Assessing operative risk and benefit in elderly patients with heart valve disease

Abstract

The prevalence of heart valve disease increases with age, with the predominance of degenerative aetiology, as shown by the Euro Heart Survey on valvular heart disease.1 Advanced age is one of the pre-eminent risk factors of mortality and major morbidity after cardiac surgery. Nevertheless, several reports (including randomized studies) showed that consistent benefits can be achieved from cardiac surgery and percutaneous valve therapies in elderly patients,2–5 in terms of quality-of-life improvement, alleviation of symptoms, prevention of major adverse events, and increased survival. The clinical decision-making and individual risk profile evaluation in this context have been so far particularly challenging, leading to the clinical question of utility vs. futility. Concomitantly, the economic burden in high-risk elderly patients could have crucial implications and should be considered before proceeding, due to reduced expectancy of life. Moreover, ethical implications should be taken into account. The therapeutic decision in this context should be reached through a shared decision-making process, not only by a multidisciplinary ‘heart team’ discussion, but also by informing the patient thoroughly, and finally by deciding with the patient and family which treatment option is optimal, taking into account the patient perspective.6 Heart valve diseases in the elderly are often associated with a consistent number of extracardiac comorbidities, mainly related to atherosclerosis.1 In the clinical routine, the two surgical risk scores mostly used are the Logistic EuroScore (http://www.euroscore.org/calcold.html) and the STS score (http://riskcalc.sts.org/STSWebRiskCalc273/). Although both of them are quite accurate in predicting mortality in low-risk surgical candidates, they present several limits in the subset of the high-risk elderly patients. These scores provide relatively good discrimination (a gross estimation of risk category), but cannot be used to estimate an accurate operative mortality in an individual patient, because of unsatisfactory calibration (comparison between predicted and observed mortality). The value …