Surgical management of aortic valve disease in the elderly: A retrospective comparative study of valve choice using propensity score analysis.

BACKGROUND AND AIM OF THE STUDY

Aortic valve dysfunction is the most common form of valvular heart disease. As the population continues to age, a greater number of patients will become candidates for aortic valve replacement (AVR); hence, prosthetic valve choice becomes of paramount importance.

METHODS

A retrospective analysis was conducted on 801 patients aged > or =65 years who underwent isolated AVR or AVR + coronary artery bypass grafting (CABG) between January 1989 and June 2003 with a Carpentier Edwards Perimount (CEP) pericardial bioprosthesis (n = 398) or a St. Jude Medical (SJM) mechanical valve (n = 403). The mean age of CEP patients was 74.5 years (range: 65-89 years), and of SJM patients 73.9 years (range: 65-90 years). The follow up was 96.2% and 96.5% complete for CEP and SJM patients, respectively. Propensity scoring was used to establish homogeneity of the groups and reduce bias.

RESULTS

The operative mortality was 4.0% (n = 16) among CEP patients and 6.5% (n = 26) among SJM patients. Predictors of hospital mortality included: peripheral vascular disease (p = 0.018), surgical urgency (p = 0.010), preoperative intra-aortic balloon pump (IABP) (p = 0.010), intraoperative perfusion time (p = 0.046) and intraoperative IABP (p = 0.001). Postoperative morbidities were similar for the two groups. The mean follow up was 72.4 and 59.2 months for CEP and SJM patients, respectively. The five-year actuarial survival was 70.9 +/- 2.3% for CEP and 71.8 +/- 2.4% for SJM patients; at 10 years the actuarial survival was 32.6 +/- 3.3% and 38.2 +/- 3.8%, respectively. Freedom from reoperation for AVR, stroke and non-fatal myocardial infarction was 98.8% (159/161), 99.4% (160/161) and 99.4% (160/161), respectively, in CEP patients, and 100.0% (220/220), 97.7% (215/220) and 97.7% (215/220), respectively, in SJM patients (p = NS). Predictors of late death (>30 days) included chronic obstructive pulmonary disease (p = 0.001) and mechanical valve replacement (p = 0.001).

CONCLUSION

In comparable elderly patients, the outcomes of CEP and SJM valves after AVR showed no significant differences in hospital morbidity, mortality, mid-term survival or late cardiac events. However, the cumulative risk of lifelong anticoagulation with a mechanical valve is a serious consideration that must be factored into the selection algorithm.