To the Editor.
—We read with interest the article by Goldberg et al1 in the February 1987 issue of the Archives in which the association between occult malignant neoplasm and deep venous thrombosis (DVT) was reviewed, along with the accompanying editorial.2 We agree with Goldhaber et al2 in the desirability of a careful screening among patients with DVT. Our experience appears to confirm the need for careful screening.Since January 1985, we have studied 104 consecutive patients who were admitted because of a deep-vein thrombosis of the lower limbs. The diagnosis of DVT was always established by bilateral phlebography. In 37 patients, the DVT developed shortly after a surgical procedure or a bone fracture, and, in 20 patients, after bed rest for more than eight days. Among the remaining 49 patients, ten had previously diagnosed cancer, six had a history of pulmonary embolism or DVT, six had