[Skip to Navigation]
February 1987

Cancer and Venous Thromboembolism

Author Affiliations

Department of Medicine; Department of Preventive Medicine and Clinical Epidemiology; Departments of Medicine and Preventive Medicine and Clinical Epidemiology Brigham and Women's Hospital Harvard Medical School Boston, MA 02115

Arch Intern Med. 1987;147(2):216. doi:10.1001/archinte.1987.00370020036030

Previous case reports, autopsy series, and another cohort study1 have demonstrated a clinically important and statistically significant association between occult cancer and pulmonary embolism. The study by Goldberg et al2 extends this finding to a far more common clinical condition than pulmonary embolism, namely, deep venous thrombosis (DVT). In this study, 370 patients with DVT (diagnosed by impedance plethysmography [IPG]) who did not have overt cancer were compared with 1073 patients suspected of DVT but who had normal IPGs (and no overt cancer). During five years of follow-up, of those with DVT, 6.3% developed cancer, compared with 2.4% of those who had negative IPGs. Thus, in the population that was investigated, the relative risk for developing cancer in the presence of DVT was 2.7, with 95% confidence limits from 1.5 to 4.7. The most frequent sites of cancer were lung and colon, and the majority developed within two