External Validation of the American College of Surgeons Surgical Risk Calculator in Elderly Patients Undergoing...

Preoperative risk stratification in the elderly surgical patient is an essential part of contemporary perioperative care and can be done with the use of the American College of Surgeons Surgical Risk Calculator (ACS-SRC). However, data on the generalizability of the ACS-SRC in the elderly is scarce. In this study, we report an external validation of the ACS-RC in a geriatric cohort. A retrospective analysis of a prospectively maintained database was performed including patients aged > 65 who underwent general surgery procedures during 2012−2017 in a Greek academic centre. The predictive ability of the ACS-SRC for post-operative outcomes was tested with the use of Brier scores, discrimination, and calibration metrics. 471 patients were included in the analysis. 30-day postoperative mortality was 3.2%. Overall, Brier scores were lower than cut-off values for almost all outcomes. Discrimination was good for serious complications (c-statistic: 0.816; 95% CI: 0.762−0.869) and death (c-statistic: 0.824; 95% CI: 0.719−0.929). The Hosmer-Lemeshow test showed good calibration for all outcomes examined. Predicted and observed length of stay (LOS) presented significant differences for emergency and for elective cases. The ACS-SRC demonstrated good predictive performance in our sample and can aid preoperative estimation of multiple outcomes except for the prediction of post-operative LOS.