Development and Validation of a Risk Score for Respiratory Failure After Cardiac Surgery

Background: Postoperative respiratory failure, defined as ventilator dependency for more than 48 hours or unplanned reintubation within 30 days, is a costly complication of cardiac surgery that increases mortality and length of stay. Stratification of patients by risk upon intensive care unit admission could identify cases requiring early measures to prevent respiratory failure. This study aimed to develop and validate a risk score for postoperative respiratory failure after cardiac surgery. Methods: This retrospective analysis of 4262 patients admitted to the cardiovascular intensive care unit after major cardiac surgery between January 2013 and December 2017, used The Society of Thoracic Surgeons database and ventilator data from the respiratory therapy department. Patients were randomly and equally assigned to development and validation cohorts. Covariates used in the multivariable models were assigned weighted points proportional to their β regression coefficient values to create the risk score, which categorized patients into low, medium, and high risk of postoperative respiratory failure. Results: In both cohorts, postoperative respiratory failure risk was significantly different between risk categories. Compared with low-risk patients, moderate-risk patients had a 2 times greater risk, and high-risk patients had a 4-7 times greater risk. Body mass index, previous cardiac surgery, cardiopulmonary bypass, cardiogenic shock, pulmonary disease presence, baseline functional status, hemodynamic instability, and number of blood products used intraoperatively were significant predictors of respiratory failure. Conclusions: This risk score can stratify patients by risk for developing postoperative respiratory failure after major cardiac surgery, which may help in the development of preventive measures.