C-reactive protein and risk of venous thromboembolism in the general population.

OBJECTIVE To examine the robustness of the association between C-reactive protein (CRP) levels and increased risk of venous thromboembolism (VTE) and to examine whether genetically elevated CRP levels cause VTE. METHODS AND RESULTS In the prospective Copenhagen City Heart Study, we observed 10 388 participants for longer than 16 years, of whom 484 developed a VTE. In the cross-sectional Copenhagen General Population Study, we studied 36 616 participants, of whom 903 previously had a VTE. Levels of CRP greater than 3 mg/L versus less than 1 mg/L were associated with a 2.3- and 2.4-fold increased risk of VTE in the Copenhagen City Heart Study and the Copenhagen General Population Study, respectively. CRP levels in tertiles, quintiles, and octiles associated with a stepwise increase in VTE risk. CRP genotypes associated with an increase in plasma CRP levels up to 59% but did not associate consistently with risk of VTE in either study. The causal odds ratio for VTE for a doubling in genetically elevated CRP on instrumental variable analyses was lower than the odds ratio for VTE observed for a doubling in plasma CRP on logistic regression ([odds ratio and 95% CI] 0.80 [0.56 to 1.12] versus 1.17 [1.08 to 1.27]; P=0.04). CONCLUSIONS We observed 47,000 participants from the general population, of whom 1387 developed VTE. Although elevated CRP levels robustly associated with increased risk of VTE, this may not necessarily be a causal association because genetically elevated CRP did not associate with VTE risk.