Abstract
BACKGROUND: Morning postoperative labs are often obtained for emergency general surgery (EGS) patients. Studies in other surgical fields indicate that routine postoperative day 1 (POD1) labs are sometimes being performed excessively and do not require intervention. The purpose of this study is to identify predictors indicating the need for POD1 labs in EGS patients based on likelihood of intervention. METHODS: This is a retrospective review of non-critically ill EGS patients from 2022 to 2023 who received POD1 morning labs. The odds of having an abnormal result and likelihood of intervention were measured through multivariate logistic regression accounting for patient characteristics and procedure. Least absolute shrinkage and selection operator (LASSO) regression analysis was performed to determine significant predictors of an abnormal result and intervention. RESULTS: 502 EGS patients were included. LASSO revealed that procedure duration, fever, lysis of adhesions, preoperative systolic blood pressure <90 mm Hg, older age, heart failure, operative blood loss, chronic kidney disease, and anticoagulation use were independent predictors for any abnormal result (area under the receiver operation curve (AUC)=0.785). Independent predictors of intervention were procedure duration, older age, higher estimated blood loss (EBL), anticoagulant use, and lysis of adhesions (AUC=0.704). Procedures >400 min carried an 84.3% chance of an abnormal lab requiring intervention. EBL >200 mL carried a 75.5% chance of an abnormal lab requiring intervention. CONCLUSION: POD1 labs for non-critically ill EGS patient rarely require intervention and can be safely omitted. Labs should be considered for longer procedures, higher EBLs, older patients, those on anticoagulation, or after lysis of adhesions.