Local variation in the tearing strength of chemically modified pericardium.

Forty squares of tissue, 40 mm X 40 mm, were harvested from bovine pericardium fixed in glutaraldehyde free of tension. Two parallel dumb-bell specimens were excised from each square. In 20 pairs of adjacent samples, two 1 mm cuts were introduced into one of the samples at the centre of the dumb-bell. The other 20 pairs were used as controls. There was no significant difference in tensile strength of the paired samples tested without cuts, but there was marked variation in the strength of the tissue in different squares. There was a significant positive correlation between the strength of the paired samples. Introduction of artificial tears produced a significant reduction in the strength of the tissue locally. However, a significant positive correlation between the sample with tears and the sample without tears was still maintained. Failure of the tissue under uniaxial load was observed dynamically by using a Panasonic video camera with zoom and magnification facilities illuminated by an oblique low-power light source placed behind the specimen. All samples, with or without tears, failed by rupture of a layer of tissue, laminate debonding and shear, followed by fibre slippage through the viscoelastic embedding matrix.