Long-term Outcome of Reexploration for Bleeding After Coronary Artery Bypass Grafting.

OBJECTIVES

This study aimed to determine the influence of reexploration for bleeding and blood product requirement after coronary artery bypass grafting (CABG) on long-term mortality.

DESIGN

A retrospective cohort study.

SETTING

A single-center institution.

PARTICIPANTS

All patients who underwent CABG between January 1998 and December 2019 were included.

INTERVENTIONS: The parameters were analyzed to assess the association between reexploration for bleeding and long-term mortality.

MEASUREMENTS AND MAIN

RESULTS

The primary endpoint was all-cause mortality up to the end of follow-up (June 1, 2021). The secondary endpoints were 30-day mortality, duration of admission, blood product transfusion, postoperative use of an intra-aortic balloon pump, deep sternal wound infection, myocardial infarction, and neurologic complications. The Cox proportional hazards model was used to assess the association between reexploration and blood product use and all-cause mortality. Median follow-up was 9.7 years (IQR 5.1-14.6). In total, 576 out of 21,346 (2.7%) patients were reexplored for bleeding. Thirty-day mortality was 6.2% v 1.6% for the reexplored versus not reexplored patients. Reexploration for bleeding was not significantly correlated with long-term mortality (hazard ratio [HR] 1.029; p = 0.068). On the other hand, blood product transfusion (HR = 1.135; p < 0.001), and in particular, packed red blood cell (pRBC) transfusion (HR = 1.139; p < 0.001), was significantly associated with higher long-term mortality. After multivariate Cox regression using ≥5 pRBC transfused as a cut-off point, reexploration for bleeding was not significantly associated with long-term mortality (HR 0.982; p = 0.813). Receiving ≥5 pRBCs was significantly associated with higher long-term mortality (HR 1.249; p < 0.001).

CONCLUSION

Reexploration for bleeding was significantly associated with higher 30-day mortality but not with long-term mortality. Poorer long-term mortality was attributed to patient characteristics and higher use of postoperative blood products.