Atypical antipsychotics in the treatment of schizophrenia: systematic overview and meta-regression analysis
BMJ 2000; 321 doi: https://doi.org/10.1136/bmj.321.7273.1371 (Published 02 December 2000) Cite this as: BMJ 2000;321:1371
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
EDITOR - In a recent study of the subjective responses of patients
stabilised on antipsychotic medication, atypical antipsychotic drugs were
found to offer better tolerability, a lower prevalence of neuroleptic
dysphoria, and improved quality of life compared with conventional
antipsychotics1. However, in a systematic review of randomised controlled
trials of acute antipsychotic treatment, Geddes et al2 concluded that
"there is no clear evidence that atypical antipsychotics are more
effective or are better tolerated than conventional antipsychotics" in the
treatment of schizophrenia. We contend that Geddes et al.’s assertion is
based on a flawed premise and is contrary to the results of other
controlled trials as well as expert guidelines in the US3 and other
countries.
In their paper, Geddes et al claim that doses of haloperidol above
12mg/day are inappropriate and that the optimal dose of haloperidol is
between 6 and 12mg/day. We concur with this observation but also point out
that there is compelling evidence that the optimal target daily dose of
risperidone is 4-6mg/day4. Thus, in their meta-analysis, only trials
comparing doses of risperidone £6mg/day with haloperidol £12mg/day should
have been included. However, of the 10 risperidone-comparative trials
cited, only one meets this criterion5. This was a dose finding trial of
risperidone using fixed dose haloperidol 10mg/day as a comparator which
showed that risperidone 4mg/day was at least as effective as haloperidol
in reducing symptoms and resulted in fewer dropouts due to lack of
efficacy. In addition, at these doses, risperidone was significantly
better tolerated in terms of fewer EPS and dropouts due to adverse
effects. Further, a post-hoc analysis of data from this trial revealed
that risperidone 4mg/day was associated with a significantly more rapid
relief of symptoms than haloperidol 10mg/day, with pronounced differences
during the first week of treatment6.
More importantly, new evidence from a recent long-term, double-blind
study comparing risperidone (median dose 4mg/day) and haloperidol (median
dose 10mg/day) has now confirmed the statistically superior symptom
improvement in patients receiving risperidone, and has shown a 36%
reduction in the rate of relapse as well as a longer time to relapse7.
This study confirms that patients receiving risperidone at a dose of
4mg/day derive significant benefits compared with haloperidol even when it
is administered at doses lower than 12mg/day. Moreover, unlike the older
evidence cited by Geddes et al., this trial meets the NHS CRD call for
‘large, long-term randomised drug trials’ that include relapse prevention
as a more relevant outcome ‘familiar to health professionals who treat
people with schizophrenia’8.
John Csernansky, MD, Department of Psychiatry
Washington University School of Medicine,
St Louis, MO63130, USA
csernanj@medicine.wustl.edu
Jonathan Rabinowitz, PhD, Professor of Social Welfare
Bar Ilan University,
Ramat Gan, Israel
rabinowz@mail.biu.ac.il
Jozef Peuskens, MD, PhD, Department of Psychiatry
University Centre St Jozef, Catholic University of Leuven,
Kortenberg, Belgium
med_sec@uc-kortenberg.be
Funding and competing interests.
JC has received funding for research projects and honoraria for
lectures from Janssen Pharmaceutica. JR has received funding for research
projects and honoraria for lectures from Janssen Pharmaceutica, Eli Lilly
and Novartis. JP has received funding for research projects and honoraria
for lectures from Janssen Pharmaceutica, Eli Lilly, Pfizer, Astra Zeneca,
Novartis, Sanofi-Synthelabo amd Lundbeck.
1. Voruganti L, Cortese L, Oyewumi L, Cernovsky Z, Zirul S, Awad A.
Comparative evaluation of conventional and novel antipsychotic drugs with
reference to their subjective tolerability, side-effect profile and impact
on quality of life. Schizophrenia Research 2000;43:135-45.
2. Geddes J, Freemantle N, Harrison P, Bebbington P. Atypical
antipsychotics in the treatment of schizophrenia: systematic overview and
meta-regression analysis. BMJ 2000;321:1371-6
3. American Psychiatric Association: Practice guidelines for the
treatment of schizophrenia. Am J Psychiatry 1997;154 (supplement 4)
4. Lemmens P, Brecher M, Van Baelen B. A combined analysis of double-
blind studies with risperidone vs placebo and other antipsychotic agents:
factors associated with extrapyramidal symptoms. Acta Psychiatr Scand
1999;97:1-41
5. Peuskens J et al. Risperidone in the treatment of patients with
chronic schizophrenia: a multi-national, multi-centre, double-blind,
parallel-group study versus haloperidol. Brit J Psychiatry 1995;166:712-26
6. Rabinowitz J, Hornik T, Davidson M. Rapid onset of therapeutic
effect of risperidone versus haloperidol in a double blind randomized
trial. J Clin Psychiatry 2001 In press
7. Csernansky JG, Okamoto A. A long-term double-blind comparison of
risperidone and haloperidol in stable outpatients with schizophrenia or
schizoaffective disorder. Int J Psychopharmacol 2000;3 (Supplement 1):S155
8. NHS Centre for Reviews and Dissemination. Drug treatments for
schizophrenia. Effective healthcare 1999;5:1-12
Competing interests: No competing interests
Editor - The impressive meta-regression analysis by Geddes et al in
the 2nd December 2000 BMJ has left us feeling a little disappointed in its
"side-stepping" of current issues. The authors believe that "cost is not a
crucial issue at present" as the majority of outcome scores for the
atypical antipsychotics are the same as those for the typical
antipsychotics. However, the predominant message of the article is that
typicals should be prescribed first-line rather than atypicals. We
strongly disagree with the notion that "cost is not a crucial issue at
present" as it is apparently the only differentiator between the new and
old antipsychotics. If cost was not a "crucial issue" why would Geddes and
his co-workers recommend which type of antipsychotic should be prescribed
first?
As trainee psychiatrists, perhaps we suffer from too much of what Kapur
describes in his editorial as "clinical hope" but we do see a key first-
line role for atypicals. We appreciate our flexibility of being able to
work collaboratively with patients and are able to discuss factors such as
weight gain, akathisia, parkinsonism, sexual dysfunction and risk of
tardive dyskinesia. In offering a choice of medication to our patients
they effectively have a choice over their potential side-effects.
Prescribing only typicals would reduce this choice.
The authors state that no single antipsychotic is to be favoured over
another and we would like to have all treatment options available in our
first line armoury.
We encourage Geddes and his colleagues to recognise that cost is a crucial
issue and to realise that in effect they are presenting a cost-
minimisation analysis.
Richard Duggins
Senior House Officer in Psychiatry
David Rhinds
Senior House Officer in Psychiatry
William Hall
Senior House Officer in Psychiatry
Division of Psychiatry, University of Nottingham,
Duncan MacMillan House, Nottingham NG3 6AA
Competing interests: No competing interests
Editor - Geddes et al1 use clinical trial data concerning mainly
chronic schizophrenic patients to propose guidelines for a different
schizophrenic patient population; namely that low dose typical
antipsychotics should be the initial treatment for patients in their first
episode of illness. This approach, in addition to its casual flouting of
the usual conventions of evidence based medicine, is we think misguided
for three further reasons.
The authors concede that atypical antipsychotics produce fewer
extrapyramidal side effects (EPS) than typical antipsychotics, but
contrast this with evidence of similar 'overall tolerability' as defined
by study dropouts for both classes of medicine. However tolerability
refers to the entire clinical trial of which the prescribed antipsychotic
forms only one component. Given the close and often intrusive supervision
characteristic of clinical trials the finding of no difference in overall
tolerability between the to treatment groups is perhaps not surprising.
EPS may be more relevant to the treatment of first onset
schizophrenic patients. Clinical trial data with 6mg risperidone daily
shows that 50% of first onset patients develop EPS compared to 22% of
chronic patients. A controlled study of haloperidol found EPS appeared at
doses as low as 2.1mg (range 0.5-4) in first episode compared to 4.3mg
(range 0.5-10) in chronically treated individuals2. Thus first episode
patients are more sensitive to such side effects, and that in this patient
group the therapeutic and EPS effects of typical antipsychotic drugs
cannot be separated by dose.
Geddes et al's1 comparison with lower dose typical antipsychotic
treatment ignores recent changes in prescribing practice for atypical
antipsychotics. Clinical trials of the atypical antipsychotic risperidone
included a wide dose range (2-16mg daily) although at doses above 6mg
daily risperidone loses its advantages in terms of tolerability. A more
pertinent comparison would be between lower dose typical and atypical
drugs.
The 'wait and see' policy concerning the emergence of EPS as
advocated in the article assumes that schizophrenia involves a static
disease process. In fact antipsychotic treatment response in schizophrenia
may decline particularly during the early course of the illness after
recurrent episodes of the illness3. Compliance with medication is the main
predictor of relapse in schizophrenia4. EPS importantly contribute to a
negative perception of antipsychotic treatment and lead to early non-
compliance. The use of toxic therapeutic agents in this early phase of the
illness may thus have profound longer-term consequences for patients with
schizophrenia.
References
1. Geddes J, Freemantle N, Harrison P, Bebbington P for the National
Schizophrenia Guideline Development Group. Atypical antipsychotics in the
treatment of schizophrenia: systematic overview and meta-regression
analysis BMJ 2000 321 : 1371-1375
2. McEvoy JP, Hogarty GE, Steingard S (1991) Optimal dose of
neuroleptic in acute schizophrenia: a controlled study of the neuroleptic
threshold and higher haloperidol dose, Arch. Gen. Psychiatry, 48, 739-745.
3. McGlashan T The profiles of clinical deterioration in
schizophrenia Journal of Psychiatric research 1998 32(3-4):133-41
4. Kapur S, Remington G Atypical antipsychotics (editorial) BMJ 2000
321: 1360
Dr Hugh M. Jones MRCPsych
Clinical research worker and Honourary Specialist Registrar
Dr Lyn S. Pilowsky MRCPsych PhD
MRC Senior Clinical Research Fellow
Section for Neurochemical Imaging and Psychiatry
Institute of Psychiatry
Corresponding author
Dr Hugh M. Jones MRCPsych
Section for Neurochemical Imaging and Psychiatry,
Institute of Psychiatry,
De Crespigny Park, Denmark Hill, London SE5 8AF
Competing interests: No competing interests
Sir,
In the recent review of antipsychotics in the treatment of schizophrenia
(1), the authors claim to have found no evidence of differential efficacy
either between conventional and atypical agents, or within the atypical
class itself. The accompanying editorial (2) did however acknowledge the
superior effects of the atypicals on aspects of mood, cognition, hostility
and in particular EPS. The increased sensitivity of first episode and
elderly patients to neurological and other side effects associated with
typical antipsychotic drugs was highlighted in subsequent correspondence
(3, 4). These observations are reflected in our own clinical experience
with the atypicals, and we have recently also found mortality to be
significantly higher in elderly patients receiving haloperidol compared
with atypical antipsychotics (5).
But even if we assume that the sole advantage of atypical
antipsychotics is lower EPS, then that would still be a powerful
justification for using them as first-line treatment based on the
malignant cascade of EPS effects that would be conflicted by conventional
neuroleptics. In addition to avoiding iatrogenic movement disorders,
minimizing EPS also means no dysphoria, no cognitive slowing, no
anticholinergic co-pharmacy, no secondary negative symptoms, and far less
(1/6) tardive dyskinesia. Patients with schizophrenia and related
psychoses are already devastated by a serious brain disease and we should,
above all, do no additional harm while treating their symptoms. The
atypicals trump the conventionals in this regard even if the efficacy is
the same (although many studies demonstrate it to be broader). Patient
satisfaction, adherence to treatment, suicide rates, and social/vocational
outcomes have all improved with atypicals and those effectiveness measures
are completely unaddressed in a meta-analysis of registration efficacy
trials.
Henry A. Nasrallah, MD
Professor of Psychiatry, Neurology, and Internal Medicine
University of Mississippi
Funding and Competing Interests
- I have received peer-reviewed funding from the NIMH, the VA, and the
Stanley Foundation.
- Non-peer reviewed funding includes grants from Abbott, Astra-Zeneca,
Janssen, Lilly, and Pfizer.
- I have received honoraria for speaking from a wide range of
pharmaceutical companies as well as from universities, teaching hospitals,
and public institutions.
- I have served as a consultant for from Abbott, Astra-Zeneca, Janssen,
Lilly, and Pfizer.
References
1 Geddes J, Freemantle N, Harrison P, Bebbington P. Atypical
antipsychotics in the treatment of schizophrenia: a systematic overview
and meta-regression analysis. BMJ 2000, 321, 1371-1376
2 Kapur S, Remington G. Atypical antipsychotics: patients value the
lower incidence of extrapyramidal side effects. BMJ 2000, 321, 1360-1361.
3 Prettyman R. Age is an important determinant of antipsychotic
tolerability. BMJ 2000 Rapid response to Ref 1 above.
4 Kerwin R. Atypical antipsychotics: going beyond the evidence. BMJ
2000. Rap[id response to Ref 1 above.
5 Nasrallah H, White T, Nasrallah A. Lower mortality in geriatric
patients receiving atypical antipsychotics compared to haloperidol.
Poster presented at the 39th ACNP meeting, San Juan, 10-14 Dec 2000.
Competing interests: No competing interests
Dear Dr Smith,
Dr Manning makes a powerful case for "narrative based" medicine.
Clinical practice is based on a continual tension between the
patient's personal account of their own experience and the doctor's wish
to achieve defined clinical outcomes.
Evidence based medicine is a useful tool but it doesnt provide an all
purpose solution. Part of the reason for this is that evidence itself is
always testing the limit of uncertainty. We know from physics that it is
impossible to be sure of both the position and velocity of atomic
particles.
If something as basic as that is open to doubt then dogma on any
subject, is nothing more or less than cant.
Of course judgements have to made but they should be against a
background of tolerance and a willingess to consider it possible that we
may be mistaken.
Yours sincerely,
Dr John Hopkins
Competing interests: No competing interests
EDITOR - The meta-analysis and meta-regression analysis of atypical
versus typical antipsychotics by Geddes et al (1)
is an important contribution to the current debate, but as they
acknowledge must be taken as the first, rather than the last, word on the
topic.
Overall they find advantages to the atypicals compared with typicals
as used in current clinical practice. This is important given the repeated
failure to get doctors to increase the dose of tricyclic antidepressants
they prescribed in spite of over years of exhortation (2); it is extremely
optimistic to imagine that typical antipsychotic doses will be reduced in
practice.
With lower typical antipsychotic doses the efficacy advantage for
atypicals diminishes rather than necessarily disappears (the 95%
confidence interval for efficacy encompasses a clinically significant
advantage to atypicals but only a small disadvantage). The tolerability
analysis is problematic. I take issue with the confident equating of
tolerability with drop outs due to all causes. The total number of
patients stopping treatment is a very 'noisy' measure comprising a mixture
of discontinuation due to protocol deviation, withdrawal of consent,
treatment failure or success, as well as adverse events. It is sensible to
analyse discontinuations attributed to adverse events alongside total drop
outs, but even then we are only looking at the 'hard end' of lack of
tolerability - using this measure alone won't detect the patients who
continue in acute treatment spite of significant side effects. This
analysis did find that atypicals do cause fewer extrapyramidal side
effects even than lower dose typicals and the strong clinical impression,
supported by the editorial in the same issue (3), is that patients choose
to avoid them if they can. What is missing from the interpretation of this
analysis is any acknowledgement that the 'user view' may be important. It
is difficult therefore to accept the authors conclusion that atypicals are
no better tolerated than typicals; on the evidence given it just depends
on what you mean.
Finally, and perhaps most importantly, the value judgements used to arrive
at the recommendations based on the evidence are not explicit; this does
us and the subject a disservice. The concluding value judgement that we
should usually use typicals first line appears, in spite of protests, to
be driven by an implicit cost-effectiveness judgement weighing a large
cost difference against a perceived minor clinical benefit. I have argued
here about the certainty of the evidence; how much more shaky and opaque
is the translation to recommendation? My interpretation is that we are
condemned on current best evidence to continue living with the uncertainty
about whether to use typical or atypical antipsychotics first line.
Perhaps this is a situation in which we should listen to our patients?
Reference List
1. Geddes J, Freemantle N, Harrison P, Bebbington P. Atypical
antipsychotics in the treatment of schizophrenia: systematic overview and
meta-regression analysis. Br.Med.J. 2000;321:1371-1376.
2. Lawrenson RA, Tyrer F, Newson RB, Farmer RD. The treatment of
depression in UK general practice: selective serotonin reuptake inhibitors
and tricyclic antidepressants compared. J Affect.Disord. 2000;59:149-157.
3. Kapur S, Remington G. Atypical antipsychotics. Patients value the
lower incidence of extrapyramidial side effects. Br.Med.J. 2000;321:1360-
1361.
Competing interests: No competing interests
Editor – we were concerned to read the paper by Geddes et al.
extolling the virtues of a typical antipsychotic haloperidol over more
expensive atypical drugs1.
This work was originally prepared for the Department of Health in
1998, hardly updated since and employs dubious hypothesis-generating
statistical methods. It has generated heated debate in the
psychopharmacology community whilst being prepared for publication. Thus,
several sources of bias are immediately apparent. It is interesting to
observe that the expertise represented by the authors is not obviously
primarily in the area of the psychopharmacology of schizophrenia.
The conclusions drawn that relate to the efficacy of haloperidol at
low doses go beyond the evidence available, the interpretation of which
seems unduly selective. For example, a low dose of a typical neuroleptic
may initially reduce PANSS scores significantly due to non-specific
effects such as anxiolysis and sedation, and these effects are not
adequately partialled out from true antipsychotic effect in the analysis.
In one study cited, a study which was designed in close collaboration with
the FDA and is usually seen as the definitive placebo-controlled dose-
ranging study of haloperidol, there was no difference in either the
positive symptom subscale score of the BPRS (itself a subscale of PANSS),
nor the Clinical Global Improvement score when comparing placebo with 4mg
haloperidol daily2. However, haloperidol did produce extra-pyramidal side
effects at this dose. One might conclude from this that it is unlikely
that an effective and tolerable dose-range of haloperidol can be found.
Further evidence of the usefulness or otherwise of low dose
haloperidol is awaited. We understand that a Cochrane review of this
issue is expected soon which should allow firmer conclusions to be drawn
on this matter.
1. Geddes J, Freemantle N, Harrison P, Bebbington P. Atypical
antipsychotics in the treatment of schizophrenia: systematic overview and
meta-regression analysis BMJ 2000; 321: 1371-6
2. Zimbroff D L, Kane JM, Tamminga CA et al. Controlled, dose response
study of sertindole and haloperidol in the treatment of schizophrenia.
Sertindole Study Group. Am J Psychiatry 1997; 154: 782-91
Declaration of interests – MK is currently involved in research into
an atypical antipsychotic drug, funded by a pharmaceutical company. AH is
on the advisory boards of many companies producing neuroleptics, both
typicals and atypicals.
Competing interests: No competing interests
Dear Editor
Well, no bias in this response from Ron Law then, who perpetuates the
myth that scientifically "evidence-guided" medicine is anything better
than measuring a sphere with a ruler.
It is also tedious to read the word "anecdotal" being used in this
continually perjorative sense by the Evidence-Based Brigade - indeed it is
a sure marker for their usual behaviour - they almost gag on the word
whenever it is used at the conferences I have attended.
To think that "science" (that well-known completely non-perceptually
driven and objective human activity conducted by unbiased, balanced
individuals which takes place in laboratories), through RCTs, or any of
its other activities, captures all of life, or captures it more
completely, and that anecdotes (or the life experience of an individual or
any number of them)can simply be marginalised is to make precisely the
same error of biased thinking.
RCTs admit bias the minute they are designed - which is why in
clinical practice, they are often woefully inadequate in guiding real
clinicians, rather than researchers, in the treatment of the types of
patient who have usually been excluded from such trials in the first
place.
Fact - patients (real people with a diagnosis) "prefer" atypical
antipsychotics - real people in real life saying how traditional rocket-
fuel derivatives make them feel. The RCTs do not include this for a start
and so they miss a major point about prescribing. A spoonful of sugar may
have no clinical efficacy in itself, but it does make a difference to
outcome, simply because the medicine gets taken.
Also, the "Evidence" is imparted by experts as some sort of gospel to
practitioners who are now too busy to think and analyse for themselves -
it has moved on from advice about best practice to the suggestion that, if
you do not practice according to "The Evidence", you are in some way
inferior, or even negligent.
Where, for example, have NiCE put the positive evidence for the
effectiveness of counselling? How can it be evidence, if we are not
considering all interventions honestly? Who decides which evidence is
going to be selected and then propagated? How can you possibly mix so many
ingredients (with different samples/cohorts/inclusion criteria) into the
meta-analysis pudding and seriously think you are doing anything other
than cooking the results?
The issue is that Evidence-Based medicine is already out of date -
driven as it is by the previous Newtonian model of science. Next we will
have a thermodynamically driven one and in about 100 years time, clinical
researchers will discover chaos and complex systems theory.
Those of us in the "real" world are already practising the "chaos"
way and allowing for anecdote (the patient's voice, view, experience,
history, perception and our own) - which is why NiCE/CHi and the EBM
movement, whilst being helpful, if kept in their place, do not and should
not have the last word. "Chaos-driven" practitioners working in a complex
system can pick and choose and not be constrained by corsets masquerading
as Wonderbras.
It is time for what Dr Antony Pelosi calls "evidence-based medicine"
(without the capitals NB) - this does encompass the fuller "chaos" reality
and includes personal experience and anecdote.
Surely,this is not about "either...or" anyway, we need a "both..and"
approach to do scientific ("knowing" in the fuller sense of that word and
probably knowing what we do not know rather than what we do)justice to the
complexity of life.
Dr Chris Manning
Conflict of interest: I'm biased against economic-fiscal mutton studies
being dressed as last word lamb.
Competing interests: No competing interests
LEVELS OF EVIDENCE (1)
Isn't it marvellous how in practice level 1++ evidence seems to in
fact be a lower level than level 4 evidence.
One anecdotal case report (written by a biased medical 'expert') of
an adverse reaction to a herbal product is deemed of such scientific
significance and declared definitive that a major public health issue is
declared.
One high quality meta analysis, systematic reviews of RCTs involving
thousands of subjects is debated and obfuscated for another decade before
a meaningful outcome is achieved.
How absurd.
Bring on scientific-based medicine.
(1) http://www.show.scot.nhs.uk/sign/levels.htm
1++
High quality meta analyses, systematic reviews of RCTs, or RCTs with a
very low risk of bias
1+
Well conducted meta analyses, systematic reviews of RCTs, or RCTs
with a low risk of bias
1 -
Meta analyses, systematic reviews of RCTs, or RCTs with a high risk
of bias
2++
High quality systematic reviews of case-control or cohort or studies
High quality case-control or cohort studies with a very low risk of
confounding, bias, or chance and a high probability that the relationship
is causal
2+
Well conducted case control or cohort studies with a low risk of
confounding, bias, or chance and a moderate probability that the
relationship is causal
2 -
Case control or cohort studies with a high risk of confounding, bias,
or chance and a significant risk that the relationship is not causal
3
Non-analytic studies, e.g. case reports, case series
4
Expert opinion
Competing interests: No competing interests
Atypical antipsychotic drugs and risk for ischemic stroke: still an open question
With regard to the ongoing discussion of safety of treatment with
antipsychotics in patients with behavioural and psychological symptoms of
dementia (BPSD) we read with great interest the study by Gill and
colleges. The retrospective data-based design reflects general practice in
the use of the medications and avoids institutional selection bias. The
large number of patients included overcomes to some extent the necessity
of standard classifications of stroke such as TOAST criteria (1). However,
we feel that this study has methodological shortcomings, which may
seriously compromise the interpretation of its results.
1. Selection of diagnoses
Diagnoses were made according to ICD-9 criteria. Patients with
intracerebral haemorrhage (ICD-9: 431), with occlusion of cerebral
arteries, with and without cerebral infarction, (ICD-9: 434) and with
acute, but ill-defined cerebrovascular disease, explicitly including
seizure and cerebrovascular accident (ICD9: 436) were admitted to the
analysis. On the other hand, the studied population comprises no patients
with extracranial stenosis, with or without cerebral infarction (ICD-9:
433).
2. Certainty of diagnoses
The surprisingly high accuracy suggested for these admission diagnoses
(quoted as 90 %) with regard to ischemic stroke as a total cannot apply
for patients with BPSD. As neurologists seeing such patients frequently,
we are convinced that diagnostic accuracy can only be raised to 90 % after
a thorough evaluation by specialists including cerebral imaging.
Dehydration, infections and metabolic disturbances frequently lead to
stroke-like symptoms in older adults, particularly in patients with BPSD.
3. Low overall incidence of stroke
The overall incidence of stroke in the studied population is much lower
than in the respective controlled trials (2,3). This may be attributed to
the omission of a major patient group (ICD-9: 433) or to other diagnostic
artefacts.
4. Comparability of patient groups
The patients treated with atypicals got a better treatment of vascular
risk factors. Clopidogrel and statines were used more frequently in this
patient group, probably reflecting a secular trend. Thus, the patients
treated with atypicals had a better stroke prophylaxis, which may explain
why no significant difference in the patient groups analysed could be
ascertained.
Having considered a letter to the editor after reading the online
version of this paper, we decided to do so after we found out that most of
the critical points we refer to have been omitted in the abridged version
published.
1 Adams HP, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL,
Marsh EE: Classification of subtype of acute stroke. Definition for use in
a multicenter clinical trail. TOAST. Trial of Org 10172 in Acute Stroke
Treatment. Stroke 1993; 24(1); 35-41.
2 Wooltorton E.: Risperidone (Risperdal): increased rate of
cerebrovascular events in dementia trails. CMAJ 2002;167:1269-70.
3 Wooltorton E.: Olanzapine (Zyprexa): increased rate of
cerebrovascular events in dementia trails. CMAJ 2004;170:1395.
Competing interests:
None declared
Competing interests: No competing interests