The research team, led by Gilberto Sousa Alves of the Federal University of Maranhão and collaborators across Brazil, Portugal, and Australia, analyzed charts from 72 outpatients aged 18 to 65 treated at three private advanced-care centers between 2023 and 2025. Twenty-six patients received the IV protocol, consisting of six sessions delivered two to three times weekly, while 46 received the SC protocol, which stretched across twelve sessions at a similar weekly rhythm. All participants had treatment-resistant depression, operationally defined as inadequate improvement after at least two adequate antidepressant trials in the current episode, and all had failed to respond sufficiently to prior treatment.

The pharmacological rationale for esketamine rests on mechanisms fundamentally different from those of classic antidepressants. Rather than targeting monoamine systems such as serotonin or norepinephrine, ketamine-based drugs antagonize N-methyl-D-aspartate (NMDA) receptors, triggering downstream glutamatergic signaling, synaptic plasticity, and the action of neuroactive metabolites. Some evidence also points to engagement of mu, kappa, and delta opioid receptors. The clinical consequence is speed: symptom relief can emerge within hours to days rather than weeks, a property that matters enormously for patients with severe or suicidal depression.

In Brazil, intranasal esketamine holds formal regulatory approval from ANVISA since 2020 for adults with major depressive disorder who have failed at least two antidepressants, and for rapid symptom reduction in major depression with acute suicidal ideation or behavior. But the IV and SC routes examined in this study remain off-label, requiring physician responsibility, written informed consent, careful risk-benefit documentation, and institutional safeguards under Brazilian Federal Council of Medicine Opinion No. 20/2024. This regulatory context makes real-world data on these parenteral routes especially valuable, because they document how clinicians actually navigate an area where evidence and approval have not fully converged.

The two protocols differed in structure as well as route. IV patients received esketamine diluted in 250 milliliters of saline and infused over one hour through a peripheral vein, with hemodynamic monitoring every ten minutes and at least one hour of post-infusion observation. SC patients received injections into the hypogastric region, a far simpler procedure that requires no pumps, venous access, or infusion infrastructure. Both protocols started at 0.5 mg/kg and titrated upward to 0.75 and then 1.0 mg/kg when patients had not achieved at least a 50 percent symptom improvement or remission. Notably, the mean initial dose was lower in the SC group, 26.9 mg versus 36.2 mg, although final doses converged at around 60 mg in both arms.

The primary outcome was categorical response on the Beck Depression Inventory-II (BDI-II), with patients classified as nonresponders, partial responders, or full responders based on percentage symptom reduction. At the shared six-session milestone, the IV group showed a clearly more favorable distribution: 57.69 percent achieved full response, compared with only 30.44 percent in the SC group, while 41.30 percent of SC patients were classified as failures versus 19.23 percent of IV patients. The overall chi-square test fell just short of conventional significance (p = 0.059), but an ordinal logistic regression that exploits the ordered hierarchy of response categories estimated roughly threefold higher odds of a superior response tier for IV treatment (odds ratio 3.06, 95 percent confidence interval 1.22 to 8.00).

By the protocol-specific endpoints, however, the picture changed. The IV endpoint remained session six, while the SC endpoint was session twelve, reflecting a longer cumulative course. At those respective endpoints, categorical response no longer differed significantly between groups (p = 0.541; odds ratio 1.44, 95 percent CI 0.57 to 3.73). The SC group’s failure rate had dropped to 19.57 percent, nearly identical to the IV group’s 19.23 percent, and 45.65 percent of SC patients achieved full response. Continuous BDI-II scores told a complementary story: the IV group showed a significantly larger median reduction by session six (a Hodges-Lehmann median difference of 7.00 points in favor of IV, p = 0.027), but the overall reduction from baseline to each protocol endpoint was statistically indistinguishable between routes.

Safety findings added an important dimension to the comparison. No serious adverse events or adverse-event-related discontinuations were documented in the analyzed cohort. Dissociation, the transient dreamlike state characteristic of ketamine-class drugs, occurred in 53.8 percent of IV patients but only 13.0 percent of SC patients, a statistically significant difference (p = 0.001). Nausea and mild transient blood pressure elevations did not differ significantly between routes. For clinicians weighing options, this tolerability profile matters: dissociation can be distressing for patients and demands staffing and monitoring resources, whereas the SC route’s gentler side-effect profile may broaden the population that can be treated in less intensively staffed settings.

Beyond the route comparison, the study contributed an exploratory psychometric analysis. Factor analysis of baseline BDI-II scores identified two symptom dimensions: a cognitive-affective factor, in which suicidal thoughts loaded most heavily (0.80) alongside guilt, sadness, pessimism, and worthlessness, and a somatic-neurovegetative factor dominated by appetite changes, concentration difficulty, loss of interest, and fatigue. The authors emphasize that this structure is hypothesis-generating rather than predictive; it does not establish that these dimensions forecast esketamine response, but it suggests a coherent organization of baseline symptoms that larger prospective studies could test as potential predictors of route-specific trajectories.

The authors are careful about what these findings do and do not prove. The retrospective, naturalistic design cannot isolate route-specific causal effects, because the administration route was tied to the center, and diagnostic composition, dosing strategies, and routine-care procedures varied across sites. The absence of an endpoint difference should not be read as formal equivalence or non-inferiority, since the study was not designed for that purpose. Still, the work stands among the few multicenter Latin American comparisons of IV and SC esketamine, and its practical message is clear: when rapid response is critical, IV remains the better-supported choice; when simplicity, tolerability, and access are priorities, subcutaneous esketamine emerges as a feasible and statistically comparable alternative. Future prospective studies with harmonized eligibility criteria, standardized dosing, and uniform assessment protocols will be needed to determine whether the speed-versus-simplicity trade-off observed here holds up as a genuine property of the routes themselves.

Subject of Research: Comparative real-world outcomes of intravenous versus subcutaneous esketamine protocols for treatment-resistant depression

Article Title: Real-world clinical outcomes under intravenous and subcutaneous esketamine protocols for treatment-resistant depression: a multicenter, retrospective study

Article References: Alves, G. S., Dos Santos, H. M. M., Damasceno, S. D., Maciel, L. M., da Silva, O. A. F. S., Fe, C. S. D. M., Santana, J. G., Freitas, M. B. L. D., Carvalho, A. F., & Monteiro, D. (2026). Real-world clinical outcomes under intravenous and subcutaneous esketamine protocols for treatment-resistant depression: a multicenter, retrospective study. Frontiers in Psychiatry, 17, Article 1963841. https://doi.org/10.3389/fpsyt.2026.1963841

Image Credits: AI Generated

DOI: 10.3389/fpsyt.2026.1963841

Keywords: esketamine, treatment-resistant depression, ketamine, intravenous administration, subcutaneous administration, NMDA receptor antagonism, BDI-II, dissociation, off-label treatment, real-world evidence, Brazil, psychiatry