The STS Tricuspid score showed no predictive value for intra-hospital mortality (AUC 0.54) compared to Euroscore II (AUC 0.765) and MELD (AUC 0.719) in isolated tricuspid valve surgery.
Does the STS Tricuspid score accurately predict mortality compared to ESII, MELD, and TRISCORE in patients undergoing isolated tricuspid valve surgery?
The newly developed STS Tricuspid score failed to predict in-hospital, 1-year, and 5-year mortality in patients undergoing isolated tricuspid valve surgery, underperforming compared to Euroscore II, MELD, and TRISCORE.
Absolute Event Rate: 0% vs 0%
Abstract Background In last decades, new scores for mortality prediction on isolated tricuspid valve surgery have been adopted: from Euroscore II and MELD to TRISCORE. A new score by Society of Thoracic Surgeon (STS), has been developed to assess specifically the mortality risk in case of tricuspid valve intervention, named STS Tricuspid score (STS-TR).The aim of the present study is, for the first time, to validate STS-TR on a large cohort, compare it with the others in use (ESII,MELD,TRISCORE) and define from all the variables of the 4 scores, which are the most relevant in real world practice to determine intra-hospital,1 year and 5 years mortality. Patient 325 patients from 2 high-volume centers, operated with isolated tricuspid valve surgery between 1997 and 2024 were retrospectively analysed and follow-up with echocardiographic and clinical outcome. Results In ROC analysis, AUC for ESII was 0.765, MELD 0.719, TRISCORE 0.69 and STS Tricuspid 0.54, for intra-hospital mortality showing no predictive value of STS-TR (Figure 2). ESII and MELD showed the best predictive power compared to TRISCORE and STS-TR At 1 year TRISCORE and ESII predict mortality at Beta value (0.081 and 0.053) more than STS-TR (0.004). At 5 years TRISCORE and ESII predict mortality at Beta value (0.129 and 0.058) more than STS-TR (0.015) We identified the intra-hospital mortality according to numbers of risk factors (Figure 5). We identified, at multivariate analysis, the 5 risk factors that better predict the intra-hospital mortality from all variable used in the 4 scores, (Figure 6). Conclusions •To the best of our knowledge, this is the first validation study of the new STS-TR score developed by STS. •STS-TR showed no predictability power for in-hospital, 1 year and 5 years mortality. •EUROSCORE II, MELD and TRISCORE showed a more accurate predictability in short and long term, with better performance of ESII and MELD for intra-hospital mortality, and TRISCORE with ESII for short and long term (1 and 5 years).For image description, please refer to the figure legend and surrounding text. For image description, please refer to the figure legend and surrounding text.
Zancanaro et al. (Sun,) reported a other. The STS Tricuspid score showed no predictive value for intra-hospital mortality (AUC 0.54) compared to Euroscore II (AUC 0.765) and MELD (AUC 0.719) in isolated tricuspid valve surgery.