Pulmonary autografts offer advantages over aortic homografts, but carry additional risks. The interim results of a prospective randomized trial of autograft versus homograft aortic valve replacement (AVR) were reviewed to determine if the greater complexity of the autograft is justified, notably with regard to hemodynamic function. A total of 182 patients (mean age 37.2 ± 14.3 years; range: 2-64 years) with isolated aortic valve disease underwent pulmonary autograft (group A, n = 97) or aortic homograft (group H, n = 85). Among patients, 42% had previous aortic valve surgery and 19% native or prosthetic valve endocarditis. Follow up was by annual examinations and echocardiography. Autograft AVR required longer cross-clamp (41%) and bypass (43%) times, but did not cause more bleeding, longer recovery or more complications. One 30-day death occurred in group A, and three deaths in group H. Median follow up was 33.9 months (range: 1-61 months). There was one late death in each group, three reoperations in group A, and three in group H. There were no autograft reoperations, and no other valve-related events. At 48 months, actuarial survival and reoperation-free survival rates were 97.8% and 94.2% in group A, and 95.3% and 87.7% in group H (p = NS). Echocardiography showed near-perfect function in all autografts, but early signs of subclinical dysfunction in many homografts. In conclusion, both autograft and homograft AVR are safe and produce good intermediate-term results. Early homograft degeneration may favor autografts in children. Echocardiographic findings may translate into superior long-term autograft durability and hemodynamics.