No abstract

The planar fibrous connective tissues of the body are composed of a dense extracellular network of collagen and elastin fibers embedded in a ground matrix, and thus can be thought of as biocomposites. Thus, the quantification of fiber architecture is an important step in developing an understanding of the mechanics of planar tissues in health and disease. We have used small angle light scattering (SALS) to map the gross fiber orientation of several soft membrane connective tissues. However, the device and analysis methods used in these studies required extensive manual intervention and were unsuitable for large-scale fiber architectural mapping studies. We have developed an improved SALS device that allows for rapid data acquisition, automated high spatial resolution specimen positioning, and new analysis methods suitable for large-scale mapping studies. Extensive validation experiments revealed that the SALS device can accurately measure fiber orientation for up to a tissue thickness of at least 500 microns to an angular resolution of approximately 1 degree and a spatial resolution of +/-254 microns. To demonstrate the new device's capabilities, structural measurements from porcine aortic valve leaflets are presented. Results indicate that the new SALS device provides an accurate method for rapid quantification of the gross fiber structure of planar connective tissues.

We undertook this study to establish a more quantitative understanding of the microstructural response of the aortic valve cusp to pressure loading. Fresh porcine aortic valves were fixed at transvalvular pressures ranging from 0 mmHg to 90 mmHg, and small-angle light scattering (SALS) was used to quantify the gross fiber structure of the valve cusps. At all pressures the fiber-preferred directions coursed along the circumferential direction. Increasing transvalvular pressure induced the greatest changes in fiber alignment between 0 and 1 mmHg, with no detectable change past 4 mmHg. When the fibrosa and ventricularis layers of the cusps were re-scanned separately, the fibrosa layer revealed a higher degree of orientation while the ventricularis was more randomly oriented. The degree of fiber orientation for both layers became more similar once the transvalvular pressure exceeded 4 mmHg, and the layers were almost indistinguishable by 60 mmHg. It is possible that, in addition to retracting the aortic cusp during systole, the ventricularis mechanically may contribute to the diastolic cuspal stiffness at high transvalvular pressures, which may help to prevent over distention of the cusp. Our results suggest a complex, highly heterogeneous structural response to transvalvular pressure on a fiber level that will have to be duplicated in future bioprosthetic heart valve designs.

Use of bovine pericardium as an engineered biomaterial in the fabrication of bioprosthetic heart valves is limited, in part, by substantial intra- and intersac variations in its fibrous structure. To quantitatively assess this variability, we determined the fiber architecture of 20 whole BP sacs. Each sac was mounted on a prolate spheroidal mold, cleared and preserved in 100% glycerol, then sectioned into four equisized quadrants. This preparation method allowed for accurate intersac comparisons and minimized tissue distortions. The fiber architecture was evaluated by small-angle light scattering (SALS) using a 2.54-mm rectilinear grid resulting in approximately 1200 SALS measurements per quadrant, along with tissue thickness measured at 55 locations per quadrant. The fiber architecture was described in terms of fiber preferred directions, degree of orientation, and asymmetry of the fiber angular distribution. The BP sac fiber architecture demonstrated substantial intra- and intersac variability, with local fiber preferred directions changing by as much as 90 degrees within approximately 5 mm. Overall, most sacs revealed potential selection areas in the apex region characterized by a high degree of orientation, high uniformity in fiber preferred directions, and uniform tissue thickness. However, the size, location, and fiber orientation of these potential selection areas varied sufficiently from sac-to-sac to question whether anatomic location alone is sufficient for consistent localization of regions of high structural uniformity suitable for improved BHV design.

In Part I of this work we used small-angle light scattering (SALS) to quantify the fiber architecture of 20 bovine pericardial sacs, along with corresponding tissue-thickness measurements, to determine optimal material selection sites. In order to determine the anatomic consistency of these sites, the fiber architecture and thickness data from all 20 sacs were averaged together using a cartographic analysis method that took advantage of the geometry of the prolate spheroid mold used to process the sacs. Optimal selection sites were determined based on a local criteria where all fiber preferred directions within a 2.54-cm circular area were within +/- 10 degrees. The largest contiguous area (LCA) for the entire BP sac was 20.54 cm2, located in the vicinity of the left ventricle of the heart. The LCA tissue thicknesses were also relatively uniform, further supporting the use of these areas. However, even within these optimal areas there was a +/- 20 degrees standard deviation in local fiber preferred directions, resulting in at best a 40 degrees spread in local preferred directions. The observed structural variability may be due to regionally heterogeneous physiologic loadings induced by the ligamentous attachments. These attachments may alter the regional fiber preferred orientation to support local mechanical loadings. Overall, given the inherent structural variability of the BP sac, we conclude that use of anatomic location alone will not consistently guarantee the selection of tissue specimens with a highly homogeneous and predictable fibrous structure. It is thus suggested that a direct fiber measurement presorting method be employed when selecting BP specimens for bioprosthetic applications where tissue structural homogeneity and uniformity is critical.

We undertook this study to establish a more quantitative understanding of the microstructural response of the aortic valve cusp to pressure loading. Fresh porcine aortic valves were fixed at transvalvular pressures ranging from 0 mmHg to 90 mmHg, and small-angle light scattering (SALS) was used to quantify the gross fiber structure of the valve cusps. At all pressures the fiber-preferred directions coursed along the circumferential direction. Increasing transvalvular pressure induced the greatest changes in fiber alignment between 0 and 1 mmHg, with no detectable change past 4 mmHg. When the fibrosa and ventricularis layers of the cusps were re-scanned separately, the fibrosa layer revealed a higher degree of orientation while the ventricularis was more randomly oriented. The degree of fiber orientation for both layers became more similar once the transvalvular pressure exceeded 4 mmHg, and the layers were almost indistinguishable by 60 mmHg. It is possible that, in addition to retracting the aortic cusp during systole, the ventricularis mechanically may contribute to the diastolic cuspal stiffness at high transvalvular pressures, which may help to prevent over distention of the cusp. Our results suggest a complex, highly heterogeneous structural response to transvalvular pressure on a fiber level that will have to be duplicated in future bioprosthetic heart valve designs.

No abstract

The purpose of this study was determine quantitative differences in collagen fiber orientation in a wound healing model in the presence of transforming growth factor-beta2 and anti-transforming growth factor-beta2,3 antibody. Full-thickness wounds were made in the paravertebral area of two young pigs. Wounds were treated once, topically, with either transforming growth factor-beta2 or anti-transforming growth factor-beta2 antibody, or with methylcellulose gel. Control wounds were left untreated. Tissue biopsies were obtained from each wound on days 7, 14 and 46 post wounding. Tissue sections were stained with hematoxylin and eosin, and collagen fiber preferred orientation was quantified using small angle light scattering. Our results indicated that wounds treated with transforming growth factor-beta2 and anti-transforming growth factor-beta2,3 antibody had a significantly higher degree of orientation of collagen fibers than normal unwounded skin on days 7, 14 and 46 (p < 0.001). Transforming growth factor-beta2- treated wounds had a higher degree of orientation of collagen fibers than control wounds on days 7 and 14 (p < 0.001), and control wounds displayed a higher degree of orientation than wounds treated with anti-transforming growth factor-beta2,3 and normal unwounded skin at all time points (p < 0.001). These results suggest that differences in the dermal collagen degree of orientation correlate with scarring, and show that small angle light scattering can be used quantitatively to assess differences in the collagen fiber architecture of dermal wounds.