Clinical Predictors of Antidepressant Effects of Ketamine and Esketamine in Treatment-Resistant Unipolar and Bipolar Depression: A Systematic Review

Background and Objective Ketamine and esketamine have emerged as effective and rapid-acting interventions for treatment-resistant depression. Nevertheless, the best-suited candidates for these treatments are not clear. As such, the ability to personalize treatment selection and parameters would likely improve therapeutic response. This systematic review synthesized the literature on clinical and demographic predictors of antidepressant response to ketamine and esketamine, including both unipolar and bipolar treatment-resistant depression.

Methods: We searched the databases of PubMed, Embase, Scopus, and APA PsychINFO on 25 March, 2025, with an update on 19 November, 2025, to identify studies on the clinical and demographic predictors of ketamine or esketamine effects in treatment-resistant depression. There were no restrictions on study design. Studies in a language other than English were excluded. Two authors, OAS and VS, independently reviewed studies and resolved conflicted judgments through a discussion. Risk of bias was assessed using the Cochrane RoB 2 for randomized studies and ROBINS-I for non-randomized studies. For each included study, the direction of effect (positive, negative, no association) for predictor variables was extracted.

Results: A total of 122 studies ( n = 12,674) were included in the review, with 75 distinct samples ( n = 6902), published between August 2006 and September 2025. There were 65 studies providing open-label ketamine, 47 secondary analyses of a trial, and 10 randomized controlled trials included. Studies most commonly treated both treatment-resistant unipolar and bipolar depression ( k = 62) and solely administered intravenous ketamine at a fixed dosage of 0.5 mg/kg ( k = 61). We examined the predictive value of 77 predictor variables, including body mass index, dissociation, and previous neuromodulation treatments. Most variables had more reports revealing no association with antidepressant outcomes than a positive or negative predictive ability. Some promising predictive factors, such as early response to treatment and a family history of substance use disorders, were identified and warrant further exploration.

Conclusions: In a comprehensive synthesis of predictor analyses for ketamine and esketamine treatments, most demographic and clinical variables did not confer differential outcomes to ketamine and esketamine treatments, although some promising predictive variables were identified for further investigation. Clinical Trial Registration PROSPERO (CRD42024554316).