The hypothesis was tested that cancer, by virtue of it being a thrombotic diathesis, may enhance ischemic heart disease. Because cancer therapy may precipitate thromboembolism, the authors focused on patients with occult neoplasia before therapy. Occult cancer was defined as a time period of 2 years before a cancer diagnosis was established. Data obtained from the files of patients with the diagnosis of malignant tumors and admitted to a general hospital during a 3-year period were reviewed for coronary risk factors, coronary events, and characteristics of cancer. Indices of coronary instability were studied: the incidence of first coronary events, the incidence of all coronary events, and the coronary events burden. These indices were calculated for 366 patients with cancer (from the Department of Medicine files of 166 consecutive patients with cancer of several primary sites and from the surgical ward files of 100 consecutive patients with colorectal cancer and 100 consecutive patients with cancer of the prostate or bladder) and for 100 patients with benign prostatic hypertrophy. The patients with benign prostatic hypertrophy served as controls. A steep and statistically significant increase in coronary instability indices was observed in all groups of patients with cancer in the 2-year period before cancer diagnosis in comparison with the coronary instability indices of control patients (P less than or equal to 0.05 to 0.0001). Patients with colorectal cancer presented the highest indices in the 2-year period before cancer diagnosis, with unstable ischemic heart disease being reported in 18% and first coronary events in 10%. The coronary events burden was 0.92. The lowest indices among patients with cancer were recorded in those with prostatic and bladder cancer. Unstable ischemic heart disease occurred in 6% and first coronary events in 4%. The coronary events burden was 0.36. The indices were several times lower for control patients than for patients with cancer. Unstable ischemic heart disease occurred in 3% of control patients, and first coronary events in 2%. The coronary events burden was 0.15. Other possible etiologic factors, particularly the known coronary risk factors and anemia, were not statistically related with an increased risk of coronary events in the 2-year period before cancer diagnosis. Based on these epidemiologic data, it appears that there may be an association between occult cancer and coronary events.