Predictors of mortality and morbidity for acute care surgery patients

As the implementation of exclusive acute care surgery (ACS) services thrives, prognostication for mortality and morbidity will be important to complement clinical management of these diverse and complex patients. Our objective is to investigate prognostic risk factors from patient level characteristics and clinical presentation to predict outcomes including mortality, postoperative complications, intensive care unit (ICU) admission and prolonged duration of hospital stay.

Retrospective review of all emergency general surgery admissions over a 1-year period at a large teaching hospital was conducted. Factors collected included history of present illness, physical exam and laboratory parameters at presentation. Univariate analysis was performed to examine the relationship between each variable and our outcomes with chi-square for categorical variables and the Wilcoxon rank-sum statistic for continuous variables. Multivariate analysis was performed using backward stepwise logistic regression to evaluate for independent predictors.

A total of 527 ACS admissions were identified with 8.1% requiring ICU stay and an overall crude mortality rate of 3.04%. Operative management was required in 258 patients with 22% having postoperative complications. Use of anti-coagulants, systolic blood pressure <90, hypothermia and leukopenia were independent predictors of in-hospital mortality. Leukopenia, smoking and tachycardia at presentation were also prognostic for the development of postoperative complications. For ICU admission, use of anti-coagulants, leukopenia, leukocytosis and tachypnea at presentation were all independent predictive factors. A prolonged length of stay was associated with increasing age, higher American Society of Anesthesiologists class, tachycardia and presence of complications on multivariate analysis.

Factors present at initial presentation can be used to predict morbidity and mortality in ACS patients.