In an era of increasing demands to provide high-quality health care, surgeons need an accurate preoperative risk assessment tool to facilitate informed decision-making for themselves and their patients. Emergency laparotomy procedures have a high risk profile, but the currently available risk-assessment models for emergency laparotomy are either unreliable (eg, small sample size or single center study), difficult to calculate preoperatively, or are specific to the geriatric population.

The American College of Surgeons National Surgical Quality Improvement Program database (2005 to 2009) was used to develop logistic regression models for 30-day mortality after emergency laparotomy. Two models were created, one with the knowledge of the postoperative diagnosis and one without. Models' calibration and discrimination were judged using the receiver operating characteristics curves and the Hosmer-Lemeshow test.

There were 37,553 patients who had undergone emergency laparotomy, with a 14% mortality rate. The American Society of Anesthesiologists classification system, functional status, sepsis, and age were the variables most significantly associated with mortality. Patients older than 90 years of age, with an American Society of Anesthesiologists class V, septic shock, dependent functional status, and abnormal white blood cell count have a <10% probability of survival.

The models developed in this study have a high discriminative ability to stratify the operative risk in a broad range of acute abdominal emergencies. These data will assist surgeons, patients, and their families in making end-of-life decisions in the face of medical futility with greater certainty when emergency surgery is being contemplated.