Discontinuation of dual antiplatelet therapy and bleeding in intensive care in patients undergoing urgent coronary artery bypass grafting: a retrospective analysis

Objectives: Our goal was to evaluate the impact of the discontinuation times of dual antiplatelet therapy with clopidogrel, prasugrel or ticagrelor on postoperative bleeding rates and the use of blood products in patients undergoing isolated urgent coronary artery bypass grafting (CABG).

Methods: We retrospectively analysed 334 patients with acute coronary syndrome undergoing urgent CABG at the University Hospital Basel. A total of 262 patients continued to take dual antiplatelet therapy during the surgery (72 received clopidogrel; 68, prasugrel; and 122, ticagrelor). They were stratified by the discontinuation time of dual antiplatelet therapy (<24 h, 24-48 h, 48-72 h and >72 h). Seventy-two patients taking acetylsalicylic acid (ASA) as monotherapy served as a comparison group.

Results: Median postsurgical bleeding rates were significantly higher with ticagrelor if it was discontinued <24 h [1220 ml, interquartile range (IQR) 978-1520 ml; P < 0.001], 24-48 h (1200 ml, IQR 800-1550 ml; P < 0.001) and 48-72 h (1100 ml, IQR 845-1245 ml; P = 0.036) but not if discontinued >72 h (700 ml, IQR 520-825 ml; P = 0.22) and with prasugrel if discontinued <24 h (1320 ml, IQR 900-1950 ml; P < 0.001) but not if discontinued 24-48 h (1050 ml, IQR 638-1438 ml; P = 0.089) or >72 h (750 ml, IQR 488-1040; P = 0.63) compared to ASA monotherapy (800 ml, IQR 593-1043 ml). The postsurgical use of blood products compared to ASA monotherapy (0, IQR 0-2 units) was significantly higher with ticagrelor and prasugrel if discontinued <24 h (2.5 units, IQR 0-6; P < 0.001 and 2 units, IQR 1-6; P < 0.001, respectively).

Conclusions: Discontinuation of ticagrelor and prasugrel for more than 72 h before urgent CABG was not associated with higher bleeding rates compared to treatment with ASA monotherapy. In contrast, discontinuation for less than 24 h was associated with higher use of blood products. For ticagrelor, this study supports evidence and recent guidelines proposing a shorter discontinuation time of 3 days and raises the question of whether the same could be true for prasugrel.