C-reactive protein and risk of venous thromboembolism - results from the MEGA study

The association between C-reactive protein (CRP) and venous thromboembolism (VTE) is currently not well explained.

To determine whether the association between CRP and VTE can be explained by concurrent presence of a procoagulant state, chronic diseases or obesity.

Using data from the Multiple Environmental and Genetic Assessment of multiple risk factors for venous thrombosis (MEGA) case-control study, data on high sensitivity CRP (hsCRP), factor VIII activity (FVIII) concentration, chronic diseases (i.e. rheumatic disease, cancer, history of arterial cardiovascular disease and chronic kidney diseases) and body mass index (BMI) were available in 2205 cases with VTE and 2777 controls. Logistic regression analysis was performed to estimate risk of VTE by categories of hsCRP (<25th, 25-50th, 50-75th, 75-97.5th and > 97.5th percentiles). The analyses were adjusted for confounding and done separately for FVIII levels, BMI and chronic diseases. We investigated whether chronic diseases could be related to concurrent high levels of hsCRP, BMI and FVIII levels by calculating the mean levels of these factors in individuals with and without the presence of chronic diseases.

HsCRP levels were associated with BMI, FVIII levels and presence of chronic diseases. The risk of VTE was higher across hsCRP categories in a dose response manner, with individuals in the highest percentile (>13.13 mg/L) having a 3.3-fold higher risk (OR: 3.3, 95% CI: 2.48-4.5) than in the lowest percentile (<0.68 mg/L). This association was strongest for unprovoked VTE (OR: 5.3, 95% CI: 3.3-8.5). Upon adjustment for BMI, FVIII and chronic diseases, the association with the overall VTE attenuated (OR: 1.6, 95% CI: 1.1-2.2).

Our findings suggest that the association between CRP and VTE risk is at least partly explained by FVIII, BMI and the presence of comorbidities. These results suggests that elevated CRP should not be considered an independent risk factor for VTE but rather as a marker of underlying inflammatory and prothrombotic states.