Heparin Bridging or Continued Warfarin for Device Surgery?

Abstract & Commentary

By John P. DiMarco, MD, PhD, Professor of Medicine, Division of Cardiology, University of Virginia, Charlottesville. Dr. DiMarco does research for Medtronic, is a consultant for Medtronic, Novartis, and St. Jude, and is a speaker for Boston Scientific.

Source: Birnie, et al, for the BRUISE CONTROL Investigators. Pacemaker or defibrillator surgery without interruption of anticoagulation. N Engl J Med 2013;368:2084-2093.

Many laboratories have changed their practice to perform some procedures without stopping warfarin anticoagulation, but trial evidence supporting this practice has been lacking. In this paper, investigators from 17 centers in Canada and one in Brazil report the results of a trial comparing continued warfarin with heparin bridging in high-risk patients undergoing device procedures. Patients at the centers scheduled for a pacemaker or implantable cardioverterdefibrillator (ICD) procedure of any type who were on chronic warfarin were eligible for enrollment. Patients were then randomly assigned in a 1-to-1 ratio to continued warfarin treatment or warfarin interruption with bridging therapy with heparin. The target INR on the day of surgery in the continued-warfarin group was 3 or lower in most patients but up to 3.5 in patients with mechanical valves. Patients in the heparin-bridging group discontinued warfarin 5 days before the procedure and then received either low molecular-weight heparin or intravenous unfractionated heparin for 3 days before the procedure. Intravenous heparin was stopped at least 4 hours before surgery. For those who received low molecular-weight heparin, the last dose was given the morning of the day before the procedure. Heparin was restarted 24 hours after the procedure and continued until a therapeutic INR had been achieved on warfarin. Clopidogrel was stopped 5 days before the procedure, except for those patients with drug-eluting or recently implanted bare-metal coronary stents. Aspirin was continued. The primary outcome measure was a clinically significant device-pocket hematoma defined as one that either required surgical drainage, prolonged hospitalization, or interruption of anticoagulation therapy.

The study was terminated after a second interim look by a data safety monitoring board. The entire cohort included 681 patients. The final comparison group was 326 patients assigned to heparin bridging and 335 patients assigned to continued warfarin. Low molecular-weight heparin was used in 89% of the patients and intravenous heparin in 11%. Postoperatively, 82% of the patients received low molecular-weight heparin and 17% intravenous heparin. In the heparin-bridging group, the median INR on the day of surgery was 1.2. In the continued-warfarin group, the median INR was 2.3. Surgery had to be postponed in eight patients in the continued-warfarin group because the INR on the day of surgery was above 3.5. In the heparin-bridging group, three patients had postponements because of INRs higher than the desired target.

Clinically significant hematomas occurred in 12 of 343 patients in the continued-warfarin group (3.5%) compared with 54 of 338 patients (16%) in the warfarin-bridging group. The most common endpoint was a non-surgical hematoma requiring interruption of anticoagulation, but an increase in prolonged hospitalizations and surgical evacuations was also noted in the heparin-bridging group. Although not statistically significant, there were also six infections related to the device system in the heparin-bridging group compared to only two in the continued-warfarin group. There were two neurologic events in the continued-warfarin group; one stroke and one TIA. Both patients, however, had unplanned subtherapeutic INRs on the day of the procedure despite continuing warfarin therapy. Patient satisfaction scores were better in the continued warfarin group.

The authors conclude that for most patients, cardiac device-related procedures can be safely performed with continued warfarin therapy avoiding the increased risk of hematoma associated with heparin bridging.


Many electrophysiology (EP) laboratories have gradually shifted their practice to perform ablation and device procedures in patients on warfarin with INRs still in the low therapeutic range. Heparin therapy after a device procedure has been associated with increased risk of hematoma and this was confirmed in this study. When a hematoma develops, the patient often has to stop anticoagulation until the hematoma has started to resolve, and this may take many days. Surgical re-exploration to drain a hematoma is also often difficult since it’s rare to identify a discrete bleeding site after the pocket has been reopened. Another factor that makes device-related hematomas a management problem is the fact that they cannot easily be controlled with direct pressure. Once a clot has developed, the device itself is floating in clot and this prevents the tissue-to-tissue contact that might stop the bleeding.

The authors have done a significant service to the EP community. This paper should lead to a revision of the current guidelines and provide benchmark data to which implanters can compare their results.