Abstract
Background
High re-rupture rates following repairs of rotator cuff tears (RCTs) have resulted in the increased use of repair grafts to act as temporary scaffolds to support tendon healing. It has been estimated that thousands of extracellular matrix repair grafts are used annually to augment surgical repair of rotator cuff tears. The only mechanical assessment of the suitability of these grafts for rotator cuff repair has been made using tensile testing only, and compared grafts to canine infraspinatus. As the shoulder and rotator cuff tendons are exposed to shearing as well as uniaxial loading, we compared the response of repair grafts and human rotator cuff tendons to shearing mechanical stress. We used a novel technique to study material deformation, dynamic shear analysis (DSA).
Methods
The shear properties of four RCT repair grafts were measured (Restore, GraftJacket, Zimmer Collagen Repair and SportsMesh). 3mm-sized biopsy samples were taken and subjected to DSA using oscillatory deformation under compression to calculate the storage modulus (G') as an indicator of mechanical integrity. To assess how well the repair grafts were matched to normal rotator cuff tendons, the storage modulus was calculated for 18 human rotator cuff specimens which were obtained from patients aged between 22 and 89 years (mean age 58.8 years, with 9 males and 9 females). Control human rotator cuff tendons were obtained from the edge of tendons during hemiarthoplasties and stabilisations.
A 1-way ANOVA of all of the groups was performed to compare shear properties between the different commercially available repair grafts and human rotator cuff tendons to see if they were different. Specific comparison between the different repair grafts and normal rotator cuff tendons was done using a Dunn's multiple comparison test.
Results
We report a significant difference in the shear moduli of all four rotator cuff repair grafts (P<0.0001, 1 way ANOVA, Kruskall-Wallis test). 2 of the grafts, Zimmer Collagen Repair and SportMesh, were not significantly different when compared to rotator cuff tendons and were found to have comparable shear mechanical properties (P > 0.05, Dunn's multiple comparison test). The other repair grafts, GraftJacket and Restore, had a significantly lower storage modulus when compared to human rotator cuff tendons.
Conclusions
With increasing numbers of repairs of rotator cuff tears, and augmentation of these repairs, there is a need to understand the mechanical and biological properties of the both repair grafts and the tendons they are designed to augment. At present there is no clear definition of the ideal mechanobiological properties of rotator cuff repair patches. Current rotator cuff repair grafts display a wide variation in their shear mechanical properties, and how closely they are matched to the mechanical properties of human rotator cuff tendons. It is hoped that this study may help to guide surgeons in deciding on the most appropriate rotator cuff tendon repair graft.
