|Home | About | Journals | Submit | Contact Us | Français|
In resource poor settings, a sleeping patient is better than one who needs constant observation
Two randomised controlled trials in this week's BMJ assess the effectiveness of different combinations of drugs for tranquillising and sedating people who are violent or agitated as a result of psychiatric disorders.1 2 Both trials were undertaken in developing countries.
The first trial, by Raveendran and colleagues, was carried out in the emergency services of a general psychiatry department in a hospital in South India. It compared the tranquillising and sedative effects of a single intramuscular administration of either olanzapine (10 mg) or a combination of haloperidol (10 mg) plus promethazine (50 mg) in 300 aggressive or agitated patients. The observation period lasted for four hours only and patients were followed up for just two weeks. This contrasts with most randomised controlled trials in psychiatry, which have treatment periods lasting for four to 12 weeks and can have several months of follow-up. The trial is important, however, because it looks at a neglected3 area—the early effects of treatment with parenteral antipsychotic drugs in patients who are violent or agitated. Without effective treatment these patients may harm themselves and their environment,3 and they are a heavy burden on resources in emergency psychiatry facilities.
Violent patients are usually psychotic and often receive antipsychotic drugs. In the Indian trial, 10% of the participants were depressed, two thirds were manic, and the remainder had other forms of psychosis. At all five assessments during the four hour study significantly more people were asleep after the haloperidol-promethazine combination than with olanzapine (number needed to treat (NNT) ranged from 5 to 8). Whereas revisits by consultants (NNT=6) and the use of additional drugs (NNT=5) were less frequent with the combination, the need for physical restraint and the adverse effects of drugs did not differ significantly between the treatment groups. No patient experienced dystonia.
Clinicians are usually satisfied if drugs tranquillise a disturbed patient. The two treatment groups did not differ significantly in the combined outcome measure of being tranquil or asleep at 15 and 30 minutes. However, at one hour significantly more people taking the combination treatment were tranquil or asleep (NNT=19). Some people taking olanzapine needed additional drugs, after which the proportion of those tranquil or asleep in the two groups was once again similar. There is a clear take home message here—if being tranquil or asleep is the desired end point, intramuscular olanzapine is as good as intramuscular haloperidol plus promethazine if the doctor is willing to take a 20% chance of being called back an hour later to give another dose.
The National Institute for Health and Clinical Excellence (NICE) guideline4 describes a hierarchy of interventions for the emergency management of violent patients; parenteral tranquillisation with antipsychotic drugs is almost a last resort, and the goal of management is to induce calm, not sleep. Regrettably, these recommendations are impractical in most emergency medical settings in developing countries, where resources are strained by heavy patient loads and understaffing, and where a sleeping patient is better than one who needs constant observation to assess the need for tranquillisation. From the patient's perspective, being asleep is also less traumatic than being physically restrained. These considerations, and the lower cost, favour the combination of haloperidol-promethazine over olanzapine for the emergency management of violent patients in the developing world. However, although the combination is popular in some settings,5 the NICE guideline4 does not recommend it because of lack of safety data from the United Kingdom.
Treatment guidelines are seldom written with developing countries in mind, so the trial by Raveendran and colleagues1 offers useful guidance relevant to these settings. Importantly, the varied treatment outcomes assessed allow clinicians in other parts of the world to make choices on the basis of the outcome that is relevant to their own settings. Other strengths of the trial are that it assessed “real world” patients who were severely disturbed and who could not provide informed consent; almost all eligible patients were recruited; the outcome measures examined were clinically relevant; proxy measures were excluded; and nearly all patients completed the four hour study.
However questions remain. Dystonia is an important adverse effect of parenteral haloperidol. Although dystonia did not develop during the four hour observation period we are not told whether any patients experienced dystonia later, when the effect of promethazine wore off. Reassuringly, the second trial in this issue, which was conducted in Brazil2 but is similar to the Indian one, found that none of the 156 patients who received intramuscular haloperidol (5-10 mg) with promethazine (25-50 mg) developed dystonia within 24 hours of treatment. In contrast, 10 of 160 patients who received only intramuscular haloperidol (5-10 mg) had dystonia.
The Brazilian study found that the combination of haloperidol-promethazine was more likely to tranquillise or induce sleep than haloperidol alone at 20 minutes (the two treatments did not differ in this or other measures at later assessments up to 24 hours).2 This suggests that at least part of the early benefit of the combination arises from the sedative action of promethazine. So, might a higher dose of intramuscular olanzapine or a combination of olanzapine and an oral or parenteral sedative have a safety-efficacy profile similar to that of the haloperidol-promethazine combination? We do not yet know, but the answer would interest those who do not wish to prescribe a neuroleptic.
Data from the United States6 and the UK7 suggest that, with regard to pragmatic long term outcomes in people with schizophrenia, at least some of the first generation antipsychotic drugs compare favourably with atypical antipsychotics. Concerns have been expressed about these conclusions.8 9 Nevertheless, the results of Raveendran and colleagues' trial suggest that similar conclusions may also apply to pragmatic early outcomes in violent and agitated patients.8 9 However, there is a caveat—the Brazilian trial reported three seizure events within 24 hours of the administration of haloperidol monotherapy (two events) or the haloperidol-promethazine combination (one event).2 Although no seizures were reported in the comparably large Indian trial, we should be mindful of the potential risk.
Rapid tranquillisation must not be confused with rapid neuroleptisation, which refers to treatment of a psychotic episode with the early administration of large doses of parenteral neuroleptics. Rapid neuroleptisation seeks to hasten recovery from psychosis, but it has not been shown to be effective for this purpose.10
This article was posted on bmj.com on 22 October 2007: http://bmj.com/cgi/doi/10.1136/bmj.39359.614387.80
Competing interests: None declared.
Provenance and peer review: Commissioned; not externally peer reviewed.