# Prediction of mortality based on the EuroSCORE II model in patients undergoing cardiovascular surgery

## TL;DR

This study evaluates how well the EuroSCORE II model predicts mortality and complications in patients undergoing cardiac surgery.

## Contribution

The study validates the EuroSCORE II model's predictive accuracy in a specific population and suggests potential improvements.

## Key findings

- The EuroSCORE II model showed good discrimination power with an AUC-ROC >0.7 for 10 out of 12 outcomes.
- The AUC for predicting mortality was 0.749, indicating moderate to good predictive accuracy.
- The model's calibration was assessed using the Hosmer–Lemeshow test, showing acceptable fit.

## Abstract

This study aims to evaluate the performance of the EuroSCORE II (European System for Cardiac Operative Risk Evaluation) model in predicting mortality and postoperative complications in adult patients undergoing cardiac surgery.

In this prospective longitudinal study, 1,173 patients who underwent cardiac surgery between August 2021 and May 2022 were included to assess the predictive accuracy of the EuroSCORE II model for mortality and 11 major cardiovascular complications. Patients were followed up for 30-day and in-hospital mortality, as well as for the occurrence of 11 major cardiovascular complications. The model’s discriminative power was evaluated using the area under the receiver operating characteristic (AUC-ROC) curve, while its calibration was assessed through the goodness-of-fit test (Hosmer–Lemeshow test).

The data analysis showed that the area under the ROC curve of the EuroSCORE II model, used to predict the post-cardiac surgery outcomes was>0.7 in 10 out of 12 outcomes, which indicates good discrimination power. The area under curve (AUC) for predicting mortality was 0.749. The model calibration was assessed through the Hosmer–Lemeshow (H-L) goodness-of-fit test. Other findings including sensitivity, specificity and cutoff were also calculated, revealing the fitness of the prediction model.

According to the findings, considering the power of differentiation and calibration of the EuroSCORE II model in the studied population, this model remains a valuable risk stratification tool, integrating additional predictive models or clinical parameters may enhance accuracy for certain postoperative outcomes.