Elevation of serum cortisol is found in many patients with major depressive disorder (MDD) and may be due to a chronic dysfunction in the feedback regulation in the Hypothalamic-Pituitary-Adrenal axis. Saliva cortisol is a valid indicator of serum cortisol. The predictive value of saliva cortisol for remission of depressive symptomatology was investigated.

Saliva cortisol was measured in a sub-sample (N=19) with unipolar MDD according to DSM-IV. Mean score on the Montgomery Aasberg Depression Rating Scale (MADRS) was 26.8 (standard deviation 3.7, range 22-32). At follow-up, two years later, mean MADRS was 13.6 (SD 10.7, range 0-37). In a linear regression model, saliva cortisol at baseline was entered as independent variable and MADRS-score at follow-up as dependent variable.

A significant correlation between the level of saliva cortisol at baseline and MADRS-score at follow-up was found (R=0.33, P=0.036). After adjustment for MADRS at baseline, the level of saliva cortisol explained 21% of the variance in MADRS at follow-up (P=0.018). After further adjustment for age, gender, and use of antidepressant medication, the model still produced significant results (R2=0.50, P=0.026).

Higher level of saliva cortisol is predictive of less improvement in depressive symptomatology over time in unipolar MDD. This finding is in line with a model in which higher secretion of cortisol is associated with a more chronic course in depression. It underlines the importance of biological correlates as predictors of outcome in psychiatric disorders.