Deep venous thrombosis (DVT) and pulmonary embolism (PE) are well recognized complications of cancer. However, our current knowledge of this association is derived from studies conducted more than a decade ago. In light of the changes in medical practice and the improvement in cancer care in recent years, a re-evaluation of the relationship between malignancy and venous thrombosis is in order. Of a total of 1041 patients with solid tumors admitted to 3 major medical centers, there were 81 (7.8%) diagnosed with DVT/PE. Patients were more likely to develop DVT/PE during chemotherapy (p = .0001). Advanced malignancies (p = .001), renal carcinoma (p = .005), pancreatic (p = .001), gastric (p = .014) and brain tumors (p = .001) were independent variables strongly associated with the occurrence of venous thrombosis. The occurrence of thrombotic events in the population tested in this study did not adversely affect survival (p = .082). The study identifies subset of patients with cancer at high risk for venous thrombosis. Early prophylaxis with anticoagulants in these patients may be warranted. Most importantly, further clinical trials are desperately awaited to detect possible new trends in the frequency and types of thrombotic events and to better define prevention strategies in cancer patients at risk for thrombosis. The association between venous thromboembolic disease (VTD) and malignant neoplastic disorders is well known and has been the subject of several reports for more than a century. There is a general agreement among investigators that thrombotic complications in patients with cancer occur at a rather high frequency, and that other circumstantial factors such as surgery or chemotherapy potentiates this risk. However, several important considerations pertaining to cancer and thrombosis continue to be shrouded in controversy. For example, there are inexplicable differences in the proportion of patients with cancer diagnosed with deep venous thrombosis (DVT) or pulmonary embolism (PE) reported in the literature. In the absence of large well-controlled studies, one may only postulate on the reasons that contributed to these differences. The inclusion of different types of VTD such as superficial, venous, arterial or vascular access device-induced thrombosis may have led to overestimation of the incidence of these events in patients with underlying malignancy. Another possible explanation for this discrepancy relates to the use of a variety of diagnostic and methodological criteria ranging from observation and clinical suspicion to more invasive procedures resulting in considerable differences in the rate of reported clotting events. Along the same line of discussion, one may argue whether VTD is a random event or constitutes a complication that occurs more commonly in patients with distinct characteristics and certain tumor types. To further investigate the association between malignancy and thrombosis, we evaluated 1041 patients with solid tumors for the risk of DVT/PE. The main objectives of the study were to determine the frequency of DVT/PE based on validated diagnostic criteria and to identify patients with cancer at high risk for developing these thrombotic episodes. Also, we evaluated the impact of VTD on the survival of these patients.