Eleven male, mongrel dogs (age, 9–13 months; weight, 23–28 kg) underwent bilateral shoulder surgery. Both shoulders received partial release of the superior 8 to 9 mm of the infraspinatus tendon, which is approximately 2/3 of the tendon width [17
]. One shoulder underwent tendon release and repair only, and the contralateral shoulder was subjected to release and repair followed by augmentation with a reinforced (human) fascia patch. Tendon retraction, cross-sectional area, stiffness, ultimate load, and biocompatibility of the repair site were evaluated at 12 weeks after surgery. In addition, seven pairs of canine cadaver shoulders underwent infraspinatus injury and repair with and without augmentation and served as Time 0 biomechanical controls. Historical biomechanical data from eight unpaired, canine cadaver shoulders were referenced as normal controls [17
]. The study received prior approval of our Institutional Animal Care and Use Committee.
This study was powered primarily to detect differences in mechanical properties between nonaugmented and augmented repairs at Time 0 and after 12 weeks of healing. We chose to detect a mean difference of 165 N in ultimate load based on the rationale that a 25% change in the properties of nonaugmented tendon repairs at Time 0 could be clinically important [17
]. Previously, the mean ± SD ultimate load of nonaugmented tendon repairs at Time 0 was 668 ± 146 N (n = 8) [17
]. A sample size of 11 allowed us to detect an effect size of 1.1 (165/146 N) with α = 0.05 and power = 0.9.
We prepared reinforced fascia patches by stitching lyophilized human fascia lata from the iliotibial tract of three human donors aged 18 to 55 years (Musculoskeletal Transplant Foundation, Edison, NJ, USA), with custom 100% PLLA fiber as described [3
]. However, given the size constraints of the canine shoulder, we scaled down the patches to 18 × 34 mm, from our previously described 50 × 50 mm [3
]. The use of smaller and rectangular-shaped patches required the stitch pattern to be modified as well. The Time 0 failure load of the patches used in the canine study was assessed by fixation of the patch to a wooden block using four simple FiberWire®
sutures (Arthrex, Inc, Naples, FL, USA) on its lateral (two), superior (one), and inferior sides (one) and pulling its medial end to failure using three simple FiberWire®
sutures. The Time 0 failure load of patches designed as in this canine study averaged 153 ± 27 N (n = 6).
Surgical methods were as described in our previous study with this animal model [17
]. Briefly, dogs were anesthetized with intravenous sodium methohexital (10 mg/kg), intubated, and maintained on isoflurane in oxygen (3%). The infraspinatus tendon was approached and the superior 2/3 was detached from its insertion. A portion of the joint capsule was excised to model an intraarticular injury (Fig. A). The tendon was repaired back to its insertion with two transosseous Number 0 FiberWire®
sutures in a modified Mason-Allen configuration (Fig. B). For augmentation, a fascia patch was laid over the tendon repair and attached to the tendon medially using three Number 0 FiberWire®
Mason-Allen sutures and tensioned across the tendon repair with four Number 0 FiberWire®
simple sutures (Fig. C). The wounds were irrigated with normal saline and closed in layers.
Fig. 1A–C Diagrams illustrate bilateral rotator cuff injury and repair with and without patch augmentation in a canine model. (A) The superior 2/3 of the infraspinatus tendon was sharply detached from its insertion at the greater tuberosity, and a 1.5- × (more ...)
Postoperatively, dogs were housed individually in 2.13- × 0.91-m cages with restricted ceilings. Dogs were given subcutaneous injections of buprenorphine (0.02 mg/kg body weight) twice daily for 3 to 5 days postoperatively for analgesia and 500 mg cephalexin orally twice daily for 7 days as a prophylactic antibiotic.
At 12 weeks, we euthanized the dogs using a lethal injection of barbiturate (1 mL/4.5 kg; Beuthanasia-D®
; WA Butler, Dublin, OH, USA). We harvested the tendon repair construct, including the infraspinatus muscle and 20 cm of the proximal humerus. It was not possible to reproducibly separate the intact portion of the tendon from the repaired portion, so the entire tendon (and reinforced fascia patch in augmented samples) constituted the repair construct. Samples were stored in saline-soaked gauze at −20° C up to 8 months until tested. This delay in testing allowed all samples to be collected and biomechanically tested in one analysis. Since both the experimental and control samples from any given dog were frozen for an identical amount of time, any effects of frozen storage, though expected to be minor [25
], would have affected each group similarly.
To assess tendon retraction distance at 12 weeks, we used visual inspection and palpation to approximate the position of the tendon stump within the fibrous tissue at the repair site [17
]. We used calipers to measure the distance between the retracted stump and osseous repair site. The somewhat subjective nature of identifying the position of the retracted tendon stump led us to report the tendon retraction data categorically in four groups: (1) 5 mm or less; (2) 5 to 10 mm; (3) 10 to 15 mm; or (4) 15 mm or more.
At mechanical testing, the humeri were potted and the muscle belly was gripped in a custom cryoclamp as described previously [17
]. We estimated the cross-sectional area of the repair construct from caliper measurements of the width and thickness. Suture markers of 4-0 braided silk were stitched through the tendon and overlying soft tissue on the repair construct 20 mm medial to the insertion site, which was medial to the reinforced fascia patch attachment to tendon. A 1.6-mm tantalum bead was placed as a bone marker at the insertion site. Samples were tested in tension along their anatomic direction of pull. Testing was conducted in air at room temperature, and samples were kept moist by spraying with saline solution. Samples underwent 100 prefailure loading cycles from 5 to 100 N and were immediately tested to failure at 30 mm/minute (MTS Systems Corp, Eden Prairie, MN, USA). Load was recorded with a 5000-N load cell (Honeywell Sensotec, Columbus, OH, USA). A custom optical system, synchronized with the load data and sampling at 30 Hz, was used to track the optical markers and calculate the local displacements across the tendon-bone repair site using custom texture correlation software [6
]. Repair stiffness was defined from failure testing as the slope of the load-displacement curve from 50 to 400 N. Ultimate load was defined as the maximum load the sample reached.
After mechanical testing, we processed the tendon-bone repair sites of three augmented repairs, which included the reinforced fascia patch, for histologic evaluation of biocompatibility. Samples were fixed in 10% neutral buffered formalin for 3 to 7 days, decalcified in 5% trichloroacetic acid solution for 1 to 2 weeks, processed routinely, and embedded in paraffin. From each specimen, 6-μm-thick sections were cut and stained with hematoxylin and eosin. One section from within the repaired region was visually inspected by a board certified pathologist (CDT) so that qualitative comments regarding fascia patch morphology, overall cellular infiltration, inflammatory cells, and neovascularization could be made.
We compared stiffness, ultimate load, and cross-sectional area between paired shoulders using a Wilcoxon signed-rank test and tendon retraction distance using a sign test. Within the nonaugmented and augmented repair groups, we compared stiffness and ultimate load with the respective Time 0 and normal controls using a Kruskal-Wallis one-way ANOVA on ranks, with pairwise Wilcoxon rank-sum post tests for significance. Since at least one data set failed the test for normality, nonparametric tests were used for all statistical comparisons. SigmaPlot® 10.0 (Systat Software, Inc, Chicago, IL, USA) was used for statistical analysis.