Rupture of the anterior cruciate ligament (ACL) is one of the most common sports-related injuries, affecting over 400,000 patients each year,1 including 38,000 high school students.2 Left untreated, patients with ACL injuries can develop cartilage damage, meniscal tears, and ultimately, arthritis.3 Unfortunately, the ACL heals poorly without surgery, and even with suture repair, non-union rates can be as high as 40–100%.4–6 As a result, ACL reconstruction remains the treatment of choice for management of ACL ruptures.
Although ACL reconstruction with tendon autograft or allograft is an effective operation for restoring the gross stability of the knee, the surgery has its limitations. One significant problem with ACL reconstruction is delayed integration and poor regeneration of the native tendon to bone insertion site. The native ACL-to-bone insertion, or direct insertion, is a highly specialized tissue which is composed of four distinct regions: ligament, non-mineralized fibrocartilage, mineralized fibrocartilage, and bone. These regions act in concert during load-bearing to minimize stress concentrations and shear stresses, thereby facilitating stress transfer between hard and soft tissue.7, 8 Studies have demonstrated that while tendon autografts heal to bone, the healing occurs by formation of an indirect type insertion, where collagen fibers blend with periosteal collagen fibers which are anchored to underlying bone. However, the non-mineralized and mineralized fibrocartilage layers are not consistently regenerated.9–11 Improper healing of the tendon in a bone tunnel can result in graft slippage which, in turn, results in concomitant increase in knee laxity and even functional failure of the ACL graft. Although multiple studies have attempted to augment the healing process to facilitate regeneration of the fibrocartilage layers at the insertion site in ACL reconstruction,12, 13 this healing process remains problematic in the early success of reconstructive surgery.
A potential advantage of primary ACL repair is that it may preserve the native, direct bone-to-ligament ligament insertion site, which may render it preferable to ACL reconstruction. To date, no studies have investigated whether the direct insertion site is actually preserved during the period of ACL healing with primary repair. In addition, no studies have yet investigated the effect of skeletal maturity on insertion site changes after primary ACL repair. However, there are studies in the response of the MCL and ACL insertion sites that guided development of the hypothesis in this study. A prior study of midsubstance MCL rupture, with no direct injury to the insertion site of the ligament, reported fibroblastic and osteoclastic remodeling of the insertion site between 2 and 6 weeks after injury.14 In another series of studies investigating ACL rupture, where ligamentous healing does not occur, the insertion sites were found to degenerate.15–17
Therefore in this study, we hypothesized that the varied biomechanical success of the ACL healing seen with age18 would be reflected in the insertion site changes in the healing process, namely that ligaments from younger animals which healed with higher maximum loads would also show signs of fibroblast and osteoclast mediated insertion site remodeling, while the ligaments of the adult animals, which had little spontaneous functional healing, would show signs of degeneration at the insertion site after a 15-week healing period.
Institutional Animal Care and Use Committee approvals were obtained prior to initiating this study. Twenty-four Yucatan mini-pigs from three different age groups were used: skeletally immature (8 ± 2.06 months, n = 8), adolescent (16 ± 2.19 months, n = 8), and adult (26.1 ± 1.06 months, n = 8). Their average weights were 33.1, 60.8, and 79.4 kg, respectively. The status of the physes was verified radiographically in all animals prior to tissue procurement. The immature animals had open physes, the adolescent animals had closed tibial and femoral but an open tibial tubercle physis, and the adult animals had all physes closed.
The initial experiment was a 15-week study to determine whether a difference in the insertion site would be seen for ligaments which would presumably heal (ACL transection treated with bioenhanced repair) and ligaments which would presumably not heal (ACL transection with no treatment). Four animals in each age group underwent bilateral ACL transections, and then one side had bioenhanced repair18 performed (see details below in the text and Fig. 1) while the other knee was closed without further treatment. After analysis of this group of animals was performed, it was found that the ACL transected knees which did not heal demonstrated only degeneration at the insertion sites, while those treated with bioenhanced repair appeared to remodel the insertion sites. To further define the timing and extent of the fibroblastic and osteoclastic remodeling, a second group of animals underwent bilateral ACL transections; however, in this group, both sides were treated with bioenhanced suture repair to minimize the number of animal lives required to study the early fibroblastic and osteoclastic processes. Four knees in each age group were procured at 1, 2, and 4 weeks after surgery to assess the early histologic changes during ACL healing. In addition, the insertion sites of the bioenhanced ligaments were compared with intact ACL insertion sites harvested from age-matched animals.
General anesthesia was induced and maintained in all animals for the surgical procedures. Pre-operative Lachman testing was performed and recorded for all knees. Both limbs were shaved, prepared with betadine and sterilely draped. An incision 4 cm in length was made at the medial border of the patellar tendon using a no. 15 blade. The medial retinaculum was divided at the patellar tendon border. The patella was gently retracted laterally and the fat pad was resected to expose the ACL.
The ACL was transected at the junction of the proximal and middle thirds of the ligament by inserting the blade parallel to the tibial plateau in the center of the ACL and moving it first laterally and then medially to cut both the posterolateral bundle and anteromedial bundle. Complete transection was verified visually and with a repeat Lachman maneuver that became positive in all knees with no significant end point detected after complete transection. The insertion site was not injured during the transection of the ligament.
For the knees undergoing enhanced anterior cruciate ligament repair, a Kessler suture of number-1 Vicryl (polyglactin; Ethicon, Somerville, NJ) was placed in the tibial stump with its ends exiting through the proximal cut of the anterior cruciate ligament (Fig. 1). The knee was then hyperflexed, and a guide-pin was drilled through the femoral anterior cruciate ligament footprint and out through the lateral cortex of the femur. The pin was overdrilled with an Endo-Button drill, and an Endo-Button (Smith and Nephew, Andover, MA) loaded with three number-1 Vicryl sutures was passed through the femoral tunnel and flipped on the lateral cortex to provide femoral fixation of the sutures. A 2.4-mm drill-pin was then used to make a tunnel in the tibia with use of a drill guide (Acufex Director Drill Guide; Smith and Nephew) to ensure that the entrance into the joint was just medial to the anterior cruciate ligament footprint. The tibial tunnel would later serve to secure the repair sutures directly to the tibia as previously described.9, 10 The resorbable sutures bridging the femur and tibia (as shown in purple in Fig. 1) served as a temporary stent within the knee. Once both suture tunnels had been drilled, a collagen scaffold (described below) was threaded onto four of the Vicryl suture ends from the Endo-Button. The remaining suture ends were used to directly tie to the anterior cruciate ligament stump. The scaffold was then passed up into the notch until femoral contact was verified visually. The sutures were passed through the tibial button and tied over the front of the tibia with the knee in 30° of flexion (which is the maximum extension angle for the porcine knee). The scaffold was then saturated with 3 ml of platelet-rich plasma (PRP) (as described below) at five times the baseline systemic platelet concentration. The variable depth suture that had been previously placed in the tibial anterior cruciate ligament stump was then tied to the remaining sutures exiting the femoral tunnel. No direct repair of the femoral and tibial ends of the ACL was performed.
The animals were not restrained post-operatively, and were allowed ab libitum activity. Animals were subsequently euthanized at the time points specified by the experimental design. Each knee was subsequently retrieved and stored at −20°C until further testing. The femur–ligament–tibia complexes from the 15-week time point were subsequently loaded in tension to failure at a rate of 20 mm per minute, as previously described.18 All knees were then transferred to neutral-buffered formalin before preparation for histology.
Histology for the 15-week specimens was performed after mechanical testing to failure. All of the repairs were examined grossly for evidence of mechanical damage (fissuring, cracks, disruption in creep patterns) at the ligament insertion sites for this group of repairs.
Collagen-Platelet Composite (CPC) Preparation for Bioenhanced Repair
At the time of surgery, 60 ml of whole blood was drawn from each animal using a large bore needle (≤18 g) and placed into 15 ml tubes containing 10% sodium citrate. The blood was centrifuged (Beckman GS-6 Centrifuge with GH3.8 Rotor, Beckman Coulter, Fullerton, CA) to isolate a platelet concentrate. The tubes were spun at 150g and the supernatant collected and spun again at 500g to form a platelet pellet. The resulting second supernatant was collected as platelet-poor plasma (PPP). The platelet pellet was resuspended in a specified amount of PPP to create a platelet concentrate with an enrichment of approximate five times the systemic platelet level. An aliquot of this was diluted with additional PPP to create a solution with an enrichment factor of five times the systemic platelet level. Initial and final platelet concentrations were determined using a VetScan HM5 Analyzer (Abaxis, Union City, CA). The platelet concentrate was kept at room temperature until use (less than 30 min).
The collagen sponge was made by solubilizing bovine connective tissue. Fresh tissue was harvested from the hind limbs, minced, and solubilized in an acidic pepsin solution. The resulting collagen solution was frozen, lyophilized, and rehydrated with a specified amount of water to create a solution with a collagen content >10 mg/ml. The slurry was neutralized using HEPES buffer (Mediates, Inc., Herndon, VA), sodium hydroxide (Fisher Scientific, Fair Lawn, NJ), PBS (Cyclone, Logan UT), and calcium chloride (Sigma-Aldrich, St. Louis, MO) and kept on ice. The neutralized collagen solution was placed into cylindrical molds with a diameter of 15 mm. The solution was then frozen and lyophilized. The resulting sponges were stored frozen and under vacuum until use.
A schematic of the bioenhanced suture repair technique is seen in Figure 1.18
At each designated time point, four ligaments from each age group were retrieved and fixed in neutral buffered formalin for 1 week, decalcified carefully, sectioned longitudinally in a sagittal plane passing through the ACL as large sections, dehydrated and embedded in paraffin. Seven micron sections were microtomed, placed onto pre-treated large section glass slides and stored at 4°C. Representative sections from each ligament were stained with hematoxylin and eosin (H&E) staining and immunohistochemical studies were carried out for each sample with the use of a mouse monoclonal antibody to von-Willebrand factor for identification of blood vessels and Cathespin K for osteoclasts (Sigma Chemical, St. Louis, MO). The tibial and femoral insertion sites were then analyzed.
Microscopic analysis at low-power magnification (10×) was used for qualitative analysis of the insertion sites for: (1) the presence of capillaries and the organization of fibroblasts and collagen within the fibrous zone; (2) chondrocyte organization in the non-mineralized cartilage and mineralized fibrocartilage zones; (3) the presence of osteoclasts at the bone–ligament junction.
Cell number density was analyzed under 40× magnification as previously described.19, 20 At each insertion site, the number of cells within three 0.1-mm2 areas were measured by two independent, blinded, reviewers and the results averaged. At each location, the total number of cells were counted and divided by the area of analysis. Quantitative proliferation of capillaries was measured with von-Willebrand factor staining and was determined by counting the total number of capillaries crossing the entire width of the section at each location under high-power view (40×). Quantitative proliferation of osteoclasts was measured with Cathespin K staining and determined by counting the total number of osteoclasts under high-power view (40×) per field. Osteoclasts were defined as multinucleated cells (>3 nuclei) that were Cathespin K-positive and adjacent to a resorbed bone surface. Three representative fields were analyzed per insertion site per animal for each age group and time of ACL harvest for cell number density, quantitative proliferation of capillaries, and quantitative proliferation of osteoclasts.
The results for each were averaged, and compared by analysis of variance (ANOVA). Post hoc analysis was performed using a Bonferroni adjusted threshold p-value.
For the 15-week study,18 the whole blood of the animals in each age group had similar red blood-cell counts (p = 0.84), white blood-cell counts (p = 0.42), and platelet counts (p = 0.21). In addition, there was no significant difference in the platelet count in the PRP for any of the age groups (mean [and standard error of the mean], 1,914 ± 140, 1,779 ± 140, and 1,443 ± 177 × 103/ml for juvenile, adolescent, and adult groups, respectively; p = 0.138).
For the second study, from 1 to 4 weeks, the systemic platelet count was not significantly different among the three age groups. The systemic platelet count (mean ± SD) in the immature group was 412 ± 127 × 103/µl, in the adolescent group was 318 ± 132 × 103/µl and in the adult group was 304 ± 102 × 103/µl (p > 0.05 for all comparisons). The platelet count in the PRP used in each group was also similar between groups. The PRP platelet count in the immature group was 736 ± 236 × 103/µl, in the adolescent group was 529 ± 210 × 103/µl and in the adult group was 615 ± 147 × 103/µl (p > 0.05 for all comparisons). The enrichment factor is defined as the ratio between the platelet concentration in the platelet rich plasma and that in the systemic whole blood of the animal. Similar enrichment factors of 1.8×, 1.8×, 2.1× were obtained for the immature, adolescent, and adult groups, respectively (p > 0.05 for all comparisons).
All animals recovered uneventfully from surgery. None had any surgical complications, difficulty walking normally, redness, swelling, fever or signs of infection. All were ambulatory within a few hours of surgery, and ambulating normally by 1 week.
After the femur–ligament–tibia complexes were biomechanically tested and loaded in tension to failure, gross analysis showed that all ligaments were found to fail within the midsubstance of the ACL tissue. There was no obvious mechanical damage (fissuring, cracks, disruption in creep patterns) noted at the ligament insertion sites, and the insertion site remained intact.
At 15 weeks, the adult (Fig. 2c) and adolescent (Fig. 3a) animals which had ACL transection with no further treatment showed degenerative changes in both the fibrous and fibrocartilage zones of the insertion site. In this group of ligaments, biomechanical healing to 7% of the intact ACL strength was seen in the adult and to 13% in the adolescent groups.18 The fibrous zone was relatively acellular and there was a loss of the parallel arrangement of the collagen fibers to the bone. The fibrocartilage zone was also relatively disorganized with loss of the columnar arrangement of the chondrocytes. The chondrocytes appeared flattened and necrotic and there was loss of the lacunae surrounding the chondrocytes (Figs. 2c and 3a). In contrast, the ligaments in the immature animals which had ACL transection only, developed a high degree of return of strength at 15 weeks, to 25% of the intact ACL strength.18 In that age group, remodeling of the insertion site was noted at 15 weeks, with changes including increased collagen organization in the fibrous zone, hypercellularity in the fibrous region, and a distinct fibrocartilage zone (Fig. 4c).
The ligaments treated with bioenhanced repair of the ACL had different histologic findings. In this group of ligaments, biomechanical healing to 32% of the intact ACL strength was seen in the immature and to 24% in the adolescent groups.18 At the insertion sites of those two groups, insertion site remodeling was seen (Figs. 3b and 4c). The fibrous zone had increased collagen organization, with a relatively large proportion of the collagen again oriented perpendicular to the insertion site in a densely packed arrangement. There was hypercellularity noted in the fibrous part of the insertion site. In the fibrocartilage zone, there was relatively good organization of the collagen perpendicular to the subchondral plate and a distinct fibrocartilage layer was again prominent. There was also a reasonable level of organization of columns of cells oriented perpendicular to the insertion site that had an appearance of normal chondrocytes with intact lacunae. However, in the adult animals undergoing bioenhanced repair, little biomechanical healing was noted at 15 weeks, with an average maximum load of only 13% of the intact ligaments.18 In these ligaments, the insertion site had changes consistent with degeneration similar to that seen in the adult and adolescent animals whose ACLs were transected but not repaired.
Early Insertion Site Remodeling: 1- to 4-Week Findings for Bioenhanced Repair
The insertion sites of the immature and adolescent animals underwent distinct histologic changes at 2–4 weeks which culminated with the partial reappearance of a four-zone insertion site at 15 weeks (Fig. 4a–c).
At 1 week, the insertion sites of the immature and adolescent animals had no significant histologic changes noted from their native morphology and appeared similar to intact non-transected controls (Fig. 4a). The ligament–fibrocartilage–bone interface remained intact. There was a distinct fibrocartilage layer. There was no evidence of osteclastic resorption of fibrocartilage or bone and there was no significant change in fibroblastic density.
However, at 2 and 4 weeks, the morphology of the insertion site changed in the immature and adolescent animals (Fig. 4b). The insertion site changed at both the original insertion site and adjacent areas, and consisted primarily of cellular proliferation and capillary infiltration within the fibrous zone as well as osteoclastic resorption of the fibrocartilage (Fig. 5). The majority of the changes were noted to occur within 100 µM of the original tidemark.
In the fibrous zone, there was a loss of collagen organization; collagen fibers were no longer arranged in a perpendicular arrangement relative to the bone (Fig. 4b). Capillaries infiltrated the fibrous tissue at these early time points in the immature and adolescent animals. At these early time points, osteoclasts were noted at the bone ligament junction and fibroblasts were noted to be present in increased numbers in areas where the non-mineralized and mineralized cartilage were resorbed (Fig. 4b). While there was a predominance of osteoclasts, osteoblasts were not present. There were no significant changes noted in the bone region located more than 100 µm from the tidemark where the osteocyte and vascular density remained unchanged.
While changes occurred in immature and adolescent insertion sites, no significant changes were noted in the adult insertion sites at 2–4 weeks of the study (Fig. 2b).
Quantitative Findings at 1–4 Weeks With Bioenhanced Repair
Skeletally immature and adolescent animals had a greater number of fibroblasts in the fibrous zone (within 100 µm of the tidemark) at the insertion site at weeks 2 and 4 when compared to the adult animals (p < 0.001) (Fig. 6). At 4 weeks, the skeletally immature and adolescent animals had a cell density in the fibrous zone more than three times that seen in the adult animals (p < 0.001).
Osteoclasts were defined as multinucleated cells (>3 nuclei) that were Cathepsin K-positive and adjacent to a resorbed bone surface. Osteoclasts were predominantly noted at the fibrous zone/bone interface, where the fibrocartilage zone previously was located. At four weeks, the skeletally immature and adolescent animals had more than five times the number of osteoclasts at the insertion site when compared to adult animals (p < 0.001) (Fig. 7).
Blood vessel density
Capillaries were defined as oval structures which stained positively for von-Willebrand's factor. Skeletally immature and adolescent animals had a greater number of capillaries at weeks 2 and 4 in the fibrous zone of the insertion sites when compared to the adult animals (p < 0.001) (Fig. 8). Capillary density was maximized at 2 weeks in the immature and adolescent animals.
In this study, we found histologic changes in the insertion site after transection and enhanced repair of the ACL, and that these changes were dependent on the degree of return of mechanical strength in the repairs. In immature and adolescent animals treated with bioenhanced repair, reasonable return of mechanical strength was seen in both groups.18 Accordingly, the initial changes of fibroblast proliferation and loss of collagen alignment in the fibrous zone, as well as osteoclastic resorption of the fibrocartilage layers, were followed by increased collagen alignment in the fibrous zone and partial restoration of the fibrocartilage zones. However, in adult animals and adolescent animals which had ACL transection only and which healed with poor mechanical strength,18 the changes over time were more gradual, with the fibrous zone becoming relatively acellular with steadily increasing disorganization of the fibrocartilage zones. The changes in these animals were more consistent with degeneration of the insertion site.
Ligaments and tendons do not consistently regenerate the mineralized and unmineralized fibrocartilage layers of the native insertion site after injury or reconstruction.9–11, 21–23
From the results of the present study, it appears that fibrocartilage layers are not completely re-formed by 15 weeks after bioenhanced repair of the ACL. By 15 weeks, while there is a reappearance of chondrocytes arranged in a columnar arrangement perpendicular to bone in layers, the organization of these layers is not as structured as in a native insertion site. Although fibroblastic proliferation and osteoclastic presence are mediators in this process, it is unclear whether they augment or hinder the development of the fibrocartilage layer in primary healing of the ACL.
The disorganization of the columnar organization within the fibrocartilage layers, as well as the apoptotic appearance of the chondrocytes and the disappearance of distinct lacunae, as observed in the insertion site of adult animals and non-treated adolescent animals at 15 weeks, are changes similar to those seen in insertion sites after rupture of the ACL. Prior studies have shown increasing chondrocyte apoptosis and decreasing thickness of both the non-mineralized and mineralized fibrocartilage layers in the ACL insertion site after substance resection by 6 weeks.15–17 Our study suggests that the insertion site of non-healing ligaments, even with bioenhanced repair, demonstrates degenerative characteristics similar to non-treated ruptured ligament insertion sites. Conversely, the insertion site of healing ligaments, as seen in the immature and adolescent animals, do not demonstrate these degenerative changes.
Prior models on insertion site healing have investigated changes after injury of the MCL,14, 24 patellar ligament,21 flexor tendons of the hand,22, 23 the rotator cuff tendons,25, 26 and during healing of an ACL graft in a bone tunnel,9, 10, 11 and collectively, these studies show that ligament or tendon-to-bone healing occurs initially by fibroblastic proliferation at the bone–ligament junction. In the present study, fibroblastic proliferation also was found to occur at the insertion site during bioenhanced repair of the ACL after transection, as demonstrated by the highly cellular fibroblastic layer which develops between 2 and 4 weeks in the fibrous zone of immature and adolescent, but not adult animals. Similar to prior investigations,9–11, 14, 21, 24 this early fibroblastic proliferation is characterized by highly disorganized collagen alignment in the fibrous layer of the insertion site. The collagen layer in the fibrous zone undergoes progressive organization, and by 15 weeks a relatively large proportion of collagen is again oriented perpendicular to the insertion site in a densely packed arrangement, although it is still hypercellular relative to a normal insertion site.
Studies on rotator cuff healing25, 26 and ACL reconstruction27, 28 demonstrate that the process of fibroblastic proliferation at the insertion site is regulated by inflammatory mediators and growth factors which direct the differential expression and remodeling of collagen within the fibrous and fibrocartilage zones. These studies show that the ability to form fibrocartilage during insertion site healing may be impaired by a reactive inflammatory response which directs fibroblasts to initially form reparative scar, rather than regenerative tissue. In the present study, although there is a predominance of fibroblastic proliferation, the inflammatory response appears to be minimal at the insertion site of the healing ACL. Thus, it is therefore unclear to what extent the fibroblastic response is reparative or regenerative during ACL healing with bioenhanced repair. The intra-articular environment, the presence of platelet-derived growth factors and the biomechanical loading in our model may affect the fibroblastic response in a manner that is different than prior models. Future studies are planned to determine the temporal expression of collagen subtypes to determine if the fibrous layer resembles scar or regenerative collagen and fibrocartilage.
Here, osteoclastic remodeling was observed in groups where functional ACL healing was most successful. However, prior studies have reported that osteoclastic activity results in impaired healing at the tendon–bone junction in a bone tunnel. Rodeo demonstrated that stimulation of osteoclastic activity with receptor activator of NF-kappa β ligand (RANKL) impaired formation of the bone–tendon insertion site while inhibition of osteoclastic activity with osteoprotegerin (OPG) facilitated enhanced healing at the insertion site during tendon to bone healing in ACL reconstruction in a tunnel.29 Similarly, Thomopoulos prevented bone loss associated with osteoclastic activity and showed improved insertion site healing in a flexor tendon healing model in a tunnel by the administration of alendronate after insertion site injury.23 Although the results in the present study seem contradictory to these prior reports, the environment of the insertion site for ACL healing is likely different from that of a healing tendon within a bone tunnel. For example, the vascular supply of the healing ACL insertion site is not completely disrupted as it is when a free graft is placed in a bone tunnel. In addition, the healing ACL insertion site experiences mechanical stress which is predominantly tensile, where a graft in a tunnel experiences stress which has a shear component as well. Thus, the two models of tendon and ligament to bone healing may represent fairly distinct biologic and mechanical situations, and the role of osteoclasts may be different in each case. However, because this study did not test a hypothesis for the mechanism of osteoclasts in healing, future studies would be necessary to better define their role in ACL healing. For these reasons, it is as yet unclear whether or not the osteoclasts impair insertion site healing as seen in other models, or whether they are, in fact, important mediators that facilitate healing during repair of the uninjured ACL insertion site.
In addition, the insertion site was not directly injured in the present study. The histologic changes seen are therefore likely related to something other than injury, possibly loss of tensile stress across the interface. Similar histologic changes occurred after midsubstance MCL rupture without direct injury to the insertion site,14 where there was osteoclastic remodeling of the insertion site between 2 and 6 weeks after rupture. However, the increased osteoclastic activity seen in the present study is likely not distinctly associated with mechanical unloading, as the adult animals did not demonstrate the types of osteoclastic changes at the insertion site that was observed in the younger animals, despite the fact that all age groups initially had their insertion sites unloaded after transection of the ACL. If the osteoclastic changes were due to unloading alone, one would expect to see osteoclastic changes in the insertion sites of the adult animals as well.
Interestingly, treatment with collagen-platelet composite resulted in all ligaments healing in continuity. While some repairs were stronger than others at the 15-week time point, none of the repairs frankly failed or formed gaps. The lower strength in the adult animals at 15 weeks may have resulted in relative unloading of the insertion sites and contributed to the degenerative changes noted in that group.
Although the results of our study reveal that insertion site remodeling during enhanced suture repair occurs initially by fibrovascular proliferation and osteoclastic remodeling, our results do not show what occurs at intermediate time points. Although prior studies have shown osteoblastic growth to be an important mediator which succeeds fibroblastic proliferation in tendon-to-bone healing in ACL reconstruction,29, 30 our study did not show the presence of osteoblasts at early or late time points. Future studies at intermediate time points will provide a better understanding of transitional steps in this process to determine if there is an antecedent osteoblastic or chondroblastic stage which directs the remodeling during enhanced suture repair.
One potential criticism of the present study was that there were more than twice as many platelets in the 15-week group as in the other time points. However, recent studies have demonstrated no difference in mechanical performance of ACL repairs with a range of 3× to 5× the systemic concentration,31 so this difference in platelet concentration on functional healing is likely less significant. Future studies of the effect of platelet concentration on histologic changes in the repair tissue would be of interest.
One limitation of our study is that it is unknown whether or not unilateral surgery and/or more protective rehabilitation of the repaired ligaments and insertion sites would have significantly increased or decreased the fibroblastic proliferation and osteoclastic activity noted here. Unfortunately, we do not have biomechanical data on the animals tested from weeks 1 to 4, however, a previously published article from our group found that there was a nadir in healing ligament yield load between 6 and 9 weeks after repair,32 after which time increases in yield load occurred up to 15 weeks post-repair.
This study has shown that specific insertion site changes occur when the ACL heals biomechanically after enhanced suture repair in skeletally immature and adolescent animals. These changes include fibroblast and osteoclast proliferation, particularly at 2–4 weeks after ACL injury, and these changes partially reverse by 15 weeks after successful repair. Further investigation into the age-related biology of the insertion site and the surrounding tissues, including evaluation of the matrix composition, the presence of stem cells adjacent to the injury site, the molecular signals, and potential molecular manipulations of the healing process, are all clinically relevant avenues of study that could potentially lead to a revolution in the treatment of ACL injuries, particularly for young patients.
This study was supported by NIH grant number R01 AR054099 as well as an OREF Resident Research Grant (BMH).