We report our initial experience from April 1992 to November 1993 with a stentless porcine valve (Toronto SPV Valve, St. Jude Medical) for aortic valve replacement (AVR) in 21 consecutive patients and compare this group to a matched cohort that underwent AVR with a Hancock II (Medtronic) bioprosthesis. There were no hospital deaths in either group. Postoperative hospitalization was 5.5 +/- 0.8 versus 7.0 +/- 2.3 days (p = 0.004). Aortic cross-clamp time was 114.5 +/- 15.7 min in the SPV group and 96.0 +/- 25.0 min in the Hancock II group (p = 0.003). Complications in the SPV group were: one patient suffered perioperative infarction, one patient required late reoperation for left main stenosis, and one patient died suddenly following femoral thrombectomy at another center. Complications in the Hancock II group included: one patient with postoperative low output syndrome, and two late deaths (one from an aortic dissection and the other from chronic liver disease secondary to alcohol abuse). Comparison data indicate that the average size valve implanted in the SPV group was higher than in the Hancock II group (26.3 +/- 1.9 vs 24.0 +/- 1.9, p = 0.001). In the SPV group, 16 patients had 0 or trivial regurgitation and 1+ regurgitation was seen in 5 patients; regurgitation did not change over a 12-month follow-up. We observed a decrease in gradients over time (p < 0.01). Our results are compatible with a hypothesis that the ventricle undergoes remodeling over time, once the obstruction is relieved. We think the stentless design is an important feature that allows this to occur.(ABSTRACT TRUNCATED AT 250 WORDS)