For the millions of people living with bipolar disorder who have exhausted multiple pharmacological options, treatment-resistant bipolar depression represents one of psychiatry's most stubborn and dangerous challenges. Unlike unipolar depression, where ketamine's antidepressant potential has been steadily validated, bipolar depression has remained an evidence vacuum — until now.
The Ket-BD trial, a double-blind, active-controlled randomized clinical trial conducted across three Ontario sites, directly tested whether serial ketamine infusions could outperform midazolam — a sedating comparator chosen to mimic ketamine's dissociative profile — in adults with treatment-resistant bipolar I or II depression. Participants carried significant illness burden: moderate-to-severe depressive episodes (MADRS ≥21) and at least two failed evidence-based pharmacotherapy trials. Critically, all participants remained on stable mood stabilizer or antipsychotic regimens throughout, addressing the long-standing clinical concern that ketamine might destabilize bipolar mood architecture. The intervention consisted of four flexibly dosed infusions over two weeks at 0.5–0.75 mg/kg, and the primary endpoint was MADRS score change at day 14.
This trial fills a meaningful gap that prior ketamine research has left conspicuously open. Most ketamine RCTs have explicitly excluded bipolar patients over switch-to-mania concerns, meaning clinical use in this population has outpaced its evidence base for years. The midazolam control is methodologically important — it is widely regarded as the gold standard blinding comparator in ketamine trials, substantially reducing placebo inflation compared to saline controls. That said, several limitations warrant caution: the three-site Canadian design limits demographic generalizability, the two-week observation window cannot speak to durability of response, and the flexible dosing protocol introduces variability that complicates dose-response interpretation. Whether the mood stabilizer co-medication requirement meaningfully reduces mania risk — or simply selects a more stable subpopulation — also remains an open question. Overall, this qualifies as a genuinely important contribution: a properly controlled trial addressing a population that clinical practice has long treated off-label.