OBJECTIVE: To describe our experience in the surgical treatment of infective, native and prosthetic aortic valve endocarditis, using cryopreserved homograft valves. METHODS: Between January 1988 and September 1995, cryopreserved homografts were implanted in 49 patients (mean age 47 +/- 15 years; range 19-79) with acute infective endocarditis of the native (21/49; 43%) or the prosthetic (28/49; 57%) aortic valve. Aortic root abscesses were found in 39/49 (80%) patients, ventriculo-aortic disconnection in 27/49 (55%). An intracardiac fistula, originating from the left ventricular outflow tract was found in 25/49 (51%) patients. Indications for emergency surgery were congestive heart failure due to severe aortic valve regurgitation in 44/49 (90%) and systemic emboli in 5/49 (10%) patients. Preoperatively, 23/49 (47%) patients were in New York Heart Association (NYHA) class IV, and 5/49 (10%) were in acute circulatory failure. Mean left ventricular ejection fraction was 53 +/- 10% (25-65). Streptococci (27%) and staphylococci (27%) were the most important microorganisms found. The homograft was implanted as a scalloped freehand valve (34/49; 70%), as an intra-aortic inclusion cylinder (4/49; 6%) or as a free-standing root replacement (12/49; 24%). Combined procedures were necessary in 11/49 (22.5%) patients. RESULTS: Hospital mortality was 8.2% (4/49): 2/49 (4.1%) patients died from endocarditis-related sepsis, one (2%) from low cardiac output and one (2%) from a cerebrovascular accident. After a mean interval of 21 +/- 15 months (2-48), 9/45 (20%) patients had to be reoperated, all reoperations except one being homograft related. After a mean follow-up of 35 +/- 22 months (2-90), 4/44 (9%) patients had their homograft replaced by a mechanical prosthesis. After 5 years, actuarial freedom from late death was 97 +/- 3%; from late reoperation 69 +/- 9%; from late endocarditis 85 +/- 8%; and from late homograft degeneration 87 +/- 6%. Explanted homografts were acellular and non-vital, containing bacteria and/or leucocytes. B-lymphocytes were found in all and in one, T-cell lymphocytes were present. CONCLUSION: Emergency aortic valve replacement with cryopreserved homografts for acute native or prosthetic aortic valve endocarditis has a low operative mortality. The late incidence of recurrent endocarditis or homograft failure up to 7 years is acceptable. Cryopreserved homografts are non-viable. The presence of T-cell lymphocytes in explanted homografts indicates that rejection may be possible.