Antianhedonic Effect of Repeated Ketamine Infusions in Patients With Treatment Resistant Depression

Anhedonia, defined as reduced ability to experience pleasure, is a core symptom of depressive episodes that correlates with suicidality and predicts poor antidepressant outcomes. Previous research implicates dopaminergic and glutamatergic circuits in anhedonia, and mechanistic data suggest ketamine—a rapid-acting N-methyl-D-aspartate receptor (NMDAR) antagonist—may exert antidepressant and antisuicidal effects via disinhibition of glutamate transmission, AMPA receptor activation, and downstream BDNF–TrkB signalling. Human and preclinical studies provide limited and sometimes inconsistent evidence that ketamine increases dopaminergic activity in reward-related brain regions, suggesting a plausible route for an antianhedonic effect. Strzelecki and colleagues set out to examine changes in anhedonia during a course of eight intravenous ketamine infusions delivered as an add-on in patients with treatment-resistant depression (TRD). The primary aim was to assess change in anhedonia measured by the Snaith–Hamilton Pleasure Scale (SHAPS). Secondary aims were to evaluate how change in anhedonia related to overall depressive symptoms (IDS‑SR 30) and to suicidal ideation (item 18 of IDS‑SR 30). The investigators also tested whether concomitant benzodiazepine (BDZ) use moderated ketamine’s effects, hypothesising that ketamine would reduce both depressive symptoms and anhedonia and that BDZ co‑treatment would attenuate these effects.

The study used a naturalistic, observational registry design of intravenous ketamine treatment for treatment‑resistant mood disorders (GDKet; NCT04226963). The extracted text reports a sample of 41 patients (26 female) with a mean age of 48.5 years (SD 14.3), diagnosed with major depressive disorder or bipolar disorder without psychotic features. Diagnosis was made by clinician psychiatrists according to DSM‑5 and confirmed with the MINI. Treatment resistance was defined as failure of at least two antidepressant trials of adequate dose and duration for MDD, and analogous criteria for bipolar TRD. Medically unstable individuals, pregnant or breastfeeding patients, and those with prior adverse reactions to ketamine were excluded. The study protocol received institutional review board approval and participants gave informed consent. Intervention consisted of eight intravenous ketamine infusions administered over 4 weeks as an add-on to patients’ ongoing psychotropic regimens. Ketamine dosing was 0.5 mg/kg (actual body weight) infused over 40 minutes (Ketalar 50 mg/ml preparation). Safety monitoring included periodic vital signs checks and psychometric safety scales (Brief Psychiatric Rating Scale and Clinician‑Administered Dissociative States Scale) at baseline and 1 hour after each infusion. Psychometric outcomes comprised the SHAPS as the primary outcome (14‑item self‑report; scores 0–14, SHAPS >2 indicating clinically significant anhedonia) and the 30‑item IDS‑SR for depressive symptoms as a secondary outcome. Suicidal ideation was assessed via IDS‑SR item 18. Assessments were conducted pre‑treatment, at the third, fifth, and seventh infusions, and 1 week after the final infusion. Statistical procedures employed IBM SPSS v26. Group changes in SHAPS and IDS‑SR 30 were analysed with repeated measures one‑way ANOVA, supplemented by Tukey post‑hoc tests. General linear models with repeated measures adjusted for sex, age, BMI, and benzodiazepine use. IDS‑SR item 18 was analysed with the Friedman test and Dunn post‑hoc tests. A moderated mediation analysis (PROCESS macro v3.5) was planned to test whether SHAPS mediated the relationship between number of infusions and IDS‑SR scores with BDZ use as a moderator; bootstrapping with 5,000 resamples was used and significance set at α = 0.05. The authors report that, for a moderate effect size, N = 41 provided >0.8 power for the selected tests.

The extracted results indicate no serious adverse events; patients experienced transient increases in blood pressure and mild‑to‑moderate dissociative symptoms. Baseline assessment found that 97.5% of participants (n = 39 of 41) met criteria for clinically significant anhedonia (SHAPS ≥ 2). On the SHAPS, repeated measures ANOVA showed a significant reduction across time: F(4,160) = 13.36, p < 0.001, partial η2 = 0.25 (Greenhouse–Geisser and Huynh–Feldt corrections p < 0.001). Pairwise comparisons indicated a significant decrease from baseline to each assessed infusion and to the post‑treatment visit (all p < 0.001). There were no significant differences between individual infusions. After adjustment for sex, age, BMI, and benzodiazepine use, a significant interaction emerged between infusion number and BDZ use: F(4,144) = 2.60, p = 0.039, partial η2 = 0.07 (Huynh–Feldt corrected p = 0.041). In stratified analyses, patients not using benzodiazepines showed significant SHAPS reductions after each infusion and at the post‑treatment visit (p values ranging from 0.002 to <0.001). Patients using benzodiazepines showed a reduction only after the seventh infusion (p = 0.036), and their SHAPS scores did not differ between baseline and the post‑treatment visit. For overall depressive symptoms (IDS‑SR 30), one‑way repeated measures ANOVA was significant: F(4,160) = 8.86, p < 0.001, partial η2 = 0.18 (G‑G and H‑F corrections p < 0.001). The difference between baseline and the third infusion reached p = 0.013, and baseline versus later infusions and the post‑treatment visit were p < 0.001. The adjusted general linear model remained significant: F(4,144) = 3.28, p = 0.013, partial η2 = 0.08, but no significant interaction effects with the covariates were reported. Analysis of IDS‑SR item 18 (suicidal ideation) used the Friedman test and showed a significant effect across time: χ2(4) = 15.53, p = 0.004, W = 0.09. Dunn post‑hoc tests indicated reductions in suicidal ideation after the fifth (p = 0.047) and seventh (p = 0.028) infusions compared with baseline; however, the post‑treatment visit did not differ from baseline. Mediation analysis initially tested a moderated mediation with benzodiazepine use as moderator but, owing to nonsignificance of the interaction, the authors report a simple mediation result. The unstandardised indirect effect of anhedonia (SHAPS) on the relationship between number of infusions and depression severity (IDS‑SR 30) was −1.90 (95% CI −2.83 to −0.91) using bootstrap estimation. The mediation model was significant: F(2,202) = 74.42, p < 0.001, explaining 42% of the variance in IDS‑SR 30 score change.

Strzelecki and colleagues interpret their findings as evidence that repeated intravenous ketamine infusions produced a statistically significant reduction in anhedonia among patients with treatment‑resistant depression, and that decreases in anhedonia mediated a substantial portion of the observed improvement in overall depressive symptoms. The authors note that the antianhedonic effect was observed primarily in patients not taking benzodiazepines; BDZ co‑treatment appeared to attenuate ketamine’s effect on SHAPS scores. Depressive symptoms measured by IDS‑SR 30 declined over the infusion course, and suicidal ideation (IDS‑SR item 18) decreased after the fifth and seventh infusions but returned to baseline by the post‑treatment visit. The discussion places these results alongside earlier reports of rapid antianhedonic effects after single ketamine infusions and retrospective series of multiple infusions. The investigators cite prior findings in which improvements in SHAPS accounted for portions of variance in reductions of suicidal ideation or depressive scores (for example, prior reports attributing 13% or 26% of variance to anhedonia change), and consider their mediation result (42% of variance explained) consistent with the notion that anhedonia reduction is an important pathway for ketamine’s antidepressant action. Regarding benzodiazepines, the authors compare their attenuation finding with case reports and post‑hoc analyses suggesting BDZs can limit ketamine response or prolong time to response/remission; they propose a mechanistic hypothesis that BDZs, as GABA‑A agonists, might counteract ketamine’s disinhibitory effects on GABAergic interneurons and downstream neurotrophic processes. The paper also remarks that regular BDZ use has been associated with treatment resistance and potential worsening of depressive symptoms in other observational work. The investigators caution about ketamine’s abuse potential and stress the need for monitoring, particularly with longer‑term regimens. Key limitations acknowledged by the authors include the open‑label, non‑randomised, non‑blinded observational design without placebo control, lack of blood measurements of ketamine/metabolites or BDZ concentrations, and pooling of unipolar and bipolar patients in a relatively small sample size which precluded analysis of different benzodiazepines separately. They state that observational studies yield a low level (level C) of evidence and that the true effect might differ due to bias, imprecision, or inconsistency. Finally, the authors conclude that their findings warrant replication in larger, placebo‑controlled randomised trials and may inform safety and tolerability considerations in real‑world ketamine practice.