A retrospective cohort study found that patients with venous thromboembolism (VTE) who received bridge therapy had a 17-fold increase in the risk of bleeding during warfarin therapy interruption for invasive procedures compared with those who did not receive bridge therapy.

The risk of recurrent VTE was similar between the groups, according to lead researcher Nathan P. Clark, PharmD, of Kaiser Permanente Colorado in Aurora. He and his colleagues published their findings online in JAMA Internal Medicine on May 26.

They defined bridge therapy as the use of a short-acting anticoagulant such as low-molecular-weight heparin during the periprocedural period in patients receiving warfarin for the secondary prevention of VTE. Patients at high or moderate risk of bleeding typically stop taking warfarin a few days before and after undergoing invasive diagnostic or surgical procedures.

In this analysis, researchers examined 1,178 patients who underwent 1,812 procedures between Jan. 1, 2006 and March 31, 2012 at Kaiser Permanente Colorado, an integrated healthcare delivery system. All of the patients received warfarin, including 56.3 percent for deep vein thrombosis and 43.7 percent for pulmonary embolism. The mean age was 66.1 years, and 45.8 percent of the procedures were performed in men.

The most common types of procedures when warfarin was stopped were gastrointestinal tract endoscopic (37.1 percent), orthopedic (13.6 percent), spinal or intracranial (9.7 percent) and nonmajor abdominal or thoracic (8.6 percent) procedures.

Of the procedures, 79 percent were considered low risk for recurrent VTE, while 17.9 percent were considered moderate risk and 3.1 percent were considered high risk. Bridge therapy was performed in 28.7 percent of the low-risk patients, 33.6 percent of the moderate-risk patients and 63.2 percent of the high-risk patients.

At 30 days, clinically relevant bleeding was found in 2.7 percent of patients undergoing bridge therapy and 0.2 percent of patients not receiving bridge therapy. Clark et al defined clinically relevant bleeding as overt bleeding resulting in hospitalization or an emergency department visit or bleeding that complicated the procedure.

Of the bleeding events in the bridge therapy group, 52.9 percent were procedure complications and 33.3 percent were related to bridging agent injections.

Recurrent VTE occurred in no patients in the bridge therapy group and three patients in the nonbridge therapy group. There were no deaths during the 30-day period following the procedures. Major bleeding occurred in 12 bridge therapy procedures and two nonbridge therapy procedures.