The letters by Sutton [1] and Banga and colleagues [2] highlight the controversy surrounding the initiation of anticoagulant therapy in patients with thromboembolism. Although some trials have compared early with late initiation of warfarin therapy [3], we could not find any information on whether cessation of heparin therapy causes a significant decrease in the international normalized ratio (INR).

During a 2-week period, all patients in a large teaching hospital who were receiving both intravenous heparin and warfarin were recruited for a prospective study. These patients were reviewed daily until 2 days after cessation of heparin therapy. We collected the following information on the patients: age, sex, indication for anticoagulation therapy, activated partial thromboplastin time, INR, warfarin dosage, and date of cessation of heparin therapy. A decrease in the INR of 10% or more was considered clinically significant.

We studied 42 patients. The indication for anticoagulant therapy was pulmonary embolism in 18 patients; deep venous thrombosis in 12 patients; cerebrovascular thrombosis in 3 patients; atrial fibrillation, left ventricular thrombus, and coronary artery stent in 2 patients each; and mesenteric artery occlusion, myocardial infarction, and aortic valve replacement in 1 patient each. The INR significantly decreased in 6 of the 42 patients (14%); in 4 of these patients, the INR decreased to below the lower limit of the designated therapeutic range.

This small study of 42 consecutive, unselected patients showed a significant decrease in INR after the cessation of heparin therapy in 6 patients. Of particular concern was that 4 patients had an INR below the therapeutic range. Although no patients developed any adverse clinical events as a result of the decrease in INR, we believe that an adverse event is possible, especially in patients with early initiation of warfarin therapy who are likely to be discharged soon after the cessation of heparin therapy.