We investigated 81 consecutive patients (median age 74 years, interquartile range [IQR]: 68 to 78) who underwent open-heart operations at our institution between July 2014 and June 2016. All patients presented for surgery while on NOAC therapy: 37 received rivaroxaban (45.7%), 35 apixaban (43.2%), and 9 dabigatran (11.1%). The calculated risk using the European System for Cardiac Operative Risk Evaluation II was 3.5% (IQR: 2.0% to 8.1%).
Surgery was performed at a median 4 days (IQR: 3 to 6) after NOAC withdrawal. Reduced renal function was predictive for length of intensive care unit stay and administration of red blood cells (p < 0.0001 and p = 0.0291, respectively). The NOAC withdrawal interval significantly influenced postoperative drainage volume (p = 0.0056). Five patients needed rethoracotomy because of relevant bleeding (6.2%), 4 after apixaban (11.4%) and 1 after rivaroxaban therapy (2.7%). Apixaban showed a borderline influence on prolonged intensive care unit stay (p = 0.0736). Prolonged cardiopulmonary bypass time was predictive for thrombocyte administration (p = 0.0249). Intensive care unit stay was 2 days after NOAC withdrawal of 10 days, compared with 4.2 days without termination. Thirty-day mortality was 3.7%.
A lengthy NOAC withdrawal period, particularly for patients with reduced renal function, is essential for safe open-heart surgery. We conclude that despite official recommendations, patients should whenever possible not be considered for elective cardiac surgery within 10 days of terminating NOAC treatment.