Temporary interruption of warfarin treatment is sometimes necessary when patients who require anticoagulant therapy need surgery or other invasive procedures. Interruption of anticoagulant therapy can result in a significant increased risk for thromboembolic morbidity and mortality. Many different strategies for bridge therapy use short-acting anticoagulants in patients taking warfarin. One strategy is to hospitalize patients and stop warfarin therapy four to five days before surgery, using intravenous unfractionated heparin as the bridge anticoagulation. Another strategy uses low-molecular-weight heparin (LMWH) as the bridging anticoagulant. Despite these suggestions and the frequency of this problem before surgery or an invasive procedure, these strategies have not been studied vigorously. Douketis and colleagues evaluated the safety and efficacy of a standardized periprocedural anticoagulation regimen using LMWH.

The study included consecutive adult patients who were on warfarin therapy for thromboembolism prevention and had scheduled an elective surgery or invasive procedure that required the normalization of the International Normalized Ratio (INR). All patients received a standard anticoagulation regimen. Warfarin was stopped five to six days before the procedure (see accompanying table). Three days before the procedure, LMWH (dalteparin) at a dosage of 100 IU per kg twice daily was started. The last dose of dalteparin was given no less than 12 hours before the procedure. On the day of the procedure, if hemostasis was adequate, warfarin was restarted, usually in the evening. If patients had a low risk for bleeding, dalteparin was restarted the following day at the preprocedural dosage and continued until the INR was at least 2.0. If the patient was at high risk for bleeding, no dalteparin was used, and the warfarin was continued. The major outcomes measured during the study included thromboembolic events, major bleeding problems, and death. A secondary outcome was increased wound-related blood loss in patients who were not at high risk for bleeding.

Of the 798 patients assessed, 650 participated in the study. Patients were followed for a mean of 13.8 days before and after the surgery or invasive procedure. Two patients who were not at high risk for bleeding had thromboembolic events (0.4 percent), and four had major bleeding episodes (0.7 percent). In this same group, there were 32 episodes (5.9 percent) of increased wound-related blood loss that precluded the use of dalteparin after the procedure. The group of patients who had high-bleeding-risk procedures had two major bleeding disorders (1.8 percent) and two deaths (1.8 percent), which possibly were related to thromboembolic events.

The authors conclude that patients who receive chronic warfarin therapy and require temporary interruption of treatment can be managed with LMWH. They note that this management strategy can be used with a low risk for thromboembolic and major bleeding complications.