Ketamine infusions are increasingly used off-label for hard-to-treat depression, including on psychiatric wards. Strong results from some saline-controlled studies raised hopes of a rapid add-on for inpatients. A pragmatic Irish trial now tests serial ketamine against a sedating psychoactive placebo while usual ward care continues; midazolam was chosen because it sedates and can protect blinding better than saline.
Researchers at Trinity College Dublin and St Patrick’s University Hospital randomly assigned adults hospitalised with a major depressive episode to up to eight twice-weekly intravenous infusions. The active arm used ketamine 0.5 mg/kg and the control used midazolam 0.045 mg/kg. The findings appear in 2025 in JAMA Psychiatry.
Of 65 randomised participants, 62 entered the final analysis. Mean age was 53.5 years, and 37 participants (59.7%) were men. End-of-treatment Montgomery-Asberg Depression Rating Scale scores did not differ significantly between groups, with an adjusted mean difference of -3.16 points and a Cohen’s d of -0.29, well short of the eight-point gap the trial was powered to detect.
Self-rated depression, cognition, cost-effectiveness, and quality of life likewise showed no clear between-group benefit for ketamine. During infusions, dissociative symptoms and mood elevation rose more with ketamine, while midazolam was more sedating at 30 minutes. Most patients and raters correctly guessed allocation.
Common adverse events included fatigue, sleep disturbance, and abnormal liver tests. Serious adverse events occurred in both arms during the study. Scores on a withdrawal symptom scale tended to fall over time in both groups rather than rise.
The trial was funded by the Health Research Board of Ireland. One author reports prior employment with Compass Pathways, and another reports speaker fees and advisory work for industry during the study. Those ties are disclosed in the paper.
Patients were followed for six months after the infusion course in a naturalistic phase. Cognitive scores on the Montreal Cognitive Assessment were similar between arms after adjustment for baseline function.
The study has clear limits. The sample was smaller than planned, blinding was hard to keep, and usual inpatient care may have left little room for an add-on effect to show. Results apply to serial adjunctive infusions on a ward, not to every ketamine protocol used in outpatient clinics; outpatient ketamine clinics use different schedules and selection rules.
The finding cools expectations for routine ward ketamine courses of this kind. It does not tell readers to seek or avoid any medicine on their own. Treatment choices for depression belong with the clinical team who know the person’s history and risks.
