A survey of direct oral anticoagulant cessation in general surgery and outcomes in patients with nonvalvular atrial fibrillation

Abstract

There is little data on management and outcomes of atrial fibrillation (AF) patients on direct oral antico-agulants (DOAC) undergoing general surgery. We retrospectively assessed 98 surgeries in 85 nonvalvular AF patients aged 73 ± 8 (59 men) receiving DOACs. Cardiac, emergency, and minimally invasive surgeries were excluded. The CHA2DS2-VASc score ranged from 0 to 8. The DOACs being given were: dabigatran, 16; rivaroxaban, 25; apixaban, 28; and edoxaban, 16. While the DOACs were not suspended in 11 cases, they were interrupted for a median of 2.0 days before surgery and restarted at a median of 3.0 days after surgery. There were 9 complications (9.2%), 3 instances of thromboembolism and 6 bleeding. Thromboembolism occurred at a mean of 3.0 postoperative days, all of which occurred before resumption of DOACs, while bleeding events occurred at a mean of 4.0 postoperative days. Two of the 3 patients with thromboembolism went into cardiopulmonary arrest during the event, but were resuscitated. There were significantly more patients with congestive heart failure or combined antiplatelets in the patients with complications. The complication group had a significantly higher HAS-BLED score and lower preoperative hemoglobin level. There were no significant differences in the management of DOAC interruption between those with complications and without. The perioperative complication rate in nonvalvular AF patients undergoing elective surgery treating with DOACs was 9.2%. Patients with congestive heart failure, receiving combined therapy with antiplatelets, a higher HAS-BLED score, or lower preoperative hemoglobin level were at higher risk. Further studies evaluating the ideal perioperative DOAC protocol are warranted.