Longitudinal hypothalamic-pituitary-adrenal axis trait and state effects in recurrent depression

Background: Hypothalamic-pituitary-adrenal (HPA)-axis hyperactivity has been observed in (recurrent) major depressive disorder (MDD), although inconsistently and mainly cross-sectional. Longitudinal studies clarifying state-trait issues are lacking. We aimed to determine whether HPA-axis (hyper)activity in recurrent MDD is: (I) reflecting a persistent trait; (II) influenced by depressive state; (III) associated with stress or previous episodes; (IV) associated with recurrence; and (V) influenced by cognitive therapy. Methods: We included 187 remitted highly recurrent MDD-patients (mean number of previous episodes: 6.3), participating in a randomized-controlled-trial investigating the preventive effect of additional cognitive therapy on recurrence. In an add-on two-staged patient-control and prospective-cohort design, we first cross-sectionally compared patients' salivary morning and evening cortisol concentrations with 72 age- and sex-matched controls, and subsequently longitudinally followed-up the patients with repeated measures after three months and two years. Results: Patients had higher cortisol concentrations than controls (p < .001), which did not change by MDD-episodes during follow-up. HPA-axis activity had no relation with daily hassles or childhood life events. Cortisol concentrations were lower in patients with more previous episodes (p = .047), but not associated with recurrence(s) during follow-up. Finally, randomly assigned cognitive therapy at study-entry enhanced cortisol declines over the day throughout the two-year follow-up (p = .052). Conclusions: Our results indicate that remitted recurrent MDD-patients have a persistent trait of increased cortisol concentrations, irrespective of stress. In combination with our finding that patients' cortisol concentrations do not change during new MDD-episodes (and thus not represent epiphenomenal or state-effects), our results support that hypercortisolemia fulfills the state-independence criterion for an endophenotype for recurrent depression. (c) 2011 Elsevier Ltd. All rights reserved.