Malignant disease is recognized as a risk factor for venous thromboembolism. A number of recent reports have suggested that cancer chemotherapy may contribute to this risk, but it was not possible to separate the role of chemotherapy from the effects of the malignant disease. We are conducting a randomized trial to determine the optimal duration of adjuvant chemotherapy in women with Stage II breast carcinoma. These ambulatory patients, with negligible tumour burden, receive either 12 weeks of chemo-hormonal therapy (cyclophosphamide, methotrexate, 5 fluorouracil, vincristine, prednisone, adriamycin and tamoxifen) or 36 weeks of chemotherapy (cyclophosphamide, methotrexate, 5 fluorouracil, vincristine and prednisone). This study has provided us with an opportunity to evaluate the thrombogenic effects of chemotherapy since patients in the 12 week group, while off chemotherapy, can be compared directly to the patients in the other group who are still on chemotherapy. This allows the confounding influence of the malignant process to be circumvented. All patients undergo screening tests for thrombosis (impedance plethysmography and Doppler ultrasound) and routine clinical assessments. Suspected venous thrombosis is confirmed by venography and suspected pulmonary embolism by either pulmonary angiography or high probability ventilation perfusion scanning. There have been 11 episodes of venous thromboembolism to date among 191 patients of whom 164 have completed the first 36 weeks of study. There were 3 episodes in each group during the first 12 weeks. During the subsequent 24 weeks there have been no events in the group whose treatment was stopped and 5 events in the group still on treatment (p 0.03). These findings demonstrate that chemotherapy per se is an important risk factor for venous thromboembolism in patients with malignant disease.
