The rapid antidepressant effectiveness of repeated dose of intravenous ketamine and intranasal esketamine: A post-hoc analysis of pooled real-world data

Treatment-resistant depression (TRD) is a common and disabling form of major depressive disorder that fails to respond to at least two adequate antidepressant trials and contributes substantially to individual and societal burden. Recent research has identified glutamatergic agents, notably intravenous racemic ketamine (KET-IV) and intranasal esketamine (ESK-NS), as rapid-acting antidepressant options for TRD. Despite randomised and real-world studies supporting both formulations, direct comparative data on their relative short-term effectiveness, safety and tolerability in routine clinical practice are limited, and meta-analyses to date have yielded mixed findings. D'andrea and colleagues set out to address this gap by performing a retrospective, post-hoc pooled analysis of two real-world TRD cohorts treated with repeat-dose KET-IV or ESK-NS. The primary aim was to compare antidepressant effectiveness over an approximately one-month interval using effect sizes, response and remission rates; secondary objectives included comparisons of anti‑suicidal effects, adverse events and discontinuation rates. The analysis focuses on adults with unipolar TRD and evaluates outcomes after the acute treatment/induction phases for each formulation.

This study pooled data from two retrospective cohorts: a KET-IV sample (n = 171) treated at the Canadian Rapid Treatment Center of Excellence (CRTCE) and an ESK-NS sample (n = 140) from the multicentre REAL-ESK real-world study in Italy. Only participants primarily diagnosed with major depressive disorder were included; subjects with bipolar disorder were excluded. To permit comparison, analyses were restricted to participants with available age and gender data and to two common assessment time-points: baseline (T0) and an approximately one-month follow-up (T1), defined as the post-acute visit for KET-IV and the end of the induction phase for ESK-NS. KET-IV was administered following American Psychiatric Association guidance: an acute phase of four infusions over two weeks, with the first two infusions at 0.5 mg/kg over 40 minutes and optional dose optimisation to 0.75 mg/kg for later infusions if response was suboptimal. Treatments were generally given adjunctive to existing psychotropic regimens; MAOIs were stopped at least two weeks before infusion, naltrexone was prohibited during ketamine treatment, benzodiazepines withheld for 12 hours pre-infusion, and other medications paused around infusion to reduce interaction risk. ESK-NS dosing followed EMA recommendations: starting at 56 mg on day one (28 mg for patients >65 years), with bi-weekly administration and clinician-determined dose adjustments including optimisation to 84 mg if needed. Depressive severity was measured with different instruments in each cohort: self-rated QIDS-SR16 for the KET-IV group and clinician-rated MADRS for the ESK-NS group. Primary outcomes were change in depression severity (effect sizes), response (≥50% reduction from baseline on the respective scale) and remission (QIDS-SR16 < 5 or MADRS < 10) at T1. Suicidal ideation was assessed via a single MADRS item (item 10) or QIDS-SR16 item 12. Safety outcomes included treatment-emergent adverse events (TEAEs) and discontinuation rates. Statistical analyses used SPSS and JASP: normality was tested with Shapiro–Wilk; repeated-measures ANCOVA (rm-ANCOVA) assessed within-subject change controlling for gender (between-factor) and age and number of failed antidepressant trials (covariates); non-parametric Wilcoxon signed-rank tests were used when distributions were non-normal. Effect sizes were reported as Cohen's d for rm-ANCOVA and Rank-Biserial Correlation for Wilcoxon tests. Response and remission comparisons used Pearson χ2 tests and binomial logistic regression estimated effects of treatment type and demographic covariates. Ethical approvals and data confidentiality measures were reported.

The pooled sample comprised 311 TRD participants: 171 in the KET-IV cohort and 140 in the ESK-NS cohort. Groups were matched on age and gender and did not differ significantly in the number of previous unsuccessful antidepressant trials. In the KET-IV group, 106 participants had dose escalation to 0.75 mg/kg at infusion three while 65 remained on 0.5 mg/kg; mean number of infusions was 4 (±0.353). In the ESK-NS group, most participants received either 56 mg (n = 66) or 84 mg (n = 69), with 5 participants receiving 28 mg. Both treatments produced statistically significant reductions in depressive symptoms from T0 to T1 after adjustment for covariates. For the KET-IV cohort, the rm-ANCOVA estimated a mean reduction of 5.65 on QIDS-SR16 (95% CI 5.59–5.71), F1,143 = 22.428, p < 0.001, with a large effect size (Cohen's d = 1.666; 95% CI 1.367–1.859). For the ESK-NS cohort, the rm-ANCOVA estimated a mean reduction of 11.41 on MADRS (95% CI 10.76–12.06), F1,126 = 5.146, p = 0.025, with a large but smaller effect size (Cohen's d = 1.244; 95% CI 1.039–1.350). Comparing categorical outcomes, KET-IV achieved higher response rates at T1 than ESK-NS (36% vs. 25%; p = 0.042), whereas remission rates were similar (13% vs. 12%; p = 0.845). Suicidal ideation scores (non-normally distributed) decreased significantly in both cohorts by Wilcoxon signed-rank tests (KET-IV: W = 2,381, p < 0.001; ESK-NS: W = 3,451.5, p < 0.001) with comparable large effect sizes (Rank-Biserial Correlation 0.916 and 0.934, respectively). In terms of safety, KET-IV had a higher incidence of reported dissociative symptoms and a higher overall rate of TEAEs; the KET-IV group also had a higher overall discontinuation rate (18% vs. 8.5%). However, discontinuations specifically for severe TEAEs were low in both groups and not significantly different (4.6% vs. 2.12%; p = 0.228). No serious or life‑threatening adverse events, including urological toxicity, were reported in either cohort.

D'andrea and colleagues interpret these findings as evidence that both repeat-dose KET-IV and ESK-NS produce rapid antidepressant effects in real-world TRD patients, including reductions in suicidal ideation within the first month. The study's rm-ANCOVA results and large effect sizes in each cohort are presented as consistent with prior literature showing expedited symptom relief with glutamatergic agents. A direct comparison of effect-size estimates and response rates suggested a greater short-term antidepressant magnitude for KET-IV than for ESK-NS at the one-month time-point, although remission rates were similar. The authors discuss possible explanations for the faster observed action of KET-IV, emphasising pharmacokinetic and pharmacodynamic differences: intravenous racemic ketamine delivers nearly 100% bioavailability versus approximately 54% for intranasal esketamine, and racemic ketamine contains arketamine, which emerging preclinical and early clinical work suggests may contribute substantially to rapid antidepressant effects. They caution, however, that late responders to ESK-NS have been reported in other real-world studies, meaning the apparent superiority of KET-IV at one month should not be interpreted as definitive evidence of overall superiority without longer follow-up. Safety-wise, both interventions were judged generally well tolerated in naturalistic settings; TEAEs were transient and no serious adverse events were recorded. ESK-NS had a lower incidence of TEAEs, while KET-IV showed more reported dissociation and higher overall discontinuation, although discontinuations for severe adverse events were rare and statistically similar between groups. The authors acknowledge several important limitations affecting interpretation: the retrospective, pooled real-world design introduces potential selection biases; different assessment instruments were used across cohorts (self‑rated QIDS-SR16 versus clinician-rated MADRS), which may affect sensitivity to change despite convergent validity between scales; precise alignment of T1 relative to last treatment differed across cohorts; and recruitment settings differed (private KET‑IV clinic versus public ESK‑NS centres), which could contribute to heterogeneity despite matching on age, gender and number of prior antidepressant trials. Given these limitations, the authors call for prospective, head‑to‑head randomised comparisons with longer follow-up to determine comparative and maintenance efficacy and to further characterise safety profiles. They also suggest exploring combined strategies—using KET‑IV for acute symptom control and ESK‑NS for maintenance—as a potential avenue for future research.

In this real-world pooled analysis, both repeat-dose intravenous ketamine and intranasal esketamine produced rapid antidepressant effects and reduced suicidal ideation in adults with treatment-resistant unipolar depression. KET-IV was associated with a larger short-term antidepressant effect and higher response rates at approximately one month, whereas ESK-NS showed a lower incidence of treatment-emergent adverse events. The authors recommend prospective head‑to‑head trials with longer-term follow-up to clarify comparative efficacy and safety and suggest investigating potential sequential or combined use of the two formulations (KET‑IV for acute relief and ESK‑NS for maintenance).