PURPOSE: Recently, different scoring systems to assess operative risk were adopted by cardiac surgeons. The EuroSCORE and the Society of Thoracic Surgeons Mortality Risk (STS) Score are widely used and validated among US patient populations. The study aim is to explore potential differences in the performance of the EuroSCORE and the STS Scores in the prediction of operative mortality following coronary artery bypass surgery cardiac surgery (CABG) with special focus on the impact of gender.
METHODS: We retrospectively reviewed 3,125 consecutive CABG cases performed at our institution between 2001 and 2004. STS and EuroSCORE (logistic [E-Log] and additive [E-Add]) operative mortality risk scores were calculated and stratified by gender (female: n=692; male: n=2,433). Mortality risk scores were compared between the STS and EuroSCORE using receiver operator curves (ROC), likelihood ratios (LR) and Pearson Correlation coefficients.
RESULTS: Stratified by gender, the E-Log and E-Add correlated very well with the STS (female: r=0.51, 0.49, p< 0.001; male: r=0.68, 0.61, p< 0.001). Using a ROC analysis across all categories, both EuroSCORE models correctly modeled operative mortality compared to the STS. Among female patients, the EuroSCORE area under curve (AUC) (E- log, 0.853; E - add, 0.855) performed much more comparable to the STS AUC (0.827) compared to male patients (E- log, 0.808; E - add, 0.809; STS, 0.796). For fixed sensitivities, female patients consistently have higher cut points for the prediction of mortality for all scores. In addition, across all scores for fixed sensitivities, female patients demonstrate LR+ scores nearly double those of their male counterparts.
CONCLUSION: Among CABG patients, the EuroSCORE and STS scores are very similar in predicting mortality. Stratified by gender, the EuroSCORE mortality estimates were more accurate among female patients. Furthermore, while the STS predicted mortality estimates are closer to future actual observed estimates; its complex calculations render it a more useful tool for administrative purposes.
CLINICAL IMPLICATIONS: The EuroSCORE, however, is an excellent predictor of mortality and provides the clinician a user-friendly tool at the bedside.
DISCLOSURE: Niv Ad, None.