To establish the risk factors and impact of reexploration for bleeding in a large modern cardiac surgical cohort.Methods:
At a tertiary referral center, baseline, index procedural, reexploration, outcome, and readmission characteristics of 16,793 consecutive adult cardiac surgery patients were prospectively entered into dedicated clinical databases. Correlates of reexploration for bleeding, as well as its association with outcomes and readmission, were examined with multivariable regression models.Results:
The mean patient age was 65.9 ± 12.1 years, and 11,991 patients (71.4%) patients were male. Perioperative mortality was 2.8% (458 of 16,132) in those who did not undergo reexploration for bleeding and 12.0% (81 of 661) in those who underwent reexploration for bleeding, corresponding to an odds ratio of 3.4 ± 0.5 (P <.001) over other predictors of mortality, including Euroscore II. Mortality was highest in patients who underwent reexploration after the day of index surgery (odds ratio, 6.4 ± 1.1). Hospital stay was longer in patients who underwent reexploration for bleeding (median, 12 days, vs 7 days in patients who did not undergo reexploration; P <.001), to an extent beyond any other correlate. Reexploration for bleeding also was independently associated with new-onset postoperative atrial fibrillation, renal insufficiency, intensive care unit readmission, and wound infection. Risk factors for reexploration for bleeding were tricuspid valve repair, on-pump versus off-pump coronary artery bypass grafting, emergency status, cardiopulmonary bypass (CPB) duration, low body surface area, and lowest CPB hematocrit of <24%.Conclusions:
Reexploration for bleeding is a lethal and morbid complication of cardiac surgery, with a detrimental effect that surpasses that of any other known potentially modifiable risk factor. All efforts should be made to minimize the incidence and burden of reexploration for bleeding, including further research on transfusion management during CPB.