SP2.9 - National Emergency Laparotomy Audit (NELA) and American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) risk prediction calculators are not accurate in predicting 30-day mortality in patients aged 80 and above

Abstract Background 30- to 50000 Emergency Laparotomies (EL) are performed annually in the UK. More than half of patients are aged 70 or over and have a high post-operative complication rate along with a 16-20% risk of 30-day mortality (30-dm). To date, the literature lacks a dedicated risk prediction tool for mortality and morbidity following EL in the elderly population. Methods A multi-center, retrospective study of patients aged ≥80 undergoing EL examined the correlation between predicted 30-dm using NELA and ACS-NSQIP and actual 30-dm. This occurred in two UK centers over a 2-year period. The primary outcome was defined as the correlation between actual 30-dm and NELA or ACS-NSQIP predicted 30-dm. GraphPad Prism 10.1.1. was used for statistical analysis. Results 141 cases were included in the final analysis (Table 1). The actual 30-dm was 17.73%, NELA predicted 30-dm 13.07% (7.21;27.84), and ACS-NSQIP 9.4% (3.8;20.6). NELA correlated positively with ACS-NSQIP predicted 30-dm (R²=0.79; p<0.0001). No association was found between actual 30-dm and predicted NELA 30-dm, R²=0.01; p<0.0001 or ACS-NSQIP, R²=0.01; p<0.0001. All data is expressed as Median + IQR. Conclusion In octo- and nonagenarians, following an emergency laparotomy, both NELA and ACS-NSQIP are not accurate in predicting 30-day mortality.Table 1Male/female ratio0.63Age (years)84 (81;87)BMI (kg/m²)24.27 (21.43;27)Lactate (most recent preoperative value) mmol/L1.7 mmol/L (1.14;3.01)ASA grade3 (2;4)Clinical frailty score3 (3;5)Charlson Comorbidity Index4 (2;5)