Vagally mediated heart rate variability (HRV) is a psychophysiological indicator of mental and physical health. Limited research suggests there is reduced vagal activity and resulting lower HRV in patients with Major Depressive Disorder (MDD); however little is actually known about the association between HRV and symptoms of depression and whether the association mirrors symptom improvement following psychotherapy. The aim of this study was to investigate the association between antidepressant therapy, symptom change and HRV in 50 inpatients (68% females; 17–68 years) with a diagnosis of MDD. Severity of depressive symptoms was assessed by self-report (Beck Depression Inventory II) and the Hamilton Rating Scale of Depression. Measures of vagally mediated HRV (root mean square of successive differences and high-frequency) were assessed at multiple measurement points before and after inpatient psychotherapeutic and psychiatric treatment. Results showed an expected negative correlation between HRV and depressive symptoms at intake. Depressive symptoms improved (d = 0.84) without corresponding change in HRV, demonstrating a de-coupling between this psychophysiological indicator and symptom severity. To our knowledge, this study is the first to examine an association between HRV and depressive symptoms before and after psychotherapy. The observed de-coupling of depression and HRV, and its methodological implications for future research are discussed.

The cortisol awakening response (CAR) is a non-invasive biomarker for hypothalamic-pituitary-adrenal axis (HPA) dysregulation, reflecting accumulated stress over time. In a previous study we reported that a blunted CAR before an inpatient treatment predicted self-reported depressive symptoms six weeks and six months after discharge [Eikeseth, F. F., Denninghaus, S., Cropley, M., Witthöft, M., Pawelzik, M., & Sütterlin, S. (2019). The cortisol awakening response at admission to hospital predicts depression severity after discharge in major depressive disorder (MDD) patients. Journal of Psychiatric Research, 111, 44-50)]. This replication study adopted an improved overall methodology with more stringent assessment protocols and monitoring. The longitudinal design included 122 inpatients from a psychosomatic hospital with a diagnosis of MDD displaying symptoms of moderate to severe major depression (n = 80 females). The CAR was measured at intake. Depression severity was assessed as Beck Depression Inventory II scores at intake, discharge, 6 weeks and 6 months following discharge. Results from the original study were replicated in terms of effect size but did not reach statistical significance (correlation between BDI-II 6 months after discharge and AUCg: r = −0.213; p = 0.054). The replication study yielded nearly identical correlation coefficients as in the original study (BDI-II 6 months and CAR, r = −0.223, p < 0.05). The replication of previously reported effect sizes with a concurrent lack of statistical significance in the more restrictive, larger and better controlled replication study may well inform research on psycho-endocrinological predictors for treatment success, but suggests a rather limited practical relevance for cortisol awakening response measures in this clinical context.