Prediction of Postoperative Outcomes and Long-Term Survival in Cardiac Surgical Patients Using the Intensive Care National Audit & Research Centre Score.

ABSTRACT

OBJECTIVES: Scoring systems used in cardiac surgery, such as the European System for Cardiac Operative Risk Evaluation and the Society of Thoracic Surgeons scoring systems, do not adjust for events that take place intraoperatively. The authors hypothesized that intensive care unit scoring systems such as the Intensive Care National Audit & Research Centre (ICNARC) could predict accurately not only in-hospital mortality, but also other significant complications, as well as long-term survival after cardiac surgery. DESIGN: Prospective cohort study using perioperative data from the ICNARC Audit and Dendrite database. SETTING: Single tertiary referral cardiac surgery center. PARTICIPANTS: A total of 4,446 consecutive cardiac surgical patients who had surgery between January 2011 and April 2018. INTERVENTIONS: Comparison of scoring systems to predict postoperative outcomes. MEASUREMENTS AND RESULTS: Receiver operating curves (ROCs) were used to evaluate how well the ICNARC scores predicted in-hospital mortality and postoperative complications (renal failure, pulmonary complications, gastrointestinal complications, and multiorgan failure). Cox regression analysis was used to determine factors affecting long-term survival. The C-indices for the ROC graphs for the ICNARC score were 0.840 for in-hospital mortality, 0.858 for renal failure, 0.665 for pulmonary complications, 0.764 for gastrointestinal complications, 0.702 for neurological complications in general and 0.654 for confusion, and 0.885 for multiorgan failure. From Cox regression analysis, the significant (p < 0.05) predictors of midterm mortality (5 years) were a higher ICNARC score, a higher age at surgery, chronic obstructive pulmonary disease, preoperative renal failure, preoperative neurological comorbidity, arteriopathy, and non-coronary artery bypass graft surgery. CONCLUSION: The ICNARC scoring system is simple and can be used as an early warning screening tool to predict which patients are at higher risk for postoperative organ failure.