Baseline insomnia as a predictor of antidepressant efficacy to repeated intravenous ketamine for unipolar and bipolar depression: A preliminary study.

OBJECTIVE: Ketamine has been demonstrated to have robust and rapid antidepressant effects, and few studies have focused on the relationship between insomnia and the efficacy of ketamine. The objective of this study was to examine whether baseline insomnia predicted the antidepressant efficacy of repeated intravenous ketamine infusions for unipolar and bipolar depression.
METHOD: Patients with high insomnia (n = 64) or low insomnia (n = 68) received six intravenous infusions of ketamine (0.5 mg/kg over 40 min) over 12 days (Monday-Wednesday-Friday). The Montgomery-Asberg Depression Rating Scale (MADRS) without sleep item was used to assess depressive symptoms. Response was defined as a MADRS total score ≥ 50%, and remission was defined as a MADRS total score ≤ 10. RESULT: There were no differences in response or remission rates between patients with high and low insomnia. However, the logistic regression model showed that high insomnia predicted an increased likelihood of response and remission. Cox proportional hazards models showed a reduced latency to respond and remit in patients with high insomnia. A linear mixed model showed that the high insomnia subgroup had greater improvement than the low insomnia subgroup (all p < 0.05). LIMITATION: The major limitation of this study is the open-label design.
CONCLUSION: When given six ketamine infusions, patients with high insomnia were more likely to respond and remit than those with low insomnia. Patients with high insomnia showed not only a shorter latency to respond and remit, but also greater improvement than those with low insomnia.