We are most thankful for the insightful comments by Nguyen et al. As mentioned, thrombomodulin levels are affected by inflammation and several clinical conditions1 and are confounders in our study.2 We have tried to adjust for inflammation by including high-sensitivity C-reactive protein (hsCRP) in our multivariate models. The inclusion of hsCRP did not affect the outcome of the study, and it is unlikely that the associations found for thrombomodulin would be explained by inflammatory processes alone. We had also adjusted for diabetes in one population of the study cohort without effect on the main results. There were no associations found between sTM and myocardial infarction or mortality. Furthermore, the findings of our study were not affected when we adjusted for the indication for warfarin treatment; patients with mechanical heart valves had the same risk for bleeding at a given thrombomodulin level as did patients with other indications for treatment.
Lind M, Boman K. A Question of the Clinical Relevance of Determining Thrombomodulin Level While Receiving Anticoagulant Therapy—Reply. Arch Intern Med. 2009;169(21):2031-2032. doi:10.1001/archinternmed.2009.423