Aims: Risk scores provide an important contribution to clinical decision-making, but their validity has been questioned in patients with valvular heart disease (VHD), since current scores have been mainly derived and validated in adults undergoing coronary bypass surgery. The Working Group on Valvular Heart Disease of the European Society of Cardiology reviewed the performance of currently available scores when applied to VHD, in order to guide clinical practice and future development of new scores.
Methods and results: The most widely used risk scores (EuroSCORE, STS, and Ambler score) were reviewed, analysing variables included and their predictive ability when applied to patients with VHD. These scores provide relatively good discrimination, i.e. a gross estimation of risk category, but cannot be used to estimate the exact operative mortality in an individual patient because of unsatisfactory calibration.
Conclusion: Current risk scores do not provide a reliable estimate of exact operative mortality in an individual patient with VHD. They should therefore be interpreted with caution and only used as part of an integrated approach, which incorporates other patient characteristics, the clinical context, and local outcome data. Future risk scores should include additional variables, such as cognitive and functional capacity and be prospectively validated in high-risk patients. Specific risk models should also be developed for newer interventions, such as transcatheter aortic valve implantation.