Prediction of Early and Late mortality outcome in patients undergoing Valvular Heart surgeries using EuroSCORE II at Skims University Teaching Hospital

Abstract

BACKGROUND: The European System for Cardiac Operation Risk Evaluation (EuroSCORE) is one of the possible tools to assess the peri-operative mortality risk in patients undergoing cardiac valvular  surgieries. The first EuroSCORE, initially based on an additive system derived by a logistic regression model, was developed on data collected in 1995 and reported in 1999 to predict in-hospital or 30-day mortality . Recently, in order to improve the poor calibration and to optimize EuroSCORE usefulness, an  updated version of this model named as EuroSCORE II, was better calibrated than the original model in detecting hospital mortality preserving high discrimination.

OBJECTIVES:1.Preoperative EuroSCORE Calculation in cardiac surgical patients undergoing valvular heart surgeries in  at SKIMS ,SOURA.

2.Analysing one week, one month, 6 months, 1 year, mortality in these patients.

3. A Co-relation of preoperative risk evaluation by  EuroSCORE with postoperative mortality at our institute and its comparison with the literature.    

STUDY DESIGN: A prospective observational study

PARTICIPANTS: A total of 56 patients of any gender, age of more than 18 years, admitted for elective or urgent cardiac valve surgeries within a period of two years were included in the study.

METHODS: Baseline and peri-operative data was  collected prospectively which included the variables needed to estimate the operative risk based on EuroSCORE II. The risk score was determined with an online interactive  calculator (http://www.euroscore.org/calc.html).Patients were followed prospectively in the post-operative period at one week, one month, at six months and at the end of one year. The primary end point was in hospital mortality and mortality upto one year. Early mortality was taken as mortality at the end of one week and one month and late mortality was  taken at the end of six months and one year. During follow up, data was obtained through outpatient clinical review.

 RESULTS: The Hosmer-Lemeshow goodness-of-fit test did not show significant difference between expected and observed mortality in accordance to the EuroSCORE II model (Chi-square = 14.23, p = 0.076) suggesting good calibration of the model in this population for early mortality. The area under the receiver operating characteristic (ROC) curve for EuroSCORE II is 0.85 (95% CI 0.712-0.921) indicating a good discriminatory power for early mortality.The area under the receiver operating characteristic (ROC) curve of EuroSCORE II for predicting late mortality is 0.493 (95% CI 0.367-0.641) indicating a poor discriminatory power.

CONCLUSIONS: From our study we concluded that EuroSCORE II has a good discriminatory power and calibration for prediction of early mortality (in hospital mortality and mortality upto 1 month). The EuroSCORE II estimated risk should not be considered a direct measure of long term mortality as its predictive power(performance) decrease significantly for mortality at a follow up longer than 30 days.

IMPLICATIONS: EuroSCORE II helps us to predict perioperative mortality of valvular heart surgery patients.