ABSTRACT
The predictors of prolonged mechanical ventilation and subsequent morbidity after cardiac surgery are ill defined. Our aim was to evaluate them. Four hundred and seventy consecutive patients undergoing coronary artery bypass grafting on cardiopulmonary bypass (CPB) between January and June 2002 were retrospectively analysed for preoperative predictors of prolonged ventilation, which included age, gender, ejection fraction (EF), renal function, diabetes, angina status, severity of the disease (New York Heart Association class), number of vessels diseased and chronic lung disease. Intraoperative variables such as prolonged CPB, aortic cross clamp time, intra-aortic balloon pump (IABP) usage, inotropes and postoperative variables like temperature on arrival at intensive care unit(ICU), IABP usage, organ dysfunction, inotropes and reintervention (reintubation and re-exploration) were also analysed. Prolonged ventilation was defined as > or = 24 hours and these patients were included in group I (n=22). Patients requiring less than 24 hours ventilation (n=448) were included in group II. Stepwise logistic regression analysis was performed. The average age of patients was 56.9 +/- 8.8 years with male predominance (88.4%). The overall perioperative mortality was 2.1% (10 patients) with Group I showing mortality rate of 36.3% (8 patients). In multivariate analysis, predictors of prolonged ventilation were found to be EF <40% (odds ratio, (OR) 13.38), preoperative renal dysfunction [OR 4.06 (serum creatinine > 1.2 mg%)], prolonged CPB, > 120 min (OR 9.6) and reintervention in the form of re-exploration or reintubation in the ICU (OR 13.8). Identification of perioperative variables, which may lead to prolonged ventilation may allow the development of strategies to optimize the patient's condition and ICU management.