Valvular surgery in the elderly.

Abstract

In previous studies from this university, advanced age was identified as an independent predictor of operative mortality for valvular surgery. Therefore, between January 1982 and December 1986, early results were compared in 469 patients greater than 70 years old (Old; 74.0 +/- 3.4 years, mean +/- SD) and in 2,040 patients less than 70 years old (Young; 53.7 +/- 12.1 years). Patients underwent single- or multiple-valve repair or replacement with or without concomitant coronary artery bypass surgery. Data consisting of 31 clinical and angiographic variables were collected prospectively and were analyzed by univariate and multivariate statistics. Old patients were characterized by having more frequent left ventricular dysfunction (left ventricular ejection fraction less than 40%: Old, 33.5%; Young, 22.4%; p less than 0.001), coronary artery disease (Old, 54.5%; Young, 24.2%; p less than 0.001), and urgent surgery (Old, 15.6%; Young, 11.8%; p = 0.02). Aortic valve procedures (Old, 59.7%; Young, 40.1%; p less than 0.001) and concomitant coronary artery bypass surgery (Old, 46.4%; Young, 20.3%; p less than 0.001) were performed more commonly in the Old patient. Operative mortality occurred in 10.0% of Old patients compared with 5.6% of Young patients (p less than 0.001), and major morbidity (low-output syndrome, perioperative myocardial infarction, intra-aortic balloon pump counterpulsation, and stroke) occurred significantly more frequently (p less than 0.001) in the Old. Stepwise logistic regression identified that urgent operation (p = 0.002), mitral or double-valve surgery (p = 0.004), coronary artery disease especially when not treated by bypass surgery (p = 0.02), female gender (p = 0.02), and left ventricular dysfunction (p = 0.05) independently predicted operative death in the Old population. The significant predictors of mortality in the Young patient were urgent operation (p less than 0.001), New York Heart Association class IV (p less than 0.001), associated tricuspid valve disease (p = 0.002), decreased left ventricular function (p = 0.01), valvular re-replacement (p = 0.02), and increasing age (p = 0.03). The predicted probability of operative mortality in Old patients ranged between 0.9 +/- 0.5% and 76 +/- 16%. Elderly patients in good risk categories should be offered surgical intervention for correction of valvular lesions. Alternative therapy may be indicated in patients with multiple risk factors.