The objective of this study was to determine how in vitro mechanical stimulation of tissue engineered constructs affects their stiffness and modulus in culture and tendon repair biomechanics $12weeks$ after surgical implantation. Using six female adult New Zealand White rabbits, autogenous tissue engineered constructs were created by seeding mesenchymal stem cells $(0.1×106cells∕ml)$ in collagen gel $(2.6mg∕ml)$ and combining both with a collagen sponge. Employing a novel experimental design strategy, four constructs from each animal were mechanically stimulated (one $1Hzcycle$ every $5min$ to 2.4% peak strain for $8h∕day$ for $2weeks$) while the other four remained unstretched during the $2week$ culture period. At the end of incubation, three of the mechanically stimulated (S) and three of the nonstimulated (NS) constructs from each animal were assigned for in vitro mechanical testing while the other two autogenous constructs were implanted into bilateral full-thickness, full-length defects created in the central third of rabbit patellar tendons (PTs). No significant differences were found in the in vitro linear stiffnesses between the S $(0.15±0.1N∕mm)$ and NS constructs ($0.08±0.02N∕mm$; $mean±SD$). However, in vitro mechanical stimulation significantly increased the structural and material properties of the repair tissue, including a 14% increase in maximum force $(p=0.01)$, a 50% increase in linear stiffness $(p=0.001)$, and 23–41% increases in maximum stress and modulus $(p=0.01)$. The S repairs achieved 65%, 80%, 60%, and 40% of normal central PT maximum force, linear stiffness, maximum stress, and linear modulus, respectively. The results for the S constructs exceed values obtained previously by our group using the same animal and defect model, and to our knowledge, this is the first study to show the benefits of in vitro mechanical stimulation on tendon repair biomechanics. In addition, the linear stiffnesses for the construct and repair were positively correlated $(r=0.56)$ as were their linear moduli $(r=0.68)$. Such in vitro predictors of in vivo outcome hold the potential to speed the development of tissue engineered products by reducing the time and costs of in vivo studies.

This study sought to determine anterior force in the porcine knee during simulated 6‐degree‐of‐freedom (DOF) motion to establish the role of the anterior cruciate ligament (ACL). Using a 6‐DOF robot, a simulated ovine motion was applied to porcine hind limbs while recording the corresponding forces. Since the porcine knee is more lax than the ovine knee, anterior tibial translations were superimposed on the simulated motion in 2 mm increments from 0 mm to 10 mm to find a condition that would load the ACL. Increments through 8 mm increased anterior knee force, while the 10 mm increment decreased the force. Beyond 4 mm, anterior force increases were non‐linear and less than the increases at 2 and 4 mm, which may indicate early structural damage. At 4 mm, the average anterior force was 76.9 ± 10.6 N (mean ± SEM; p < 0.025). The ACL was the primary restraint, accounting for 80–125% of anterior force throughout the range of motion. These results demonstrate the ACL dependence of the porcine knee for the simulated motion, suggesting this model as a candidate for studying ACL function. With reproducible testing conditions that challenge the ACL, this model could be used in developing and screening possible reconstruction strategies. © 2010 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 29:641–646, 2011

Our group has shown that numerous factors can influence how tissue engineered tendon constructs respond to in vitro mechanical stimulation. Although one study showed that stimulating mesenchymal stem cell (MSC)–collagen sponge constructs significantly increased construct linear stiffness and repair biomechanics, a second study showed no such effect when a collagen gel replaced the sponge. While these results suggest that scaffold material impacts the response of MSCs to mechanical stimulation, a well-designed intra-animal study was needed to directly compare the effects of type-I collagen gel versus type-I collagen sponge in regulating MSC response to a mechanical stimulus. Eight constructs from each cell line ($n=8$ cell lines) were created in specially designed silicone dishes. Four constructs were created by seeding MSCs on a type-I bovine collagen sponge, and the other four were formed by seeding MSCs in a purified bovine collagen gel. In each dish, two cell-sponge and two cell-gel constructs from each line were then mechanically stimulated once every $5min$ to a peak strain of 2.4%, for $8h∕day$ for 2 weeks. The other dish remained in an incubator without stimulation for 2 weeks. After 14 days, all constructs were failed to determine mechanical properties. Mechanical stimulation significantly improved the linear stiffness $(0.048±0.009$ versus $0.015±0.004$; $mean±SEM$ (standard error of the mean ) N/mm) and linear modulus ($0.016±0.004$ versus $0.005±0.001$; $mean±SEM$ MPa) of cell-sponge constructs. However, the same stimulus produced no such improvement in cell-gel construct properties. These results confirm that collagen sponge rather than collagen gel facilitates how cells respond to a mechanical stimulus and may be the scaffold of choice in mechanical stimulation studies to produce functional tissue engineered structures.

Background: Clinically, one-bundle posterior cruciate ligament reconstructions frequently result in the return of abnormal posterior translation. We hypothesized that the return of posterior translation is caused by a nonuniform distribution of load among the graft fibers. The purpose of the present study was to determine how the femoral attachment location of the second bundle of a two-bundle posterior cruciate ligament reconstruction affects the anterior bundle tension and the load distribution between the graft bundles. Methods: One and two-bundle posterior cruciate ligament reconstructions (one one-bundle type and three two-bundle types) were performed in nineteen cadaveric knees. The grafts were tensioned to restore posterior translation to within ±1 mm of that of the intact knee at 90° of flexion while a 100-N posterior force was applied to the proximal part of the tibia. For each reconstruction, the total graft tension was a minimum of 2.3 times larger than the applied posterior force. Bundle tension and knee motions were measured as the knee was cycled from 5° to 120° of flexion while a 100-N posterior force was applied. Analysis of variance was used to compare the four reconstructions, and post hoc testing was performed with use of Fischer's protected least significant difference method. Results: Two-bundle reconstructions involving a middle-distal or middle-middle second bundle significantly reduced the tension in the anterior bundle in comparison with the tension in the one-bundle (anterior-distal) reconstruction. The peak anterior-bundle tensions with the middle-distal and middle-middle second bundles were 43% and 37% less than the peak bundle tension for the one-bundle reconstruction (p < 0.001 and p = 0.002, respectively). With the exception of the average bundle tension, the tension parameters calculated for the middle bundle decreased as the distance from the articular cartilage increased. The peak tensions for the middle-middle and middle-proximal bundles were 32% and 61% less than that for the middle-distal bundle (p = 0.028 and p = 0.001, respectively). Conclusions: The femoral position of the second bundle significantly affected the tension in the anterior bundle and the load distribution. A second bundle placed in a middle or distal position resulted in a significant reduction in anterior bundle tension and in cooperative load-sharing (with the bundles functioning together). A proximal second bundle resulted in reciprocal loading (with one bundle functioning in flexion and one in extension), but the tension in the anterior bundle was not different from the tension in the one-bundle reconstruction. Clinical Relevance: The present study demonstrated that a proximal-to-distal change in the femoral position markedly affects bundle tension and function. A middle placement of the second bundle appears to be the most ideal because the bundles exhibit cooperative load-sharing and the peak loads are significantly less when compared with other reconstructions.