The behavior of pericardial versus porcine valve xenografts in the growing sheep model.

The comparative long-term behavior of the pericardial versus the porcine bioprostheses is not yet known. The need for long follow-up times to answer this question makes the growing sheep model an attractive alternative, given its ability to induce early valve degeneration. Sixty-three sheep, 12 to 16 weeks old, were operated on and received 39 porcine (11 Xenomedica, 10 Carpentier-Edwards S, nine Hancock I standard, and nine Hancock I T6-treated) and 24 pericardial (14 Mitroflow and 10 Ionescu-Shiley low profile) prostheses of clinical quality in the tricuspid position. Of the 52 operative survivors (32 received porcine valves and 20 received pericardial bioprostheses), six animals (five pericardial and one porcine) were eliminated because of bioprosthetic infection. Late sudden death before the scheduled killing occurred significantly more often (p less than 0.0001) in the pericardial (8/15 or 53%) than in the porcine group (1/31 or 3%). Calcium content of the explanted valves was significantly correlated with time in the pericardial group and the Xenomedica porcine prostheses (p less than 0.05) but not in the Hancock I and Carpentier-Edwards S valves, where it was only marginally significant (0.1 greater than p greater than 0.05). Linear regression analysis of tissue calcium content showed a similar slope for the pericardial group and Xenomedica porcine valves, in comparison with the remaining porcine valves. Comparison between the two lines using covariance analysis demonstrated a statistically significant difference between them, which shows that the pericardial and Xenomedica porcine valves appear to reach higher levels of calcification in a shorter follow-up time than the Hancock I, standard and T6-treated, and the Carpentier-Edwards S valve in this animal model.