Hospital for Special Surgery 2010 Resident and Fellow Research Presentations Award-Winning Abstracts

Hospital for Special Surgery 2010 Resident and Fellow Research Presentations Award-Winning Abstracts TheLewis ClarkWagner Award inOrthopaedic Excellence Recipient: Carolyn M. Hettrich, MD, MPH The Role of Mechanical Load in Tendon to Bone Healing Carolyn M. Hettrich, MD, MPH; Selom Gasinu, BS; Mark Stasiak, ME; Brandon S. Beamer, MD; Alice Fox, MS; Lilly Ying, DVM; Xiang-Hua Deng, MD; Scott A. Rodeo, MD Introduction Structural failure after rotator cuff surgery has been reported as between 11% and 94%, with increased function in patients with intact rotator cuffs. Improved strength and long-term outcomes after rotator cuff repair are ultimately predicated upon successful tendon-to-bone healing. Joint motion is commonly prescribed following tendon repair surgeries such as rotator cuff repairs; however, the ideal rehabilitation program to optimize tendon-to-bone healing is unknown. Prior studies have demonstrated the detrimental effects of immediate load, as well as both the detrimental and positive effects of immobilization; however, there is no literature to date describing the in vivo effect of immobilization versus delayed and immediate load in tendon-to-bone healing. Objective This study aims to develop an animal model to study the effect of mechanical load in surface tendon-to-bone healing, such as rotator cuff tears. Hypothesis Delayed loading after the resolution of acute inflammation would result in a mechanically stronger and better organized tendon-to-bone interface than the immobilized and immediately loaded animals. Materials and methods A total of 278 Sprague Dawley rats underwent unilateral patellar tendon acute detachment and repair followed by placement of a custom designed external fixator. Rats were assigned either to be immobilized, to begin loading immediately postoperatively, or to begin loading after a 4or 10-day delay. Tendon loading was controlled using a specially designed motorized device to apply 3 N (low load) or 6 N (high load) of axial tensile load to the healing bone– tendon complex at 0.17 Hz for 50 cycles per day, with immobilization for the remainder of the day. Rats were sacrificed at 4 weeks post-operatively for histomorphometric, immunohistochemical, radiographic, and biomechanical analyses of the bone–tendon complex. Biomechanical testing was performed in a materials testing system to allow uni-axial tensile testing in line with the repaired tendon. The specimen was loaded to failure and the failure site was recorded. The load-to-failure data were recorded and stiffness was calculated from the load deformation curves. Results In the 6-N immediate load group, early failures through the tendon–bone insertion site were at 80%; so, this group was discontinued early in the study. The load-tofailure was significantly higher in the immobilized group than in the immediate or delayed loading groups (p<0.05). Immobilized animals had less formation of fibrocartilage at the tendon–bone insertion site. MicroCT analyses demonstrated less volume and density of cortical bone in the immediately loaded animals, but no difference in trabecular bone. In the immobilized animals, all formed bone at the insertion site, in comparison to 78% in the immediately loaded group and 89% in both the 3and 6-N 10-day delayed loading groups. None of the 6-N, 4-day delayed animals formed bone at the insertion site. Conclusion In our rat tendon-to-bone healing model, the immobilized group had significantly less formation of fibrocartilage and was able to consistently form new bone at the insertion site. They also had superior mechanical properties to the immediate and delayed loading groups. Consistent with our in vivo study, in vitro studies of tendon cells have demonstrated that, after a 48-h period of stress deprivation prior to cyclic loading, inhibition of MMP-13 was decreased as compared to immediately loaded controls and that, with increasing periods of immobilization prior to loading, there are corresponding increases in MMP-13 and a decrease in the TIMP/MMP-13 ratio. Funding acknowledgment: This study was funded by grants from the OREF and NFL Charities. HSSJ (2010) 6: 240–242 DOI 10.1007/s11420-010-9181-0