There is increasing experimental evidence that endothelial dysfunction represents an important component of atherosclerotic cardiovascular disease (CVD).Thrombomodulin (TM) is an endothelial cell surface receptor for thrombin and exerts anti-coagulatory effects through thrombin-mediated activation of protein C which, in the presence of protein S, degrades factor (F)Va and FVIIIa, thereby restraining the coagulation reactions and restricting fibrin formation [1]. In spite of being mostly located on endothelial cells of arteries, there is, however, a small amount of circulating, soluble TM (sTM). The physiological role of sTM is not fully known, but it is thought to indicate endothelial damage [2]. Data from epidemiological studies, aiming to evaluate the potential association between circulating levels of sTM and incident CHD, are sparse and controversial [3]. Therefore, we sought to further elucidate whether sTM is a predictor of CHD in a large prospective population-based cohort study of middle-aged men and women.We conducted a case-cohort study using data from the population-based MONICA/KORA Augsburg cohort study. The study design has been described previously [4,5]. Briefly, three independent surveys, with a total number of 13 427 participants aged 25-74 years, were conducted. A combined endpoint that included incident fatal/non-fatal myocardial infarction (MI) and sudden cardiac death occurring before the age of 75 years was used as the outcome variable. Until 2000, the diagnosis of a major non-fatal MI event was based on the MONICA algorithm taking into account symptoms, cardiac enzymes and electrocardiograph (ECG) changes. Since 2001 MI was diagnosed according to European Society of Cardiology (ESC) and American College of Cardiology (ACC) criteria. Deaths from MI were validated by autopsy reports, death certificates, chart review and information from the last treating physician. As a result of the low incidence of CHD under 35 years, the present study was limited to 10 718 persons (5382 men and 5336 women) between 35-74 years at baseline. After exclusion of participants with self-reported prevalent CHD, the source population for the present study comprised 9300 subjects. For the case-cohort study, a random sample of the source population was selected stratifying by sex and survey. Participants with missing values for sTM or any of the covariables used in the present analysis were excluded. The present analysis comprised 1378 participants (197 men and 49 women with CHD, and 602 men and 530 women without CHD). The follow-up time (mean ± SD) was 7.2 ± 4.0 years.Levels of sTM were measured in plasma using ELISA (Diaclone, Besanc¸on, France). The intra-and inter-assay CVs were 10.3% and 16.8%, respectively, which may have attenuated the association between sTM concentration and CHD slightly.All statistical analyses have been described in detail elsewhere [4]. Cox proportional hazards analysis was used to assess the association between sTM and incident CHD using different adjustments for classic cardiovascular...