Relationship between depression and C-reactive protein in a screening population

Background: Both depression and C-reactive protein (CRP) are markers of increased risk for cardiovascular events. This study examined the relationship between CRP and depression in a cohort of participants undergoing a periodic physical to assess potential for interaction as either mediation or confounding of effect on cardiovascular risk. Methods: We conducted a cross-sectional study of a cohort of 696 consenting, active duty US Army personnel undergoing a periodic physical. We measured depression using the Patient Health Questionnaire-9, the depression module of the self-administered version of the Primary Care Evaluation of Mental Disorders (PRIME-MD). We used a highly sensitive assay to measure CRP. Results: The mean age in the cohort was 44 years (SD +/- 3; 82% male). The mean CRP level was 1.7 mg/l (range, 0.3-9.9; SD +/- 1.6 mg/1). Depression scores ranged from 0 to 26 with a mean of 2 (SD +/- 3). Depression scores correlated with prevalences of major depressive disorder and of any depressive disorder of 3.3% and 15%, respectively. Depression scores correlated positively with CRP levels (r = 0.085;p = .028), as did other variables known to be associated with CRP: body mass index (BMI; r = 0.36), insulin levels (r = 0.22), mean arterial pressure (r = 0.21), triglycericles (r = 0.18), exercise (r = -0.12), female sex (r = 0.097), current smoking status (r = 0.08), and high density lipoprotein (r = -0.09). After controlling only for BMI, the relationship between depression and CRP lost statistical significance among women (adjusted r = 0.08; p = .37), among men (adjusted r = -0.11; p = .8), and overall (adjusted r = 0.047; p = .219). Conclusion: Depressive symptoms are only weakly correlated with CRP. However, after adjusting for BMI, we found no significant relationship between CRP and depression. The relationship between depression and clinical coronary disease is unlikely to be explained through direct effects on CRP levels, but may be mediated by BMI.