Pain mediates the improvement of social functions of repeated intravenous ketamine in patients with unipolar and bipolar depression

Depressive disorders, including major depressive disorder (MDD) and bipolar depression, cause marked impairments in psychosocial functioning and are a major contributor to global disease burden. A substantial proportion of patients are treatment resistant and conventional antidepressants can be slow to act; chronic pain is common in depressive disorders and is associated with worse psychosocial outcomes. Over the past two decades, subanaesthetic doses of the NMDA receptor antagonist ketamine have shown rapid antidepressant effects and preliminary evidence suggests ketamine may also relieve pain and improve functional outcomes, but the pathways linking symptom change to functional recovery are unclear. Wu and colleagues set out to test whether improvements in psychosocial functioning following a course of repeated intravenous ketamine are mediated by changes in depressive symptoms and/or by reductions in pain. Specifically, the authors hypothesised that total depressive symptom severity (MADRS) and the affective component of pain would act as important mediators of improvements in both subjective (SDS) and objective (GAF) measures of social functioning after six ketamine infusions.

This was a single-centre, open-label clinical trial of adults with DSM-5 diagnoses of MDD or bipolar depression who had failed at least two antidepressant trials or were experiencing suicidal ideation. The analytic sample comprised 103 patients who completed baseline and day 13 assessments; 19 patients (18.4%) did not complete the day 26 follow-up. Key exclusions included alcohol or drug dependence, recent mania/hypomania (past 6 months for bipolar disorder), and unstable medical or laboratory findings. All participants received six intravenous infusions of ketamine hydrochloride at 0.5 mg/kg, administered via an infusion pump in 0.9% sodium chloride over 40 minutes, on a Monday–Wednesday–Friday schedule across two weeks. Concomitant psychotropic medications (for example benzodiazepines, mood stabilisers or antipsychotics) were permitted but were required to remain at a stable dose during the two-week treatment window. Outcome assessments took place at baseline, the day after the sixth infusion (day 13), and 14 days later (day 26). Depression severity was measured with the Montgomery–Åsberg Depression Rating Scale (MADRS). Subjective psychosocial functioning was assessed by the Self-Rating Depression Scale (SDS) total score and its three domains (work/school, social life, home/family responsibilities). Objective functioning was measured by the Global Assessment of Functioning (GAF). Pain was evaluated with the Short Form McGill Pain Questionnaire (SF-MPQ), which yields sensory and affective indices and a present pain intensity (PPI) score. For statistical analysis the investigators used SPSS (v22) and conducted two-sided tests with significance set at P < 0.05. Longitudinal changes in MADRS, SDS and GAF were analysed using linear mixed models with group (pain versus no pain at baseline) and time (baseline, day 13, day 26) as factors; baseline differences were entered as covariates where appropriate and Bonferroni corrections were used for pairwise contrasts. Mediation analysis was performed using an SPSS mediation program (v2.15) with ketamine treatment as the predictor, SDS and GAF as dependent variables, the three pain indices as primary mediators (M1) and MADRS total score as a secondary mediator (M2). Covariates in the mediation models included age, sex, education, illness duration, BMI, initial diagnosis and concomitant medication status. Unstandardised coefficients (Coef) and P values were reported.

Of the 103 patients analysed, 48 (46.7%) reported pain at baseline. Nineteen patients (18.4%) lacked the day 26 follow-up. Across the whole sample there were statistically significant reductions in pain measures after six ketamine infusions. Mean sensory index scores fell from 2.18 ± 3.17 at baseline to 0.35 ± 0.85 at day 13 and 0.30 ± 1.03 at day 26 (F = 27.607, P < 0.001). The affective index declined from 2.75 ± 3.53 to 0.53 ± 1.50 at day 13 and 0.29 ± 0.87 at day 26 (F = 33.482, P < 0.001). PPI decreased from 1.29 ± 1.60 at baseline to 0.33 ± 0.80 at day 13 and 0.29 ± 0.87 at day 26 (F = 23.514, P < 0.001). Depressive symptoms (MADRS) showed a significant main effect of time (F = 158.936, P < 0.01), with lower scores at day 13 and day 26 compared with baseline regardless of baseline pain status; there were no significant main effects of pain group or group-by-time interaction on MADRS. Subjective functioning (SDS total and each SDS domain) similarly exhibited significant time effects (SDS-T: F = 70.195, SDS-1: F = 58.922, SDS-2: F = 52.983, SDS-3: F = 52.333; all P < 0.001), with improvements at day 13 and day 26 versus baseline and no significant group differences or interactions. Objective functioning (GAF) showed significant main effects of group (F = 11.469, P = 0.001) and time (F = 124.089, P < 0.001); patients without baseline pain had higher GAF scores at baseline (t = 2.929, P < 0.004), but by day 13 and day 26 the GAF scores no longer differed significantly between the pain and no pain groups. Mediation analyses indicated a significant total effect of ketamine on SDS scores (unstandardised Coef = -5.171; the extracted text does not clearly report the confidence interval). Both direct and indirect effects of ketamine on subjective psychosocial functioning were reported as statistically significant. The MADRS total score and the affective index of the SF-MPQ emerged as important mediators of the relationship between ketamine treatment and reductions in SDS total scores; the sensory index and PPI did not show consistent mediating effects. A comparable pattern was observed for objective functioning (GAF): improvements were mediated by reductions in depressive symptom severity (MADRS) and by changes in the affective pain index, while the sensory index and PPI were not significant mediators. Exact confidence intervals and some coefficient estimates for the mediation pathways were not clearly reported in the extracted Results text.

Wu and colleagues interpret their findings as demonstrating that six subanaesthetic ketamine infusions were associated with rapid improvements in depressive symptoms, reductions in pain, and gains in both subjective and objective psychosocial functioning among patients with MDD or bipolar depression. The authors highlight that mediation modelling suggested psychosocial recovery was partially explained by reductions in overall depressive severity and, specifically, reductions in the affective component of pain rather than sensory pain or present pain intensity. The investigators situate their results within prior work showing ketamine’s antidepressant and analgesic properties and note concordance with earlier reports that pretreatment pain relates to antidepressant response. They discuss potential neurobiological mechanisms invoked in the paper, including recruitment of cortical and limbic circuits such as the anterior cingulate cortex (ACC), alterations in molecular markers (for example Mkp-1), NMDA receptor blockade on inhibitory interneurons leading to increased glutamatergic transmission, downstream modulation involving KCNQ2, plus possible involvement of endogenous opioid pathways and anti-inflammatory effects; these mechanisms are presented as plausible links between ketamine’s analgesic and antidepressant actions and consequent functional improvements. The authors acknowledge important limitations: concomitant psychotropic medications were continued and could confound outcomes, the trial was open-label without placebo control or blinding, some participants missed the day 26 assessment which may introduce bias, and social and mental functioning were not assessed continuously beyond the scheduled visits. They therefore caution against generalising the findings to drug‑naïve patients and note the need for controlled studies to confirm mediation pathways and mechanisms. Finally, the paper suggests future research should examine ketamine’s effects on psychosocial functioning in patients with pronounced affective pain and further probe underlying biological mediators.

After six subanaesthetic ketamine infusions, patients with unipolar or bipolar depression showed rapid improvements in depressive symptoms, reductions in pain measures and better subjective and objective psychosocial functioning. The authors conclude that improvements in social functioning were partially mediated by reductions in depressive severity and, notably, by reductions in the affective dimension of pain. They propose that these effects may relate to anti-inflammatory actions, NMDA receptor blockade on GABAergic interneurons with resultant glutamatergic effects, and possibly indirect opioid system activation, and they recommend further controlled studies to examine psychosocial outcomes in patients with prominent affective pain.