KETAMINE VERSUS ELECTROCONVULSIVE THERAPY FOR SEVERE TREATMENT-RESISTANT DEPRESSION: A SYSTEMATIC REVIEW AND META-ANALYSIS OF EFFICACY, SPEED OF RESPONSE, COGNITIVE OUTCOMES, AND SAFETY

Abstract

Introduction: Treatment-resistant depression (TRD) affects approximately one-third of patients with major depressive disorder and accounts for disproportionate disability and mortality. Electroconvulsive therapy (ECT) has long been the definitive intervention, whereas subanaesthetic intravenous ketamine has emerged as a rapid-acting alternative. Direct head-to-head evidence has accumulated rapidly but remains discordant. This review resolved this discordance and quantified comparative efficacy, speed of response, neurocognitive burden, and safety.

Methods: The study strictly adhered to the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) 2020 guidelines. Eligible designs included RCTs, cohort, case-control, and cross-sectional studies. Risk of bias assessed with RoB 2 and ROBINS-I. Dichotomous outcomes pooled as risk ratios (RR) with 95% CI using DerSimonian-Laird estimator; heterogeneity quantified by I²; prediction intervals calculated; exploratory summary ROC analysis performed.

Results: Eighteen studies (1,388 participants) were included: 12 head-to-head reports (10 independent samples; 1,009 participants) and 6 adjunctive/mechanistic studies (379 participants). Five studies contributed response data (n=760). Pooled response did not differ between modalities (RR 1.13; 95% CI 0.85–1.49; P=0.393; I²=80.0%; 95% PI 0.47–2.72). Remission likewise showed no difference (RR 1.16; 95% CI 0.77–1.76; P=0.468; I²=80.9%). Heterogeneity was overwhelmingly attributable to the ELEKT-D trial: leave-one-out omission yielded significant ECT advantage (RR 1.27; 95% CI 1.07–1.51) with near-resolution of inconsistency (I²=25.5%). Among exclusively inpatient samples, ECT was directionally superior (RR 1.21; 95% CI 0.98–1.48; I²=26.1%). Relapse (RR 1.20; 95% CI 0.67–2.15) and discontinuation (RR 1.79; 95% CI 0.15–21.36) did not differ. Summary AUC was 0.536 for response and 0.532 for remission—negligible discrimination by modality. Ketamine demonstrated consistently superior neurocognitive outcomes, most markedly on delayed verbal recall (−0.9 ± 1.1 versus −9.7 ± 1.2 T-score points), and earlier symptomatic separation in less severe populations. ECT was associated with musculoskeletal pain and transient amnesia; ketamine with dose-dependent dissociation and transient haemodynamic effects. Certainty was low to very low for efficacy outcomes, downgraded for inconsistency and risk of bias.

Discussion: The apparent contradiction is not statistical artefact but reflects genuine effect modification. ECT and ketamine are not interchangeable across the whole spectrum; relative advantage is conditioned by illness severity, treatment setting, psychotic features, and age. In hospitalised, severely ill, psychotic populations—the KetECT population—ECT retains clear efficacy advantage. In ambulatory, non-psychotic populations—the ELEKT-D population—ketamine is non-inferior with superior neurocognitive profile and greater acceptability. Ketamine's principal liabilities are brevity of effect and uncertainties regarding long-term cystitis and misuse potential; ECT's liabilities are amnesia, anaesthetic exposure, and stigma.

Conclusion: Intravenous ketamine and ECT produce broadly comparable response and remission rates in severe TRD, with no significant pooled difference and negligible discrimination between modalities. Clinically meaningful ECT superiority persists in hospitalised and more severely ill populations, whereas ketamine offers faster separation and superior short-term cognitive profile. Treatment selection should be individualised according to severity, setting, psychotic features, cognitive vulnerability, urgency, and patient preference. Adequately powered, longer, stratified randomised trials with standardised continuation strategies and harmonised neurocognitive batteries are required.