Real‐World Effectiveness and Cost‐Differential of Intranasal Esketamine Versus Intramuscular Ketamine

Treatment-resistant depression is associated with substantial functional impairment, suicide risk, and mortality. Ketamine has been shown to have rapid antidepressant effects, and intranasal (IN) esketamine is an FDA-approved formulation for treatment-resistant depression. However, the main advantage of IN esketamine is ease of administration, while its major drawback is high cost. The authors note that evidence for intramuscular (IM) ketamine has been limited, with no published head-to-head comparisons against IN esketamine, leaving an important gap for clinics considering whether IM ketamine could offer similar benefits at lower cost. Bhatt and colleagues set out to test whether IM ketamine is non-inferior to IN esketamine during the induction phase of treatment for treatment-resistant depression, and to compare safety and healthcare costs between the two approaches. They focused on a real-world Veteran population treated in a VA outpatient neuromodulation clinic, where the clinic had sequentially moved from IN esketamine to IM ketamine, creating an opportunity for comparison.

Bhatt and colleagues conducted a retrospective, observational sequential cohort study at the Neuromodulation Program within the Jennifer Moreno VA Medical Center in San Diego. The clinic served Veterans receiving interventional psychiatric care. Because the clinic changed its formulary over time, patients treated from January 2020 to March 2022 received IN esketamine, whereas those treated from April 2022 to April 2025 received IM ketamine. This temporal sequencing was intended to reduce allocation bias that might occur if both treatments were offered concurrently. The sample included 179 adults with treatment-resistant depression, defined as non-response to at least two adequate antidepressant trials. Patients had either major depressive disorder or bipolar disorder, and comorbid PTSD was common. Exclusion criteria included active substance use disorders, psychosis, and unstable medical illness. For the main effectiveness analysis, the authors restricted the sample to treatment completers, defined as those who finished the full induction course of eight treatments; 92% of the initial sample completed treatment. The study was reviewed and deemed exempt by the local IRB. Both groups underwent an induction phase of eight sessions over 4 weeks, given twice weekly. IN esketamine started at 56 mg and could be increased to 84 mg according to response and tolerability. IM ketamine began at 0.5 mg/kg and could be titrated up to 1.0 mg/kg. Patients were encouraged to remain on their usual oral medications. All treatments were delivered by certified nursing staff under psychiatrist supervision, with 120 minutes of monitoring afterwards. The primary outcome was change in depressive symptoms measured by the Patient Health Questionnaire-9 (PHQ-9). Secondary clinical outcomes included PTSD symptoms measured with the PTSD Checklist for DSM-5 (PCL-5), response defined as at least 50% improvement in PHQ-9 scores, and remission defined as PHQ-9 ≤5. Assessments were collected at baseline and at weekly intervals during induction, with a final post-treatment assessment after the eighth treatment. Safety outcomes were severe events within 6 months of first treatment, based on chart review, including emergency department visits, Code Green and Code Blue activations, and psychiatric and non-psychiatric hospitalisations. The cost analysis estimated staffing, pharmacy labour, and drug acquisition costs from the VA perspective. Statistical analyses used linear mixed-effects models for the primary outcome, with non-inferiority assessed against a pre-specified margin of -1.25 PHQ-9 points. The paper states that groups were compared using standard tests for continuous and categorical variables, and that the study was not randomised or blinded.

Baseline characteristics were similar between the IN esketamine and IM ketamine cohorts. Mean age was essentially identical, the proportion of men was similar, and there were no significant differences in prior ketamine exposure, baseline PHQ-9 scores, baseline PCL-5 scores, or racial and ethnic distribution. For the primary outcome, IM ketamine was non-inferior to IN esketamine. The baseline-adjusted estimated marginal mean difference in final PHQ-9 scores was 0.04 points, with a 95% confidence interval from -1.21 to 1.29. Because the lower bound stayed above the non-inferiority margin of -1.25, the authors concluded non-inferiority. PTSD symptoms were also non-inferior, with a baseline-adjusted difference in final PCL-5 scores of -1.05 points (95% CI -4.81 to 2.71), within the pre-specified margin of five points. Both groups improved substantially over the induction course. In the IN esketamine group, mean PHQ-9 scores fell from 19.5 to 14.2, a reduction of 5.3 points, with response and remission rates of 19.1% and 7.9%. In the IM ketamine group, PHQ-9 scores also fell from 19.5 to 14.2, again a 5.3-point reduction, with response and remission rates of 22.2% and 5.6%. Response and remission did not differ significantly between groups. PTSD symptoms improved in both groups as well: PCL-5 scores decreased from 52.9 to 38.9 with IN esketamine and from 53.6 to 38.9 with IM ketamine. Safety outcomes were broadly similar. Emergency department visits did not differ between groups, no Code Blue events occurred in either group, and non-psychiatric hospitalisations were uncommon and similar. Psychiatric code activations were rare and only occurred in the esketamine group. Psychiatric hospitalisations were seen in five esketamine-treated patients and none of the IM ketamine group, but this difference was not statistically significant. The cost analysis showed a marked difference in favour of IM ketamine. The per-treatment cost was $758.67 for IN esketamine and $80.95 for IM ketamine. Staffing costs were the same across groups, but esketamine had much higher drug acquisition costs. Across the full eight-treatment course, total cost was $6069.36 for IN esketamine versus $647.60 for IM ketamine, a difference of $5421.76 per course.

The authors interpret this study as the first direct comparison of IM ketamine with FDA-approved IN esketamine for clinical effectiveness, safety, and cost. They argue that the findings support three main points: IM ketamine produced non-inferior antidepressant and PTSD symptom outcomes, had a similar safety profile, and did so at roughly one-tenth of the cost. They present IM ketamine as a pragmatic and potentially high-value alternative that could improve access to ketamine-based treatment. Bhatt and colleagues position their findings as consistent with earlier research showing rapid antidepressant effects for IV ketamine and IN esketamine, while extending the evidence base to IM ketamine. They also highlight the improvement in PTSD symptoms, which they consider clinically important in a Veteran population with high rates of comorbid PTSD and depression. They suggest that these results support further randomised controlled research on PTSD outcomes. With respect to safety, the authors state that there were no significant differences in sentinel adverse events between groups. They note that emergency department use was fairly high in both cohorts, which may reflect the severity of illness rather than treatment-specific harm. They also remark that the absence of significant differences supports the practicality of IM ketamine in outpatient settings that may not be able to meet the administrative requirements associated with esketamine. The largest difference was economic. The authors emphasise that the cost gap was driven almost entirely by medication acquisition, not by staffing or overhead. They interpret this as especially relevant to the VA and similar systems where patients often face little direct cost, but they also note that the situation may differ in commercial insurance settings, where coverage patterns can affect patient access and out-of-pocket expense. They suggest that broader insurance coverage for IM ketamine could create payer savings and improve access. The main limitations acknowledged by the authors are the observational, non-randomised, and unblinded design, which leaves room for unmeasured confounding. They also note that the sequential treatment periods could have introduced temporal bias, including changes in referral patterns, clinician experience, and external conditions such as the COVID-19 period during which IN esketamine was used. Additional limitations were reliance on self-reported PHQ-9 outcomes, focus on the induction phase only, lack of durability or relapse data, and the fact that the cost analysis was specific to the VA setting. They also caution that findings from a Veteran population may not generalise to other healthcare environments.

The authors conclude that IM ketamine appears to be a safe, effective, and much less expensive alternative to IN esketamine for treatment-resistant depression. They suggest that IM ketamine could be an important way to expand access to ketamine-based treatment for patients and healthcare systems.