Background: The risk of mortality after emergency general surgery (EGS) in elderly patients is prolonged beyond initial hospitalization. Our objective was to develop a preoperative scoring tool to quantify risk of 1-year mortality.
Methods: Three hundred ninety EGS patients aged 70 years or more were analyzed. Risk factors for 1-year mortality were identified using stepwise-forward logistic multivariate regression and weights assigned using natural logarithm of odds ratios. A geriatric emergency surgery mortality (GEM) score was derived from the aggregate of weighted scores. Leave-one-out cross-validation was performed.
Results: One-year mortality was 32%. Risk factors and their weights were: acute kidney injury (2), American Society of Anesthesiology class greater than or equal to 4 (2), Charlson Comorbidity Index greater than or equal to 4 (1), albumin less than 3.5 mg/dL (1), and body mass index (less than 18.5 kg/m2 [1]; 18.5 to 29.9 kg/m2 [0]; ≥30 kg/m2 [-1]). One-year mortality was: GEM 0 to 1 (0% to 7%); GEM 2 to 5 (32% to 68%); GEM 6 to 8 (94% to 100%). C-statistics were .82 and .75 in training and validation data sets, respectively.
Conclusions: A simple score using 5 clinical variables predicts 1-year mortality after EGS with reasonable accuracy and assists in preoperative counseling