Randomized trial of ketamine masked by surgical anesthesia in patients with depression

Abstract

Ketamine may have antidepressant properties, but its acute psychoactive effects complicate successful masking in placebo-controlled trials. We present a single-center, parallel-arm, triple-masked, randomized, placebo-controlled trial assessing the antidepressant efficacy of intravenous ketamine masked by surgical anesthesia (ClinicalTrials.gov, NCT03861988). Forty adult patients with major depressive disorder who were scheduled for routine surgery were randomized to a single infusion of ketamine (0.5 mg/kg) or placebo (saline) during usual anesthesia. All participants, investigators, and direct patient care staff were masked to treatment allocation. The primary outcome was depression severity measured by the Montgomery-Åsberg Depression Rating Scale (MADRS) at 1, 2, and 3 days post-infusion. After all follow-up visits, participants were asked to guess which intervention they received. A mixed-effects model showed no evidence of effect of treatment assignment on the primary outcome (-5.82, 95% CI -13.3 to 1.64, p=0.13). 36.8% of participants guessed their treatment assignment correctly; both groups allocated their guesses in similar proportions. In conclusion, a single dose of intravenous ketamine delivered during surgical anesthesia had no greater effect than placebo in acutely reducing the severity of depressive symptoms in adults with major depressive disorder. This trial successfully masked treatment allocation in moderate-to-severely depressed patients using surgical anesthesia. Although this masking strategy is impractical for most placebo-controlled trials, future studies of novel antidepressants with acute psychoactive effects should make efforts to fully mask treatment assignment in order to minimize subject-expectancy bias.

Extended Data Figure 1.

Depression Ratings Reanalyzed According to Patient Guess On day 14, the final day of patient assessments, patients were asked the following questions: “What treatment do you think you received?” MADRS scores were reanalyzed according to their guess, irrespective of their true group allocation. Mean and standard deviation (SD) MADRS scores are shown using the alternate grouping: “Ketamine”, n=17; “Placebo”, n=10; “I don’t know”, n=11.

Figure 1.. CONSORT Flow Diagram

CONSORT, Consolidated Standards of Reporting Trials. PI, principal investigator. ITT, intention-to-treat. Participants were randomized to a single intravenous dose of either ketamine or saline, given during surgical anesthesia. *The severe adverse event refers to an unexpected death that occurred 2 days after the patient was discharged home without complications on postoperative day 3; this patient experienced a witnessed cardiac arrest, attributed to the participant’s medical factors and not resulting from study procedures. †One participant was withdrawn by the PI due to an unanticipated surgical revision of an implanted device, which took place on postoperative day 3 after study assessments were completed.

Figure 2.. Depression Severity, Masking Assessment and Other Outcomes

Panel A shows the mean and standard deviation (SD) scores by group on the Montgomery-Åsberg Depression Rating Scale (MADRS; scores range from 0 to 60, with higher scores indicating greater depression); the screening baseline visit occurred on average 5 days before infusion on day 0. Total N=39 independent participant responses per day, on post-infusion days 1 through 3 (see Extended Data Table 1 for group-specific counts). Panel B shows the distribution of guesses as a percentage of each group (n=19 per group) made by participants when asked to guess which treatment they received after the last follow-up visit. Panel C shows the difference in MADRS scores relative to pre-infusion baseline scores obtained on day 0 (same sample size as reflected in Panel A). Panels D shows the proportions of clinical response, respectively as a percentage of each group (n=20 per group), within the first 3 days. Panel E shows the cumulative opioid consumption in MME by group, represented as median and interquartile range (IQR); both inpatient and outpatient opioids were included in the total (n=20 per group).