Ruptured abdominal aortic aneurysm: a novel method of outcome prediction using neural network technology

Background: reported survival following emergency surgery for ruptured abdo minal aortic aneurysm (RAAA) varies widely between institutions. This is la rgely attributable to differences in case mix. The aim of this study was to identify and evaluate a set of prognostic variables that would accurately predict outcome for individual patients from perioperative indices. Methods: perioperative factors associated with subsequent mortality at our institution were identified by retrospective review of 102 consecutive oper ations for RAAA over a 7-year period (January 1990 to January 1997). Logist ic regression analysis was used to select the most significant variables as sociated with subsequent mortality. These were used to construct, train, an d validate a neural network designed to predict survival from surgery in in dividual cases on a prospective basis. Results: the 30-day mortality rate was 53%. Multivariate analysis identifie d four highly significant independent predictors of mortality; preoperative hypotension, intraperitoneal rupture, preoperative coagulopathy, and preop erative cardiac arrest. Using these inputs, the neural network correctly pr edicted outcome in 82.5% of individual cases. Conclusion: a neural network based on just four perioperative variables can accurately predict outcome of RAAA. Prognostic variables should be reporte d in studies as a measure of the effect of case mix on survival data. Neura l networks have potential to aid decision-making relating to outcome for in dividual cases.