Effect of Repeated Intravenous Esketamine on Adolescents With Major Depressive Disorder and Suicidal Ideation: A Randomized Active-Placebo-Controlled Trial

This was a double-blind, randomised, active-placebo-controlled trial conducted at the Affiliated Brain Hospital of Guangzhou Medical University between December 2020 and April 2022. Q11 and colleagues enrolled an adolescent cohort from a larger trial and obtained written informed consent from participants and their parents. The trial protocol was approved by the institutional ethics committee, and the present report uses the adolescent-specific data. Adolescents aged 13–18 years admitted to the inpatient ward were eligible if they met DSM-5 criteria for major depressive disorder without psychotic features (assessed with K-SADS-PL), had baseline moderate-to-severe depressive symptoms (17-item HAM-D score ≥17), and had suicidal ideation of at least 1 on the clinician-rated Columbia Suicide Severity Rating Scale (C-SSRS) Ideation and elevated scores on the Beck Scale for Suicide Ideation (SSI) at screening. Exclusion criteria included substance use, primary psychotic or bipolar disorder, certain neurodevelopmental or neurological disorders, significant medical illness, and suicidal behaviour (an actual attempt) on presentation or within the prior 6 months. Participants were randomly assigned 1:1 to receive three intravenous infusions of either esketamine (0.25 mg/kg) or midazolam (0.045 mg/kg) on days 1, 3 and 5, in addition to routine inpatient nursing care and clinician-managed oral antidepressant therapy (monotherapy or combination). Midazolam was chosen as an active comparator to mimic nonspecific sedative effects and help preserve blinding. Infusions were given in 50 mL saline over 40 minutes with clinical monitoring of pulse and blood pressure before infusion and at 20, 40 and 70 minutes; infusions could be stopped for intolerable adverse events. The primary outcome was change in suicidal ideation from baseline (day 0) to 24 hours after the final infusion (day 6), assessed by the C-SSRS Ideation and C-SSRS Intensity (the former grades suicidal thoughts from wishing to be dead up to intent with plan; the latter addresses frequency, duration and controllability). Key secondary outcomes included change in depressive symptoms measured by the Montgomery–Åsberg Depression Rating Scale (MADRS) and the first five items of the SSI (SSI-5), assessed at the same time points: 24 hours after each infusion (days 2, 4, 6) and at follow-ups on days 12, 19 and 33 (1, 2 and 4 weeks after the third infusion). Clinical response and remission were defined a priori (anti‑suicidal response = ≥50% improvement in C-SSRS Ideation; remission = C-SSRS Ideation = 0; antidepressant response = ≥50% improvement in MADRS; MADRS remission = score ≤12). All participants who received at least one infusion were included in efficacy and safety analyses. The planned sample size was at least 25 per group (allowing for a 10% dropout) based on a meta-analytic effect size estimate of 0.85 for clinician-rated suicidal ideation after a single ketamine dose, giving 80% power at α = .05. Changes over time in primary and secondary continuous outcomes were analysed using linear mixed models including fixed effects for time, drug and time-by-drug interaction, and covariates for oral antidepressant class (SSRI v non-SSRI), augmentation therapy (yes/no) and baseline scores. Bonferroni-corrected post hoc tests and two-sample proportion tests for categorical outcomes were used. Safety measures included systematic adverse event reporting, brief psychiatric rating items, the Clinician Administered Dissociative Symptoms Scale (CADSS), Young Mania Rating Scale Item 1, and assessments for probable dependence (Severity of Dependence Scale, SDS) and craving (visual analogue scale) during follow-up. Statistical analyses were conducted in SPSS v27 with significance set at .05.

Fifty-four adolescents (mean age 14.7 years, SD 1.4; 48 female participants, 88.9%) underwent randomisation and received at least one infusion. The treatment groups were reported as similar on baseline demographic and clinical measures. Primary outcomes: From baseline to day 6 (24 hours after the third infusion), the mean change in C-SSRS Ideation score was greater in the esketamine group (mean change −2.6, SD 2.0) than in the midazolam group (mean change −1.7, SD 2.2). The mixed-model analysis yielded a significant main effect of drug (F = 11.172, p = .002) and a significant time effect (F = 30.424, p < .001); the time-by-drug interaction for Ideation was not significant (F = 1.239, p = .298). For C-SSRS Intensity, the mean change to day 6 was −10.6 (SD 8.4) with esketamine versus −5.0 (SD 7.4) with midazolam; there were significant main effects for drug (F = 9.346, p = .004) and time (F = 26.634, p < .001) and a significant time-by-drug interaction (F = 4.171, p = .007). Group differences favouring esketamine were apparent by day 2 and persisted through day 6. Secondary outcomes: For MADRS, the linear mixed model found a significant time-by-drug interaction (F = 2.970, p = .011) and a time main effect (F = 42.501, p < .001) but no overall drug main effect (F = 2.089, p = .154). Between-group differences in MADRS were numerically greater for esketamine from day 2 onward but reached statistical significance only at day 6. The SSI-5 change from baseline to day 6 was reported as greater after esketamine (extracted text gives an incomplete numeric value). More participants in the esketamine group were free of suicidal ideation at day 4, and this advantage continued through days 6 and 12 (p < .05). Reported 4-week clinical outcomes in the abstract indicated anti‑suicidal and antidepressant response rates of 69.2% and 61.5% after esketamine, versus 52.5% and 52.5% after midazolam; the main text states that antidepressant responder proportions were greater for esketamine on days 12, 19 and 33 (p < .05), and remitter proportions were greater for esketamine on days 6, 12, 19 and 33 (p < .05). The extracted text references figures and supplementary tables for additional numeric detail. Safety: No deaths or serious cardiorespiratory adverse events were reported. Discontinuation because of treatment-emergent adverse events was reported for 2 participants in the esketamine group (nausea, hallucinations) and for 3 participants in the midazolam group (agitation in two, suicidal behaviour in one); the percentage figures in the extraction appear inconsistent and are not clearly reported. Common adverse events with esketamine included nausea, dissociation, dry mouth, sedation, headache and dizziness; most adverse events occurred on infusion days and were transient. Physiological effects included transient mean increases on infusion days in the esketamine group for systolic blood pressure (mean max increase ≈ 5.4 ± 9.0 mm Hg), diastolic blood pressure (≈ 5.8 ± 8.2 mm Hg) and pulse (≈ 5.8 ± 7.0 beats/min), whereas the midazolam group showed small transient decreases. Eight infusions (in six patients) were associated with transient hypertension (systolic ≥140 mm Hg and/or diastolic ≥90 mm Hg). Psychiatric adverse measures showed higher BPRS-5 scores after esketamine on days 1 and 3 (day 1: F = 5.102, p = .028; day 3: F = 7.410, p = .009). Dissociative symptoms measured by CADSS were higher at 30 minutes after each esketamine infusion compared with midazolam (day 1: F = 16.026, p < .001; day 3: F = 11.628, p = .001; day 5: F = 10.906, p = .002) but resolved by 24 hours and during follow-up. Dissociation severity declined across repeated infusions in the esketamine group (mean CADSS at 30 minutes: 7.9 on day 1 vs 2.8 on day 5). There were no manic episodes reported and no cases meeting the study’s SDS cutoff for probable drug dependence during the 4-week follow-up; differences in SDS and craving VAS scores between groups were not statistically significant.

Q11 and colleagues interpret their findings as evidence that repeated intravenous esketamine added to routine inpatient care produced a rapid reduction in suicidal ideation and significant improvement in depressive symptoms among adolescents with MDD and suicidal ideation. They highlight that clinician-rated measures of ideation and intensity showed greater reductions with esketamine than with midazolam, with significant benefits detectable as early as 24 hours after the first infusion and persisting through at least one week; antidepressant benefits were more apparent by 24 hours after the third infusion. The investigators place these results alongside adult ketamine/esketamine literature, noting similarities in anti‑suicidal effects but also differences in timing and magnitude relative to some adult intranasal esketamine trials and single‑dose intravenous ketamine studies. They suggest that differences in formulation, dose, frequency and patient age may account for discrepant findings across studies. The authors also note the possibility that combined treatment with oral antidepressants may have contributed to mood improvement and that ketamine augmentation could have had a synergistic effect. Regarding safety, the study team reports an adverse-event profile broadly consistent with adult data: transient increases in blood pressure and pulse, short-lived dissociation that diminished with repeated dosing, and no persistent dissociation, emergent mania, psychosis or observed drug dependence within the 4‑week follow-up. The authors caution, however, that the short follow-up precludes firm conclusions about longer-term risks such as cognitive effects or recreational use. Key limitations acknowledged by the authors include the concomitant and variable routine inpatient psychiatric medications (including benzodiazepines in up to 68.8% of patients), which could have altered esketamine's effects and complicated interpretation; potential unblinding because characteristic dissociative effects of esketamine are difficult to mask even with active placebo midazolam; exclusion of individuals with recent suicidal behaviour, which limits generalisability to higher‑risk populations; and use of intravenous rather than intranasal esketamine, which affects comparability with trials conducted where intranasal formulations are available. The authors conclude that larger, longer-term studies are needed to confirm clinical efficacy and to assess long-term safety outcomes in adolescents, such as cognitive impairment and potential for recreational use.