mirtazapine both have potent effects on antagonising a 2-adrenergic auto- and heterorecepters. This leads to an increase in noradrenaline release. These drugs have a lo w affinity for m uscarinic, cholinergic and dopam ine receptors, and this in turn is related to reduced side-effects. M irtazapine has effects on both the noradrenergic and serotonergic system s. It enhances central noradrenergic and 5H T!-receptor-m ediated serotonergic neurotransm ission. It is an antagonist at a 2-adrenergic presynaptic autoreceptors, and it also blocks 5HT2 and 5HT3 receptors. This results in increased noradrenergic release. Differences in pharm acokinetics w ith age are considered to be m inor . 26 M ianserin has fallen out of favour because of the small risk of agranulocytosis necessitating blood m onitoring. The m ain adverse effects of m irtazapine are sedation and w eight gain. Reboxetine is a specific noradrenaline reuptake inhibitor. It has not b een recom m ended for use in older people in the UK because there are insufficient efficacy data for the elderly. There have b een som e uncontrolled efficacy studies, and it appears to be w ell tolerated and safe in this patient group .27

Efficacy o f antidepressants There have b een a lim ited num ber of double-blind, placebo-controlled trials of antidepressants involving elderly people, and m ost of these are of tricyclics. M ost of the efficacy data in volvin g double-blind, placebo-controlled trials is for tricyclic drugs .28,29 N ew er drugs usually involve head-to-head com parisons w ith estab­ lished drugs. In all age groups the use of tricyclic drugs to treat depression has b een declining in favour of SSRIs, although view s differ as to w h eth er SSRIs should be used as first-line treatm ent in depression . 30 The assessm ent of efficacy has b een m ade easier by the publication of tw o recent system atic review s of antidepressants in older people, nam ely a Cochrane R eview 29 and a system atic review of num ber needed to treat (NNT) by Katona and L ivingston . 31 The NNT is the num ber of patients w h o n eed to be given the treatm ent in question com pared w ith another treatm ent or placebo in order for

12 4

Practical old age p sy c h o p h arm ac o lo g y

one further patient to gain benefit. The low er the NNT, the more efficacious the drug. In the Cochrane R eview by W ilson et a l ,29 the NNT for tricyclics was 4 (a figure as good as m any other treatm ents in m edicine), and for SSRIs it was 7, although only tw o trials m et the criteria for the analysis (both of fluoxetine). In the study by Katona and Livingston , 31 m ost antidepressant trials show ed efficacy, but this w as less clear for m oclobem ide and flu oxetine. In head-to-head com parisons b etw een antidepressants there w as a trend for the superiority of SSRIs and venlafaxine over tricyclics, but m ost of the trials w ere under-pow ered statistically. There is som e evidence that in m elancholic depressive disorder the tricyclics are superior to SSRIs, but there are no data specific to older patients . 32

Efficacy in special patient groups Depression associated with systemic disease

A m eta-analysis by Gill and Hatcher 33 has sh ow n that antidepressants are effective in the m anagem ent of depression com plicating a variety of system ic illnesses. The NNT w as the sam e as for tricyclic antidepressants in the m eta­ analysis of older patients conducted by W ilson et al.29 SSRIs are preferred because of their favourable side-effect profile in depression that is comorbid w ith medical disorder. Cole et al.34 calculated that 38-87% of elderly m edically ill inpatients have contraindications to the use of older heterocyclic antidepressants, but only 4% have contraindications to SSRIs. Dementia

Studies sh ow a high rate of spontaneous recovery in m ild to m oderate depression com plicating dem entia. Current advice suggests providing support to the patient and caregiver w ith a review in 2 to 4 w eeks, rather than initiating antidepressants . 6 For m oderate to severe cases of depression com plicating dem entia there is evidence of antidepressant efficacy. For exam ple, citalopram w as found to im prove a variety of em otional sym ptom s, including depression in dem entia , 35,36 w h ile m oclobem ide significantly im proved depressive sym ptom s in those w ith d em entia . 37 Fluoxetine treatm ent w as sh ow n to be effective in patients w ith severe physical illn ess . 38 In an open-label study by Zimmer 39 involving 18 geriatric patients w ith concurrent cardiovascular and cerebrovascular illness, depression w as successfully and cost-effectively treated (78% of patients responded) w ith venlafaxine at dosages of 5 0 -2 6 2 .5 m g/day. The logistic prob­ lem s inherent in such studies m ean that large placebo-controlled studies of such patients are unlikely to be feasible. Stroke

In a review of antidepressants in stroke, Gustafson et al.40 do not recom m end the use of the older tricyclic antidepressants because of the high rate of contra­ indications and side-effects, including delirium . As w ith dem entia, there is a high rate of spontaneous recovery, but this dim inishes rapidly after 6 w eek s .41 Citalopram 42 and flu oxetin e have b een sh ow n to be m ore effective than placebo .43 Nursing home residents

Few studies have been conducted on this population. This represents an im port­ ant gap in know ledge, as these patients are likely to encom pass the m ost difficult to treat, nam ely the very old and the very frail. Tricyclic antidepressants have

M a n ag em en t o f geriatric m o o d d iso rd e rs

125

b een sh o w n to be effective but associated w ith unacceptably low tolerability . 13 In an open study of SSRIs by Trappier and C ohen , 44 the response rate was poor if the patient had associated dem entia, but otherw ise it w as good. Likewise, in an open pilot study of sertraline in patients w ith m inor (sub-threshold) depression, outcom e and tolerability w ere g ood .45 Current best practice suggests that the patient should be treated w ith an SSRI or m oclobem ide, starting w ith a lo w dose and titrating upwards. An alternative is to deploy psychosocial interventions. In one study a socialisation program me was associated w ith short-term b en efit . 46

Treatment o f major depressive episode The pharm acological m anagem ent of depressive disorder is divided into three phases, n am ely acute treatm ent, continuation therapy and m aintenance treat­ m ent. These w ill be discussed in turn.

Acute treatment phase Antidepressants are usually the first-line treatm ent in patients w ith m oderate and severe depressive episode. In general, no antidepressant drug is clearly m ore effective than another, but rather antidepressants should be tailored to the patient, taking account of the likely side-effects and tolerability. Older anti­ depressants should be avoided in patients at risk of suicide. Therapeutic response m ay occur w ith in 2 w eeks. Likewise, if there has been no response w ithin 4 w eeks recovery is u n lik ely . 47 Once recovery has started it m ay take up to 1 2 w eeks for a full response (i.e. longer than in younger patients).

Continuation phase C ontinuation drug therapy reduces the risk of relapse after rem ission. It is not a fixed period, but in older people a 1 2 -m onth period of continuation w ith antidepressants is recom m ended, in contrast to a 6 -m onth period for younger patients . 6 H ow ever, patients w ith recurrent depression can be treated for 2 years . 6 For patients w ith delusional depression on antipsychotic m edication, it is recom m ended that this be continued for 6 m onths before being tailed off.

Maintenance phase Major depression often follow s a recurrent course, and older people benefit from m aintenance therapy even after a first episode of depression. Evidence com es from several studies. A placebo-controlled trial of dothiepin (dosulepin) show ed that over a 2 -year period the relapse rate w as reduced by a factor of 2.5 in the actively treated group . 48 In another study, by Reynolds et al . , 14 m ost protection was conferred by a com bination of nortriptyline, a secondary am ine tricyclic, and interpersonal therapy. Patients w h o received placebo and usual care had a very high rate of relapse/recurrence. Citalopram, an SSRI, prevented recurrence over a period of 1-2 years, suggesting that a protection effect is not confined to tricyclic antidepressants . 49 Extrapolating from studies of younger patients, the risk of relapse is increased if there are residual sym ptom s or if chronic life stresses exist . 50 W ilson et al.29 did

126

Practical old age p sy c h o p h arm ac o lo g y

not find any benefit from sertraline in prophylaxis. Possibly the dose used was too small, otherw ise this study cautions against assum ing that all antidepressants are equally effective in prevention. The case for long-term antidepressant treatm ent needs to be balanced against adverse effects, w h ich for the older antidepressants can include troublesom e w eight gain, tooth decay and cardiovascular disturbance.

Resistant depression At least a third of patients fail to respond to first-line antidepressant therapy . 51 tn such patients it is im portant to first review the diagnosis, rule out undetected system ic disease or dem entia, ensure treatm ent adequacy in terms of duration and com pliance, and attend to the possibility of psychosocial reinforcers. This has b een described as a stepped care approach, m eaning that each step of treatm ent follow s a logical sequence rather than a chaotic m ishm ash . 52 This has b een reinforced in a study by Flint and Rifat, 53 in w h ich over 80% of patients recovered if a logical process w as adopted - for exam ple, starting w ith a trial of anti­ depressant m onotherapy and even tu ally m oving up to electroconvulsive therapy (ECT). The various strategies for treating resistant depression include the follow ing. • Increasing the dose. This m ay be beneficial for TCAs and venlafaxine, but there is little evidence that increasing the dose significantly alters the outcom e in the case of SSRIs. 54 • Prolonging the length of treatment. Here the therapeutic trial is continued beyond w hat is usually considered to be an 'adequate' period of tim e. There is evidence that an exten sion to 9 w eek s or even longer m ay im prove recovery from a depressive episode in older patients , 16 but not if there has been no im prove­ m ent in the first 4 w e e k s .47 • Switching antidepressant class. Sw itching b etw een classes of antidepressants has b een sh ow n to have a m odest benefit, although there has b een little system atic evaluation of this. The m ost popular strategy is a change from one antide­ pressant to another from a different class (e.g. from a TCA to an SSRI, or vice versa ) . 55 Use of the traditional MAOIs has b een m entioned. V enlafaxine show s som e prom ise in resistant depression, but evaluation is m ostly confined to clinical experien ce 56 or open trials . 57 It has been estim ated that the overall success rate for class sw itching is about 50% .54 • Augmentation. M ost of the evidence is for lithium augm entation . 54 Austin et a l conducted a m eta-analysis of five small double-blind trials and found that 18 out of 50 of patients augm ented w ith lithium responded, com pared w ith 6 out of 49 of those w h o w ere given a placebo . 58 In another study , 59 venlafaxine and lithium augm entation w as studied in an open trial in outpatients aged 18-70 years. At the end of the 7 -w eek study, 35% of the patients show ed more than a 50% reduction in depressive sym ptom s, and the com bination was w ell tolerated. Other augm entation approaches include the use of neuroleptics, anti-epileptics, thyroid horm one and L-tryptophan, but the evidence base is too small to allow recom m endations to be made.

M a n ag em en t o f geriatric m o o d d iso rd e rs

12 7

Prevention Depression in old age frequently follow s a chronic and relapsing course. As endorsed by the US N ational Institute of H ealth , 60 a major goal of the treatm ent is to prevent relapse, recurrence and chronicity. Antidepressant pharm acotherapy is the m ainstay of this therapeutic goal. Studies published to date have established the m aintenance efficacy of TCAs and som e SSRIs, as discussed above.

Bipolar affective disorders Introduction Bipolar disorder affects approxim ately 0.1% of individuals over the age of 65 years, com pared w ith 0.4% of those aged 4 5 -6 4 years, and 1.4% of those aged 1 8 -4 4 years . 61 Little is kn ow n about the course of illness in elderly patients, but it is associated w ith significant m orbidity, high rates of mortality and considerable use of m ental health services . 62 Elderly people w ith bipolar disorders fall into tw o clinically distinct subgroups, n am ely early and late onset, each w ith different aetiologies and o u tco m es . 63 The first group has an initial onset w ith depressive episode in m iddle age, w hich subsequently converts to m ania in late life, w hereas late-onset cases are som etim es term ed secondary m ania. Krautham mer and K lerm an 64 described secondary m ania as a syndrom e w ith m ultiple causes, including drugs (steroids) and m etabolic, infectious and structural cerebro­ vascular diseases such as stroke and tum our. Certainly the onset of m ania in later life should prom pt a search for underlying, often cerebral pathology. The course and presentation of the illness m ay be less typical, w ith a picture of pseudodem entia, confusion or even depressed m ood . 65 Several studies suggest that bipolar disorder m ay becom e m ore persistent and less responsive to treatm ent in later life, predisposing patients to frequent relapses . 66

Treatment As w ith depressive disorder, establishing and m aintaining a therapeutic alliance is vital. Providing education about bipolar disorders, enhancing treatm ent com pli­ ance, identifying n e w episodes early on and reducing the morbidity and sequelae of bipolar disorder are all im portant goals.

Pharmacological treatments M edication for patients w ith bipolar disorders includes those drugs that decrease sym ptom s of m ania or depression, those that prevent further episodes and those that are helpful at various tim es during the course of the disorder. These m edications m ay be categorised as mood stabilisers (e.g. lithium , carbamazepine, sodium valproate, lam otrigine), neuroleptics (typicals and atypicals), benzodiaze­ pines and antidepressants. Lastly, there are n ovel and adjunctive treatm ents such as calcium -channel blockers, clozapine, psychostim ulants, thyroxine and ECT.

128

Practical old age p sy c h o p h arm ac o lo g y

These, together w ith benzodiazepines and antidepressants, w ill not be considered in detail here. There have been few prospective studies of the treatm ent of m ania in older patients. A num ber of issues relating to efficacy, side-effects and optim al blood levels of m ood stabilisers rem ain unresolved and await further study. M eanw hile, w e have to rely on studies conducted on a younger population. Because the treatm ent of m ania is m ade difficult by reduced tolerability of drugs such as haloperidol, n e w treatm ents such as atypical antipsychotics and the an ticon vu l­ sants take on a m ore im portant role . 67 The three m ain m ood stabilisers currently available, nam ely lithium , valproate and carbamazepine, have b een studied and used in the treatm ent of all phases of bipolar disorder.

Lithium

Lithium has b een sh ow n to block the m etabolism of the intracellular second m essenger in o sito l-1 ,4 ,5 -triphosphate, w hich is involved in the increase in ionic intracellular calcium concentration, and w h ich in turn triggers neurotransm itter release and other cellular changes in secretory cells. Its narrow therapeutic w in d o w necessitates frequent blood m onitoring, and discontinuation m ay pre­ cipitate recurrence, a serious problem in the poorly com pliant patient. Lower dosages are required in older people because of the com bination of a reduced volu m e of distribution and decreased renal clearance . 68 There have been no controlled prospective studies of the treatm ent of m anic behaviour in the elderly, but in case studies 69 and retrospective trials 70 lithium has b een found to be effective. It has b een suggested that lithium m ay be m ore efficacious than anticonvulsants (see below ) for the treatm ent of classic m ania w ith m inim al or no neurological signs . 71 It is generally held that older patients m anifest toxic effects at low er co n cen ­ trations . 72 Sym ptom s include polyuria, gastrointestinal abnorm alities, n eu ro­ m uscular effects, ataxia, cognitive im pairm ent and tremors (see Table 8.4). In the longer term, w eight gain, oedem a and psoriatic rashes can be troublesom e. B ow d en 73 concludes that patients w h o respond best to lithium have mild and uncom plicated illness, pure m anic sym ptom atology and infrequent episodes, w hereas those w ith secondary m ania or substance abuse, and adolescent and elderly groups respond less w ell. The optim um lithium plasma concentration for m aintenance therapy for the elderly using prospective studies has not been definitely determ ined, although a lithium range of 0 .4 -0 .7 m E q /l has b een proposed . 54 Prior to com m encing lithium , a typical range of tests (see Table 8.4) should include renal function, electrolytes, electrocardiogram (ECG/EKG), thyroid function and blood glu cose . 74 Typical starting doses are 1 0 0 -2 0 0 mg daily.

Valproate

In a random ised, double-blind, placebo-controlled study of 179 younger m anic patients, B ow den et al.75 dem onstrated that valproate is effective in the treatm ent of m anic episodes in bipolar patients, although possibly it is not as effective as lithium . Valproate seem s to have a broad spectrum of activity, and has significant anti-m anic action in m ixed affective state, rapid-cycling bipolar disorders and

M a n ag em en t o f geriatric m o o d d iso rd e rs

I 29

secondary m ania, w h ich is particularly relevant to the elderly population . 76 It has therefore been suggested that valproate (either as sodium valproate or as divalproex) is m ore effective than lithium in the treatm ent of non-classic m ania and secondary m an ia . 71 In a few open and retrospective case studies, valproate w as w ell tolerated and efficacious . 77 Prior to com m encing valproate, liver enzym e tests, a full blood count w ith platelets and an electrocardiogram should be performed. There is potential for the interaction of valproate w ith aspirin, p henytoin, flu oxetin e and carbam azepine . 74 Starting dosages are typically 1 2 5 -2 5 0 m g daily, w ith a therapeutic range of 5 0 0 -1 0 0 0 mg to achieve blood levels of 6 5 - 9 0 /xg/ml.74 Possible adverse effects include sedation, nausea, tremor and w eight gain. Recently, in North A m erica , 78 there has been criticism of the w h olesale change aw ay from lithium and towards valproate in older patients, on the grounds that this has little evidence base.

Carbamazepine

There are case reports of the effectiveness of carbem azepine in the treatm ent of late-life m an ia . 70 As w ith valproate, there is som e evidence that carbamazepine is m ore effective than lithium in the treatm ent of non-classic m ania and secondary m an ia .71 The data suggest that it m ay be m ore efficacious as part of com bined drug therapy rather than as m on otherapy . 73 H owever, elderly patients on m ultiple m edications are prone to increased side-effects, as carbamazepine can influence the m etabolism of m any drugs, such as warfarin, erythrom ycin, diltiazem, verapam il and valproate. Prior to com m encing carbam azepine, liver enzym e tests, electrolytes, a full blood count w ith platelets and electrocardiogram should be undertaken. Sideeffects include sedation, ataxia, nystagm us, blurred vision, agranulocytosis, anticholinergic effects and hyponatraem ia secondary to inappropriate anti­ diuretic horm one secretion. Starting dosages are typically 100 m g once or tw ice daily, w ith average daily dosages of 4 0 0 -8 0 0 mg to achieve blood levels of 6 to 1 2 /xg/1.74

Lamotrigine

Lamotrigine is a phenyltriazine derivative, and is a w ell-established an ticon vu l­ sant agent that has sh ow n efficacy in the prevention of m ood episodes in younger patients w ith bipolar 1 disorder . 79 No drug trial data in elderly populations are available so far. The m echanism of action of the drug m ay be related to the inhibition of sodium and calcium channels in presynaptic neurons and sub­ sequent stabilisation of the neuronal m em brane. Lamotrigine is generally w ell tolerated, the m ost com m on side-effects being headache, nausea, infection and insom nia. It does n ot cause bod y-w eight gain, and the incidence of diarrhoea and tremors is significantly low er than w ith lithium , w ith an incidence of serious rash of about 0.1% . It generally does n ot require m onitoring of serum levels. The dosage should be titrated over a 6 -w eek period to 2 0 0 m g/day in order to m inim ise the incidence of serious rash.

13 0

Practical old age p sy c h o p h arm ac o lo g y

T able 8.4 M ood stabilisers used in bipolar disorder in older people Drug

Investigations prior to prescribing

Side-effects

Average daily dosages

Target serum levels

Lithium

Renal function Electrolytes ECG/EKG Thyroid function Blood glucose (preferably fasting)

Polyuria, gastrointestinal upset, tremor, neurological disturbance (toxicity), weight gain, oedema, psoriatic rash

400 mg (slow release)

0.4-0.7 mEq/1

Valproate

Liver enzymes Full blood count ECG/EKG

Sedation, nausea, tremor, weight gain

500-1000 mg

65-90 /xg/ml

Carbamazepine

Liver enzymes Full blood count ECG/EKG

Sedation, ataxia, nystagmus, blurred vision, agranulocytosis, anticholinergic effects, hyponatraemia

400-800 mg

6—12 /xg/1

Other treatments Other anticonvulsants such as gabapentin and topiram ate are used in younger patients w ith bipolar disorder, but as yet there are no controlled trials concerning older people. C onventional high -p oten cy antipsychotics such as haloperidol are increasingly being replaced by the new er atypical antipsychotics such as risperidone, olanza­ pine and quetiapine. H owever, there is still a place for the use of high-potency typical antipsychotics, w ith or w ith ou t a benzodiazepine, in the treatm ent of the acutely disturbed m anic patient . 74 A typical regim e is haloperidol, 5 m g intra­ m uscularly, w ith 1 mg lorazepam repeated in an hour if necessary. M any hospitals have their ow n rapid tranquillisation drug protocols, w hich should be consulted. O lanzapine has a licence for the treatm ent of acute bipolar m ania, and clozapine is used as m onotherapy for refractory bipolar disorder in younger adults . 74 H ow ever, the latter requires regular m onitoring for agranulocytosis, and cardiovascular effects can lim it its use. There are reports of the efficacy of all of the atypical antipsychotics in the treatm ent of older adults w ith bipolar disorder, but there have been no prospective random ised controlled trials . 74

Conclusion D epression in elderly patients often goes unrecognised, and the diagnosis m ay be com plicated by m edical com orbidity and cognitive im pairm ent. The co n se­ quences of untreated depression include increased disability, morbidity and mortality. D epressive episode in elderly patients responds readily to appropriate treatm ent, w ith pharm acotherapy being the m ainstay of treatm ent. Because of the favourable adverse effect profile, SSRIs have in m ost cases replaced TCAs as first-line therapy. In addition, the advent of other n ew antidepressants such as venlafaxine and m irtazepine has provided clinicians and patients w ith more tolerable alternatives to tricyclics and m onam ine oxidase inhibitors. The

M a n ag em en t o f geriatric m o o d d iso rd e rs

13 I

prevention of relapse and recurrence in the continuation and m aintenance phases of treatm ent, respectively, are just as im portant goals as the rem ission of depressive sym ptom s in the acute phase of treatm ent. Adjunctive treatm ent w ith psychotherapy or psychosocial interventions is effective in both acute and continuation therapy. E lectroconvulsive therapy is both effective and w ell tolerated in severe non-responsive depression affecting older people. U nfortunately, there have b een no random ised controlled studies of the treatm ent of bipolar disorder in old age. H owever, the lim ited data that are available support the usefulness of lithium in acute and prophylactic treatm ent of m ania in old age, although w ith an increased risk of side-effects and toxicity. Increasingly drugs such as sodium valproate, divalproex, carbamazepine and lam otrigine are being used as safe, although not necessarily m ore effective, alternatives to lithium .

Key points • The treatm ent of major depression in older people is com plicated by m edical com orbidity, yet outcom es in the acute phase of treatm ent are as good as at other tim es of life. • There is no on e 'right' antidepressant treatm ent, but rather treatm ent should be tailored to the individual patient. • The goal in acute treatm ent is full rem ission of sym ptom s. Partial recovery is a risk factor for chronic depression and/or increased risk of relapse. • K eeping the patient w ell is as im portant as acute treatm ent, and evidence exists for prophylaxis w ith both tricyclics and SSRIs. • The principles of treatm ent of m ania and bipolar disorder are the same as at other tim es of life. The m ain difficulties are w ith side-effects, especially w ith lithium salts.

References 1 M acdonald AJD (1986) Do general practitioners 'miss' depression in elderly patients? BMJ. 292: 13 6 5 -7 . 2 Copeland JRM, B eekm an ATF, D ew ey ME et a l (1999) D epression in Europe:

geographical distribution am ong older people. B r J P sych iatry. 174: 312-21. 3 H am m erlein A, D erendorf H and L ow enthal DT (1998) Pharm acokinetic and pharm a­ codynam ic changes in th e elderly: clinical im plications. Clin Pharmacokinet. 35: 4 9 -6 4 . 4 B aldw in RC (2001) Depressive disorder. In: R Jacoby and C O ppenheim er (eds) Psychiatry in the Elderly (3e). Oxford U niversity Press, Oxford. 5 Lebowitz BD, Oearson JNL, Schneider LS et a l (1997) D iagnosis and treatm ent of depression in late life. JAM A. 278: 1 1 8 6 -9 0 . 6 B aldw in RC, Chiu E, Katona C and Graham N (2002) Guidelines on Depression in Older People: practising the evidence. M artin Dunitz, London. 7 W illiams JW, Barrett J, O xm an T et a l (2000) Treatm ent of dysthym ia and m inor depression in primary care: a random ised controlled trial in older adults. JAM A. 284: 1 5 1 9 -2 6 .

13 2

Practical old age p sy c h o p h arm ac o lo g y

8 Folstein MF, Folstein SE and M cHugh PR (1975) 'M ini-M ental State': a practical m ethod for grading the cognitive state of patients for the clinician. J Psychiatr Res. 12: 189-98. 9 Yasavage JA, Brink TL, Rose TL et al. (1983) D evelop m en t and validation of a geriatric depression screening scale: a prelim inary report. J Psychiatr Res. 17: 3 7 -4 9 . 10 M ontgom ery SA and Asberg M (1979) A n e w depression scale in ten d ed to be sensitive to change. Br J Psychiatry. 134: 3 8 2 -9 . 11 Ham ilton M (1960) A rating scale for depression. J Neurol Neurosurg Psychiatry. 23: 5 6 -6 2 . 12 M ottram P, W ilson K and Copeland J (2000) Validation of the H am ilton D epression Scale and M ontgom ery and Asberg rating scales in term s of AGECAT depression cases. Int J Geriatr Psychiatry. 15: 111 3 -1 9 . 13 Chiu E, A m es D, Draper B and S n ow d on J (1999) D epressive disorders in the elderly: a review . In: I Maj and N Sartorious (eds) Depressive Disorders. John W iley and Sons, Chichester. 14 R eynolds CF, Frank E, Perel JM et al. (1999) Nortriptyline and interpersonal p sy ch o ­ therapy as m ain ten an ce therapies for recurrent major depression: a random ized controlled trial in patients older than 59 years. JAM A. 281: 3 9 -4 5 . 15 Norm an TR (1993) Pharm acokinetic aspects of antidepressant treatm ent in the elderly. Prog Neuropsychopharmacol Biol Psychiatry. 17: 3 2 9 -4 4 . 16 Georgotas A and M cCue R (1989) The additional b enefit of exten d in g an a n ti­ depressant trial past seven w eek s in the depressed elderly. In t J Geriatr Psychiatry. 4: 191-5. 17 W augh J and Goa KL (2003) Escitalopram: a review of its use in the m an agem en t of major depressive and an xiety disorders. CNS Drugs. 17: 3 4 3 -6 2 . 18 B aum ann P (1998) Care of depression in the elderly: com parative pharm acokinetics of SSRIs. Int Clin Psychopharmacol. 13 (S u p p l. 5): S 3 5 -4 3 . 19 Stahl S, Lindquist M, Pettersson M et al. (1997) W ithdrawal reactions w ith selective serotonin reuptake inhibitors as reported to the WHO system . Eur J Clin Pharmacol. 53: 163-9. 20 Stahl S, Zivkov M, Reim itz PE et al. (1997) M eta-analysis of random ised, double-blind, placebo-controlled, efficacy and safety studies of m irtazipine versus am itriptyline in major depression. Acta Psychiatr Scand. 96 (S u p p l. 39): 2 2 -3 0 . 21 R osenbaum JF, Fava M, Hoog SL et al. (1998) Selective serotonin reuptake inhibitor discontinuation syndrom e: a random ised clinical trial. Biol Psychiatry. 44: 7 7 -8 7 . 22 Greenblatt DJ, vo n M oltke LL, Harmatz JS et al. (1998) Drug interactions w ith n ew er antidepressants: role of h u m an cytochrom es P450. J Clin Psychiatry. 59 (S u p p l. 15): 1 9-27. 23 Von M oltke LL, G reenblatt DJ, Giancarlo GM et al. (2001) Escitalopram (S-citalopram) and its m etabolites in vitro: cytochrom es m ediating biotransform ation, inhibitory effects, and com parison to R-citalopram. Drug Metab Dispos. 29: 1 1 0 2 -9 . 24 Spin E and Scardo MG (2002) Clinically significant drug interaction w ith an ti­ depressants in the elderly. Drugs Aging. 19: 2 9 9 -3 2 0 . 25 Kirby D, Harrigun S and A m es D (2002) Hyponatraem ia in elderly psychiatric patients treated w ith SSRIs and venlafaxine: a retrospective controlled study in an inpatient unit. Int J Geriatr Psychiatry. 17: 2 3 1 -7 . 26 Timmer CL, Paanakker JE and van Hal HJM (1996) Pharm acokinetics of m irtazipine from orally adm inistered tablets: in flu en ce of age, gender and treatm ent regim en. H um Psychopharmacol. 11: 4 9 7 -5 0 9 . 27 Aguglia E (2000) R eboxetine in the m ain ten an ce therapy of depressive disorder in the elderly: a long-term open study. Int J Geriatr Psychiatry. 15: 7 8 4 -9 3 . 28 M ittm an N, Herrmann N, Einarson TR et al. (1997) The efficacy, safety and tolerability of antidepressants in late-life depression: a m eta-analysis. J Affect Disord. 46: 1 9 1 -2 1 7 .

M a n a g e m e n t o f geriatric m o o d d iso rd e rs

13 3

29 W ilson K, M ottram P, Sivanranthan A and N ightingale A (2001) A ntidepressant versus placebo for depressed elderly (Cochrane R eview ). In: The Cochrane Library. Issue 2. Update Software, Oxford. 30 Edwards JG (1995) Drug choice in depression. Selective serotonin reuptake inhibitors or tricyclic antidepressants? CNS Drugs. 4: 1 4 1 -5 9 . 31 Katona C and Livingston G (2002) H ow w ell do antidepressants w ork in older people? A system atic review of num ber n eed ed to treat. J Affect Disord. 69: 4 7 -5 2 . 32 Perry PJ (1996) Pharm acotherapy for major depression w ith m elan ch olic features: relative efficacy of tricyclic versus selective serotonin reuptake inhibitor a n ti­ depressants. J Affect Disord. 39: 1-6. 33 Gill D and Hatcher S (1999) A ntidepressant drugs in depressed patients w h o also h ave a physical illness (C ochrane R eview ). In: The Cochrane Library. Issue 3 . Update Software, Oxford. 34 Cole MG, Elie LM, M cCusker J et al. (2000) Feasibility and effectiveness of treatm ents for depression in elderly m edical inpatients: a system atic review . In t Psychogeriatr. 12: 4 5 3 -6 1 . 35 N yth AL and Gottfries CG (1990) The clinical efficacy of citalopram in treatm ent of em otional disturbances in d em entia disorders. B r J Psychiatry. 157: 8 9 4 -9 0 1 . 36 N yth AL, Gottfries CG, Lyby K et al. (1992) A controlled m ulti-centre clinical study of citalopram and placebo in elderly depressed patients w ith and w ith o u t concom itant dem entia. Acta Psychiatr Scand. 86: 1 3 8 -4 5 . 37 Roth M, M ountjoy CQ and A m rein R (1996) M oclobem ide in elderly patients w ith cognitive decline and depression: an international double-blind placebo-controlled trial. Br J Psychiatry. 168: 1 4 9 -5 7 . 38 Evans M, H am m ond M, W ilson K, Lye M and Copeland J (1997) Placebo-controlled treatm ent trial of depression in elderly physically ill patients. In t J Geriatr Psychiatry. 12: 8 1 7 -2 4 . 39 Zimmer B (1999) Direct and indirect cost of ven lafaxin e treatm ent of depression in the elderly w ith com orbid m edical disorders. A n n Long-Term Care. 7: 4 0 5 -9 . 40 Gustafson Y, N ilsson I, M attsson M et al. (1995) E pidem iology and treatm ent of p o st­ stroke depression. Drugs Aging. 7: 2 9 8 -3 0 9 . 41 A ndersen G, Vestergaard K and Lauritzen L (1994) Effective treatm ent of post-stroke depression w ith th e selective serotonin reuptake inhibitor citalopram. Stroke. 25: 1 0 9 9 -1 0 4 . 42 A nderson IM (1998) SSRIs versus tricyclic antidepressants in depressed inpatients: a m eta-analysis of efficiency and tolerability. Depression Anxiety. 1 (Suppl. 1): 1 1-17. 43 Wiart L, Petit H, Joseph PA, M azaux JM and Barat M (2000) F lu oxetin e in early post-stroke depression: a double-blind placebo-controlled study. Stroke. 31: 1 8 2 9 -3 2 . 4 4 Trappier B and C ohen Cl (1998) Use of SSRIs in V ery old' depressed nursing h o m e residents. A m J Geriatr Psychiatry. 6: 8 3 -9 . 45 R osen J, M ulsant BH and Pollock BG (2000) Sertraline in the treatm ent of m inor depression in nursing h o m e residents: a pilot study. In t J Geriatr Psychiatry. 15: 177 -8 0 . 46 R osen J, Rogers JC, M arin RS et al. (1997) C ontrol-relevant in terven tion in the treatm ent of m inor and major depression in a long-term care facility. A m J Geriatr Psychiatry. 5: 2 4 7 -5 7 . 47 M ottram P, W ilson KCM, A shw orth L and A bou-Saleh M (2002) The clinical profile of older patients' response to antidepressants - an open trial of sertraline. In t J Geriatr Psychiatry. 17: 5 7 4 -8 . 48 Old Age D epression Interest Group (1993) H ow long should the elderly take an ti­ depressants? A double-blind placebo-controlled study of continuation/prophylaxis therapy w ith dothiepin. B r J Psychiatry. 162: 1 7 5 -8 2 .

134

Practical old age p sy c h o p h arm ac o lo g y

49 Klysner R, B ent-H ansen J, H ansen HL et al. (2002) Efficacy of citalopram in the prevention of recurrent depression in elderly patients: placebo-controlled study of m ainten ance therapy. Br J Psychiatry. 181: 2 9 -3 5 . 50 Paykel ES (1998) R em ission and residual sym ptom atology in major depression. Psychopathology. 31: 5 -1 4 . 51 Flint AJ (1995) A ugm entation strategies in geriatric depression. Int J Geriatr Psychiatry. 10: 1 3 7 -4 6 . 52 Guscott R and Grof P (1991) The clinical m ean in g of refractory depression: a review for the clinician. A m J Psychiatry. 148: 6 9 5 -7 0 4 . 53 Flint AJ and Rifat SL (1996) The effect of sequential antidepressant treatm ent on geriatric depression. J Affect Disord. 36: 9 5 -1 0 5 . 54 Baldw in RC (1996) T reatm ent-resistant depression in the elderly: a review of treatm ent options. Rev Clin Gerontol. 6: 3 4 3 -8 . 55 Akiskal HS (1985) An approach to chronic and 'resistant' depressions: evaluation and treatm ent. J Clin Psychiatry. 46: 3 2 -6 . 56 B ow skill RJ and Bridges PL (1997) T reatm ent-resistant affective disorders. B r J Hosp Med. 57: 1 71 -2 . 57 N ierenberg AA, Feighner JP, Rudolph R et al. (1994) V enlafaxine for treatm entresistant unipolar depression. J Clin Psychopharmacol. 14: 4 1 9 -2 3 . 58 A ustin MPV, Souza FGM and G oodw in GM (1991) Lithium au gm en tation in an ti­ depressant-resistant patients: a quantitative analysis. Br J Psychiatry. 159: 5 1 0 -1 4 . 59 H oencam p E, Haffmans PMJ, Dijken W A et al. (2000) Lithium au gm en tation of venlafaxine: an op en -lab el trial. J Clin Psychopharmacol. 20: 5 3 8 -4 3 . 60 N ational Institute of Health (1992) NIH C onsensus panel on diagnosis and treatm ent of depression in late life JAM A. 268: 1 0 1 8 -2 4 . 61 W eissm an MM, Leaf PJ, Tischler GL et al. (1988) Affective disorders in five U nited States com m unities. Psychol Med. 18: 1 4 1 -5 3 . 62 Bartels SJ, Forester B, M iles KM et al. (2000) M ental health service use by elderly patients w ith bipolar disorder and unipolar depression. A m J Geriatr Psychiatry. 8: 16 0 -6 . 63 Shulm an I< (1994) M ania in late life: conceptual and clinical issues. In: E Chiu and D A m es (eds) Functional Psychiatric Disorders o f the Elderly. Cambridge U niversity Press, Cambridge. 64 K rautham m er C and Klerman GL (1978) Secondary mania: m anic syndrom e associated w ith anteced en t physical illness or drugs. Arch Gen Psychiatry. 35: 1 3 3 3 -9 . 65 Fortin L (1990) M anic disorder in the aged: a review of th e literature. Can J Psychiatry. 35: 6 7 9 -8 3 . 66 Am eblas A (1987) Life even ts and m ania. Br J Psychiatry. 150: 2 3 5 ^ 0 . 67 M cD onald W M (2000) E pidem iology, etiology, and treatm ent of geriatric m ania. J Clin Psychiatry. 61 (Suppl. 13): 3 -1 1 . 68 Sproule BA, Hardy BG and Shulm an KI (2000) Differential pharm acokinetics of lithium in elderly patients. Drugs Aging. 16: 1 6 5 -7 7 . 69 Charron M, Fortin L and Paquette I (1991) De novo m ania am ong elderly people. Acta Psychiatrica Scand. 84: 5 0 3 -7 . 70 M irchnadani I and Young R (1993) M anagem ent of m ania in the elderly: an update. A nn Clin Psychiatry. 5: 2 1 5 -2 2 . 71 Chen ST, A ltshuler LL, M elnyk KA et al. (1999) Efficacy of lith iu m vs. valproate in the treatm ent of m ania in the elderly: a retrospective study. J Clin Psychiatry. 60: 18 1 -6 . 72 Hirschfield R (1996) Practice Guidelines fo r the Treatment o f Patients with Bipolar Disorder. A m erican Psychiatric A ssociation, W ashington, DC. 73 B ow den C (1995) Predictors of response to divalproex and lithium . J Clin Psychiatry. 56 (Suppl. 3): 2 5 -3 0 . 74 Sajatovic M (2002) Treatm ent of bipolar disorder in older adults. In t J Geriatr Psychiatry. 17: 8 6 5 -7 3 .

M a n ag em en t o f geriatric m o o d d iso rd e rs

135

75 B ow d en C, Brugger AM, Sw ann AC et al. (1994) Efficacy of divalproex vs. lithium and placebo in the treatm ent of m ania. JAM A. 271: 9 1 8 -2 4 . 76 K houzam H, Emery P and R eaves B (1994) Secondary m ania in late life. J A m Geriatr Soc. 42: 8 5 -7 . 77 Kando JC, T ohen M, Castillo J et al. (1996) The use of valproate in an elderly population w ith affective sym ptom s. J Clin Psychiatry. 57: 2 3 8 -4 0 . 78 Shulm an KI, R ochon P, Sykora K et a l (2003) Changing prescription patterns for lithium and valproic acid in old age: shifting practice w ith o u t evid en ce. BMJ. 326: 9 6 0 -1 . 79 Goldsmith DR, W agstaff AJ, Ibbotson T et a l (2003) Lamotrigine: a review of its use in bipolar disorder. Drugs. 63: 2 0 2 9 -5 0 .

Chapter 9

The treatment of psychosis in the elderly Roger Bullock

Introduction A ge-related deterioration of significant cortical areas w ith associated n eu roch em ­ ical changes, com orbid physical illnesses, social isolation, sensory deficits and polypharm acy leads to an increased risk of psychosis in the elderly . 1 Thus the prevalence of neuropsychiatric disorders requiring treatm ent is expected to increase significantly in people above the age of 65 years over the next 20 years, perhaps quadrupling. The Cache C ounty population study found that 5% of all older people had had delusions or hallucinations in the preceding m o n th . 2 A ntipsychotic agents are the first-line treatm ent for psychosis, the sym ptom s that respond best being hallucinations, delusions, anxiety, verbal and physical agitation and hostility . 3 Despite this, the elderly population has traditionally been neglected in clinical research, especially w here antipsychotics are concerned. M uch of the prescribing is technically unlicensed and often unm onitored, especially in primary care. H ow ­ ever, the focus of attention has changed, partly due to concern about em erging side-effects, but also because the population is ageing and the increasing presence of psychiatric sym ptom s in nursing h om es has led to widespread use of neuroleptic m edication, not alw ays ju d iciou sly . 4 Understanding the aetiology of the psychi­ atric syndrom es in w h ich the psychosis occurs in older people, along w ith the introduction of n ew er and safer treatm ents, has helped to im prove prescribing.

Psychotic disorders in late life The conditions m ost frequently seen in the elderly w h o present primarily w ith psychosis or have a psychotic elem en t to their presentation include the following: • • • • • • • •

schizophrenia delusional disorder dem entia (every type) neurological conditions, especially Parkinson's disease delirium m ood disorder substance abuse m etabolic disturbances 137

I 38

Practical old age p sy c h o p h arm ac o lo g y

• chronic m edical conditions • drug-induced psychosis. A ntipsychotic m edication is alw ays the first-line treatm ent w here psychosis presents in each of these conditions. H ow ever, w here there are comorbid medical or m etabolic conditions contributing to the sym ptom s (e.g. thyroid disease), these m ust also be treated to ensure a better recovery.

Schizophrenia

Chronic early-onset schizophrenia Classical schizophrenia occurring before 45 years of age runs a variable course, w ith around 50% of patients show ing im provem ent or attenuation of sym ptom s as th ey get older . 5 D elusional beliefs and hallucinations becom e less frequent as the patient ages, and even formal thought disorder m ay becom e less bizarre . 6 These patients w ill have b een on antipsychotic m edications for m any years, as th ey are the m ost effective treatm ent, im proving acute sym ptom s and controlling relapses. M ost chronic patients w h o have b een treated for m ore than 15 years w ill probably be on typical antipsychotics. These m ay be at relatively low doses com pared w ith the earlier stages of illness. If these patients are free from u n w an ted side-effects, they should be left alone, and in particular no attem pt should be m ade to stop w hat look like ineffective doses. If they do have any u n w anted side-effects, especially extrapyramidal effects, th en a sw itch to atypicals m ay be warranted. This strategy is endorsed by recent guidance from the National Institute for Clinical E xcellen ce . 7

Schizophrenia in later life Schizophrenia can occur at any tim e in life. M ost patients are diagnosed early in life, but 15% of patients w ith schizophrenia or schizophrenia-like illness present after the m iddle period of life. As these patients age they should be treated in the sam e w ay as you nger-onset patients.

Late-onset schizophrenia Around 3% of patients w ith schizophrenia present after 60 years of age . 8 Lateonset schizophrenia is probably related to cortical degeneration, and so m ay be a different illness to earlier-presenting schizophrenia. There has been considerable debate about this since the term 'paraphrenia' was dropped. D elusional disorder is a separate diagnosis, but is often used w here paraphrenia had b een used previously. A nother consensus group coined the term 'very-late-onset schizo­ phrenia-like psychosis'. W hatever it is called, the disease follow s different dem ographics to schizophrenia (e.g. a m ale:fem ale ratio of 8 : 1 ), and generally follow s a m ore chronic course . 9 Social isolation and hearing im pairm ent are also associated w ith the develop m ent of late-onset schizophrenia . 10 Com plex, w ellform ed persecutory delusions are especially com m on, w ith hallucinations com ­ paratively less frequent . 11 These delusions are often very difficult to treat, partly because the patient usually has little insight and cannot see the need for treatm ent. In som e instances treatm ent m ay not be the best option, and this n eeds to be considered prior to any m edication use. The side-effects of typical antipsychotics traditionally alerted suspicious patients, m aking com pliance a

T h e t r e a tm e n t o f psychosis in th e elderly

139

problem . This has im proved w ith the use of atypicals, for w hich it m ay be expected that low er doses w ou ld be n eeded com pared w ith early-onset patients . 12 H ow ever, it is not un com m on for them to require and tolerate equivalent doses.

Psychosis and agitation in elderly patients with dementia D em entia occurs in approxim ately 10% of individuals in their seventh decade, and in 20-30% in the eighth and nin th decades . 13 Sym ptom s of psychosis and agitation such as delusions, hallucinations, aggression and disinhibition occur in about 50% of these patients, and are often challenging to caregivers. Both traditional and atypical antipsychotic agents are effective in treating psychotic sym ptom s and agitation in elderly patients w ith dem entia. Relatively lo w dosages of the antipsychotics can be effective , 14 but a m eta-analysis of double-blind trials published b etw een 1954 and 1989 found that although traditional antipsychotic agents w ere significantly m ore effective than placebo, the effect size w as very small (r = 0.18) in agitated patients w ith d em entia . 15 No one agent w as found to be superior to another. A m ore recent m eta-analysis of random ised, controlled, double-blind studies in dem entia of Alzheim er's type evaluated clinical im provem ent, side-effects and dropout rates . 16 This included studies published b etw een 1966 and 1995 that w ere 4 w eeks or longer in duration and had a m easurable outcom e based on a specific rating scale. Again, this study dem onstrated a m odest effect in the treatm ent of dem entia (61% for treatm ent com pared w ith 34% for placebo). Dropout rates w ere the sam e b etw een groups, w ith the m ost frequently reported adverse effects being sedation, m ovem en t disorders and orthostasis. A recent review of literature published b etw een 1960 and 2000 again supports the view that conven tion al antipsychotics are m odestly effective . 17 Fewer studies of atypical antipsychotics have b een published. Data from clinical trials support the efficacy of risperidone , 18-20 olanzapine , 21 and quetiapine 22,23 in this population. These are discussed in Chapter 14.

Delirium Delirium is an organic syndrom e characterised by acute onset, fluctuating levels of consciousness and global im pairm ent of cognitive functioning. It is more com m on in the elderly , 24 w here 14-56% of those in hospital experience an episode during their stay. It is m ore prevalent w here there is pre-existing cognitive im pairm ent , 25 but other com m on causes include infections (43% ), prescribed m edications (20-40% ) and endocrine, fluid and electrolyte im balance. Constipation can also trigger an episode. N on-detection rates are in the range of 3 3 - 6 6 % , 25,26 and this increases m ortality , 27 hospital length of stay and subsequent risk of residential care . 25 Treatment of any underlying physical illness and attention to environm ental factors are im portant . 27 If behavioural disturbance such as agitation and halluci­ nations is prom inent, psychotropic m edication m ay be indicated. Antipsychotics are effective w here rapid onset of action and little respiratory depression occur. Haloperidol has been w idely used in small doses , 25 but there are few data on the use of atypical antipsychotics in delirium . Olanzapine and haloperidol were com pared in an uncontrolled case series of 1 1 patients w ith delirium due to

14 0

Practical old age p sy c h o p h arm ac o lo g y

diverse aetiology . 24 O lanzapine had a similar response, but w as better tolerated. This w as replicated in a case-report series of post-operative delirium in geriatric patients . 28 There is also lim ited evidence of clinical benefit w ith risperidone (1 .5 -4 m g ) . 25

M ood disorders with psychosis Depression is the com m onest psychiatric disorder in the elderly, and psychotic sym ptom s frequently coexist . 1 Older patients w ith a psychotic depression have increased rates of relapse, m ore persistent sym ptom s, m ore suicide attempts, and m ore hospital adm issions than those w ith a non-psychotic depression. A nti­ psychotic m edications are useful in the treatm ent of psychosis associated w ith major depression, and also m anic episodes in patients w ith bipolar disorder . 29 In the elderly, often both an antidepressant and an antipsychotic are needed to successfully treat a truly psychotic depression. H owever, there are few data to support this. In fact, a non-random ised study of 25 patients that com pared a tricyclic antidepressant plus antipsychotic w ith a tricyclic antidepressant and electrocon vulsive therapy (ECT) revealed that the former group responded m ore slo w ly . 30 Elderly patients w ith m anic sym ptom s have a higher prevalence of cognitive dysfunction, w ith persistence of sym ptom s and greater m ortality . 31 No prospec­ tive studies in the elderly have been undertaken, so extrapolations from younger patients' data are all that can be used. Risperidone is reported to reduce psychotic sym ptom s in elderly patients w ith m ania, although other reports suggest that it m ay induce m ania in som e patients . 32 Clozapine appears to be safe and effective in the treatm ent of m ania in the elderly, but does not seem to help the depressed phase in bipolar patients . 33 O lanzapine has b een approved for the treatm ent of m ania, so its future use in the elderly should inform practice.

Antipsychotic medication in the elderly In the elderly, the recom m ended doses for traditional antipsychotics are m uch low er than in younger patients. A ge-related pharm acokinetic and pharm acody­ nam ic factors, coexisting m edical illnesses and concom itant m edications increase the risk for both adverse effects and drug interactions in the older patient taking any antipsychotic . 34 Behavioural sym ptom s tend to respond to low er doses, so the appropriate starting dose of a typical antipsychotic is approxim ately a quarter of the usual adult d ose . 35 To m inim ise adverse effects (e.g. postural hypotension and sedation), antipsychotics can be given in divided doses and titrated slow ly up to the optim um therapeutic effect. For the older patient w ith schizophrenia, the required dose of antipsychotic correlates inversely w ith both the patient's current age and the age of onset of illness . 36 In chronically ill patients w ith schizophrenia, dose requirem ents often decrease over time. Patients w ith schizophrenia have high rates of m edical com orbidity and drugdrug interactions because of the high rates of polypharm acy due to coexisting physical illn ess . 37 Respiratory, cardiovascular, gastrointestinal and neoplastic disorders are m ore com m on than in the general p opulation . 38

T h e tr e a tm e n t o f psychosis in th e elderly

14 1

Typical antipsychotics The dopam ine hypothesis originated in the 1950s. D elay and D eniker 39 discov­ ered the potential therapeutic effects of chlorprom azine in suppressing hallucina­ tions and delusions in schizophrenic patients. N eurochem ical experim ents in anim al brain dem onstrated the dopam ine-antagonist properties of n eu rolep ­ tics .40”42 Currently, five subtypes of dopam ine receptors have b een distinguished by gen e clon in g . 43 They are G -protein-coupled receptors that activate an enzym e or ion channel either to produce an intracellular second m essenger or to cause ion flu x es .44,45 D opam ine receptors have b een grouped into tw o m ain classes, nam ely D ^ lik e (Di and D 5) and D2-like (D2, D 3 and D 4 ) . 46 D x receptors are m ore abundant in the brain, and seem to be involved in the control of m otor, cognitive and cardiovascular fu nction s .47,48 H ow ever, D i-receptor antagonists do not sh ow any antipsychotic activity. D2-receptor antagonists are key to the treatm ent of positive sym ptom s of schizophrenia. Excessive blockade of these receptors m ay lead to extrapyram idal sym ptom s (EPS) and tardive dyskinesia (TD). All of the current antipsychotics have D2-antagonist activity. M oderate blockade of central D 2 receptors reduces positive sym ptom s of schizophrenia, but too high a degree of D2-receptor occupancy gives rise to EPS. Blockade of D 2 receptors in the anterior pituitary causes elevation of blood prolactin lev els .49,50 Interestingly, clozapine has a higher affinity for D 4 than for D 2 receptors. These m ay play a role in the m odulation of gam m a-am inobutyric acid (GABA)-ergic neuronal activity by dop am in e . 51 C onventional antipsychotics such as haloperidol, chlorprom azine and prom a­ zine are D2-receptor antagonists that inhibit dopam inergic neurotransm ission in a dose-related m anner. They virtually decrease the intensity of all psychotic sym p­ tom s, although not necessarily to the sam e extent and w ith the sam e tim e course. N egative sym ptom s m ay persist to a m uch m ore striking degree than delusions, hallucinations and thou ght disorders, and there is a dose-related incidence of EPS.

Adverse effects associated with typical antipsychotics Older patients are particularly vulnerable to the adverse effects associated w ith conven tional antipsychotic agents, such as postural hypotension, sedation and anticholinergic side-effects . 34 In particular th ey are at risk of developing adverse m otor effects and EPS. Typical antipsychotics are associated w ith both acute and late-occurring m o v e ­ m ent disorders, particularly EPS and TD. Dystonia, parkinsonism and akathisia usually occur w ithin a few days or w eeks, w hereas TD has a delayed onset, occurring m onths to years after com m encing treatm ent. M ovem ent disorders often impair a patient's quality of life, and TD can lead to speech im pairm ent, falls, feeding problem s and depression . 52 Dystonia, dyskinesia, bradykinesia and akathisia can cause gait disturbances, increasing the risk of falls and consequent fractures. The incidence and prevalence of antipsychotic-induced parkinsonism and TD are significantly increased in older patients (up to 50% in those over 60 years of a g e ) . 53,54 After 1 year of treatm ent w ith a typical antipsychotic, 26% of elderly patients (m ean age 6 6 years) had TD, com pared w ith 4 % of younger patients . 55 This increased risk of m ovem en t disorders can be attributed to age-related loss of

14 2

Practical old age p sy c h o p h arm ac o lo g y

dopam inergic function. Typically, ascending dopam inergic neurons and post­ synaptic dopam ine receptors degenerate w ith age, w h ich m eans that the dopam ine content in the striatum has declined by as m uch as 50% by the age of 60 years . 56,57 This m akes an older individual m ore susceptible to dopam inergic blockade. In general, this risk is reduced w ith atypical antipsychotics. 58~60 H owever, these drugs are not free from side-effects . 61 Typical antipsychotics are m ore likely to exhibit a cum ulative effect if taken over long periods. Concerns over QT-interval abnormalities, w ith subsequent ventricular tachycardia and sudden death in patients of all ages, has led to restrictions in the use of thioridazine 62 and withdraw al of droperidol.

Treatment of common movement disorders Mild cases of antipsychotic-induced m ovem en t disorders can be left if the sym ptom s do n ot trouble the patient. In m ore marked cases, reducing the dose to the low est effective level, choosing a low er-potency antipsychotic or sw itching to an atypical antipsychotic m ay reduce or alleviate the sym ptom s . 53 A nticholinergic drugs, w h ich have traditionally been used for som e EPS, are associated w ith w orse adverse effects in the elderly, including tachycardia, urinary retention, constipation, confusion and m em ory im pairm ent. They rem ain the treatm ent of choice for acute dystonia, w hich fortunately is rare in the elderly. H owever, anticholinergic treatm ent does not usually help acute akathisia, and its traditional use in parkinsonism is n o w under question. Betablockers can help w ith akathisia, but can cause hypotension and bradycardia. TD is usually a persistent m ovem en t disorder associated w ith antipsychotic use, but it som etim es occurs spontaneously. M inim ising the risk of TD is important, as there is no definitive treatm ent. Treatment options include using the low est possible doses, especially in those at high risk of developing TD, and regular review . The incidence of TD appears to be low er in elderly patients w h o receive atypical antipsychotics rather than typicals . 61

Atypical antipsychotics Atypical antipsychotics are drugs w ith no or m inim al EPS and greater efficacy in positive and particularly negative sym ptom s (and in the original definitions, they should n ot raise prolactin levels or induce narcolepsy). They have a characteristic m ode of action, n am ely a com bination of D2-receptor antagonism , coupled w ith 5HT2A blockade, w hich paradoxically reverses the D 2 blockade, especially in the striatum . 63 The en h an ced efficacy of atypical antipsychotics is attributed to this high ratio of 5HT2A to D2-receptor blockade, and a greater specificity for the m esolim bic rather than the striatal dopam ine system . 64 For exam ple, risperidone show s very p otent 5HT2A antagonism w ith w eaker D 2 antagonism , and is effective against the positive and negative sym ptom s of schizophrenia. Its action suggests that negative sym ptom s m ay actually derive from overactivity of these excitatory 5HT2A receptors, w h ich occur m ainly in the telencephalon, and m uch less in the midbrain and hindbrain. Atypical antipsychotics also show antagonism at D 3 and D 4 receptors. A lthough the precise functional significance of these m ultiple receptors has yet to be

T h e tr e a tm e n t o f psychosis in th e elderly

14 3

clarified, this high D 3 and D 4 selectivity m ay additionally contribute to the low EPS rates observed w ith these drugs . 65 The selectivity of these atypical agents w ith regard to the lim bic system has been dem onstrated in electrophysiological studies (percentage of silent dopam ine neurons w ith chronic treatm ent) and anim al tests (low catalepsy in rodents and lo w EPS in n o n -h u m a n prim ates ) . 66 The biggest issue is the disparity in cost b etw een the older and new er drugs, w ith som etim es a 30-fold difference in price .67 This m eans that cost-effectiveness is a major issu e . 68 Because of the im proved treatm ent profile, there seem to be low er relapse rates and few er adm issions to hospital on atypical antipsychotics, coupled w ith im proved quality of life, w h ich m ake them a more popular choice for patients and clinicians alike.

Adverse events with atypical antipsychotics The reduced risk of EPS and TD associated w ith atypical antipsychotics is offset by apparent increases in m edical morbidity, including w eight gain, abnorm al lipid concentrations, hyperglycaem ia and consequent type 2 diabetes m ellitus . 69 The evidence suggests that som e atypical antipsychotic agents, particularly clozapine and olanzapine, m ay significantly impair glucose m etabolism , increasing this diabetic risk. Patients w ith specific risk factors for diabetes should be closely m onitored, particularly if th ey are on either of these tw o drugs. Periodic screening for m etabolic abnorm alities, especially high blood sugar levels, w ou ld be best practice regardless of the choice of atypical . 70 As w ell as w eight gain, any atypical antipsychotic agent is capable of causing cardiovascular problem s and orthostatic h ypotension, w h ile som e have anticholinergic effects (e.g. olanzapine) or cause sedation. More recently, risperidone and olanzapine have been sh ow n to be associated w ith an up to fourfold higher risk of cerebrovascular adverse events in patients w ith dem entia, as a result of w h ich the UK C om m ittee for Safety of M edicines does not recom m end their use in this patient group. Thus the profile of som e atypical antipsychotics is no better and can often be w orse than that of typical antipsychotic drugs .71 Table 9.1 show s the receptor-binding profiles of the com m on atypicals com pared w ith haloperidol, and the side-effects that this binding w ill produce. Table 9.1 Receptor-binding affinities as predictors of side-effects of antipsychotic drugs Receptor

Risperidone

Olanzapine

Quetiapine

Haloperidol

M uscarinic M t

-

+++++

4- + +

-

Histam ine Hi

4- +

++++

+

a i-adrenergic

4- + +

+++

++++

D opam ine D 2

++++

+++

++

+++++

Serotonin 5-HT2

+++++

++++

+

+

4 - 4- +

++

5-HT = hydroxytryptamine; —= no affinity; + = very low affinity; 4 + = low affinity; -(—l—l- = moderate affinity; +4-4-+ = high affinity; + + + + + = very high affinity.

144

Practical old age psychopharmacology

Drug compliance Poor uptake of prescribed m edication is estim ated to occur in up to 50% of any patient population. W ith antipsychotics this figure can be as high as 7 0 -8 0 % .72 B reen and Thornhill 73 gave three reasons for this. 1 Medication-related factors - e.g. adverse effects, com plex schedules of the drug itself or polypharm acy (especially in the elderly), cost and perceived lack of efficacy. Sw itching m edication or treatm ent of side-effects are both com m on corrective strategies. 2 Patient-specific factors - e.g. psychiatric sym ptom s, severity of illness and age. These all need to be considered. Older people com m only have visual and auditory im pairm ents, som etim es in com bination w ith cognitive deficit . 74 3 Patients' attitudes - psychosocial rehabilitation, im proved patient-clinician relationship (w here actively in volving patients in treatm ent decisions can help) and the use of depot antipsychotics m ay be useful. A survey by the National Schizophrenia Fellow ship in the UK show ed that the atypicals are preferred by the people w h o take them , and this finding w as backed up by a survey of European psychiatrists w h o stated that this is w hat they w ould w ant them selves or their fam ily to receive. Because of this, together w ith the lack of superior efficacy and the relative toxicity of the typical antipsychotics, the latter are used infrequently in the elderly. Here the cost of the side-effects outw eighs the cost of the drugs, and apart from graduates w h o rem ain on treatm ent, the drugs are rarely initiated as firstline treatm ent. This has n o w been endorsed in both the Older Persons' National Service Fram ework 75 and the National Institute for Clinical Excellence (NICE) guidance on atypical antipsychotics , 7 both of w hich recom m end that n ew patients start on atypicals. The NICE guidance suggests that those on typicals w ith no side-effects should be left unchanged, but those w ith side-effects, especially EPS, should sw itch to atypicals.

Atypical antipsychotics in the elderly Further discussion of the use of atypical antipsychotics in the elderly will be restricted to the m ost com m only prescribed atypical antipsychotics. Each drug w ill be subdivided as follows: • schizophrenia and related psychosis • psychosis in idiopathic Parkinson's disease, w here 20-30% of such patients develop psychotic sym ptom s 76 • the effect of these drugs on cognitive functioning in the elderly population. Preclinical and clinical studies indicate that an inhibitory effect on dopam i­ nergic, cholinergic and histam inergic neurochem ical system s m ay be respons­ ible for the cognitive effects . 77 This explains the evidence that the use of antipsychotics in the elderly population can cause deleterious cognitive effects in som e elderly p atients .78 Atypical antipsychotics m ay thus possess a more favourable cognitive profile than traditional agents in the elderly population. The behavioural and psychological sym ptom s of dem entia are discussed in Chapter 14.

The treatment of psychosis in the elderly

145

Risperidone Psychosis in the elderly Risperidone is a benzisoxazole derivative licensed in the UK for the treatm ent of psychosis, including elderly patients. It follow ed clozapine on to the m arket and w as rapidly accepted, as it did n ot carry the risk of agranulocytosis or the need to m onitor the leukocyte count frequently. Technically it is not a classical atypical, as it elevates prolactin levels due to inhibition of dopam ine in the tuberoinfundibular system . The clinical im plications of this effect in the elderly are unclear. Early retrospective review s and uncontrolled prospective studies indicated the benefits of risperidone treatm ent for both positive and negative sym ptom s in latelife schizophrenia . 79 The m ean dose varied across the studies, w ith patients w ith schizophrenia tending to receive 4 - 6 m g/day (a very similar dose to the younger adult population). No controlled studies of risperidone in this treatm ent group have been published. Jeste et al.80 and M adhusoodanan et al.81 reported the inability of elderly patients w ith schizophrenia to tolerate rapid titration of risperidone ( 1 m g/day to 3 m g/day tw ice daily over 3 days) and increased hypotensive episodes, respectively. Risperidone binding potently to a - l and a -2 adrenergic receptors accounts for these reports of hyp oten sion in som e elderly patients. Davidson et a l 82 reported decreased sym ptom s, continuous im provem ent and im proved EPS in an open-label study of graduate patients sw itched from typicals to risperidone. The major m etabolite 9-hydroxy-risperidone is active w ith a similar receptor profile to its parent drug. It rem ains present in plasma at concentrations 3 to 5 tim es higher than those of risperidone itself. This m eans that an elderly patient requires one-third of the risperidone dose n eeded in younger patients to achieve similar plasm a concentrations of active drug. Peak plasma concentrations are reached after oral adm inistration w ithin 1 - 2 hours, and they are dose propor­ tional . 83,84 M etabolism is via the CYP2D6 hepatic m icrosom al isoenzym e, w h ich m ay predict potential drug-drug interactions in clinical use. Risperidone and 9-hydroxy-risperidone have half-lives of about 3 and 24 hours, respectively. A steady state occurs after 1 day for risperidone, and after 4.5 days for 9-hydroxyrisperidone. Excretion is m ainly via the kidneys. The half-life of both active com pounds is prolonged w ith age and in patients w ith hepatic or renal disease. C onsequently, the usual starting dosage in the elderly patient is 0 .2 5 -0 .5 m g/day given as a single daily dose. This should be titrated to the m inim um effective dosage, as anything higher in a susceptible individual m ay result in loss of atypicality and possible EPS or drug-induced parkinsonism . Both risperidone and its m etabolite 9-hydroxy-risperidone can prolong the QTc interval, so concom itant drugs that are also k n ow n to prolong this interval should be either avoided or adm inistered very carefully.

Psychosis associated with idiopathic Parkinson’s disease Psychosis associated w ith Parkinson's disease poses particular treatm ent prob­ lem s. Dopam inergic agents are usually indicated in the treatm ent of Parkinson's

14 6

Practical old age psychopharmacology

disease, but these agents frequently cause or exacerbate psychotic sym ptom s. On the other hand, agents that block dopam ine can w orsen m obility and tremor. One of the tw o uncontrolled studies investigating the use of risperidone in Parkinson's disease-related psychosis, by M eco et al.,85 show ed it to be safe and effective w ith ou t w orsening EPS in the dose range 0 .2 5 -1 .2 5 m g/day. Case reports by Rich et a l 86 dem onstrated that higher doses (in the range 0 .5 - 4 m g/day) exacerbated parkinsonism . H ow ever, the authors noted that the low er doses w h e n slow ly titrated m ay have sh ow n better tolerability. A m ore recent review supports the use of low -d ose risperidone in Parkinson's disease . 87 Relatively little inform ation is available to guide physicians in the use of risperidone in dem entia w ith Lewy bodies (DLB). Patients w ith DLB have been reported to both tolerate and not tolerate risperidone treatm ent. M cKeith et a l 88 have reported rigidity in on e patient w ith DLB.

Cognitive effects in the elderly

Controlled studies The large double-blind trials by De D eyn et a l } 9 and Katz et al.20 in behavioural and psychological sym ptom s of dem entia (BPSD) exam ined the cognitive effects of risperidone in patients w ith dem entia. The first flexible dose study reached a m ean dose of 1 . 1 m g/day of risperidone and the second had three fixed doses of risperidone (0.5, 1 or 2 m g/day). N one of them found any adverse effect on the cognition of patients w ith m ixed diagnoses of dem entia. Berm an et a l 89 conducted a prospective, random ised, double-blind com parison study of risperidone and haloperidol in elderly patients w ith schizophrenia in order to exam in e the cognitive effects of these tw o drugs. They did so because earlier uncontrolled studies of risperidone indicated that the latter im proved cognition in elderly patients. Berm an et a l concluded that patients w h o w ere receiving risperidone im proved on the B oston Nam ing Test (BNT) and M iniM ental State Exam ination (MMSE), but not on other cognitive tests. Patients w h o w ere receiving haloperidol sh ow ed no im provem ent. This positive cognitive profile of risperidone can be attributed to its lack of significant anticholinergic effects.

Uncontrolled studies A n open-label study by Berm an and M erson 90 in elderly patients w ith schizo­ phrenia treated w ith risperidone revealed a statistically significant im provem ent on MMSE scores and also on three of six additional cognitive m easures. An openlabel study of risperidone by W orkm an et a l 91 in patients w ith dem entia and idiopathic Parkinson's disease, reaching a m ean dose of 1.9 m g/day of risperidone, sh ow ed no significant changes in m ean MMSE scores. H owever, reports of the w ay in w h ich risperidone affects cognition in DLB patients have not show n consistent o u tcom es . 71

Conclusion Risperidone is effective for psychosis in the elderly, but often needs doses equivalent to those used in younger patients. The therapeutic dose/side-effect

The treatment of psychosis in the elderly

147

ratio determ ines its use here in individual patients, but for m ost it is an effective treatm ent. In Parkinson's disease and DLB it has a role, but it m ust be started at 0.25 m g daily and titrated slow ly, w ith m onitoring for side-effects.

Olanzapine Psychosis in the elderly O lanzapine is one of the clozapine-derived antipsychotics, and initially looks to be a particularly suitable drug for the elderly, having very few drug-drug inter­ actions and a favourable m etabolic profile. It is a thienobenzodiazepine, prin­ cipally m etabolised by hepatic glucuronidation, and there is little evidence of a clinically significant reduction in this m etabolic process w ith age (except in the very old and in patients w ith Alzheim er's disease). Renal or hepatic im pairm ent does not significantly affect the pharm acokinetics of olanzapine, and dose adjustm ents are not usually required. Furthermore, w ith a m ean half-life of 30 hours, it can be given once a day, w hich m ay be an important factor in aiding com pliance. Elderly patients w ith schizophrenia and bipolar disorders require doses that are usually similar to those of their younger counterparts. The drug is w ell absorbed, and peak serum concentrations are reached in approxim ately 6 hours. There is exten sive first-pass m etabolism , w ith about 40% of a dose being m etabolised before it reaches the system ic circulation. Steady state occurs after a w eek. O lanzapine is 93% bound to plasma proteins, w ith a half-life averaging 30 hours. The com m onest side-effects are som nolence (26% ), dizziness, agitation, constipation, akathisia, postural hypotension and w eight gain. Tardive dyskinesia m ay occur in about 1% of patients and can be irreversible. Orthostatic h ypoten sion occurs in about 5.5% of patients. Hyperglycaemia, diabetes m ellitus and ketoacidosis have b een w idely reported as drug-induced effects . 92,93 A potential drawback of olanzapine in the elderly m ay be its anticholinergic propensity. This is real, although it is unclear h o w important it is clinically. There are claims that it has agonist activity on the cholinergic receptors as w ell, but m ore research is needed to confirm this. In general, drugs w ith anticholinergic activity are m inim ised in the elderly, so this m ay be a factor in som e clinicians' judgem ent. A prospective, open-label, 8 -w eek trial of olanzapine therapy in elderly schizophrenic patients of m ean age 70.6 years show ed that olanzapine therapy (at a m ean dose of 4.2 m g/day) w as effective and w ell tolerated . 94 A post-hoc analysis of efficacy and safety data from a double-blind, pivotal schizophrenia study com paring olanzapine w ith haloperidol was conducted on 59 patients aged 65 years or older . 95 This analysis again confirm ed the overall efficacy and safety of olanzapine in these geriatric patients. The favourable risk-benefit balance seen at the daily dose of olanzapine used in this geriatric population (12.4 m g/day) w as comparable to that seen at the dose used for younger patients w ith schizophrenia ( 1 2 . 6 m g/day).

Psychosis associated with idiopathic Parkinson’s disease A random ised, double-blind, parallel com parison of olanzapine and clozapine in patients w ith Parkinson's disease w ith chronic hallucinations revealed that

148

Practical old age psychopharmacology

olanzapine aggravated parkinsonism to such an extent that safety-stopping rules w ere in v o k ed .96 The researchers concluded that olanzapine should not be regularly used in the m anagem ent of hallucinations in patients w ith Parkinson's disease. M anson et al.97 conducted a random ised, placebo-controlled, double-blind, cross­ over trial to assess the usefulness of low -d ose olanzapine (2 .5 -7 .5 m g/day) for levodopa-induced dyskinesias in patients w ith Parkinson's disease. This study concluded that low -d ose olanzapine w as effective in reducing dyskinesias in Parkinson's disease, but even at very lo w doses olanzapine could result in unacceptable increases in parkinsonism and 'off' tim e. In an open-label study by Friedman et al.,98 clozapine was substituted by olanzapine in psychiatrically stable patients w ith Parkinson's disease. This study revealed that 9 of the 12 patients w ith Parkinson's disease w ere unable to make this transition because of w orsened parkinsonism w ith olanzapine. M ost of the subjects preferred taking clozapine, despite the onerous m onitoring procedure. A nother open-label study by Graham et a l." concluded that olanzapine was effective in suppressing hallucinosis in patients w ith Parkinson's disease, but resulted in unacceptable exacerbation of m otor disability. Conversely, an openlabel, uncontrolled study by W olters et al.100 concluded that olanzapine w ith a m ean dosage of 6.5 =b 3.9 m g/day w as w ell tolerated in n on -d em en ted patients w ith psychotic sym ptom s associated w ith Parkinson's disease. In this study, patients dem onstrated a significant im provem ent in the psychotic sym ptom s w ith ou t w orsening of parkinsonism . They w ere also able to tolerate further increases in the dosages of dopam ino-m im etic drugs, w ith a subsequent im prove­ m ent in m otor functions and no exacerbation of psychosis. Similar results w ere reported by Aarsland et al . 101 W alker 102 found that 37.5% of patients w ith DLB w h o w ere given olanzapine dem onstrated sensitivity reactions, but for the rem ainder a low dosage (2 .5 -7 .5 mg) w as effective.

Cognitive effects in the elderly A study using olanzapine (5, 10 or 15 m g/day) in elderly patients w ith dem entia looked at the cognitive effects of olanzapine. It reported that MMSE scores w ere not adversely affected clinically am ong any of the three groups receiving olanzapine com pared w ith placebo . 21 Similar findings of unchanged MMSE scores com pared w ith baseline w ere reported in a prospective, open-label study of elderly schizophrenic patients conducted by Sajatovic et al.94 Other u n c o n ­ trolled reports suggest that som e elderly patients m ay experience adverse cognitive effects w hile taking olanzapine . 102 103

Conclusion Olanzapine appears to be very effective in elderly psychosis and schizophrenia­ like syndrom es, using the sam e treatm ent regim ens as in younger patients. It is also given once daily, w h ich can aid com pliance. There are concerns about its anticholinergic profile, and m ixed reports on cognition enhance this. In Parkinson's disease it appears to be particularly poorly tolerated, and in DLB it is potentially effective if tolerated. In sum m ary, current evidence m akes it one of the first-line choices for primary psychoses in the elderly, but it is perhaps less

The treatment of psychosis in the elderly

149

useful for behavioural problem s or psychosis secondary to other organic disease. No controlled studies com paring risperidone w ith olanzapine in elderly psychotic patients have b een published. M adhusoodanan et al.81 reported data on 151 hospitalised patients treated w ith risperidone (n = 114) and olanzapine (n = 37). The response to treatm ent, interruption of treatm ent and side-effects w ere similar in the tw o groups.

Quetiapine Psychosis in the elderly Q uetiapine is a dibenzothiazepine that is structurally similar to clozapine, and was introduced into clinical practice in 1997. Its claim w as to be an agent efficacious in the treatm ent of schizophrenia, w ith exceptionally lo w potential for EPS even at higher doses and w ith little anticholinergic or prolactin-elevating action. Therapeutic indications are treatm ent of psychotic and behavioural disorders. Q uetiapine is adm inistered orally, and is rapidly absorbed, w ith plasma levels peaking at about 1.5 hours. It is 83% protein bound and m etabolised by the liver. Its half-life is about 6 hours, and clearance is reduced in the elderly. The principal adverse effects include drow siness and postural hypotension. To date, several abstracts but only lim ited published literature are available regarding the use of quetiapine in elderly patients. M adhusoodanan et a l.104 dem onstrated the efficacy of quetiapine in 4 out of 7 elderly hospitalised patients aged b etw een 61 and 72 years, w ith schizophrenia-related psychotic sym ptom s, schizoaffective disorder and bipolar disorder. There are n o published controlled studies of quetiapine, but its efficacy in treating both the positive and negative sym ptom s of schizophrenia has b een extrapolated from a num ber of controlled com parative clinical trials in younger patients. A 5 2 -w eek, open-label, m ulti-centre trial enrolling 184 elderly patients (m ean age 76.1 years) was conducted by Tariot et al.23, to assess the safety and efficacy of long-term quetiapine use in the elderly population. This trial enrolled elderly patients w ith a variety of neuropsychological disorders. In total, 72% had a psychosis due to general m edical conditions such as A lzheim er's disease and 28% had other forms of psychosis, m ainly schizophrenia. They w ere residing in a variety of settings, w ith a host of com orbid m edical conditions and receiving num erous concom itant m edications, thus reflecting real-world conditions. W ith a m ean total daily dose of 137.5 m g (range 1 2 .5 -8 0 0 m g/day), this study concluded that quetiapine w as safe and effective for the long-term treatm ent of psychosis in elderly populations, across all causes. The conclusions from the trial recom ­ m ended an initial dosage of 25 m g/day, w ith increases in dose to a target of 1 0 0 -1 5 0 mg, w ith individualised adjustm ents as needed. Analysis of the dropouts suggested the use of slow and steady dose escalation in elderly patients, especially w h en long-term use w as anticipated. Q uetiapine was associated w ith negligible EPS, no clinically im portant changes in laboratory parameters or ECG rates or intervals, and m inim al w eight gain. To date, the results extrapolated from the controlled studies in younger patients, in com bination w ith this open-label study in elderly patients, suggest that quetiapine m ay be beneficial in terms of both efficacy and tolerability in the treatm ent of elderly patients w ith psychotic sym ptom s. On the basis of the clinical

150

Practical old age psychopharmacology

literature on quetiapine, Jeste et a l 61 recom m ended that the starting dose of quetiapine in elderly patients should be 12.5 m g to 25 m g daily, w ith the optimal target dose being 75 m g to 125 mg per day.

Psychosis associated with idiopathic Parkinson’s disease Like other atypical antipsychotics, quetiapine has a clinical advantage in the treatm ent of psychosis associated w ith Parkinson's disease. It has a higher affinity for serotonin receptors than for dopam ine D x and D 2 receptors, and has more affinity for m esolim bic than for nigrostriatal dopam ine blockade. The dopam ine blockade is only m odest. This clinical advantage is reflected in encouraging results from studies using quetiapine in psychosis associated w ith Parkinson's disease. An open-label study of the use of quetiapine in 40 patients (m ean age 72.6 years) from the Tariot 1 -year study w h o had psychosis associated w ith Parkinson's disease w as reported by Juncos et al.105 These patients w ere follow ed up for 1 year w ith a m ean daily dose of 75 m g/day, show ing that quetiapine was effective in im proving psychotic sym ptom s in this patient group. This im prove­ m en t w as m aintained throughout the 1 -year trial period w ithout w orsening the m otor sym ptom s of Parkinson's disease. In fact, surprisingly, a significant short-term im provem ent in m otor perform ance in Parkinson's disease was observed. Similar results have b een replicated in other open-label studies w here the m ean daily dose of quetiapine ranges from 37.5 to 70 m g/day. In DLB, Parsa 106 reported im proved psychiatric sym ptom s w ith no w orsening of confusion or m otor sym ptom s.

Cognitive effects in the elderly There are no published controlled trials of quetiapine in the elderly. An openlabel study of quetiapine in elderly patients by M cM anus et al.22 did not find any evidence of cognitive im pairm ent in the first 12 w eeks of the trial. Similar findings have b een reported from case series and case reports.

Conclusion The benign side-effect profile of quetiapine, along w ith the ability to use it in very lo w dosage, has m ade it an attractive choice in the elderly, but there is relatively little evidence to support it. It has w orked w ell in open-labelled studies, notably the Tariot study, and has earned an anecdotal reputation for being 'm otor friendly' - h en ce it is often favoured in Parkinson's disease and DLB. It seem s to do little harm, but m ore controlled data are needed to sh ow just w hat its benefits m ay prove to be.

Clozapine Psychosis in the elderly Clozapine is a dibenzodiazepine derivative w ith greater specificity for the limbic system and a subsequent low incidence of extrapyramidal side-effects. It was the first atypical antipsychotic to be launched, and it changed perceptions about w hat

The treatment of psychosis in the elderly

15 1

antipsychotic drugs could do in schizophrenia. Its adverse effects profile prevents it from being prescribed for first-episode patients, but at the tim e of writing it rem ains the gold standard for treatm ent-resistant patients. Follow ing oral adm inistration, it is rapidly absorbed, but only 27-50% of a dose reaches the system ic circulation, due to an important first-pass m etabolism . 107108 Various individual factors can alter the response, such as sm oking, hepatic m etabolism , gastric absorption, age and sex. Clozapine is rapidly and extensively distributed, crossing the blood-brain barrier, and 95% is bound to plasma proteins. Steady state is reached after 7 -1 0 days, w ith m axim um antipsychotic effect occurring several m onths later. M etabolism occurs primarily via the CYP1A2 and CYP3A4 hepatic m icrosom al en zym es and leads to tw o m etabolites, nam ely nor-clozapine and clozapine-N -oxide. The half-life ranges from 6 to 33 hours, and about 50% is excreted in the urine and 30% in the faeces. Clozapine low ers the seizure threshold and m ay precipitate grand-m al seizures, particularly in higher doses. It is less likely to cause extrapyramidal side-effects, but can exacerbate prostatic hypertrophy and closed-angle glaucom a due to anticholinergic activity. Tachycardia, orthostatic hypotension and raised glucose levels, w ith w eight gain and diabetes, have all b een reported . 109 Excessive sedation and som n olence frequently occur. M ost studies on the use of clozapine in elderly patients are based on retro­ spective chart review s, and each seem s to agree its place in treating psychosis in the elderly. Clozapine is licensed for this, but side-effects and safety issues limit its use. Elderly patients are not able to tolerate the rapid titration of clozapine dosage (target dose 3 0 0 m g/day in 3 w eek s). It is recom m ended that a low er dose should be used to initiate treatm ent (12.5 mg once on the first day), and to restrict subsequent dose increm ents cautiously to 25 m g/day. The dosage m ust be adjusted individually, and for each patient the low est effective dose should be used. W ith a m ean clozapine dose range of 5 3 -2 0 8 m g/day, a marked treatm ent response w as reported in 33-90% of patients receiving the drug by Chengappa et al . , 110 Salzm an et al.u l and Richards et a l 112 Clozapine is a potent antagonist at a-adrenergic, m uscarinic and histam inergic receptors, w h ich explains its ability to cause sedation and delirium in elderly patients. Orthostatic hyp oten sion can occur, and there are reports of tachycardia in elderly patients, w h ich m ay be sustained. Elderly patients, particularly those w ith com prom ised cardiovascular function, m ay be m ore susceptible to these effects. In such settings clozapine should be used at the low est possible dosage, as all of the cardiac effects appear to be dose related . 113 Elderly patients m ay also be particularly susceptible to the anticholinergic effects of clozapine, such as constipation and urinary retention. Unlike clozapine's action at the other m uscarinic receptor subtypes, clozapine is a full agonist at m uscarinic M 4 receptors, w h ich m ay partly explain its ability to cause hypersalivation and salivary gland sw elling. Special attention m ust be given to the issue of agranulocytosis and the probability of subsequent m ortality in elderly patients. Reports indicate that increasing age and fem ale gender are both independently associated w ith an increased risk of agranulocytosis . 114 As neither clozapine dosage nor plasma drug concentration have been found to be associated w ith the agranulocytosis risk, it appears that ageing-related changes in clozapine clearance m ay account for this

152

Practical old age psychopharmacology

increased rate of agranulocytosis in the elderly . 115 Thus the elderly appear to be m ore likely to develop agranulocytosis, and they m ay be m ore likely to die if agranulocytosis occurs.

Psychosis associated with idiopathic Parkinson’s disease Case reports, retrospective chart review s, open-label trials and double-blind, placebo-controlled trials 116 have sh o w n clozapine to be effective overall in the treatm ent of psychosis associated w ith Parkinson's disease. The doses required are generally substantially low er than those required for the treatm ent of patients w ith a primary psychiatric diagnosis. Several studies have also sh ow n that clozapine can be used to m anage psychosis w h ile allow ing optim isation of antiparkinsonian treatm ent and m anagem ent of m otor disability. There is little reported use of the drug in DLB, w here caution w ould be necessary because of clozapine's particular anticholinergic activity. Burke et a l.117 have reported increasing confusion in tw o cases w ith in 48 hours of use.

Cognitive effects in the elderly

Controlled studies Two large, double-blind, placebo-controlled trials have evaluated the cognitive effects of low -dose clozapine (m ean dose 24.7 m g/day and 36 m g/day) given for 4 w eeks in patients w ith Parkinson's disease and drug-induced psychosis, respect­ ively. N either study found detrim ental cognitive effects. Both studies found no differences in MMSE scores b etw een clozapine- and placebo-treated groups . 116,118 A double-blind study of elderly patients w ith schizophrenia reported half the sedation rate com pared w ith that of a chlorprom azine-treated group . 119

Uncontrolled studies In contrast, uncontrolled studies and case reports of clozapine treatm ent in patients w ith idiopathic Parkinson's disease and DLB have reported problem atic sedation and/or confusion.

Conclusion In the elderly, clozapine is less frequently utilised, m ainly because of its side-effect profile, particularly the anticholinergic activity. It has a place in treatm entresistant psychosis in the elderly, but at m uch low er doses ( < 2 0 0 mg) than in younger patients. For Parkinson's disease and BPSD there is evidence that it m ay help individual cases, but the risk-benefit cost is high, and other treatm ents should be preferentially tried first. Thus clozapine w ill usually be used as a last strategy in defined cases.

Amisulpiride Am isulpiride is a substituted benzam ide. No specific data on its use in the elderly h ave b een published, but it is not anticholinergic and it can be initiated in very lo w doses. It is a pure D 2 antagonist w ith relative selectivity for the m esolim bic

The treatment of psychosis in the elderly

153

system . This clean profile and the lack of drug-drug interactions are often cited in support of its being a suitable drug for use in older people. It should perform similarly in psychosis in the elderly to that in younger patients , 120 and be safe in the treatm ent of BPSD. It also has a 15-year record of use in Europe in all age groups. H ow ever, the bottom line is that m ore research needs to be done if specific recom m endations are to be m ade. The sam e is true of its parent com pound, sulpiride.

Aripiprazole Aripiprazole is an interesting com pound that was licensed in the USA in late 2002 and is n o w available in Europe. It is perhaps the first of a n ew type of antipsychotic, as it has a different m ode of action to anything that has preceded it. It is a partial agonist at the D2 receptor as w ell as having the expected 5HT2a serotonergic blockade. H ow ever, it also has 5HT1A agonism , w hich gives an increased anxiolytic action and m ay offer som e additional cognitive advantages. This profile should have an effect that serves to stabilise dopam inergic function, reducing the effect of dopam ine at the receptor through com petitive inhibition w h en dopam ine levels are too high, and m aintaining adequate dopam ine function through its agonist action w h en levels are lo w . 121 It has a mild affinity for adrenergic and histam inic receptors, but no m uscarinic affinity at all. This profile w ou ld m ake it attractive w ith regard to all aspects of elderly psychosis. Aripiprazole is w ell absorbed, reaching peak plasma concentration at 5 hours. Its oral bioavailability is 87% and it can be taken w ith or w ithout food. Its major m etabolite is dehydro-aripiprazole, w hich has similar activity to the parent m olecule. The half-life of aripiprazole is 75 hours, so it is taken daily. Steady state occurs at 14 days. Elim ination is m ainly through the hepatic enzym es CYP2D6 and CYP3A4, w hich m eans that there are som e drug interactions, notably w ith paroxetine and flu oexetin e, w hich m ay raise the levels of aripiprazole. Aripiprazole does not n eed any dose alteration in renal or hepatic disease. In the elderly, single-dose pharm acokinetic studies show ed a 2 0 % low ering in the clearance of aripiprazole, yet m ultiple-dose studies show ed no difference overall. Thus no dose changes are required in older patients. Studies in schizophrenia have sh o w n favourable and consistent results across m any studies . 122 Therefore elderly patients w ith schizophrenia or related psychotic episodes can be treated in the sam e w ay as those described. To date, patients treated w ith aripiprazole have sh ow n a very lo w incidence of EPS, w eight gain, sedation, hyperprolactinaem ia, m e ta b o lic d iso rd er or QTc c h a n g e s .

Currently no specific studies have b een published concerning aripiprazole use in the elderly. A placebo-controlled study of its use in 208 com m unity-dw elling dem entia patients w ith BPSD over a period of 10 w eeks has b een conducted. The m ean age of patients w as 81.5 years, and the m ean baseline MMSE was 13.6. A m ean dose of 1 0 m g/day sh ow ed that it was significantly more effective than placebo on both the total score and the psychotic subscore ( - 1 . 9 3 vs. —1.27; P = 0.03) of the Brief Psychiatric Rating Scale . 123 D iscontinuation rates w ere 8 % w ith aripiprazole and 7% w ith placebo. Som nolence was mild and not associated w ith falls or accidental injury. There w ere no differences in vital signs, laboratory tests, w eigh t gain or ECG. The Neuropsychiatric Inventory was also used as an

154

Practical old age psychopharmacology

outcom e in the study, but reduction of the total score at 1 0 w eeks w as similar in both groups. The data suggest a favourable effect on psychosis in Alzheim er's disease, as w ell as schizophrenia. W ith its daily use, good tolerability profile and lack of any m uscarinic activity, this m ay becom e an im portant choice in elderly patients w ith psychosis.

Ziprasidone Ziprasidone is a n ew antipsychotic agent w ith a high 5HT2 to D2receptor blockade ratio. It is available orally and in a short-acting injectable form (neither of w hich is yet available in the UK), and is effective on both positive and negative sym ptom s. Given orally, it is rapidly absorbed, w ith steady-state serum levels being reached w ithin 2 -3 days , 124 and age does not affect its pharm acokinetics. It does not cause w eight gain or orthostatic hypotension, but it has been suggested that it m ay increase the QTc interval . 125 There are no published reports of ziprasidone treatm ent specifically in the elderly. However, as this drug is nearly devoid of anticholinergic activity, it appears that it m ay have favourable effects in patients w ith cholinergic dysfunc­ tion, including those w ith Alzheim er's disease. It should be able to be used to treat psychosis in the elderly in a similar fashion to that in younger patients.

Zotepine Zotepine is a dibenzothiepine licensed for use as an atypical for schizophrenia . 126 It is m ore difficult to use in the elderly, in w h o m som e severe side-effects have b een reported, including tachycardia, hypotension, prolongation of the QTc interval, som n olence and sleep disorders. Parkinsonian side-effects w ere reported in about 5.3% of the patients treated, and seizures w ere observed at higher d oses . 127 There seem s to be little evidence to support its use in the elderly.

Long-acting injectable antipsychotics To date, low -dose fluphenazine has been the preferred typical depot antipsychotic for elderly patients because it produces less sedation, postural hypotension and anticholinergic effects . 38 Long-acting atypical antipsychotic form ulations have long been aw aited because of the im proved efficacy of the base com pounds 128 and the reduced expectancy of EPS and tardive dyskinesia. Risperidone is n o w the first agent to be approved in a long-acting injectable form ulation. The long delay in form ulating this depot has been due to the fact that, because of its lack of a hydroxyl group, risperidone cannot be esterified so it cannot be oilbased. H ow ever, the n ew form ulation encapsulates risperidone in 'microspheres' m ade of a biodegradable polym er that is delivered in a w ater-based vehicle. This provides a steady release of risperidone, w ith the polym er dissolving to form water and carbon end-products. There have been very few studies of the use of depot antipsychotics in elderly patients, but those that exist suggest positive outcom es. A total of 20 patients w ith chronic schizophrenia (ranging in age from 60 to 73 years) w ere treated w ith

The treatment of psychosis in the elderly

155

fluphenazine decanoate, 12.5 m g adm inistered every 21 days for 6 m on th s . 129 This effectively treated em otional w ithdrawal, blunted affect, suspiciousness and thought disturbances, but not hallucinations. Raskind et com pared 5 m g fluphenazine every 2 w eeks w ith oral haloperidol 2 mg three tim es daily for 6 w eeks in elderly patients w ith late-onset schizophrenia. In total, 11 out of 13 fluphenazine patients im proved, com pared w ith 3 out of 13 in the haloperidol group. This w as considered to be due to im proved com pliance. Low -dose depot zuclop en th ixol w as effective over a period of 1 2 w eeks in treating persistent psychom otor agitation and psychosis in six patients w ith Alzheim er's disease . 131 Low doses of fluphenazine enanthate in 10 elderly patients also im proved BPSD . 132 One study in volvin g risperidone in long-acting injectable form has been reported in stable psychotic elderly patients . 133 Patients of m ean age of 70.9 years w ere treated w ith long-acting risperidone 25 mg (n = 24), 50 mg (n = 17) or 75 m g ( n ~ 2) for up to 50 w eeks. Even though patients w ere clinically stable at entry to the study, 49% show ed a 20% reduction in their total positive and negative sym ptom s in schizophrenia (PANSS) score and 54.5% show ed an im provem ent on the clinical global im pression (CGI) . 134 In these elderly patients there w ere no cases of tardive dyskinesia as assessed by defined research criteria . 135 The m ost com m on adverse events w ere in som ­ nia (10.5% ), constipation (10.5% ) and bronchitis (12.3% ). W eight gain was 0.3 kg on average, and changes in ECG and vital signs w ere not clinically relevant. This suggests that this n e w m ethod of delivery is safe and effective over a period of a year in this com paratively frail patient group, even above the current recom m endation of 25 mg.

Short-acting injectable antipsychotics C lopenthixol acetate has b een used in the elderly, but no research evidence supports this. It exposes patients to all the risks of typical antipsychotics seen w ith other form ulations. Lorazepam is also used for rapid tranquillisation - again w ith no controlled data, and there is an increased risk of falls and sedation. A shortacting injectable preparation of olanzapine has been com pared w ith intram us­ cular lorazepam and placebo in the elderly . 136 Both proved superior to placebo, w ith olanzapine producing a faster and m ore sustained response. W hen available, this m ay prove a therapeutic choice w h en rapid tranquillisation is needed, and it m ay offer a m ore sensible alternative to the large haloperidol doses that are often prescribed in acute hospitals.

Conclusion B ecause the use of typical antipsychotics has fallen out of favour w ith geriatric psychiatrists, so too has the use of depot m edication over the last 1 0 years, except w here there has b een an absolute n eed to trade off potential side-effects against the n eed for efficacy. The availability of a long-acting injectable atypical anti­ psychotic w ill be a w elcom e addition to the formulary, especially for the lateonset schizophrenia patients w h o are notoriously resistant to treatm ent. These patients can tolerate doses similar to those used in younger adults, so are able to

156

Practical old age psychopharmacology

tolerate all three doses involved. The low est dose available delivers a 2 m g/day serum level. H owever, this m ay be too high for patients w ith BPSD, and further research is needed to clarify this.

K ey p o in ts The increasing use of atypical antipsychotics in the elderly has highlighted a num ber of im portant considerations. • The use of these drugs needs to be put into individual context. It is not alw ays necessary to treat a patient w ith late delusional disorder, and cases of BPSD should have non-pharm acological m ethods tried first. The use of antipsychotics as a panacea has n o w becom e m uch less com m on. • The mantra of 'start low , go slow' n o w has som e caveats. Low doses of typicals do not alw ays prove efficacious , 137 138 and low -dose olanzapine has proved no m ore effective than placebo in one study . 139 Elderly patients w ith schizophrenia tolerate similar doses to younger ones. Starting too low m ay not be therapeutic, so rapid escalation to similar doses m ay be necessary to control sym ptom s. • The real issue in using antipsychotics is side-effects. These w ill determ ine the outcom e of the treatm ent selected, and the w h ole rationale of m aking the choice involves balancing efficacy against side-effects. In an individual patient this m eans regular review , in the early stages to look for know n side-effects in order to m inim ise them or change the drug, and in the later stages either to look for later-developing side-effects or to review w heth er continuation is necessary. Clearly, the com plexity of titration and subsequent review w ill also affect clinical preference and choice. • The cytochrom e P450 system can determ ine drug-drug interactions. How m uch this affects clinical practice is uncertain, but all attempts to reduce adverse events, especially cardiological ones, should be encour­ aged. Elongated QTc intervals are a problem w ith m any of the anti­ psychotics. Care m ust be taken w ith patients w h o have risk factors, especially those w h o are on polypharm acy. •

M o st o f th e s e d ru g s are te c h n ic a lly b e in g u s e d o ff-lic e n c e , as ap art fr o m

risperidone, w h ich is licensed for psychosis, they are all usually licensed for schizophrenia. This increase in indications is being addressed by research. H owever, as a rule of thum b it is safe to assum e that an antipsychotic should treat the com m on psychotic sym ptom s, and pro­ vided that this remains the rationale for their use, medical opinion will support it. • Atypical antipsychotic agents have a significant role to play in the treatm ent of schizophrenia and other psychosis in late life. H owever, further studies in the elderly population are still required to make certain issues clearer, and until then these agents should be introduced carefully at relevant doses in elderly patients. Conversely, typicals seem to be relatively ineffective in BPSD, problematic w ith Parkinson's disease and DLB, and for primary psychosis their indefinite use w ill increase the

The treatment of psychosis in the elderly

157

risk of TD. Therefore in principle it w ould seem that their use in the elderly is n o w severely lim ited. • Elderly patients w ith major com orbidity, such as dem entia or other serious physical illnesses, w ill require doses of atypicals at the low er end of their respective ranges. C onversely, patients w ith schizophrenia­ like illnesses w ith ou t significant com orbidity w ill need and tolerate higher doses, similar to the doses used in younger patients.

References 1 Targum SD (2001) Treating psychotic sym ptom s in elderly patients. J Clin Psychiatry. 3: 1 5 6 -6 3 . 2 Lyketsos CG, Steinberg M, Tschanz JT et al. (2000) M ental and behavioural dis­ turbances in dem entia: findings from the Cache C ounty study on m em ory and aging. A m J Psychiatry. 157: 7 0 8 -1 4 . 3 M aletta GJ (1990) Pharm acologic treatm ent and m an agem en t of th e aggressive dem ented patient. Psychiatr A nn. 20: 4 4 6 -5 5 . 4 Furniss L, Craig SK and Burns A (1998) M edication use in nursing h om es for elderly people. In t J Geriatr Psychiatry. 3: 4 3 3 -9 . 5 Schneider LS (1993) Efficacy of treatm ent for geropsychiatric patients w ith severe m ental illness. Psychopharmacol B u ll 29: 5 0 1 -2 4 . 6 Raskind M A and Risse SC (1986) A ntipsychotic drugs and the elderly. J Clin Psychiatry. 47 (S u p p l.): 1 7 -2 2 . 7 N ational Institute for Clinical E xcellence (2002) Guidance on the Use o f Newer (Atypical) Antipsychotic Drugs fo r the Treatment o f Schizophrenia. NICE T echnology Appraisal No 43. N ational Institute for Clinical E xcellence, London. 8 Harris MJ and Jeste DV (1988) L ate-onset schizophrenia: an overview . Schizophr Bull. 14: 3 9 -5 5 . 9 Levy R and Naguib M (1985) Late Paraphrenia. B r J Psych. 146: 451. 10 Am eida OP, Howard RJ, Levy R and David AS (1995) Psychotic states arising in late life (late paraphrenia): the role of risk factors. Br J Psychiatry. 166: 2 1 5 -2 8 . 11 Tran-Johnson TK, Krull AJ and Jeste DV (1992) Late-life schizophrenia and its treatm ent: pharm acologic issues in older schizophrenia patients. Clin Geriatr Med. 8: 4 0 1 -1 0 . 12 Jeste DV, Harris MJ, Peralson GD et al. (1988) L ate-onset schizophrenia: studying clinical validity. Psychiatr Clin North A m . 11: 1-1 4 . 13 Evans DA, Funkenstein HH, Albert MS et al. (1989) Prevalence of A lzheim er's disease in a com m u n ity population of older persons: higher than previously reported. JAM A. 262: 2 5 5 1 -6 . 14 Wragg R and Jeste DV (1989) O verview of depression and psychosis in A lzheim er's disease. A m J Psychiatry. 146: 5 7 7 -8 7 . 15 Schneider LS, Pollock VE and Lyness SA (1990) A m eta-analysis of controlled trials of neuroleptic treatm ent in dem entia. J A m Geriatr Soc. 38: 5 5 3 -6 3 . 16 Lanctot KL, Best TS, M ittm ann N et a l (1998) Efficacy and safety of neuroleptics in behavioral disorders associated w ith dem entia. J Clin Psychiatry. 59: 5 5 0 -6 1 . 17 K inderm ann SS, D older CR, B ailey A, Katz IR and Jeste DV (2002) Pharm acological treatm ent of psychosis and agitation in elderly patients w ith dem entia: four decades of experience. Drugs Aging. 19: 2 5 7 -7 6 .

158

Practical old age psychopharmacology

18 Brodaty H, A m es D, Snow don J et al. (2003) A random ized placebo-controlled trial of risperidone for the treatm ent of aggression, agitation and psychosis of dem entia. J Clin Psychiatry. 64: 1 3 4 -4 3 . 19 De D eyn PP, Rabheru K, R asm ussen A et al. (1999) A random ized trial of risperidone, placebo, and haloperidol for behavioral sym ptom s of dem entia. Neurology. 53: 8 9 9 -9 0 1 . 20 Katz IR, Jeste DV, M intzer JE, Clyde C, N apolitano J and Brecher M (1999) Comparison of risperidone and placebo for psychosis and behavioral disturbances associated w ith dem entia: a random ized, double-blind trial. Risperidone Study Group. J Clin Psychiatry. 60: 1 0 7 -1 5 . 21 Street JS, Clark WS, G annon I