For Perioperative Anticoagulation in Atrial Fibrillation: Burn Those Bridges

When patients with atrial fibrillation (AF) being treated with warfarin need to interrupt anticoagulation to undergo an elective operation or invasive procedure, the common practice is to “bridge” the patient with low-molecular-weight heparin (LMWH) during the perioperative period. According to a recent study published in New England Journal of Medicine, that practice is not only unnecessary, but it may be dangerous.

In a randomized, double-blind, placebo-controlled trial, James D. Douketis, MD, from St. Joseph’s Healthcare Hamilton and the Department of Medicine at McMaster University in Hamilton, Ontario, and colleagues found that bridging anticoagulation offered no benefit over no bridging, but it actually increased the risk of major bleeding nearly threefold.

“Each year, this common clinical scenario affects approximately one in six warfarin-treated patients with atrial fibrillation,” Dr. Douketis and co-authors wrote. “Multiple observational studies have assessed the timing and dosing of perioperative bridging with low-molecular-weight heparin. However, the fundamental question of whether bridging anticoagulation is necessary during perioperative warfarin interruption has remained unanswered.”

The Bridging Anticoagulation in Patients who Require Temporary Interruption of Warfarin Therapy for an Elective Invasive Procedure or Surgery (BRIDGE) trial compared the risk of arterial thromboembolism or bleeding events with bridging anticoagulation or no anticoagulation in 1,884 patients – 950 assigned to the bridging therapy group and 934 to the no-anticoagulation group.

Patients were enrolled from 108 sites in the United States and Canada between July 2009 and December 2014. All patients had chronic AF or atrial flutter, had received warfarin therapy for three or more months, were undergoing an elective operation or invasive procedure, and had at least one of the following CHADS2 stroke risk factors:

  • Congestive heart failure or left ventricular dysfunction
  • Hypertension
  • 75 years of age or older
  • Diabetes mellitus
  • Previous ischemic stroke, systemic embolism, or transient ischemic attack

The mean age of the patients was 71.7 years, and 73.4 percent were male. Patients were generally at lower risk for stroke, with a mean CHADS2 score of 2.3 (from a scale of 1 to 6); 38.3 percent of patients had a CHADS2 score of 3 or higher. A total of 34.7 percent of patients were also taking aspirin, and 7.2 percent were taking a different antiplatelet drug, in addition to the warfarin.

Patients in the bridging therapy group received either LMWH or dalteparin sodium (in the no-anticoagulation group) at a dose of 100 IU/kg of body weight administered subcutaneously twice daily. Warfarin was discontinued five days before the procedure and two days before the bridging therapy was administered. Patients received bridging treatment for three days prior to 24 hours before the procedure and then for five to 10 days following the procedure.

A total of 1, 813 completed the study, and follow-up assessments were completed by 37 days post-procedure.

As seen in the TABLE, at 30 days post-procedure, the risks for arterial thromboembolism, the study’s primary efficacy endpoint, were similar between the two therapy groups.

The two groups diverged, though, in terms of safety: The incidence of major bleeding rose from 1.3 percent in the non-bridging group to 3.2 percent in the bridging group (relative risk = 0.41; 95% CI 0.2-0.78; p=0.005 for superiority). None of the instances of major bleeding were fatal. The non-bridging group was also at a significantly lower risk for minor bleeding compared with the bridging group.

Incidence of the study’s primary safety endpoints (incidence of acute myocardial infarction, deep-vein thrombosis, pulmonary embolism, or death) were also increased in the bridging group, although these findings were non-significant.

“Taken together, these findings show that there is a net clinical benefit in favor of a strategy of forgoing bridging, as compared with perioperative bridging with low-molecular-weight heparin,” the researchers wrote.

One of the main limitations of the BRIDGE trial, Dr. Douketis and authors noted, was the underrepresentation of certain patient populations, including those with higher CHADS2 scores and patients undergoing major surgical procedures, which are associated with higher rates of arterial thromboembolism and bleeding. Also, the authors added, the overall rate of arterial thromboembolism was lower than expected – potentially affecting the power of the trial to detect a benefit associated with bridging.


Douketis JD, Spyropoulos AC, Kaatz S, et al. Perioperative bridging anticoagulation in patients with atrial fibrillation. N Engl J Med. 2015 June 22. [Epub ahead of print].

TABLE. Study Outcomes

No Bridging (N=918) (%)

Bridging (N=895)
p Value
(for superiority)
Arterial thromboembolism

4 (0.4%)

3 (0.3%)

0.01  (for noninferiority)



2 (0.2%)

3 (0.3%)

Transient ischemic attack

2 (0.2%)


Systemic embolism


Major bleeding

12 (1.3%)

29 (3.2%) 0.005

5 (0.5%)

4 (0.4%) 0.88

Myocardial infarction

7 (0.8%) 14 (1.6%)


Deep-vein thrombosis

0 1 (0.1%)


Pulmonary embolism

0 1 (0.1%)


Minor bleeding

110 (12.0%) 187 (20.9%)