Undiagnosed malignancy in patients with deep vein thrombosis -: Incidence, risk indicators, and diagnosis

BACKGROUND. The reported incidence of a subsequent diagnosis of malignancy in patients presenting with deep vein thrombosis (DVT) varies from 2-25%. Risk indicators and diagnostic procedures to be performed in these patients are controversial. METHODS. Four hundred consecutive patients with confirmed DVT included in a randomized clinical trial were followed prospectively for 6 months. The incidence of a subsequent diagnosis of malignancy was calculated and compared between patients with unexplained DVT and patients with secondary DVT. Potential risk indicators for subsequent malignant disease were evaluated. RESULTS. Of the 400 patients, 70 already had been diagnosed with malignancy; another four patients were lost to follow-up. Of the remaining 326 patients, 10 new malignancies were diagnosed among 137 patients with unexplained DVT (7.3%) and 3 new malignancies were diagnosed in 189 patients with secondary DVT (1.6%). The relative risk was 4.6 (95% confidence interval, 1.3-16; P = 0.009). Age, gender, or location of the DVT had no significant effect on the incidence of diagnosis when adjusted for unexplained DVT. Ten of these 13 patients (77%) had abnormal clinical findings suggestive of malignancy at the time of presentation with DVT. CONCLUSIONS. Unexplained DVT is a significant risk indicator of underlying malignancy. The majority of patients with undiagnosed malignancy have some clinical abnormality suggestive of underlying malignancy at the time of presentation with unexplained DVT. A simple clinical evaluation comprised of medical history, physical examination, routine laboratory tests, and chest X-ray can detect such patients. Extensive screening of all patients presenting with unexplained DVT does not appear to be justified. (C) 1998 American Cancer Society.