Response to Intravenous Racemic Ketamine After Switch From Intranasal (S)-ketamine on Symptoms of Treatment-Resistant Depression and PTSD in Veterans: a Retrospective Case Series

Racemic (R,S)-ketamine produces rapid antidepressant effects in randomised trials, and its mechanism is commonly linked to NMDA receptor antagonism. Because the (S) enantiomer is more potent at NMDA receptors and intranasal formulations have only partial bioavailability, a nasal spray of (S)-ketamine (Spravato) was developed and approved for treatment-resistant depression (TRD). Despite this approval, it remains unclear whether intravenous (IV) racemic ketamine provides greater clinical benefit than intranasal (IN) (S)-ketamine, given differences in dosing limits, bioavailability and the potential activity of (R)-ketamine metabolites. Bentley and colleagues conducted a retrospective analysis to examine clinical outcomes in Veterans with comorbid TRD and post-traumatic stress disorder (PTSD) who had received at least six IN-(S)-ketamine treatments and were subsequently switched to IV racemic (R,S)-ketamine. The primary aim was to compare changes in depression (PHQ-9) and PTSD (PCL-5) symptoms across the two formulations, with the hypothesis that IV-(R,S)-ketamine would produce larger symptom improvements than IN-(S)-ketamine. This investigation addresses a practical clinical question about next-step options for patients who do not adequately respond to the FDA-approved intranasal formulation.

This work is a retrospective chart review of Veterans treated in the VA San Diego Neuromodulation Clinic between January 2020 and March 2021. Inclusion required receipt of at least six IN-(S)-ketamine treatments followed by at least two IV-(R,S)-ketamine treatments; 15 Veterans met these criteria (one additional Veteran was excluded for having only one IV treatment and no follow-up PHQ-9). All included Veterans had diagnoses of treatment-resistant depression (failure of at least two antidepressant trials) and PTSD. IN-(S)-ketamine treatment in these patients was titrated up to a maximum intranasal dose (reported as 84 mg for at least three sessions) prior to switching. IV-(R,S)-ketamine was started at 0.5 mg/kg, increased to 0.75 mg/kg at the second IV session and to 1.0 mg/kg at the third session; subsequent IV sessions continued at 1.0 mg/kg. Clinical assessments comprised the PHQ-9 for depression and the PCL-5 for PTSD, administered prior to each treatment as part of routine care. Additional chart-extracted variables included demographics, years in mental health care, suicide attempts, hospitalisations, diagnostic history (including bipolar vs unipolar depression), chronic pain, substance-use disorders, and prior neuromodulation treatments. The team did not systematically collect side-effect data from charts and therefore did not report adverse-event frequencies. For statistical analysis, the investigators used repeated-measures ANOVA (rmANOVA) to evaluate PHQ-9 and PCL-5 across four prespecified time-points, selecting the 6th IV treatment as a common induction end-point. Three patients discontinued before the 6th IV and the authors applied a last-observation-carried-forward (LOCF) approach for those missing data; a mixed-effects model with missing data was also tested and gave similar results. Greenhouse–Geisser correction was applied to rmANOVA p-values and Tukey's HSD was used for post hoc pairwise comparisons. Additional analyses included paired t-tests comparing IN- versus IV-related change from baseline, an rmANOVA across the first six IV treatments to assess speed of effect, and a paired t-test comparing scores at the 6th IV versus the last IV session to assess durability. Normality checks were performed before t-tests/ANOVAs; analyses used SPSS and GraphPad Prism. The authors note that a detailed Table described participant characteristics but full table contents are not present in the extracted text.

Across the treatment course (IN-(S)-ketamine followed by IV-(R,S)-ketamine), repeated-measures ANOVA indicated significant overall reductions in both depression and PTSD scores. Specifically, PHQ-9 scores showed a significant effect over time (rmANOVA F(14,42) = 12.6, p < 0.0001) and PCL-5 scores likewise decreased significantly (rmANOVA F(13,39) = 5.9, p = 0.006). Post hoc comparisons reported by the authors showed that, relative to pretreatment baseline, PHQ-9 scores were reduced on average by 2.4 ± 1.2 points after the IN-(S)-ketamine phase (p = 0.1) and by 5.6 ± 1 points after the IV-(R,S)-ketamine phase (p = 0.0003). For PTSD symptoms, PCL-5 scores declined by an average of 4.3 ± 3.3 points after IN-(S)-ketamine (p = 0.6) and by 11.8 ± 3.5 points after IV-(R,S)-ketamine (p = 0.03) compared to baseline. The temporal pattern of response differed between domains. Antidepressant effects after switching to IV-(R,S)-ketamine emerged rapidly, within two IV treatments, and then remained relatively stable. Improvements in PTSD symptoms appeared more gradual and continued during the transition to maintenance (weekly or less frequent) dosing. Dose-related observations included no significant antidepressant effect at the starting IV dose of 0.5 mg/kg, with symptom change becoming significant after escalation to 0.75 mg/kg. Sample and dosing context are relevant to interpretation: 15 Veterans contributed data, with three discontinuing before the 6th IV treatment (handled via LOCF). The IV dosing employed from the third session averaged 1 mg/kg, corresponding in these Veterans to approximately 89 mg total racemic ketamine (about 45 mg each of (S)- and (R)-ketamine). Given estimated intranasal bioavailability of roughly 50%, the authors note that the highest approved IN-(S)-ketamine dose would correspond to an effective S-enantiomer dose of about 40.5 mg, implying that the IV racemic regimen provided an overall higher amount of active compound and the addition of the R-enantiomer.

Bentley and colleagues interpret their findings as preliminary evidence that IV racemic (R,S)-ketamine may yield additional symptom benefit for Veterans with TRD and comorbid PTSD who showed suboptimal response to intranasal (S)-ketamine. They highlight the relatively rapid antidepressant response after switching to IV treatment (within two sessions) and a slower, continuing improvement in PTSD symptoms during maintenance. Two mechanistic explanations are proposed: a simple dose-response effect, with IV dosing delivering a larger effective ketamine dose, and a potential unique contribution of the (R)-enantiomer or its metabolite hydroxynorketamine, which preclinical work suggests may act via AMPA-receptor modulation. The authors emphasise multiple limitations that constrain causal inference. This was a small, retrospective, non-randomised and unblinded case series drawn from a specific clinical population, and some Veterans had prior IV/IM ketamine exposure, which could affect expectations and placebo/nocebo responses. The lack of systematic side-effect recording is noted as an additional limitation. The investigators acknowledge that expectancy effects (for example, a stronger placebo response to IV treatments or to the experience of having an IV placed) and selection biases related to which patients were switched to IV could have contributed to the observed differences. Given these constraints, the authors refrain from concluding that IV-(R,S)-ketamine is definitively superior to IN-(S)-ketamine on average. In practical terms, the study team suggests that IV racemic ketamine could be considered as a next-step option for individuals who show limited improvement with the highest tolerated doses of intranasal (S)-ketamine. They call for a properly powered, double-blind randomised controlled trial directly comparing IV racemic ketamine (up to 1 mg/kg) with intranasal (S)-ketamine to clarify comparative efficacy, dose-response relationships, and mechanistic differences between enantiomers. The authors conclude that their results provide initial clinical data to support such trials but that confirmation will require prospective, controlled research.