The purpose of this study was to determine the effect of anticoagulation on the incidence of thrombotic propagation and pulmonary embolism in patients with calf vein thrombosis after total hip or total knee arthroplasty. Patients undergoing arthroplasties had prospective surveillance for postoperative deep vein thrombosis by both bilateral contrast venography and venous duplex scanning. Calf vein thrombosis was documented by venography in 42 patients (50 limbs), including 29 of 253 patients undergoing total hip arthroplasty (11.4%) and 13 of 99 patients undergoing total knee arthroplasty (13%). Of patients on whom follow-up duplex scans were performed, heparin followed by warfarin anticoagulation was used in 11 (13 limbs) and withheld in 21 (25 limbs). Propagation of thrombosis to the popliteal or superficial femoral vein or both was detected by serial duplex scanning in 3 of 13 treated limbs (23%) and 2 of 25 untreated limbs (8%), (p = 0.43). All thrombus propagations were detected within 2 weeks of the operative procedure. There were no pulmonary emboli or deaths. Propagation of asymptomatic calf vein thrombosis after arthroplasty was not influenced by anticoagulation, suggesting that postoperative calf vein thrombosis need not be routinely treated. Serial venous duplex scanning is useful to identify the occasional patient in whom thrombotic propagation requiring anticoagulation develops.