Non-improvement predicts subsequent non-response to repeated-dose intravenous ketamine for depression: a re-analysis of a 2-week open-label study in patients with unipolar and bipolar depression.

There is insufficient evidence to guide dose and frequency optimization with repeated-dose ketamine for depression. This study assessed the value of symptomatic non-improvement after the first few ketamine infusions as a predictor of overall non-response in depression for early decision-making to discontinue treatment. A total of 135 individuals with major depressive disorder or bipolar disorder experiencing a current major depressive episode were administered six repeated doses of intravenous ketamine. Depressive symptoms were assessed using the Montgomery-Åsberg Depression Rating Scale (MADRS) at baseline, 4 h after the first infusion, and 24 h after each infusion. Improvement, partial response, and response were defined as a reduction rate of ≥ 20%, 30%, and 50% in MADRS scores, respectively. This study examined the relationship between improvement (as opposed to non-improvement after each infusion or consecutive non-improvements after the first few infusions) and partial response and response after the sixth infusion. This analysis was summarized using sensitivity, specificity, and other diagnostic test parameters. The sensitivities of improvement at 24 h post-infusion 4 and improvement at 24 h post-infusion 3, vs. three consecutive non-improvements, as predictors for overall partial response and response exceeded 90%. No significant reduction in depressive symptoms was seen in non-improvers following the remaining infusions after the above-identified point. Our study suggests that non-improvement after four infusions, or more conservatively three consecutive non-improvements after three infusions, could serve as a signal of overall non-response to repeated-dose intravenous ketamine for depression and that subsequent treatments would not be warranted.