Body mass index, but not physical activity, is associated with C-reactive protein

Purpose: To longitudinally examine the effects of BMI and both current and previous-year physical activity on hs-CRP in healthy men and women (N = 109). Methods: BMI and hs-CRP were measured five times (baseline and quarterly) over 1 yr. Current physical activity was assessed 12-15 times during the study via 24-h recall. Previous-year physical activity was assessed using the Baecke questionnaire at baseline. Results: Mean BMI and hs-CRP were unchanged over the course of the study, but current physical activity increased on visit 3. Average hs-CRP was not related to average current physical activity or to natural changes in current physical activity across the five visits. Additionally, current physical activity on any given visit was not associated with hs-CRP on the following visit. When current physical activity, BMI, age, gender, and smoking were included in the statistical model, only BMI was significantly related to hs-CRP (P < 0.001). Average hs-CRP was significantly correlated with average BMI (r = 0.50; P < 0.001) but was not related to previous-year (Baecke) physical activity levels (r = 0.02; P = 0.89). When subjects were grouped by BMI (< 25 kg.m(-2), 25-29.9 kg.m(-2), > 30 kg.m(-2)) hs-CRP was significantly greater in obese (3.2 +/- 1.9 mg.L-1) and overweight (2.1 +/- 1.7 mg.L-1 1) than normal weight (1.1 +/- 1.0 mg.L-1) subjects (ANOVA P < 0.05). Current physical activity was similar between the three BMI groups at all times, and was unrelated to hs-CRP in all groups, throughout the study period. Conclusion: These data indicate that BMI, but not previous-year or current physical activity, predicts hs-CRP.