Résumé : Background: Some markers of chronic inflammation have been recognized as predictors of cardiovascular risk in apparently healthy subjects and in patients with coronary heart disease (CHD). High sensitivity C-reactive protein (CRP) appears to be the most useful marker in clinical settings. Several studies reported associations between inflammatory markers and other cardiovascular risk factors, such as age, obesity, cholesterol levels, the presence of diabetes mellitus, physical activity, social level and smoking habits. We focussed on the association between C-reactive protein, serum amyloid A (SAA), fibrinogen and leisure time physical activity (LTPA). Methods: This report deals with the results observed in a sub-sample of the BELSTRESS study. 892 male subjects, free from clinical CHD and major ECG abnormalities, working in the same environment, aged 35-59 years, were selected. A questionnaire was used to estimate the level of leisure time physical activity. Associations between CRP, SAA, fibrinogen and leisure time physical activity were evaluated through univariate and multivariate methods. Subjects taking statins or other lipid lowering medication were excluded from the study. Results: Regular leisure time physical activity is associated with reductions of several cardiovascular risk factors, such as body mass index (BMI), waist hip ratio and the lipid profile. Smokers and low educated subjects had a lower physical activity status. Age adjustment did not alter the means of inflammatory parameters according to the levels of leisure time physical activity. After correction for personal characteristics (BMI, current smoking status, educational level, presence of diabetes and alcohol consumption) no significant relation was found between leisure time physical activity and levels of inflammatory markers. The differences of CRP and fibrinogen according to the level of physical activity, found in bivariate analysis, seem to be explained by linked differences in BMI, or related to current smoking habits. Leisure time physical activity, as reported in this study, is not significantly related to C-reactive protein, serum amyloid A or fibrinogen levels, after correction for other cardiovascular risk factors. Conclusion: These data indicate that leisure time physical activity, as reported in our study, is not an independent predictor of C-reactive protein, serum amyloid A or fibrinogen levels. Possible interactions of physical activity and other cardiovascular risk factors might explain the (indirect) relation we found in the bivariate analysis. © 2004 Elsevier Ireland Ltd. All rights reserved.