Effect of Early and Delayed Mechanical Loading on Tendon-to-Bone Healing After Anterior Cruciate Ligament Reconstruction

Abstract

Modulation of the mechanical environment may profoundly affect the healing tendon graft-bone interface. The purpose of this study was to determine how controlled axial loading after anterior cruciate ligament reconstruction affects tendon-to-bone healing. Our hypothesis was that controlled cyclic axial loading after a period of immobilization would improve tendon-to-bone healing compared with that associated with immediate axial loading or prolonged immobilization. One hundred and fifty-six male Sprague-Dawley rats underwent anterior cruciate ligament reconstruction with use of a flexor digitorum longus autograft. A custom-designed fixture was used to apply an external fixator across the knee parallel to the anterior cruciate ligament graft. Animals were randomly assigned to be treated with immobilization (n = 36) or controlled knee distraction along the long axis of the graft to achieve approximately 2% axial strain beginning (1) immediately postoperatively (n = 36), (2) on postoperative day 4 ("early delayed loading," n = 42), or (3) on postoperative day 10 ("late delayed loading," n = 42). The animals were killed at fourteen or twenty-eight days postoperatively for biomechanical testing, micro-computed tomography, and histomorphometric analysis of the bone-tendon-bone complex. Data were analyzed with use of a two-way analysis of variance followed by a post hoc Tukey test with p < 0.05 defined as significant. Delayed initiation of cyclic axial loading on postoperative day 10 resulted in a load to failure of the femur-anterior cruciate ligament-tibia complex at two weeks that was significantly greater than that resulting from immediate loading or prolonged immobilization of the knee (mean and standard deviation, 9.6 ± 3.3 N versus 4.4 ± 2.3 N and 4.4 ± 1.5 N, respectively; p < 0.01). The new-bone formation observed in the tibial tunnels of the delayed-loading groups was significantly increased compared with that in the immediate-loading and immobilization groups at both two and four weeks postoperatively (1.47 ± 0.11 mm(3) [postoperative-day-10 group] versus 0.89 ± 0.30 mm(3) and 0.85 ± 0.19 mm(3), respectively, at two weeks; p < 0.003). There were significantly fewer ED1+ inflammatory macrophages and significantly more ED2+ resident macrophages at the healing tendon-bone interface in both delayed-loading groups compared with the counts in the immediate-loading and immobilization groups at two and four weeks (2.97 ± 0.7 [postoperative day 10] versus 1.14 ± 0.47 and 1.71 ± 1.5 ED2+ cells, respectively, per high-power field at two weeks; p < 0.02). The numbers of osteoclasts in the delayed-loading groups were significantly lower than those in the immediate-loading and immobilization groups at two and four weeks postoperatively (0.35 ± 0.15 [postoperative-day-10 group] versus 1.02 ± 0.08 and 1.44 ± 0.2 cells, respectively, per high-power field at two weeks; p < 0.01), and the delayed-loading groups also had significantly reduced interface tissue vascularity compared with the other groups (p < 0.003). Delayed application of cyclic axial load after anterior cruciate ligament reconstruction resulted in improved mechanical and biological parameters of tendon-to-bone healing compared with those associated with immediate loading or prolonged postoperative immobilization of the knee.