Although the Food and Drug Administration (FDA) has not approved atypical antipsychotics for use in patients with dementia, they are commonly prescribed in this population. Recent concerns about increased risk of cerebrovascular events and mortality have led to warnings. A systematic review was conducted to assess the benefits and harms of atypical antipsychotics when used in patients with behavioral and psychological symptoms of dementia. Electronic searches (through March 2005) of the Cochrane Library, Medline, Embase, and PsycINFO were supplemented with hand searches of reference lists, dossiers submitted by pharmaceutical companies, and a review of the FDA Website and industry-sponsored results database. Using predetermined criteria, each study was assessed for inclusion, and data about study design, population, interventions, and outcomes were abstracted. An overall quality rating (good, fair, or poor) was assigned based on internal validity.
The evidence for olanzapine and risperidone supports their effectiveness compared with placebo. Short-term adverse events were similar to placebo. Risperidone had no advantage over haloperidol on efficacy measures in the better-quality studies. Risperidone had an advantage over haloperidol on some measures of extrapyramidal symptoms. Evidence for the other atypical antipsychotics is too limited to assess efficacy and safety. Trials were short term and conducted in highly selected populations.
The potential for increased risk of cerebrovascular adverse events and mortality is a serious concern. To make judgments about when the benefits of atypical antipsychotics outweigh the potential harms, clinicians need more information. Additional data from existing trials and more-complete reporting of trial results could provide this information.