Modified Early-Warning Score combined with early-warning symptoms and electrocardiographic findings in predicting in-hospital cardiac arrest in critically ill patients: a retrospective cohort study

Abstract

Background We aimed to investigate the clinical value of the Modified Early-Warning Score (MEWS) combined with early-warning symptoms and electrocardiogram (ECG) findings in predicting in-hospital cardiac arrest (IHCA) in critically ill patients, to assess and reduce the occurrence of IHCA. Methods This retrospective cohort study examined critically ill patients who were enrolled in a hospital from January 2019 to March 2023 and divided into an IHCA group and NO-IHCA group. The critically ill patients were randomly divided into 2 sets at the ratio of 7:3, for the training set and test set. The training set used to develop the model and the test set used to test the model. Univariate and multivariate logistic regressions were used to determine the independent predictors. The generated prediction models were evaluated using 10-fold cross verification, and the areas under the curve (AUCs), accuracy, sensitivity, and specificity were reported. Hosmer-Lemeshow goodness of fit test was used to compare the calibration degree of the model and Delong test was used to compare the AUC. Results Multivariate logistic analysis showed that MEWS, early-warning symptoms, and ECG findings were independent risk factors for IHCA in critically ill patients (P < 0.05). The AUC values for MEWS, early-warning symptoms, and ECG findings were 0.671, 0.527, and 0.723, respectively. The AUC value for the combination of MEWS, early-warning symptoms, and ECG findings was 0.792 (P < 0.001), which was higher than MEWS. Conclusion MEWS combined with early-warning symptoms and ECG findings can predict IHCA in critically ill patients, which may help reduce IHCA in this population.