Characterizing the underlying diagnoses for exploratory laparotomies to improve risk-adjustment models of postoperative mortality

BACKGROUND Surgeons perform emergent exploratory laparotomies (ex-laps) for a myriad of surgical diagnoses. We characterized common diagnoses for which emergent ex-laps were performed and leveraged these groups to improve risk-adjustment models for postoperative mortality. METHODS Using American Association for the Surgery of Trauma criteria, we identified hospitalizations where the primary procedure was an emergent ex-lap in the 2012 to 2014 (derivation cohort) and 2015 (validation cohort) Nationwide Inpatient Sample. After tabulating all International Classification of Diseases-9th Rev.-Clinical Modification diagnosis codes within these hospitalizations, we divided them into clinically relevant groups. Using two stepwise regression paradigms-forward selection and backward elimination-we identified diagnostic groups significantly associated with postoperative mortality in multivariable logistic regressions. We evaluated the addition of these groups as individual covariates in risk-adjustment models for postoperative mortality using the area under the receiver operator characteristic curve. All regressions additionally adjusted for clinical factors and hospital clustering. RESULTS We identified 4127 patients in the derivation cohort (median age, 50 years; 46.0% female; 62.1% white), with an overall mortality rate of 13.4%. Among all patients, we tabulated a total of 164 diagnosis codes, of which 27 (16.5%) may have led to an emergent ex-lap. These 27 codes clinically represented seven diagnostic categories, which captured a majority of the patients (70.4%). Backward elimination and forward selection led to four common diagnosis categories associated with mortality: bleeding, obstruction, shock, and ischemia. Adjusting for these four diagnostic groups in a multivariable logistic regression assessing postoperative mortality increased the area under the receiver operator characteristic curve from 74.5% to 88.2% in the derivation cohort and from 73.8% to 88.2% in the validation cohort. CONCLUSION Seven diagnostic groups account for the majority of the emergent ex-laps. Adjusting for four groups may improve the accuracy of risk-adjustment models for mortality and validating such analytic standardization may optimize best research practices for EGS procedures. LEVEL OF EVIDENCE Prognostic and epidemiologic, Level III.