669 Mortality Risk Prediction for Emergency Laparotomy: Are We Utilising the Best Tool?

Abstract Aim Emergency laparotomy is one of the common operations performed in UK. To aid more objective decision making, several scoring systems have been formulated. In this project, we aimed to explore predictive power both NELA and P-Possum mortality scores against 30 days 90 observed for emergency patients. Method Patient details from two large district general hospitals were extracted database over a period three years. Pre-operative NELA, post-operative P-POSSUM predicted calculated compared with entire cohort. Model discrimination (statistical accuracy) was tested by calculating area under receiver operating characteristic curve (AUC), which used assess how accurately model could discriminate. Results There 378 patients eligible inclusion median age 64. 39 (10.3%) died within 52 (13.8%) days. score, pre-operative as (2.7%, 3.7%, 2.4%) (2.9%, 4.8%, 4%) respectively. The discriminative highest score (AUC 0.870, CI: 0.824 – 0.916), 0.826, 0.769 0.884) Conclusions Both underpredicted actual mortality. It however noted that showed accurate than other 2 tools.