Thromboembolic (TE) events preceding cancer have been observed. Some studies failed to find this correlation. We retrospectively examined the cancer incidence following thromboembolic events in patients at our medical center. Medical records of 183 patients with established thromboembolic events documented in their records were selected and reviewed. Time interval between primary, secondary, and recurrent TE events preceding cancer diagnosis was analyzed. Two hundred age-and sex-matched controls seen during the same period and without any evidence of TE were randomly selected and charts reviewed for malignancy. Cancer occurred after TE in 48 of 183 patients (26.2%). In controls, cancer was diagnosed in 23 (11.5%). This was statistically significant with an odds ratio of 2.736 (1.586, 4.720). In the 64 primary TE patients, the cancer incidence was 37.5%. The 63 patients with recurrent TE had an incidence of 35.4%, and 56 secondary TE patients had an incidence of 27.1%. Time between initial TE and cancer diagnosis was <6 months in 27 (56.3%) patients, between 6 months and 1 year in 12 (25.0%), 1-5 years in 5 (10.4%), and >5 years in 4 (8.3%). Fourteen (31.1%) TE patients presented with metastatic cancer. This study indicates that thromboembolic events are important predictors of cancer. Cancer in this population occurs within a year in the majority of patients. Cancer screening in patients without identifiable risk factors for thrombosis could be helpful for early detection, diagnosis, and management of cancer. Am. J. Hematol. 72: 109-114, 2003.