Comparison of techniques for implantation of aortic valve allografts

Background. Various implantation techniques for allograft aortic valve replacement have evolved over the years. Our objective was to examine the effects of different implantation methods on subsequent valve performance and durability. Methods. Between May 1985 and January 1994, 137 patients underwent allograft aortic valve replacement. The first 59 aortic valve allografts were inserted by the freehand scalloped technique with removal of the aortic sinuses, and the last 78 valves were inserted by the cylinder technique, in which the aortic sinuses and sinotubular junction were retained. The mean age of the 91 men and 46 women was 53.7 years (range, 18 to 83 years). Preoperative diagnoses were aortic stenosis (n = 57), aortic regurgitation (AR, n = 40) and aortic stenosis/AR (n = 40); 27 patients had prior aortic valve operations and 1 patient had a previous heart transplantation. Active endocarditis was present in 29 patients. Associated procedures included coronary artery bypass (n = 33), ascending aneurysm repair (n = 4), left ventricular aneurysmectomy (n = 3), repair of atrial septal defect (n = 2), mitral valve repair or replacement (n = 6), and aortic root enlargement (n = 24). Follow-up was complete in 133 patients (97%) a mean of 4.9 years (range, 1 day to 9.8 years) after allograft aortic valve replacement. Results. Operative mortality was 6.5% for all patients but only 1.9% for patients without infection having isolated aortic valve replacement. Early echocardiography (mean of 8.4 days postoperativety) demonstrated no AR or mild AR and a mean gradient of 10.6 ± 6.2 mm Hg in all patients. The cumulative risk of development of grade III or IV AR at 7 years postoperatively was 26.2% ± 6.3% in the scallop group and 12.4% ± 5.6% in the cylinder group (p = 0.4). Late postoperatively, transvalvular gradient by echocardiography was 13.1 ± 9.4 mm Hg, and was similar in the two study groups. Late AR led to reoperation in 13 patients (22%) who had initial implantation with the scallop method and only 4 patients (5.4%) who had the valve inserted with the cylinder method. However, because duration of follow-up was longer for patients in the scallop group, cumulative risk of reoperation was similar at 5 years postoperativety (scallop, 13.7% [95% confidence interval, 76.7% to 95.8%]; cylinder, 11.5% [95% confidence interval, 75.5% to 99.1%]). Conclusions. The insertion of an aortic valve allograft as a cylinder, retaining the sinotubular junction, appears to result in less aortic regurgitation at 7 years postoperativety, and with additional follow-up may result in less reoperation for AR.