Aortic valve replacement in the elderly: a comparative assay of potential risk factor modification.

From January 1989 to October 1992, 208 consecutive patients underwent isolated aortic valve replacement for calcified aortic stenosis in our department. Since the mean age of this patient population was 70 +/- 9 years, a retrospective clinical study was completed to assess the potential influence of advanced age on the independent predictors of early and late mortality. Hospital mortality was 6.2% (13 patients). Total follow up was 422.5 patient-years with a mean of 26 months. Nineteen patients died during the follow up period, equivalent to 4.5% per patient-year late mortality rate. Survival including hospital death was 88 +/- 2%, 86 +/- 2% and 79 +/- 4% at one, two and three years respectively. Eighteen variables as potential predictors of early and late mortality were studied. Predictors of hospital mortality were determined by logistic regression analysis, and those of late mortality by Cox proportional hazard model. Results were expressed as odds ratio (OR) or relative risk (RR). Age greater than 70 years (OR = 9.8, 95% CI = 1.2 to 80) and emergency surgery (OR = 8, 95% CI = 2.1 to 31) appeared as independent predictors of hospital mortality in multivariate analysis. Age above 75 years (RR = 3, 95% CI = 1.1 to 8.3), preoperative acute pulmonary edema (RR = 2.9, 95% CI = 1.1 to 7.7) and emergency surgery (RR = 4.2, 95% CI = 1.2 to 15) were independently associated with decreased late survival. Advanced functional class (NYHA III-IV) was shown to be an independent predictor of early or late mortality only in univariate analysis.(ABSTRACT TRUNCATED AT 250 WORDS)