The duration of oral anticoagulation after a first episode of venous thromboembolism depends on the risk of recurrence. Patients with transient risk factors can be treated for shorter periods than those with continuing risk factors. Recommendations are less clear in patients with pulmonary embolism, a manifestation of the same disease. Because patients with pulmonary embolism are more likely to develop fatal recurrent venous thromboembolism, the length of oral anticoagulation after an initial event requires clarification. Agnelli and associates performed a multicenter, randomized study to evaluate the benefit of extending the usual three-month course of oral anticoagulation to six months in patients with a first pulmonary embolism and temporary risk factors, and extending to one year in patients with an idiopathic first pulmonary embolism. The outcome was symptomatic, confirmed recurrence of venous thromboembolism.

Patients with a first confirmed, symptomatic pulmonary embolism were categorized as having transient risk factors or idiopathic pulmonary embolism (i.e., no known cancer, no known thrombophilia, no transient risk factors). Patients with permanent risk factors (e.g., known cancer, known thrombophilia) were excluded from the study. After completing three months of anticoagulation, patients were assigned randomly to discontinue therapy or to continue therapy for three months (presence of transient risk factors) or nine months (idiopathic pulmonary embolism). The dosage of warfarin was adjusted to maintain the International Normalized Ratio between 2.0 and 3.0.

Of 326 patients enrolled in the study, 33 patients had confirmed recurrent venous thromboembolism (10.1 percent; 3.6 percent per patient-year; average follow-up, 33.8 months). The incidence of recurrence was 12.2 percent in patients with idiopathic pulmonary embolism and 7.6 percent in those with pulmonary embolism associated with transient risk factors. Among the patients assigned to continue therapy, the recurrence rate for venous thromboembolism was 3.1 percent per patient-year, compared with 4.1 percent per patient-year among the patients assigned to discontinue therapy. In the patients who continued anticoagulation therapy, the incidence of recurrence after treatment discontinuation increased to 3.8 percent per patient-year. The rates of bleeding and other adverse events were low.

The authors conclude that the optimal duration of anticoagulation therapy after pulmonary embolism depends on the risk for recurrent venous thromboembolism after anticoagulation is discontinued. Although the best way to assess this risk is uncertain, patients with idiopathic pulmonary embolism are at higher risk for recurrence than are those with pulmonary embolism associated with transient risk factors. The authors note that the risk analysis probably will involve clinical evaluation, as well as assessment for residual pulmonary hypertension, screening for genetic thrombophilia, and a D-dimer assay at the end of the treatment period. Indefinite anticoagulation may be required in patients with idiopathic pulmonary embolism who are at high risk for recurrence of venous thromboembolism.