Directionality of the relationship between depressive symptom dimensions and c-reactive protein in patients with acute coronary syndromes

Objective: Previous theoretical models predict that elevated inflammation may predict later depressive symptoms, but bidirectional associations are possible. We examined whether depressive symptoms or inflammation predicts change in the other for a 3-month period in a sample of adults with acute coronary syndromes (ACS). Methods: During hospitalization for their index ACS event (baseline) and then again 1 and 3 months later, 163 post-ACS patients completed the Beck Depression Inventory, a measure of depressive symptom severity with cognitive-affective and somatic-affective subscales. C-reactive protein (CRP) was also assessed at each visit; known correlates of depression and CRP were assessed at baseline. Path analyses were conducted to evaluate prospective associations between depressive symptoms and log-transformed CRP values and whether strength and/or directionality varied by specific depressive symptom dimensions. Results: Baseline total depressive symptom severity predicted a smaller decrease in CRP from baseline to 1 month (unstandardized parameter estimates [B] = 0.04, p < .001) controlling for all covariates, as did baseline cognitive-affective depressive symptom severity (B = 0.10, p = .02). Baseline somatic-affective depressive symptom severity did not predict change in CRP (B = -0.002, p = .94). CRP did not predict 1- or 3-month change in total, cognitive-affective, or somatic-affective depressive symptom severity. The results did not differ for men and women. Conclusions: Greater cognitive-affective and total depressive symptom severity at the time of a cardiac event predicts a smaller decrease in CRP 1 month later, but there was no evidence in this study that CRP predicts change in depressive symptoms.