Skip to content

Non–Vitamin K Antagonist Oral Anticoagulants Are Safe for Selected Patients with Postoperative AF

Key findings

  • This retrospective study examined 30-day outcomes for 26,522 patients who developed nonvalvular atrial fibrillation (AF) after coronary artery bypass or valve surgery and received a non–vitamin K antagonist oral coagulant (NOAC, 37%) or warfarin (63%)
  • Compared with warfarin, NOACs were not linked to any significant difference in the three primary short-term outcomes (30 days): rehospitalization for major bleeding complications, stroke/transient ischemic attack or all-cause mortality
  • These findings remained unchanged across strata of renal function and type of cardiac surgery, and NOACs were also similar to warfarin with regard to secondary outcomes

Non–vitamin K antagonist oral anticoagulants (NOACs) are now recommended as first-line therapy, in preference to warfarin, as thromboembolism prophylaxis for individuals with nonvalvular atrial fibrillation (AF). However, that guideline is based on trials that excluded patients with postoperative AF, so the optimal oral anticoagulant for that population has been unclear.

Now, the largest multicenter study of this issue to date has established that NOACs are a safe alternative to warfarin for selected patients who develop nonvalvular AF after cardiac surgery. In collaboration across the Corrigan Minehan Heart Center at Massachusetts General Hospital, Steven A. Lubitz, MD, MPH, cardiac electrophysiologist in the Telemachus & Irene Demoulas Family Foundation Center for Cardiac ArrhythmiasPatrick T. Ellinor, MD, PhD, director of the Telemachus & Irene Demoulas Family Foundation Center for Cardiac Arrhythmias, Thor Sundt, MD, chief of the Division of Cardiac Surgery, and colleagues, and colleagues published the details in The Annals of Thoracic Surgery.

Study Methods

The researchers searched the Adult Cardiac Surgery Database of the Society of Thoracic Surgeons for patients who developed new-onset nonvalvular AF after undergoing coronary artery bypass graft, surgical valve repair or valve replacement with a bioprosthesis between July 2017 and December 2018. They identified 26,522 patients, of whom 37% received a NOAC for thromboprophylaxis. The others received warfarin.

Three Primary Outcomes

Within 30 days:

  • Rehospitalization for a major bleeding complication: 0.37% of the NOAC group vs. 0.50% of the warfarin group; P=NS
  • Stroke/transient ischemic attack (TIA): 0.23% vs. 0.30%; P=NS
  • Postdischarge all-cause mortality: 0.98% vs. 0.76%; P=NS

In multivariable analysis, there was no significant association between type of oral anticoagulant therapy and any of these short-term outcomes.

The results were similar across strata of renal function and regardless of type of cardiac surgery.

Secondary Outcomes

  • Postoperative length of stay was significantly shorter by half a day in the NOAC group vs. the warfarin group
  • There was no difference between groups in 30-day rates of rehospitalization for pericardial effusion/cardiac tamponade, myocardial infarction, venous thromboembolism or wound infection

30-Day Outcomes vs. No Anticoagulation

Compared with no anticoagulation, NOACs or warfarin were associated with:

  • Significantly higher risk of the composite endpoint of major bleeding complication/cardiac tamponade
  • Higher rates of rehospitalization for wound infection (significant for NOACs but not warfarin)
  • Nonsignificantly reduced rates of stroke/TIA
  • Significantly lower risk of rehospitalization for myocardial infarction or venous thromboembolism
  • No difference in all-cause mortality

Applying the Findings to the Clinic

These data support the broader use of NOACs in preference to warfarin in patients with new-onset nonvalvular AF after cardiac surgery who have an elevated risk of stroke. Careful evaluation of bleeding risk is necessary before initiating anticoagulation.

Learn more about the Corrigan Minehan Heart Center

Refer a patient to the Corrigan Minehan Heart Center


A Massachusetts General Hospital study is testing whether a novel software algorithm can be used to identify atrial fibrillation using wearable devices.


A study led by a Massachusetts General Hospital cardiologist used single nucleus sequencing to create a novel, detailed cellular map of the heart.