Should We Rethink Warning Against Conventional Antipsychotics in Elderly?

Conventional antipsychotics, including haloperidol, may not increase the risk of death in elderly patients after all, says a team of Dutch researchers.¹ Although a number of observational studies point to an association between conventional antipsychotic agents and increased risk of death in older patients, experimental evidence has not been reviewed before. Meanwhile, the US Food and Drug Administration and the UK Commission for Drug Safety have warned against use of these agents for management of dementia symptoms such as hallucinations, delusions, and aggression.

The observational studies may have been flawed, the study authors contend. They pointed out, for example, that although the risk differences varied between 4.2% and 7.3% for the first 6 months of antipsychotic use, none of the studies adjusted for severity of illness. The researchers conducted a systematic review and meta-analysis of randomized clinical trial (RCT) data. Their goal was to establish the mortality risk of conventional antipsychotics in elderly patients with dementia or delirium.

Included in the meta-analysis were RCTs that compared a conventional antipsychotic agent with placebo in patients aged 65 years or older who had dementia or delirium or who were at risk for delirium. Excluded were RCTs that included patients with schizophrenia, advanced cancer, or terminal illness and trials in which the intervention included multiple drugs. The RCTs were gleaned from electronic databases, online trial registries, and hand-searched references of published reviews. Seventeen trials, inclusive of 2387 patients, met the study criteria and were ultimately included in the meta-analysis.

Among these 17 RCTs, a total of 32 deaths were recorded. Seventeen deaths occurred in the active treatment groups and 15 in the placebo groups. No difference was seen in the rounded mortality rates, which were 1.4% for the active treatment groups and 1.3% for the placebo groups. The pooled risk difference (RD) was 0.1%, which was not statistically significant. When the researchers turned their attention to the 11 RCTs in their sample that compared haloperidol to placebo, inclusive of 1799 patients, the pooled RD was 0.4%-again, not statistically significant.

The researchers pointed out flaws and confounding factors in the studies reviewed in relation to the GRADE recommendations. They concluded that available data does not support the notion that conventional antipsychotics in general or haloperidol in particular increase the risk of death in elderly patients and that the warning against use of these agents may be unwarranted.

References:

1. Hulshof TA, Zuidema SU, Ostelo RW, Luijendijk HJ. The mortality risk of conventional antipsychotics in elderly patients: a systematic review and meta-analysis of randomized placebo-controlled trials. J Am Med Dir Assoc. 2015; S1525-8610(15)00229-7. http://www.jamda.com/article/S1525-8610(15)00229-7/abstract. Accessed June 4, 2015.