Article Text
PDF
Statistics from Altmetric.com
Question
Question What is the comparative efficacy, risk of all-cause discontinuation and major side effects of antipsychotics used for the treatment of schizophrenia?
Outcomes The mean overall change in symptoms (primary outcome), all-cause discontinuation, weight gain, extrapyramidal side effects, prolactin increase, prolonged QT interval and sedation.
Methods
Design Systematic review and multiple-treatment meta-analysis.
Data sources Cochrane Schizophrenia Group's specialised register, MEDLINE, EMBASE, PsycINFO, CENTRAL and ClinicalTrials.gov were searched up to September 2012, supplemented by review of Food and Drug Administration and pharmaceutical company websites and reference lists of reviews.
Study selection and analysis Randomised controlled trials (single-blind at minimum) of 15 orally administered antipsychotic drugs used as monotherapy for the acute treatment of schizophrenia or related disorders (schizoaffective, schizophreniform or delusional disorder). Trials in stable patients, those with predominant negative symptoms, concomitant medical illnesses or treatment resistance were excluded. A multiple-treatment meta-analysis was performed. The standardised mean differences were calculated for continuous outcomes and ORs for binary outcomes.
Main results
A total of 212 studies, with 43 049 participants were included (mean age 38.4 years, mean duration of illness 12.4 years). All drugs were more effective than placebo, and clozapine was significantly more effective than all other drugs. The standardised mean differences with 95% credible intervals (CrI), in the order of effectiveness of 15 antipsychotics, were clozapine, amisulpride, olanzapine, risperidone, paliperidone, zotepine, haloperidol, quetiapine, aripiprazole, sertindole, ziprasidone, chlorpromazine, asenapine, lurasidone and iloperidone. The odds of all-cause discontinuation were significantly lower with all drugs compared with placebo (ORs ranging from 0.43 for amisulpride to 0.80 for haloperidol), with the exception of zotepine. Apart from haloperidol, ziprasidone and lurasidone, all drugs produced more weight gain than placebo. Examining the frequency of extrapyramidal side effects, compared with placebo, it was significantly fewer with clozapine; there were no significant differences with sertindole, olanzapine, quetiapine, aripiprazole, iloperidone, amisulpride and asenapine; and all other drugs caused significantly more extrapyramidal side effects than placebo. Aripiprazole, quetiapine, asenapine, chlorpromazine and iloperidone did not cause significantly increased prolactin concentrations compared with placebo; all other drugs caused significantly more prolactin (amisulpride, clozapine and zotepine not analysed). QT interval was not significantly prolonged with lurasidone, aripiprazole, paliperidone and asenapine compared with placebo, but it was significantly prolonged with all other drugs (clozapine, chlorpromazine and zotepine not analysed). Amisulpride, paliperidone, sertindole and iloperidone were not significantly more sedating than placebo; all other drugs were more sedating than placebo. For effect sizes, 95% CrI and rankings of all 15 antipsychotic drugs compared with placebo for all outcomes see Webextra table.
Conclusions
This multiple-treatment meta-analysis provides evidence-based hierarchies for the efficacy and tolerability of antipsychotic drugs.
Abstracted from
Commentary
The analysis by Leucht and colleagues is an impressive work, and the authors are to be commended on describing their sophisticated analyses in such a cogent manner. It is evident that first-generation and second-generation antipsychotics overlap substantially.1 However, in evidence-based practice we need to take into account the values and preferences that individual patients may have, and should also consider the penalty imposed on patients when benztropine or other anticholinergic medicine is required.2 In addition, rank-ordering medications based on data gathered from group comparisons can be problematic as there can be substantial heterogeneity that prevents the reliable prediction of effects on individual persons, except when there are large differences in the outcome in question such as, for example, effects on prolactin increase with aripiprazole vs. risperidone/paliperidone.1
Of particular concern is how the Leucht study is reported in the professional media (http://alert.psychiatricnews.org/2013/06/comparison-of-15-antipsychotics-finds.html), with statements such as “The researchers found that … respectively, the greatest efficacy, while … showed the least,” which if taken in isolation may lead clinicians to avoid medications at the bottom of the list without fully considering other relevant outcomes such as weight gain, extrapyramidal side effects and sedation that are well-described and quantified elsewhere in the paper. Matching up the right medication to the right patient will require careful consideration of the tolerability and safety profile of the medication which will impact on the patient’s adherence to treatment and on the clinician's willingness to prescribe. This study has the potential to positively impact clinical decision-making, provided that appropriate attention is paid to both efficacy and tolerability.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online supplement
Footnotes
-
Competing interests LC in the past 36 months was engaged in collaborative research with, or received consulting or speaking fees from Alexza, Alkermes, AstraZeneca, Avanir, Bristol-Myers Squibb, Eli Lilly, Envivo, Forest, Genentech, Janssen, Lundbeck, Merck, Mylan, Novartis, Noven, Otsuka, Pfizer, Reckitt Benckiser, Reviva, Shire, Sunovion, Takeda and Valeant.