Attributable harm of severe bleeding after cardiac surgery in hemodynamically stable patients

Background: We sought to quantify the effect of severe postoperative bleeding in hemodynamically stable patients following cardiac surgery. Methods: We reviewed the charts of all cardiac surgery patients operated on at our institution between 2010 and 2014. After excluding patients with tamponade or MAP <60, we propensity matched patients having chest tube output >300 mL in the first postoperative hour, >200 mL in the second, and >100 mL in the third (“bleeding” group) with patients having <50 mL/h of chest tube output (“dry” group). The primary outcome was a composite of morbidity or mortality (excluding reexploration). Results: 5016 patients were operated on between 2010 and 2014; of these, we included the records of 84 bleeding and 498 dry patients. Propensity matching resulted in 68 pairs of patients well-matched on baseline and operative variables. As compared to matched dry patients, bleeding patients were more likely to experience the primary outcome of any morbidity/mortality (36.8 vs. 13.2 %, p = 0.002), as well as ventilation >24 h (33.8 vs. 7.4 %, p < 0.001) and 30-day mortality (11.8 vs. 1.5 %, p = 0.02). Of the 84 bleeding patients, 46 underwent reexploration for bleeding within 24 h of surgery. A subgroup analysis propensity matching bleeding patients who were or were not reexplored <24 h demonstrated similarly poor outcomes in each group (primary outcome, 44.7 % reexplored vs. 50.0 % non-reexplored, p = 0.65), though reexplored patients were far less likely to require hematoma evacuation/washout >24 h after surgery (0 vs. 18.4 %, p = 0.005). Conclusions: Even among hemodynamically stable patients, severe bleeding is associated with markedly worse outcomes following cardiac surgery.