Abstract

Objective: To determine the predictive accuracy of Modified Early Warning Score (MEWS) Model for surgical re-exploration and mortality following emergency laparotomy.
Methodology: This prospective observational study was conducted in Department of Surgery, Civil Hospital/Dow University of Health Sciences Karachi from January 2022 to July 2022 with non-probability consecutive sampling technique. Patients aged 20 to 80 years, of either gender who were operated secondary to secondary peritonitis were included. The primary variables derived from vital signs were enumerated daily after admission in ward and outcome variables were computed electronically for up to 2 weeks. The predictive accuracy of the model in terms two outcomes including re-exploration and incidence of mortality was assessed by binary logistic regression of the model using ROC in SPSS version 23.
Results: Out of 139 cases, males comprised 53.8% and females 46.2% with a male-to-female ratio of 1.2:1. Model predicted re-exploration in 26.9% of case, with sensitivity of 89.29% and specificity as 86.84%, while in-hospitality mortality was 4.8% with sensitivity and specificity of 80.0% and 81.82%, respectively. The overall predictive accuracy was estimated 87.50% for re-exploration and 81.73% for in-hospital mortality. The AUC was derived as 0.8807 and 0.8091 in terms of surgical re-exploration and mortality respectively.
Conclusion: MEWS can serve as a valuable model in surgical ward for identifying patients at risk re-exploration or in-hospital mortality allowing healthcare professionals to provide targeted and timely interventions which could support early clinical decision-making in resource limiting settings.

Key words: Early Warning Score, laparotomy, peritonitis, predictive value, re-operation, hospital mortality.