Aortic valve replacement in octogenarians: utility of risk stratification with EuroSCORE

With the advent of percutaneous valve implantation, an increasing amount of interest is being expressed in outcomes of conventional aortic valve replacement (AVR) in elderly patients. We evaluated characteristics and outcomes of elderly patients undergoing isolated AVR with a particular focus on the European System for Cardiac Operative Risk Evaluation (EuroSCORE) risk stratification.

All patients aged 80 years or older (n = 282) undergoing isolated AVR between November 1995 and June 2006 at our institution were reviewed according to logistic EuroSCORE (ES(log)) risk stratification. Surgical risk was defined as low risk (ES(log) < or = 10% [n = 107]), moderate risk (10% < ES(log) < 20% [n = 103]), and high risk (ES(log) > or = 20% [n = 72]). Patient age was 82 +/- 2 years (low risk), 82.7 +/- 2.7 years (moderate risk), and 83.6 +/- 3.1 years (high risk), respectively (p < 0.05). Mean ES(log) predicted risk of mortality was 7.3% +/- 1.4% (low risk), 13.7% +/- 2.5% (moderate risk), and 33.0% +/- 11.5% (high risk; p < 0.05). Follow-up was 99.7% complete.

In-hospital mortality was 7.5% (low risk), 12.6% (moderate risk), and 12.5% (high risk; p = 0.4). One-year survival was 90%, 78%, and 69% (p = 0.002); 5-year survival was 70%, 53%, and 38% (p = 0.05); and 8-year survival was 38%, 33%, and 21% (p = 0.017), for low-, moderate-, and high-risk patients, respectively. Independent predictors for in-hospital mortality were pulmonary hypertension and urgent indication for surgery. Cox regression predictors of medium-term survival were congestive heart failure, urgent timing, previous stroke or transient ischemic attack, and EuroSCORE stratum.

Aortic valve replacement can be performed in the elderly population with acceptable outcomes. EuroSCORE risk stratification is imprecise for prediction of perioperative mortality among octogenarian AVR patients, but may be useful for predicting mortality during medium-term follow-up.