Aortic valve replacement in the elderly. Risk factors and long-term results.

OBJECTIVE: The current study was undertaken to determine long-term results of aortic valve replacement (AVR) in the elderly, to ascertain predictors of poor outcome, and to assess quality of life. SUMMARY BACKGROUND DATA: Aortic valve replacement is the procedure of choice for elderly patients with aortic valve disease. The number of patients aged 70 and older requiring AVR continues to increase. However, controversy exists as to whether surgery devoted to this subset reflect a cost-effective approach to attaining a meaningful quality of life.
METHODS: This study reviews data on 247 patients aged 70 to 89 years who underwent isolated AVR between 1980 and 1995; there were 126 men (51%) and 121 women (49%). Follow-up was 97% complete (239/247 patients) for a total of 974.9 patient-years. Mean age was 76.2 +/- 4.8 years. Operative mortality and actuarial survival were determined. Patient age, gender, symptoms, associated diseases, prior conditions, New York Health Association class congestive heart failure, native valve disease, prosthetic valve type, preoperative catheterization data, and early postoperative conditions were analyzed as possible predictors of outcome. Functional recovery was evaluated using the SF-36 quality assessment tool.
RESULTS: Operative mortality was 6.1% (15/247). Multivariate logistic regression showed that poor left ventricular function and preoperative pacemaker insertion were independent predictors of early mortality. After surgery, infection was predictive of early mortality. Overall actuarial survival at 1, 5, and 10 years was 89.5 +/- 2% (198 patients at risk), 69.3 +/- 3.4% (89 patients at risk), and 41.2 +/- 6% (13 patients at risk), respectively. Cox proportional hazards model showed that chronic obstructive pulmonary disease and urgency of operation were independent predictors of poor long-term survival. Postoperative renal failure also was predictive of poor outcome. Using the SF-36 quality assessment tool, elderly patients who underwent AVR scored comparably to their age-matched population norms in seven of eight dimensions of overall health. The exception is mental health.
CONCLUSIONS: Aortic valve replacement in the elderly can be performed with acceptable mortality. Significant preoperative risk factors for early mortality include poor left ventricular function and preoperative pacemaker insertion. Predictors of late mortality include chronic obstructive pulmonary disease and urgency of operation. These results stress the importance of operating on the elderly with aortic valve disease; both long-term survival and functional recovery are excellent.