Pulmonary autograft aortic root replacement (Ross' operation) is now associated with low operative risk. Recent series suggest that patients with primary aortic insufficiency have diminished autograft durability and that patients with large discrepancies between pulmonary and aortic valve sizes have a low but consistent rate of mortality. Therefore, Ross' operation in these patients has been avoided when possible at Texas Children's Hospital. Our objective was to report outcomes of Ross' operation when selectively employed in pediatric patients with aortic valve disease. Between July 1996 and February 2006, 55 patients (mean age 6.8 +/- 5.5 years) underwent Ross' procedure. Forty-seven patients (85%) had a primary diagnosis of aortic stenosis, three (5%) patients had congenital aortic insufficiency, and five (9%) patients had endocarditis. Forty-two (76%) patients had undergone prior aortic valve intervention (23 [55%] percutaneous balloon aortic valvotomies, 12 [29%] surgical aortic valvotomies, 12 [29%] aortic valve replacements, 2 [5%] aortic valve repairs). Fourteen (25%) patients had >or=2 prior aortic valve interventions. Thirty-two patients (58%) had bicuspid aortic valves. Follow-up was 100% at a mean of 3 +/- 2.5 years. Hospital and 5-year survival were 100% and 98%, respectively. Morbidity included one reoperation (2%) for bleeding. Median length of hospital stay was 6 days (3 days-3 months). Six (11%) patients needed a right ventricular to pulmonary artery conduit exchange at a median time of 2.3 years. Freedom from moderate or severe neoaortic insufficiency at 6 years is 97%. Autograft reoperation rate secondary to aortic insufficiency or root dilation was 0%. By selectively employing Ross' procedure, outcomes of the Ross procedure in the pediatric population are associated with minimal autograft failure and mortality at mid-term follow-up.