The Importance of Initial Heparin Treatment on Long-term Clinical Outcomes of Antithrombotic Therapy

Abstract

Recent clinical trials of venous thromboembolism treatment suggest inadequate initial heparin therapy predisposes patients to late recurrence of thromboembolism. However, a recent review article was unable to demonstrate a relationship between initial heparin therapy and late recurrence. To evaluate the relationship between initial heparin treatment and long-term clinical outcome in 3 consecutive, randomized, double-blind trials that used similar study designs and patient populations and objective documentation of recurrent venous thromboembolism. The trials were performed sequentially and compared the use of continuous intravenous with subcutaneous heparin, continuous intravenous heparin for 10 or 5 days, and continuous intravenous heparin with once-daily subcutaneous low-molecular-weight heparin. All patients were followed up for 3 months to assess the a priori hypothesis that inadequate initial heparin therapy could lead to recurrent venous thromboembolism during long-term therapy with warfarin sodium. The following were the observed rates of recurrent venous thromboembolism: continuous intravenous heparin, 3 (5.2%) of 58 patients vs subcutaneous heparin, 11 (19.3%) of 57 patients; continuous intravenous heparin for 10 days, 7 (7.0%) of 100 patients or for 5 days, 7 (7.1%) of 99 patients; and continuous intravenous heparin, 15 (6.9%) of 219 patients vs low-molecular-weight heparin, 6 (2.8%) of 213 patients. Pooled analysis of the patients treated with continuous intravenous heparin showed that of the total 32 patients with recurrent venous thromboembolism, in 6 patients thromboembolism occurred early (< 10 days) and 26 patients thromboembolism occurred late. Of these patients, the majority (20/32 [62.5%]) had therapeutic prothrombin time or international normalized ratio values before or at the time of the recurrent thromboembolic event. Our findings demonstrate that the initial heparin treatment affects the long-term outcome. This conclusion applies when these data are analyzed for each individual study by treatment group, observed difference in outcome, and pooled analysis.