ANZ J Surg 2020 05 29;90(5):746-751. Epub 2020 Apr 29.
Department of Surgery and Surgical Education Research and Training Institute, Royal North Shore Hospital, Northern Sydney Local Health District, Sydney, New South Wales, Australia.
Background: The American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) surgical risk calculator provides an estimate preoperatively of operative risks including mortality; however, its utility is not known in Australian emergency general surgical patients. This study sought to determine accuracy of the calculator in predicting outcome of high-risk patients in an Australian acute surgical unit to establish if this calculator could be a useful tool to identify high-risk patients in an Australian setting.
Methods: Retrospective analysis of patients admitted to the acute surgical unit at a tertiary referral centre between 2018 and 2019 was conducted. High-risk patients were defined as those who underwent an emergency operation with an ACS-NSQIP surgical mortality score ≥5%. Post-operative outcomes assessed included mortality and return to operating theatre, readmission and discharge to nursing home. External validation of the calculator was performed using discrimination and calibration statistics.
Results: Over a 14-month period, 58 patients were high risk, with an average age of 75 years, 93% were classified as functionally independent/partially dependent and 91.4% underwent a laparotomy. Overall 30-day mortality rate was 20.7%. The ACS-NSQIP calculator was a reliable predictor of mortality, with c-statistic of 0.835 (0.654-0.977), Brier score of 0.125 (0.081-0.176) and Hosmer-Lemeshow statistic of 0.389. The calculator was less accurate in its prediction of other outcomes assessed.
Conclusion: The ACS-NSQIP calculator accurately approximated mortality in high-risk Australian patients requiring emergency surgery. This study has demonstrated that in this patient population, the calculator could reliably be applied in the multidisciplinary care of emergency surgical patients.