Ketamine versus ECT for hard-to-treat depression: two trials, two answers

Whether intravenous ketamine can stand in for electroconvulsive therapy in severe, treatment-resistant depression depends on which trial you read: a large US study found ketamine at least as effective as ECT, while a Swedish study found ECT clearly more likely to bring on remission [s1][s2]. Both were open-label head-to-head randomised trials, and the disagreement traces mostly to how ill the patients were and what counted as success [s1][s2].

ECT is the most effective treatment psychiatry has for severe depression, but its anaesthesia, its logistics and its short-term effect on memory keep both patients and clinicians wary of it. Ketamine, a rapid-acting antidepressant given by infusion, raised the possibility of matching ECT without those costs. Two trials put the two treatments directly against each other and reached opposite headline verdicts — a contrast that is more instructive than either result alone.

The US trial: ketamine held its own

ELEKT-D was an open-label, non-inferiority trial in 403 patients referred to ECT clinics for treatment-resistant major depression without psychosis; 200 were assigned to ketamine and 203 to ECT [s1]. During a three-week phase, patients received ketamine (0.5 mg per kilogram over 40 minutes) twice a week or ECT three times a week [s1]. Success was a response — a drop of at least 50% on a self-reported depression scale — and the margin for declaring ketamine non-inferior was 10 percentage points [s1].

Ketamine cleared the bar and then some: 55.4% of the ketamine group responded versus 41.2% of the ECT group (a difference of 14.2 percentage points; 95% CI, 3.9 to 24.2; P < 0.001 for non-inferiority) [s1]. The memory contrast was stark. On a delayed-recall test, scores fell by 0.9 points with ketamine and by 9.7 points with ECT after three weeks, recovering gradually over follow-up [s1]. ECT brought musculoskeletal side effects; ketamine brought dissociation [s1].

The Swedish trial: ECT came out ahead

KetECT, also open-label, enrolled 186 hospitalised inpatients with unipolar depression and randomly assigned them to thrice-weekly racemic ketamine infusions (0.5 mg/kg) or ECT [s2]. Here the target was remission — a near-absence of symptoms, defined as a score of 10 or below on a clinician-rated scale — a higher bar than response [s2].

On that measure ECT won: 63% of ECT patients remitted versus 46% on ketamine (P = 0.026; 95% CI for the difference, 2% to 30%), so ketamine did not meet the non-inferiority criterion [s2]. Both treatments needed a median of six sessions to induce remission [s2]. The side-effect ledger again split by treatment: serious and long-lasting effects, including cases of persisting amnesia, were more common with ECT, while treatment-emergent side effects caused more dropouts in the ketamine group [s2]. Among those who remitted, relapse within 12 months was similar — 70% with ketamine and 63% with ECT (P = 0.52) [s2].

Reconciling the two

The trials disagree less than their headlines suggest. ELEKT-D studied outpatients referred for ECT, excluded psychotic depression, and scored response; KetECT studied more severely ill hospitalised patients and scored the tougher endpoint of remission [s1][s2]. A treatment can look non-inferior on "meaningfully improved" and inferior on "nearly symptom-free," especially in a sicker group. Both trials were also open-label, so patients and clinicians knew which treatment was given — a design that can inflate subjective symptom ratings and that neither trial could avoid [s1][s2].

For related evidence on ketamine's real-world use, see our coverage of a trial of repeated subcutaneous ketamine for treatment-resistant depression, and of magnetic seizure therapy and other approaches to targeting the depressed brain.

What it means

For outpatients with non-psychotic depression who want to avoid ECT's memory cost, ketamine looks like a reasonable alternative on the evidence of ELEKT-D [s1]. For the most severely ill, hospitalised patients, ECT still produced more remissions in KetECT [s2]. Neither trial addresses psychotic depression, where ECT's advantage is best established, so the choice remains one calibrated to severity and setting rather than a single winner [s1][s2].

What to watch

Blinded comparisons, head-to-head trials scored on remission rather than response, and longer follow-up on durability — the questions these two open-label trials leave open.

This article describes trial evidence and is not medical advice. Treatment decisions belong with a clinician who knows the individual case.

Sources

1. Ketamine versus ECT for Nonpsychotic Treatment-Resistant Major Depression — New England Journal of Medicine, 24 May 2023
2. Racemic Ketamine as an Alternative to Electroconvulsive Therapy for Unipolar Depression (KetECT) — International Journal of Neuropsychopharmacology, 4 December 2021