Investigating associations between cortisol and cognitive reactivity to sad mood provocation and the prediction of relapse in remitted major depression

Abstract

Overview In remitted depressed patients, an increase in dysfunctional thoughts following a sad mood induction can predict relapse over 18 months. The current analysis examined whether salivary cortisol levels could also predict relapse in these same individuals. Method 99 subjects with major depression were first treated to full remission using either antidepressant medication or cognitive behavioural therapy. While in the remitted state, subjects were exposed to sad music to trigger dysfunctional thoughts. In a subset of 55 subjects, salivary cortisol levels taken before and after the mood challenge were also obtained. Results Unexpectedly, cortisol levels tended to decrease rather than increase following the mood challenge, suggesting that anticipation of the mood challenge was more stressful than the challenge itself. We thus used pre-challenge cortisol levels as the main grouping variable. Based on Kaplan–Meier survival curves, among subjects with low pre-challenge cortisol levels, those with a history of three or more prior episodes had significantly higher rates of relapse than did subjects with two or less prior episodes. In subjects with high pre-challenge cortisol levels, there was no significant difference in rates of relapse based on the number of prior episodes. Conclusion In depressed patients with few prior episodes, assessing risk of relapse and thus establishing the duration for treatment can be a difficult clinical problem. Pending replication, the current results suggest that high anticipatory cortisol levels may have utility in predicting relapse even in patients with few prior episodes.