No abstract

We studied the mechanical behavior of membranes of calf pericardium, similar to those employed in prosthetic valve leaflets, when subjected to tensile fatigue. The objective was to assess its durability, as a fundamental property of cardiac bioprosthesis, and analyze the energy consumption. For this purpose, the authors built a hydraulic simulator to subject a spherical valve leaflet made of calf pericardium to cyclic stress mimicking cardiac function. A total of 522 assays were performed in 40 samples, subjected to cyclic pressures greater than 6 atm, and 482 subjected to pressures ranging between 2 and 6 atm. The mathematical expression that establishes the relationship between the pressure exerted and the frequency was obtained. If we assume that the function is continuous, this equation provides the range of fatigue tolerated for a given number of cycles. Using the optimal values (the five highest values per series), the expression was found to be y = 9.95x(-0 1214) (R(2) = 0.955), where x represents the frequency in cycles per second and y the pressure in atmospheres. In addition, we established the mathematical relationship between the energy consumed and the frequency, which was a function of the pressure exerted, regardless of the region or zone from which the samples had been obtained. The methods of manual and morphology-based selection employed produced widely dispersed results. When a mechanical criterion was included, the similarity in the energy consumed during fatigue testing markedly improved the correlation, with a coefficient of determination between paired samples of R(2) = 0.7477. A mechanical criterion, such as energy consumption, can help to improve sample selection and produce more consistent results. Finally, we obtained the mathematical expression that relates the energy consumed to the pressure exerted and the number of cycles per second to which the valve leaflet was subjected. This procedure enables us to establish the limit to the energy that a biomaterial can consume over a period of time during which it is subjected to a working pressure and, thus, calculate more precisely its durability.

We designed a composite-structure (laminar-reticular) prosthesis using expanded polytetrafluoroethylene (ePTFE) as the biomaterial in an attempt to improve the biomechanical resistance of the same biomaterial in the form of a single lamina. Defects (7 x 5 cm) were created in the abdominal wall of male white New Zealand rabbits (n = 24). The defects, which comprised all the wall layers except the skin, were then repaired with one of two types of ePTFE prosthesis. One was a latest generation laminar prosthesis (DualMesh, or Gore-Tex) and the other was a composite of in-house design (CV-4 mesh composite) made by suturing a mesh woven out of ePTFE thread to an ePTFE lamina. After sacrificing the animals at 14 or 90 days after surgery, implant specimens were subjected to morphologic analysis (light microscopy and scanning electron microscopy); and adhesion formation, neoperitoneal thickness, and biomechanical strength were evaluated. No significant differences were recorded between the two prosthesis regarding the consistency of adhesions or the area occupied by adhesions (DM 0.17 +/- 0.06; CV-4 mesh composite 0.18 +/- 0.08 cm2) (p> 0.05). Notably, improved tissue integration was achieved using the composite prosthesis; its reticular side became infiltrated by dense connective tissue that enveloped the mesh filaments. In contrast, the DM prosthesis became encapsulated by host tissue. The neoperitoneum induced by both prostheses was homogeneous and orderly, with a layer of typical mesothelial cells lining its inner surface. The thickness of the neoperitoneum was similar (p > 0.05) for the two implants (385.0 +/- 3.4 vs. 390 +/- 3.1 microm), although significantly higher (p < 0.05) mechanical resistance values were recorded for the composite prosthesis (26.75 +/- 3.71 vs. 14.11 +/- 3.71 N). Our findings suggest that the use of a reticular and a laminar ePTFE layer in the same prosthesis leads to better repair and biomechanical behavior compared to the use of a single-structure laminar implant.

We compared the mechanical resistance of 18 samples of calf pericardium bonded with a 100 mm2 overlap, by two types of glues: a cyanoacrylate (Loctite 4011) and a bioadhesive (BioGlue). Comparative tensile testing was also carried out in 40 paired samples, 20 bonded with the cyanoacrylate and 20 unbonded controls. The findings at rupture showed a greater resistance of the calf pericardium glued with cyanoacrylate, with a mean tensile strength of 0.15 MPa vs. 0.04 MPa for the biological glue (p= 0.000). They also demonstrated a loss of resistance of the samples bonded with cyanoacrylate when compared with that of the unbonded other halves of the pairs: 0.20 MPa and 0.27 MPa vs. 19.47 MPa and 24.44 MPa (p < 0.001). The method of selection by means of paired samples made it possible to establish the equations that relate the stress and strain, or deformation, with excellent coefficients of determination (R2). These equations demonstrate the marked elastic behaviour of the bonded samples. Moreover, these findings show the cyanoacrylate to be superior to the biological glue, leading to the examination of the compatibility, inalterability over time and mechanical behaviour of the cyanoacrylate in sutured samples, as well as the study of the anisotropy of the biomaterial when bonded with a bioadhesive.

Careful selection of the biological tissue to be used in cardiac bioprostheses and a thorough knowledge of its mechanical behavior, facilitating both the prediction of this behavior and the interactions between the tissue and the other materials employed, is the best approach to designing a durable implant. For this purpose, a study involving uniaxial tensile testing of calf pericardium was carried out. Two sets of three contiguous strips of tissue were obtained from each pericardial membrane, to perform a total of 144 trials. Two samples were sewn with one of four commercially available suture materials: Gore-Tex, nylon, Prolene and silk. In each set of three samples, the center strip remained intact and unsutured to serve as a control, while the left-hand strip was sutured at a 45 degrees angle with respect to the longitudinal axis and the right-hand strip was sewn at a 90 degrees angle. All the samples were tested until rupture. The results demonstrated a significant loss of mean load (p<0.01) in the sutured samples at rupture. The angle of the suture had no influence on these results, although the stress/strain curves showed that, as the tensile stress increased, the mechanical behaviors were less uniform. The rupture of the collagen fibers could explain this phenomenon. The pericardium sutured with Gore-Tex presented a greater strain, or deformation (elongation), at lower levels of stress, regardless of the angle of the suture. The tissue selection criteria, based on the use of paired samples, enabled a correct prediction of the mechanical behavior of the tissue, with excellent correlation coefficients (>0.98) and a high degree of homogeneity in the results.