ThP6.15 - 1, 3, and 5-year mortality following emergency laparotomy, can the National Emergency Laparotomy Audit (NELA) or the American College of Surgeons National Quality Improvement Program (ACS-NSQIP) risk calculators predict it? – a retrospective, multi-center, observational, cohort study

Abstract

Abstract Background Emergency laparotomy (EL) comprises a high-risk group of surgical procedures, associated with a significantly increased risk of death in the following 5 years. EL are performed on a heterogeneous population, of which at least 50% are ≥ 70 years old. Accurate preoperative risk prediction is vital for planning care, patient shared decision, and improving outcomes. To date, no prediction tool can reliably estimate long-term outcomes, including mortality, following an EL. Methods This is a multicenter, retrospective, observational, cohort study that investigated the correlation between actual mortality at 1, 3, and 5 years and predicted 30-day mortality (30-dm) using NELA and ACS-NSQIP risk prediction tools, following EL, in adults aged ≥18, over 2 years, in two UK centers. The sample size was calculated at 385 cases. GraphPad-Prism 10.1.1. was used for statistical analysis. Results 626 cases were analyzed. The male/female ratio was 0.85, age 70 years (55;78); BMI 24.94kg/m² (22;29); preoperative lactate 1.7 mmol/L (1.2;2.6), clinical frailty score 3 (2;4); and Charlson Comorbidity Index 4(2;5) (all data is expressed as median+IQR). The 1, 3, and 5-year survival were 76%, 67%, and 61%, respectively. A perfect positive correlation between NELA and ACS-NSQIP 30-dm and the 1-year actual mortality was found (R²=1). At 3 years only NELA correlated positively to actual mortality (R²=0.23, p<0.05). The 5-year mortality correlated positively with both predictors - NELA: R²=0.61, p<0.001; ACS-NSQIP: R²=0.31, p< 0.05. Conclusion Both NELA and ACS-NSQIP risk calculators can predict mortality at 1,3 and 5 years after an emergency laparotomy, with NELA expressing a stronger correlation.