Résumé
Objective: To estimate the association between high-sensitivity C-reactive protein (hs-CRP) and cardiovascular events as well as all-cause mortality events. Methods: During 2009- 2010, out of the 11 623 individuals, 1 000 participants aged 35-64 years, were recruited and divided into 12 age-groups, to have received a study on CVD risk factors. Information on the risk factors of cardiovascular diseases was also collected. Fasting blood sample was gathered for all the participants, with hs-CRP tested. Participants in 7 out of the 12 sites were followed, with 6.21 years (36 075 person-years) as the median follow-up period. Cardiovascular and all-cause mortality events were collected. A total of 6 177 participants had been followed after excluding participants who had baseline infections, or did not take hs-CRP test/physical examination at the baseline. Finally, 5 984 participants were included for analysis. Participants were categorized into three groups based on the hs-CRP (mg/L) values: <1, 1-3 and >3, respectively. Cox proportional hazards regression model was used to analyze the relationships between hs-CRP with cardiovascular events or all-cause mortality events, after adjusting for confounding factors. Results: Mean age of the participants was 50.2 years. The incidence rates of cardiovascular disease events were 3.6/1 000 person-years, 7.1/1 000 person-years,and 10.4/1 000 person-years among three groups and 3.0/1 000 person-years, 5.7/1 000 person-years, 9.1/1 000 person-years for all-cause mortality events, respectively. After adjusting for confounding factors, the hazard risks (HR) for cardiovascular events were 1.33 (95%CI: 0.95-1.84) in the hs-CRP 1-3 mg/L group and 1.76 (95%CI: 1.20-2.60) in the hs-CRP>3 mg/L group when comparing with the hs-CRP<1 mg/L group (trend test P=0.003). The HRs for all-cause mortality events were 1.76 (95%CI: 1.23-2.54) and 2.64 (95%CI: 1.74-4.01) (trend test P<0.001), respectively. Conclusion: Hs-CRP appeared an independent predictor for cardiovascular events and all-cause mortality events.