Review: limited evidence on effects of haloperidol alone for rapid tranquillisation in psychosis-induced aggression
Question: Is haloperidol alone effective in reducing psychosis-induced agitation or aggression?
Outcomes: Primary outcomes: not being asleep or tranquil; needing repeated rapid tranquillisation.
Design: Systematic review with meta-analysis.
Data sources: The Cochrane Schizophrenia Group Trials Register was searched for randomised controlled trials (RCTs) from inception to 1 June 2011. The register is populated from searches major databases, hand searches of journals and conference proceedings. Reference lists of identified studies were hand searched. Authors of included studies were contacted for information on unpublished trials.
Study selection and analysis: Two reviewers appraised and selected RCTs in adults with acute exacerbation of psychosis-induced agitation or aggression that compared rapid use of haloperidol (any dose or route) with other antipsychotics, benzodiazepines, anticonvulsants, drug combinations, placebo or no intervention. Random effect meta-analyses were conducted using RevMan software. Trials were excluded if 50% or more data were lost. Heterogeneity was assessed using the I2 statistic.
Thirty-two RCTs met inclusion criteria (n=3877). Most trials were small, had considerable risk of bias, and did not reflect real life clinical practice. Haloperidol and comparator drugs were administered intramuscularly in all but one study, doses were not well reported. Most studies lasted for 72 h or less. Haloperidol reduced the number of people getting sleep of 2 h compared with placebo (2 RCTs, n=220; RR 0.88, 95% CI 0.82 to 0.95). Haloperidol also reduced the requirement for repeat rapid tranquilisation compared with placebo (4 RCTs, n=660; RR 0.51, 95% CI 0.42 to 0.62). Haloperidol reduced the need for repeat rapid tranquilisation compared with aripiprazole (2 RCTs, n=473; RR 0.78, 95% CI 0.62 to 0.99). Chlorpromazine increased numbers a sleep by 2 h compared with haloperidol in one small RCT (n=39; RR 1.93, 95% CI 1.04 to 3.60). Olanzapine increased the number of people getting sleep of 2 h compared with haloperidol (1 RCT, n=257; RR 1.16, 95% CI 1.02 to 1.32). Haloperidol increased the need for rapid tranquilisation compared with zuclopenthixol acetate (1 RCT, n=70; RR 2.54, 95% CI 1.19 to 5.46). Lorazepam increased the number of people getting sleep of 3 h compared with haloperidol (1 RCT, n=66; RR 1.93, 95% CI 1.14 to 3.27), but the groups did not differ at 1 h (1 RCT, n=60; RR 1.05, 95% CI 0.76 to 1.44). Haloperidol plus lorazepam increased number of people getting sleep of 3 h compared with haloperidol alone (1 RCT, n=67; RR 1.83, 95% CI 1.11 to 3.02). Haloperidol plus lorazepam reduced the number of people getting a sleep of 30 min compared with risperidone plus lorazepam (1 RCT, n=162; RR 0.84, 95% CI 0.74 to 0.95). In one better quality trial found that more participants did not get sleep of at least 20 min with haloperidol alone compared with haloperidol plus promethazine (1 RCT, n=316; RR 1.60, 95% CI 1.18 to 2.16). In this trial haloperidol alone increased the risk of acute dystonia, resulting in the trial being stopped at the interim analysis (RR 19.48, 95% CI 1.14 to 331.92).
Evidence on the effects of haloperidol alone for reducing psychosis-induced agitation or aggression is limited, and its use alone may be unethical if drugs to reduce its side effects are available. Haloperidol plus promethazine is potentially the most promising combination. Good-quality trials are still needed in this area.
Sources of funding: University of Nottingham, Nottinghamshire Healthcare NHS Trust, National Institute for Health Research, UK.
Haloperidol is widely available in a variety of formulations, is relatively inexpensive, and is generally efficacious in reducing psychotic symptoms. However, it invariably is associated with extrapyramidal adverse effects. Moreover, the use of anticholinergic medication to counteract tremour and rigidity can further impair the patient's already compromised cognitive abilities. Nonetheless, haloperidol remains in common use, especially when managing agitation. In their review, Powney and colleagues conclude that if additional drugs to offset the adverse effects are available, sole use of haloperidol could be considered unethical. Although they state that the use of an alternative antipsychotic drug is only partially supported by fragmented and poor-grade evidence, it remains evident that some of these alternatives, namely intramuscular aripiprazole, olanzapine and ziprasidone are efficacious in reducing agitation, do not result in excessive sedation and are not associated with extrapyramidal adverse effects, as observed in their respective registration trials.1 Access to these second-generation antipsychotics is becoming significantly less costly as their patent protections lapse, but there remains an enduring experience-base promoting the use of intramuscular haloperidol with or without lorazepam, which, so far, the available evidence-base has failed to reverse. Once the acute episode of agitation is resolved, the goals of treatment include the decrease in frequency and intensity of future episodes. Excluded from the meta-analysis is evidence that clozapine has the most efficacious profile in this respect, followed by olanzapine, with both clozapine and olanzapine superior to haloperidol.2 However, aggression and violence in persons with schizophrenia is aetiologically heterogeneous. We should not expect that any given pharmacological treatment will be equally effective in reducing violent behaviour caused by psychosis, impaired impulse control or psychopathy.3
Competing interests In the past 36 months, LC has 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, Shire, Sunovion and Valeant.