Perioperative determinants of morbidity and mortality in elderly patients undergoing cardiac surgery.

OBJECTIVE: To determine perioperative predictors of morbidity and mortality in patients > or =75 yrs of age after cardiac surgery.

DESIGN: Inception cohort study.

SETTING: A tertiary care, 54-bed cardiothoracic intensive care unit (ICU).

PATIENTS: All patients aged > or =75 yrs admitted over a 30-month period for cardiac surgery.

INTERVENTION: Collection of data on preoperative factors, operative factors, postoperative hemodynamics, and laboratory data obtained on admission and during the ICU stay.

MEASUREMENTS AND MAIN RESULTS: Postoperative death, frequency rate of organ dysfunction, nosocomial infections, length of mechanical ventilation, and ICU stay were recorded. During the study period, 1,157 (14%) of 8,501 patients > or =75 yrs of age had a morbidity rate of 54% (625 of 1,157 patients) and a mortality rate of 8% (90 of 1,157 patients) after cardiac surgery. Predictors of postoperative morbidity included preoperative intraaortic balloon counterpulsation, preoperative serum bilirubin of >1.0 mg/dL, blood transfusion requirement of >10 units of red blood cells, cardiopulmonary bypass time of >120 mins (aortic cross-clamp time of >80 mins), return to operating room for surgical exploration, heart rate of >120 beats/min, requirement for inotropes and vasopressors after surgery and on admission to the ICU, and anemia beyond the second postoperative day. Predictors of postoperative mortality included preoperative cardiac shock, serum albumin of <4.0 g/dL, systemic oxygen delivery of <320 mL/ min/m2 before surgery, blood transfusion requirement of >10 units of red blood cells, cardiopulmonary bypass time of >140 mins (aortic cross-clamp time of >120 mins), subsequent return to the operating room for surgical exploration, mean arterial pressure of <60 mm Hg, heart rate of >120 beats/min, central venous pressure of >15 mm Hg, stroke volume index of <30 mL/min/m2, requirement for inotropes, arterial bicarbonate of <20 mmol/L, plasma glucose of >300 mg/dL after surgery, and anemia beyond the second postoperative day. During the study period, the study cohort used 6,859 (21.5%) ICU patient-days out of a total 31,867 ICU patient-days. Nonsurvivors used 2,023 (30%) ICU patient-days and patients with morbidity used 5,903 (86%) ICU patient-days.

CONCLUSIONS: Severe underlying cardiac disease (including shock, requirement for mechanical circulatory support, hypoalbuminemia, and hepatic dysfunction), intraoperative blood loss, surgical reexploration, long ischemic times, immediate postoperative cardiovascular dysfunction, global ischemia and metabolic dysfunction, and anemia beyond the second postoperative day predicted poor outcome in the elderly after cardiac surgery. Postoperative morbidity and mortality disproportionately increased the utilization of intensive care resources in elderly patients. Future efforts should focus on preoperative selection criteria, improvement in surgical techniques, perioperative therapy to ameliorate splanchnic and global ischemia, and avoidance of anemia to improve the outcome in the elderly after cardiac surgery.