Background: Haloperidol is worldwide one of the most frequently used antipsychotic drugs with a very high market share. Previous narrative, unsystematic reviews found no differences in terms of efficacy between the various first-generation ("conventional", "typical") antipsychotic agents. This established the unproven psychopharmacological assumption of a comparable efficacy between the first-generation antipsychotic compounds codified in textbooks and treatment guidelines. Because this assumption contrasts with the clinical impression, a high-quality systematic review appeared highly necessary.
Objectives: To compare the efficacy, acceptability, and tolerability of haloperidol with other first-generation antipsychotics in schizophrenia and schizophrenia-like psychosis.
Search methods: In October 2011 and July 2012, we searched the Cochrane Schizophrenia Group's Trials Register, which is based on regular searches of CINAHL, BIOSIS, AMED, EMBASE, PubMed, MEDLINE, PsycINFO, and registries of clinical trials. To identify further relevant publications, we screened the references of all included studies and contacted the manufacturers of haloperidol for further relevant trials and missing information on identified studies. Furthermore, we contacted the corresponding authors of all included trials for missing data.
Selection criteria: We included all randomised controlled trials (RCTs) that compared oral haloperidol with another oral first-generation antipsychotic drug (with the exception of the low-potency antipsychotics chlorpromazine, chlorprothixene, levopromazine, mesoridazine, perazine, prochlorpromazine, and thioridazine) in schizophrenia and schizophrenia-like psychosis. Clinically important response to treatment was defined as the primary outcome. Secondary outcomes were global state, mental state, behaviour, overall acceptability (measured by the number of participants leaving the study early due to any reason), overall efficacy (attrition due to inefficacy of treatment), overall tolerability (attrition due to adverse events), and specific adverse effects.
Data collection and analysis: At least two review authors independently extracted data from the included trials. The methodological quality of the included studies was assessed using The Cochrane Collaboration`s 'Risk of bias' tool.We analysed dichotomous outcomes with risk ratios (RR) and continuous outcomes with mean differences (MD), both with the associated 95% confidence intervals (CI). All analyses were based on a random-effects model and we preferably used data on an intention-to-treat basis where possible.
Main results: The systematic review currently includes 63 randomised trials with 3675 participants. Bromperidol (n = 9), loxapine (n = 7), and trifluoperazine (n = 6) were the most frequently administered antipsychotics comparator to haloperidol. The included studies were published between 1962 and 1993, were characterised by small sample sizes (mean: 58 participants, range from 18 to 206) and the predefined outcomes were often incompletely reported. All results for the main outcomes were based on very low or low quality data. In many trials the mechanism of randomisation, allocation, and blinding was frequently not reported. In short-term studies (up to 12 weeks), there was no clear evidence of a difference between haloperidol and the pooled group of the other first-generation antipsychotic agents in terms of the primary outcome "clinically important response to treatment" (40 RCTs, n = 2132, RR 0.93 CI 0.87 to 1.00). In the medium-term trials, haloperidol may be less effective than the other first-generation antipsychotic group but this evidence is based on only one trial (1 RCT, n = 80, RR 0.51 CI 0.37 to 0.69).Based on limited evidence, haloperidol alleviated more positive symptoms of schizophrenia than the other antipsychotic drugs. There were no statistically significant between-group differences in global state, other mental state outcomes, behaviour, leaving the study early due to any reason, due to inefficacy, as well as due to adverse effects. The only statistically significant difference in specific side effects was that haloperidol produced less akathisia in the medium term.
Authors' conclusions: The findings of the meta-analytic calculations support the statements of previous narrative, unsystematic reviews suggesting comparable efficacy of first-generation antipsychotics. In efficacy-related outcomes, there was no clear evidence of a difference between the prototypal drug haloperidol and other, mainly high-potency first-generation antipsychotics. Additionally, we demonstrated that haloperidol is characterised by a similar risk profile compared to the other first-generation antipsychotic compounds. The only statistically significant difference in specific side effects was that haloperidol produced less akathisia in the medium term. The results were limited by the low methodological quality in many of the included original studies. Data for the main results were low or very low quality. Therefore, future clinical trials with high methodological quality are required.