DOI: 10.1192/j.eurpsy.2026.10595 ISSN: 0924-9338

Dissociation Does Not Mediate Antidepressant Response to Esketamine: A Real-World Longitudinal Study with 6-Month Follow-Up

M. Olivola, V. Martiadis, K. La Monica, C. M. Angeletti, A. Frediani, V. Casati, R. Anniverno, G. Versaci, T. Prodi, N. Brondino, B. M. Dell’Osso

Introduction

The role of dissociation in mediating antidepressant response to intranasal esketamine remains debated. While dissociative experiences are often observed during treatment, their predictive value for clinical outcomes is uncertain.

Objectives

We aimed to assess whether (i) baseline dissociative traits (DES-II) and (ii) early changes in dissociation (ΔDES T0→T2) predict short-term antidepressant response, and to explore whether dissociation mediates depressive symptom improvement across 6 months of follow-up.

Methods

This multicenter naturalistic cohort included 31 patients with treatment-resistant depression (mean age 44.7±11.6) treated at academic community clinics in Pavia and Milan. Clinical assessments included MADRS at baseline (T0), 2 weeks (T1), 1 month (T2), 2 months (T3), 3 months (T4), and 6 months (T5) (Montgomery & Åsberg, Br J Psychiatry 1979). Dissociation was measured with DES-II at T0 and T2 (Carlson & Putnam, Dissociation 1993). Outcomes: % MADRS change and responder status (≥50% reduction). Predictors: DES-II at baseline and ΔDES T0→T2. Covariates: baseline MADRS, age, RMET (Baron-Cohen et al., J Child Psychol Psychiatry 2001), Psychache . Analyses included linear/logistic regressions and exploratory mediation (Preacher & Hayes, Behav Res Methods 2008).

Results

Mean MADRS decreased from 31.9±7.9 to 16.7±9.6 at T2 (−47%, responders 48%). Baseline DES-II (mean 0.185±0.12) showed no independent association with response (OR 0.18; 95%CI 0.00–15.5; p = 0.38). ΔDES T0→T2 did not predict response (OR 0.01; 95%CI 0.00–80.5; p = 0.36) nor % improvement (β = 12.7; p = 0.33). Mediation analysis revealed no significant indirect effect (a*b ≈ 10.7%; 95%CI −21.7–79.9). By DES tertiles, % improvement was comparable (Low 42.6%; Mid 51.0%; High 46.5%). Longitudinal follow-up (T1–T5) confirmed sustained MADRS reduction independent of dissociation. (Figure 1 Madrs reduction across timepoints, Figure 2 %madrs improvement by baseline DES, Figure 3 mediation model)

Image 1:

Image 1: Long description.

Image 2:

Image 2: Long description.

Image 3:

Image 3: Long description.

Conclusions

In this real-world cohort, dissociative traits and early changes in dissociation did not predict or mediate antidepressant response to intranasal esketamine. However, in clinical practice patients frequently report intense dissociative experiences, and psychotherapists describe esketamine as facilitating emotional processing of unelaborated traumatic memories (DeCou et al., Behav Ther 2019). This suggests that standard trait-based measures such as DES-II may introduce bias and fail to capture clinically meaningful dissociative phenomena. Future studies should employ session-level state measures (e.g., CADSS, ASC-5), cumulative dosing parameters, and longitudinal mixed-effects models to disentangle trait versus state contributions and clarify the therapeutic relevance of dissociation.

Disclosure of Interest

None Declared

More from our Archive