REVIEW OF: Flint AJ et al, JAMA 2019;322(7):622–631

TYPE OF STUDY: Randomized, placebo-controlled trial

Psychotic features in depression indicate a more severe form of the disease, with a higher risk of hospitalization and double the rate of disability compared with non-psychotic depression. A combination of an antipsychotic and an antidepressant is the mainstay of treatment, but how long to continue the antipsychotic is an unanswered question.

This study enrolled patients 18–85 years of age with severe major depression and at least one delusion; hallucinations were optional. Dementia and unstable medical illness were part of the exclusion criteria, so the patients may not have been as ill as some whom we see in clinical practice. Average age was 55 years.

Researchers first treated 269 patients with open-label olanzapine and sertraline. Next, 162 patients who achieved remission or near-remission entered an open-label 8-week stabilization phase. Of the 147 who remained well after the stabilization, 126 were randomized to continue on olanzapine or have the antipsychotic replaced with a placebo for 36 weeks. The design was double blind, and the antipsychotic taper took place over 4 weeks. All patients remained on sertraline throughout the trial.

The primary outcome was risk of relapse, which included relapses into depression or psychosis as well as psychiatric hospitalization or suicidality. 55% of sertraline-placebo patients relapsed, compared to 20% of sertraline-olanzapine patients. The number needed to treat (NNT) to keep patients well with continued antipsychotic therapy was 2.8.

The majority of the relapses (79%) occurred within the first 20 weeks of the 36-week randomization phase. In a letter to the editor, Klaus Munkholm and colleagues argued that these relapses may have been a withdrawal phenomenon. The authors of the study countered that their criteria for relapse shared little in common with known symptoms of antipsychotic withdrawal.

Weight gain was the main side effect of continued olanzapine. The placebo group lost weight while the olanzapine group continued to gain, with a difference of 9 pounds between them at the end of the study. Falls were also greater in the olanzapine-continuation group (31% vs 18%).

TCPR’S TAKE
When a patient recovers from psychotic depression on an antidepressant and antipsychotic, we should continue both medications for at least 2 months as long as the medication is reasonably tolerable. After 6 months of remission (28 weeks), we might consider a slow taper of the antipsychotic, weighing the severity of the episode, side effects, and the patient’s preferences.