Esketamine for treatment-resistant depression: comparing two maintenance protocols in an observational real-world study

Treatment-resistant depression (TRD) is associated with a substantial clinical burden and functional impairment. Intranasal esketamine has been shown to be an effective treatment for TRD in clinical trials and real-world studies. However, evidence on optimal maintenance scheduling is limited. We therefore sought to investigate whether two maintenance schedules used in routine clinical care, differing in their inter-dose intervals, were associated with different depressive symptom trajectories.

In this retrospective single-site observational study, we analysed 65 patients with TRD undergoing two different intranasal esketamine treatments as part of routine outpatient care at the Psychiatric University Hospital Zurich between January 2021 and April 2025. Treatment consisted of an induction phase of eight esketamine sessions over four weeks, followed by one of two maintenance schedules, planned according to clinical availability. A total of 58 patients entered the maintenance phase. Both maintenance schedules were comprised of four sessions. Maintenance schedule 1 (n = 39) consisted of weekly sessions that ended after a total treatment period of two months. These patients were assessed again after an additional two months without treatment had passed. Schedule 2 (n = 19) comprised progressively increasing inter-dose intervals with a total treatment period of four months. Depressive symptoms were assessed using the clinician-rated Montgomery–Åsberg Depression Rating Scale (MADRS) and the self-rated Beck Depression Inventory-II (BDI). Response was defined as ≥50% reduction from baseline, and remission as MADRS ≤12 or BDI ≤13.

Depressive symptoms decreased significantly during the induction phase on both the MADRS (mean difference = 7.74, 95% CI [5.78, 9.70], p < .001) and BDI (mean difference = 7.37, 95% CI [4.81, 9.93], p < .001), with a large effect size for the MADRS (d = 1.03) and a medium effect size for the BDI (d = .72). At the end of the acute treatment phase, 21.1% of patients met response criteria on both instruments, while remission rates were 22.8% on the MADRS and 21.1% on the BDI. During the maintenance phase, contrasts of linear mixed-effects model estimates showed no significant differences between the two maintenance schedules on either outcome, and ANCOVA models comparing schedules four months after treatment initiation confirmed these findings. By the end of maintenance, response rates were 35.3% on the MADRS and 26.0% on the BDI, while remission rates were 31.4% and 38.0%, respectively.

In this real-world TRD cohort, intranasal esketamine was associated with statistically significant and clinically meaningful reductions in depressive symptoms during the induction phase, with some additional benefit observed during maintenance. This reduction was largely consistent across instruments, with slightly smaller improvements on the BDI. The two maintenance schedules were overall comparable, suggesting that spaced lower-frequency maintenance dosing may achieve comparable outcomes to those of weekly treatments, although prospective randomised studies are needed to confirm these findings.

References

No references