To examine the effect of high baseline anxiety on response to ketamine versus midazolam (active placebo) in treatment‐resistant depression (TRD).

In a multisite, double‐blind, placebo‐controlled trial, 99 subjects with TRD were randomized to one of five arms: a single dose of intravenous ketamine 0.1, 0.2, 0.5, 1.0 mg/kg, or midazolam 0.045 mg/kg. The primary outcome measure was change in the six‐item Hamilton Rating Scale for Depression (HAMD6). A linear mixed effects model was used to examine the effect of anxious depression baseline status (defined by a Hamilton Depression Rating Scale Anxiety‐Somatization score ≥7) on response to ketamine versus midazolam at 1 and 3 days postinfusion.

N = 45 subjects had anxious TRD, compared to N = 54 subjects without high anxiety at baseline. No statistically significant interaction effect was found between treatment group assignment (combined ketamine treatment groups versus midazolam) and anxious/nonanxious status on HAMD6 score at either days 1 or 3 postinfusion (Day 1: F(1, 84) = 0.02, P = 0.88; Day 3: F(1, 82) = 0.12, P = 0.73).

In contrast with what is observed with traditional antidepressants, response to ketamine may be similar in both anxious and nonanxious TRD subjects. These pilot results suggest the potential utility of ketamine in the treatment of anxious TRD.