Implants for Tendon Repair

The rotator cuff consists of four muscles and their associated tendons that attach to the proximal humerus through direct tendon-to-bone insertions. The rotator cuff acts to stabilize the glenohumeral joint and is prone to injury, with cuff tears being the most common form of injury afflicting the shoulder. Each year more than 250,000 cuff repairs are performed in the United States alone. The incidence of rotator cuff repair is also increasing because of an aging, yet active, population. Primary tendon-to-bone repair and healing are the goals of these surgical procedures. In some cases, augmentation with commercially available patches is performed, with the patch applied to the superficial tendon surface to enhance the repair. However, because of a variety of factors including graft degeneration, poor vascularization, muscle atrophy, and the lack of graft-to-bone integration after surgery, failure rates between 20% and 90% have been reported after primary repair of chronic rotator cuff injuries.

To improve healing, biologic or synthetic polymer-based tendon implants or augmentation devices have been explored to reconstruct large rotator cuff tears. To date, the most commonly used implants are derived from extracellular matrix (ECM), including TissueMend, Restore, and Permacol ( Table 4-1 ). Restore is made from porcine small intestine submucosa, whereas TissueMend and Permacol are derived from fetal bovine dermis and porcine dermis, respectively. Because of the different sources, these implants have distinct mechanical properties. Barber et al. evaluated mechanical properties of the major commercial implants (Restore, Permacol, TissueMend, and GraftJacket, an allograft not discussed here) under tension. It was reported that products derived from human skin are the strongest, followed by porcine and bovine skins, with the small intestine submucosa–based implant being the weakest. This difference in mechanical properties could be the potential cause of the difference in performance of the implants in patients.

Restore was the first implant on the market for tendon repair and was widely used in the clinic. This graft is decellularized and consists of 90% collagen with small amounts of lipids and carbohydrates. Clinical outcomes for Restore have been mixed. Sclamberg et al. performed a 6-month follow-up study of 11 patients (5 women and 6 men between the ages of 52 and 78 years) who underwent large rotator cuff repair augmented with the Restore implant and found that the repair failed in 10 of the 11 patients. Of even greater concern were the results reported by Iannotti et al. ; in a study of 15 patients (4 women and 11 men with mean age of 58 years), those treated with the Restore patch had a lower healing rate and a lower postoperative PENN score compared with members of the control group, who underwent repair without the patch. In a later study, Walton et al. compared healing results of patients (5 women and 10 men with a mean age of 60.2 years) who had their rotator cuff repaired with or without the Restore implant. It was observed that 2 years after surgery, patients who received the Restore implant reported higher rates of repeat tears, and their measured mechanical properties were significantly weaker. The patients also had more impingement and decreased levels of sports participation. In another study, Malcarney et al. examined the postoperative responses of 25 patients who underwent rotator cuff repair augmented with the Restore implant and found that four of them experienced an acute inflammatory response shortly after surgery (a mean of 13 days). The inflammatory responses were nonspecific, and all four patients required implant removal. Based on these results, Restore is not currently recommended for rotator cuff repair augmentation.

The next generation of ECM-based implants focused on improving matrix mechanical properties. TissueMend is derived from fetal bovine dermis and is decellularized through a series of chemical processes. Magnussen et al. used TissueMend to repair cadaveric Achilles tendons and tested the tendon mechanical properties under cyclic tensile loading. Results showed decreased gapping in the implant-augmented group, and the ultimate failure load increased significantly from a mean of 392 N (range, 322 to 481 N) to 821 N (range, 613 to 1021 N) when the control group was compared with the augmented group. DiDomenico et al. used TissueMend to repair a rupture of the tibialis anterior tendon in a diabetic patient, and at 6-month follow-up, they observed that the patient appeared to have regained nearly full strength. Although currently the literature does not include any reports on the use of the TissueMend implant for rotator cuff repair, Seldes and Abramchayev demonstrated, in a cadaver, the feasibility of using this implant to repair a massive rotator cuff tear.

Another implant available for cuff repair is the Permacol (or Zimmer) patch, which is derived from porcine dermis. In addition to being decellularized, this patch is cross-linked with hexamethylene diisocyanate to increase its stability. Proper et al. used the Zimmer patch to repair massive rotator cuff tears in 10 patients (5 women and 5 men between the ages of 46 and 80 years) and found that the implant caused no major postoperative complications and no adverse reactions. These investigators also found that the pain score and interval function score improved at 1 year after surgery; similarly, better range of movement and strength were reported. In the medium-term follow-up (a duration of 3 to 5 years), Badhe et al. found that although two implants had detached after the original surgery, pain was reduced and abduction power and range of motion improved significantly, with no adverse effects reported. However, when Soler et al. used the Zimmer patch to repair massive rotator cuff tears in four patients (3 women and 1 man between the ages of 71 and 82 years), all four grafts failed between 3 and 6 months after surgery, leading to recurrent tears.

Because of the aforementioned challenges associated with ECM-derived implants for rotator cuff repair, a demand exists for new technologies to better meet the needs of functional tendon regeneration and soft tissue-to-bone integration. For integrative rotator cuff repair, the scaffold should match the mechanical properties of the native tendon, which is the major limitation of available biologic implants. The ideal implant should also mimic the ultrastructural organization of the native tendon. In addition, the implant should be biodegradable so it can be gradually replaced by new tissue while maintaining its physiologically relevant mechanical properties. Lastly, the graft must integrate with the host tendon and surrounding bone tissue by promoting the regeneration of the native tendon-to-bone interface. Several groups have explored synthetic grafts and tissue engineering methods for the development of tendon implants. It is common to use scaffolds composed of nanofibers based on a variety of synthetic polymers, such as poly- l -lactic acid (PLLA), polylactide- co -glycolic acid (PLGA), polycarbonate polyurethane, and biologic materials, such as collagen and silk. In addition to being biodegradable, both PLLA and PLGA are materials already used in FDA-approved devices.

The nanoscale architecture of the collagen-rich tendon matrix can be readily recapitulated with nanofiber scaffolds, which exhibit high surface area-to-volume ratio, low density, high porosity, variable pore size, and mechanical properties approximating those of the native tissues. Moffat et al. first reported on the fabrication of PLGA nanofiber scaffolds with physiologically relevant structural and mechanical properties for rotator cuff repair. It was observed that human rotator cuff fibroblast morphology and growth on aligned and unaligned fiber matrices were dictated by fiber alignment, with distinct cell morphology and integrin expression profiles. Upregulation of α2 integrin, a key mediator of cellular attachment to collagenous matrices, was observed when the fibroblasts were cultured on aligned fibers, and upon which types I and III collagen-rich matrices were deposited. More recently, Xie et al. developed a continuous PLGA nanofiber scaffold that transitioned from aligned to random orientation to examine the effects of this transitional region on rat tendon fibroblasts in vitro . After 1 week of culture, the cells proliferated on both aligned and random nanofiber orientations. Although a rounded morphology was found on unaligned nanofibers, cells cultured on aligned nanofibers appeared long and spindlelike and were aligned along the long axes of the fibers.

The biologic response to polymeric nanofibers may also be enhanced by additional surface modifications. For example, Rho et al. electrospun aligned type I collagen nanofiber scaffolds with a mean fiber diameter of 460 nm and evaluated the response of human epidermal cells after coating the scaffolds with several adhesion proteins. It was found that cell proliferation was enhanced by coating the scaffolds with both type I collagen and laminin. Recently, Park et al. applied a plasma treatment to polyglycolic acid (PGA), PLGA, and PLLA nanofibers and grafted a surface layer of hydrophilic acrylic on these scaffolds. It was found that NIH-3T3 fibroblasts seeded on these modified scaffolds spread and proliferated faster than those on unmodified control scaffolds. Nanofibers have also been used to improve existing scaffold design, resulting in a graft with a more biomimetic surface for eliciting desired cell response. For example, Sahoo et al. electrospun PLGA nanofibers directly onto a woven microfiber PLGA scaffold to increase cell seeding efficiency while maintaining a scaffold that was mechanically competent. The attachment, proliferation, and differentiation of porcine bone marrow stromal cells was evaluated on these scaffolds and, when compared with scaffolds seeded via a fibrin gel delivery, it was found that seeding the cells onto nanofiber-coated scaffolds enhanced proliferation and collagen production and upregulated the gene expression of several tendon-related markers, namely decorin, biglycan, and type I collagen.

In addition to being used as synthetic tissue–engineered implants, polymer fibers can be used to modify allogeneic nontendon tissue and make it more suitable for rotator cuff augmentation. Aurora et al. sutured PLLA and PLLA/PGA braids (diameter = 400 µm) to a human fascia patch to increase its suture retention properties. Results showed that all reinforced patches withstood 2500 cycles of 5 to 150 N cyclic loading, with the PLLA/PGA patch withstanding 5000 cycles of loading. In addition, the suture retention properties and the maximum construct load of the reinforced patch were observed to be greater than those of human rotator cuff tendon, even after 3 months in vivo. Although more foreign body giant cells were seen in the reinforced patches, it is expected that this response will decrease after the polymer has fully degraded. These promising results suggest that the polymer fiber–reinforced tendon patch could be a more functional alternative for rotator cuff augmentation.

To address the challenge of regenerating the tendon-to-bone insertion site, several groups have evaluated the feasibility of integrating tendon implants with bone or biomaterials through the formation of anatomic insertion sites. Fujioka et al. examined the effects of reattaching the bone and tendon in a rat model for Achilles tendon avulsion. After 4 weeks, surgical reattachment of tendon to bone increased type X collagen deposition and allowed tissue to maintain distinct regions of calcified and noncalcified fibrocartilage tissue. Additionally, Inoue et al. promoted supraspinatus tendon integration with a metallic implant using a bone marrow–infused bone graft. These early attempts demonstrate that the direct tendon-bone insertion may be regenerated. To this end, the ideal implant for tendon-to-bone interface tissue engineering must exhibit a gradient of structural and mechanical properties mimicking those of the multitissue insertion. Thus a scaffold recapturing the nanoscale interface organization, with preferentially aligned nanofiber organization and region-dependent changes in mineral content, would be highly advantageous. Building on the functional PLGA nanofiber scaffold designed for tendon tissue engineering, Moffat et al. designed a biphasic scaffold, with the top layer consisting of nanofibers of PLGA and the second layer consisting of composite nanofibers of PLGA and hydroxyapatite nanoparticles. The biphasic design is aimed at regenerating both the nonmineralized and mineralized fibrocartilage regions of the tendon-to-bone insertion site, while promoting osteointegration with PLGA-HA nanofibers. The response of tendon fibroblasts, osteoblasts, and chondrocytes have been evaluated on these nanocomposite scaffolds with promising results in vitro. When tested in vivo subcutaneously, as well as in a rat rotator cuff repair model, the biphasic scaffold supported regeneration of continuous noncalcified and calcified fibrocartilage regions, demonstrating the potential of a biodegradable nanofiber-based implant system for integrative tendon-to-bone repair.

In summary, commercially available, ECM-derived tendon grafts and augmentation devices have exhibited mixed outcomes in clinical studies. In general, tendon implants made from dermis performed better than implants made from other types of ECM, as reflected in the reported lower repeat tear rates and improved postoperative scores. Alternative treatment options such as synthetic and tissue engineered grafts are being developed, with promising results for tendon regeneration and tendon-to-bone integration. However, several challenges remain to be overcome before the widespread clinical application of the tissue-engineered tendon implants can be realized. For example, one challenge is the scale-up of the tissue-engineered implants from small animal models to humans. Currently most of the tissue-engineered implants are being evaluated in vitro and are prepared in small batches. High-throughput fabrication and delivery processes need to be developed for them to augment their commercial applicability. The other challenge is that the fabrication process of the tissue-engineered implants generally utilizes a variety of toxic solvents or several steps of chemical reactions to dissolve polymers, which may have undesired effects on biomolecules or cells once they are implanted in humans.

Implants for Ligament Reconstruction

The anterior cruciate ligament (ACL) is one of the major ligaments that connect the femur and tibia, and it is the primary stabilizer in the knee. The ACL inserts into bone through a direct transition consisting of spatial variations in cell type and matrix composition, resulting in three distinct tissue regions of ligament, fibrocartilage, and bone. Tearing of the ACL is among the most common knee injuries afflicting the young and active population, with more than 100,000 ACL reconstructions performed annually in the United States alone. Because of the inherent poor healing potential of the ligament, ACL reconstruction based largely on biologic grafts is required. Because of the relative scarcity and donor site morbidity of autografts, as well as the inherent risks associated with allografts, significant interest has been expressed regarding synthetic and tissue engineered alternatives for ACL reconstruction. In the late 1980s, the FDA approved synthetic materials for use as an alternative implant for ACL reconstruction. However, because of significant complications and failures in humans, all of the synthetic implants were withdrawn from the market by the late 1990s, and at the present, only biologic grafts such as hamstring tendon or bone-patellar tendon-bone grafts are used clinically for ACL reconstruction.

One of the commercial implants tested for ACL reconstruction was the Gore-Tex implant, which is based on polytetrafluoroethylene (PTFE) ( Table 4-2 ). This implant is composed of solid PTFE nodes interconnected by PTFE fibers at each end for bony attachment. In 1987 Ahlfeld et al. used the Gore-Tex implant to treat 30 patients with unstable knees. During the follow-up period (average 24 months), the group treated with the Gore-Tex implant showed improved satisfaction compared with patients who underwent reconstruction with a different material (ProPlast ligament) (83% satisfactory vs. 52%). Glousman et al. used this implant in 82 patients (23 women and 59 men, ages 16 to 51 years) and at 18 months follow-up found that a number of patients had complications (15) and repeat operations (14). On the other hand, the subjective scores improved in all evaluation categories, including pain, swelling, giving way, locking, and stair climbing. Although the authors suggested that the results were positive, they also recommended longer-term follow-up before definitive conclusions could be determined. In another study, Indelicato et al. reported that of 39 patients (12 women and 27 men whose ages ranged between 17 and 42 years) who received the Gore-Tex implant, 87% had satisfactory results 2 years postoperatively. In many of these studies, however, some complications were noticed, such as tears of the implant and sterile effusions. In addition to the failure of the implant itself, use of Gore-Tex led to other complications in the knee, including bone tunnel widening. Muren et al. examined 17 patients at 13 to 15 years after they had ACL reconstruction with the Gore-Tex implant. It was found that six patients underwent revision surgery as a result of implant rupture and pain; moreover, 15 patients had tibia bone tunnel widening. In another study, surgery with the Gore-Tex implant was performed in 123 patients, and complete rupture of the graft was seen in 26 patients. Additionally, half of the patients exhibited graft loosening, 62% experienced osteoarthritic change, and bone tunnel osteolysis at both ends was identified in most cases. Consequently, the Gore-Tex implant was no longer recommended for ACL reconstruction and eventually was withdrawn from the market in 1993. Similarly, for other synthetic ACL implants such as the Lars Ligament and Leeds-Keio ligament implants that were used in ACL reconstruction with short-term satisfactory results, many long-term complications were reported, including repeat rupture, bone tunnel widening, severe synovitis, and inflammatory responses. Therefore these implants were likewise withdrawn from the market, and currently no synthetic implants have been approved by the FDA for ACL reconstructions.

To overcome limitations of the failed synthetic implants and the inherent shortcoming of allografts currently in use, tissue-engineered implants have been investigated. Such implants involve implantation of a bioactive material that regenerates tissues with material properties that are comparable with those of the native ACL. Similar to the requirement for tendon repair, the ACL implant should be biodegradable, match the mechanical properties of the native ACL, and mimic the ultrastructural organization of the native ACL. Finally, the implant must integrate with both the femoral and tibial bone tunnels to promote the regeneration of the native ligament-to-bone interface. To this end, most of the studies have investigated the use of synthetic polymers such as PLLA, PLGA, and polyurethane, as well as biologic materials, such as collagen and silk.

Dunn et al. were among the first investigators to evaluate a biomimetic ACL replacement in vivo. Studies were performed using a type I collagen fiber–based prosthesis with polymethylmethacrylate bone fixation plugs on the ends. Although neoligamentous tissue formation was reported, the majority of scaffolds were reported to have ruptured after 20 weeks in a rabbit model, demonstrating that although it was biomimetic, this system was insufficient to support long-term knee stability. A series of studies also were performed that focused on the development of a silk-based ACL replacement both in vitro and in vivo . Specifically, a novel silk-fiber based scaffold was designed, and several studies were performed to assess the impact of chemical and mechanical stimulation on the differentiation of seeded mesenchymal stem cells (MSCs). It was demonstrated that individually, soluble chemical factors such as basic fibroblast growth factor, as well as tensile and torsional loading, could drive matrix elaboration and fibroblastic differentiation of MSCs on the silk scaffold. In addition, chemical and mechanical stimulation, when applied concomitantly and controlled temporally, have been reported to enhance MSC response. The system was implanted in vivo in a goat model with promising results reported based on histologic and mechanical outcomes such as significantly increased knee stiffness and ultimate tensile strength after 12 months of implantation. This prosthesis is currently undergoing clinical trials.

Also using a silk-based scaffold, Liu et al. performed a series of studies to optimize a knitted graft combined with a microporous silk sponge for ACL reconstruction. This biphasic system was designed to mimic the ECM of native tissue and provide sufficient mechanical strength for ligament replacement. The scaffold was implanted in vivo using a rabbit model and later a pig model, with substantial ligament-like tissue observed on the scaffold after 24 weeks. In addition, several in vitro studies were performed to further optimize the scaffold, including the use of silk cables to increase tensile strength, the incorporation of basic fibroblast growth factor–releasing PLGA and RADA16 peptide nanofibers to enhance cell biosynthesis, and the addition of a silk-based aligned nanofiber topography to direct cell orientation and matrix production.

Progressing from single-phase systems, Cooper et al. designed and optimized, both in vitro and in vivo, a braided, α-hydroxyester, microfiber-based scaffold for ligament engineering. The architecture and porosity was standardized in vitro using a braiding technique to recapitulate the native ligament mechanical properties and included a denser fiber arrangement at each end of the construct for bone formation. Scaffold composition was also optimized based on in vitro degradation and cell response, with PLLA selected because of its ability to maintain structural integrity over the 8-week culturing duration. Subsequently, the optimized scaffold was evaluated in vivo using a rabbit ACL reconstruction model. It was demonstrated that cell seeding of the implanted scaffold resulted in a marked improvement in functional outcomes, but scaffolds in both groups ruptured after 12 weeks of implantation. In a follow-up study, Freeman et al. evaluated the effect of both braiding and twisting on the mechanical properties of the PLLA microfiber system, demonstrating that twisting coupled with braiding could increase both ultimate tensile strength and the toe-region length of the graft. Recently, Barber et al. reported on the development of a braided nanofiber-based scaffold for ACL replacement. Scaffold architecture was optimized by varying the number of braided bundles and evaluating mechanical properties, with minimal differences in toe-region or elastic modulus as a function of braid number. The scaffold was seeded with MSCs in vitro, and both cell viability and proliferation were observed.

Also in development are scaffold systems that target the regeneration of the ligament-to-bone interface, which is critical for biologic fixation of either biologic or synthetic grafts. Spalazzi et al. reported on the design and evaluation of a triphasic scaffold for the regeneration of the ACL-to-bone interface both in vitro and in vivo . The scaffold consists of three distinct yet continuous phases, each engineered for a specific tissue region found at the interface: Phase A is designed with a PLGA mesh, phase B consists of PLGA microspheres, and phase C is composed of a sintered PLGA and 45S5 bioactive glass composite. They are intended for ligament, interface, and bone formation, respectively. When this stratified scaffold was evaluated in a subcutaneous athymic rat model, abundant tissue formation was observed on phases A, B, and C. Cell migration and an increased matrix production were also observed in the interfacial region, and the phase-specific controlled matrix heterogeneity was maintained in vivo. Once ligament fibroblast, chondrocyte, and osteoblast triculture were established on their respective phase of the scaffold, the formation of both anatomic ligament- and bone-like matrices was observed on the triphasic scaffold (phases A and C, respectively), as well as the deposition of a fibrocartilage-like tissue (phase B). At 2 months after implantation, the interface-like region consisted of chondrocyte-like cells embedded within a matrix containing collagen types I and II, as well as glycosaminoglycans, indicating the formation of interface-like tissue.

In addition to ACL reconstruction, a new approach emphasizing ligament repair has recently been developed. Through a series of in vitro and in vivo experiments, a graft system that combines collagen-based implants with platelet-rich plasma (PRP) for ACL repair has been designed and optimized. This system was first evaluated in an ACL central defect model, in which the collagen-PRP implant was only used to fill in the defects, not to replace the ACL. In a canine model, Murray et al. used collagen with or without PRP to repair an ACL central defect and evaluated histologic and mechanical properties of the repaired ACL over a period of 6 weeks. Results indicate that the collagen gel with PRP showed a significantly higher percentage of defect filling and strength compared with the control group. In addition, it was found that the PRP-augmented collagen resulted in regenerated tissue that had similar properties to those of the medial collateral ligament, which has a better healing ability than the ACL. This collagen-platelet composite (CPC) is also used as a supplement to the standard allograft repair approach. In one study, Joshi et al. performed unilateral ACL reconstruction procedures in pigs with a bone-patellar tendon-bone allograft with or without the addition of CPC on the surgery site, and outcomes were evaluated at 4 and 6 weeks and 3 months after surgery. Although at 6 weeks a temporary decrease in yield load and stiffness was seen in both groups, by 3 months the group that underwent repair with CPC had improvements in yield load and linear stiffness of the repair tissue. In another study, using the same animal model, Fleming et al. evaluated the effect of the addition of CPC over a period of 15 weeks. Results confirmed the previous findings that in the long term, yield and maximum failure load of CPC-supplemented groups were greater than that of the group that underwent standard ACL reconstruction. In addition, histologic analysis revealed that the graft structure properties were also improved by the addition of CPC. Furthermore, the composite can be used by itself as an implant to enhance the suture repair procedure on the ACL. Two studies were conducted to evaluate the individual effects of collagen or PRP in ACL regeneration in a porcine model, and it was found that neither one improved the functional properties. More recently, Mastrangelo et al. combined collagen with PRP and used CPC with different PRP concentrations (×3 and ×5 of the baseline, respectively) to bridge the ACL stump and the femur tunnel and evaluated the mechanical properties of the repaired ACL over 13 weeks. It was found that regardless of PRP concentration, CPC increased the mechanical properties of repaired ACL. Results from these studies collectively demonstrated that the use of PRP-loaded collagen has great potential for ACL repair.

In summary, because of the inherent poor healing potential of ligaments such as the ACL, implants are used for ligament reconstructions, with autografts and allografts being the most common. In the 1980s, synthetic ACL such as Gore-Tex, Lars Ligament, and Leeds-Keio were popular and were used widely in ACL reconstruction surgeries. Although satisfactory short-term results were reported, long-term outcomes were poor and eventually resulted in most of the synthetic implants being extracted and withdrawn from the market. To address the unmet market demand for ligament grafts that were an alternative to biologic grafts, tissue engineering has arisen as a promising method by which to regenerate the ACL. A variety of polymeric materials were tested in vitro and in vivo; different methods such as incorporation of growth factors and active loading of the scaffolds were used to enhance graft performance. In addition, stratified implants were designed and showed promising results in vitro and in vivo for ligament-to-bone integration. Although they are promising, most of the tissue engineering options are still at the in vitro and small animal in vivo evaluation stages, and their true clinical potential remains to be demonstrated in clinical trials.

Implants for Meniscus Repair

The meniscus is a fibrocartilaginous tissue in the knee that functions to dissipate compressive and shear stresses during normal activity, as well as to distribute synovial fluid throughout the knee during loading and unloading. With a high water content (78 weight percent), the meniscus is 80% avascular and is composed primarily of type I and II collagen with other minor types of collagen, as well as proteoglycans. Meniscal tears can either be traumatic in origin or degenerative in nature. The current treatment options include both nonoperative (e.g., physical therapy, nonsteroidal antiinflammatory medications, and cortisone injections) and operative procedures, which include meniscal repairs and partial meniscectomies. Meniscal repairs, which consist of closing up a tear via suturing, are typically only performed for tears located in the vascular region of the meniscus that is associated with the meniscosynovial junction. Because of the specific nature of the types of injuries that can be repaired with this technique, partial meniscectomies are more common. In the United States alone, 690,000 partial meniscectomies were performed in 2006. A partial meniscectomy consists of removing the damaged tissue from the meniscus while leaving behind as much normal meniscal tissue as possible. Because the removed tissue extends into the avascular region, the limited access to blood flow results in a low repair rate, resulting in a hole in the meniscus that leaves a loss of functionality.

Consequently, meniscal allografts and tissue-engineered implants have been researched extensively to restore functionality to the knee. Although allografts have been shown to decrease pain and improve knee function in patients in the short term, inherent shortcomings of allogeneic tissue include limited supply, potential disease transmission (such as human immunodeficiency virus or hepatitis), and risk of infection or an immunologic response. The grafts have also been shown to demonstrate some shrinkage via magnetic resonance imaging (MRI) and a reduction in mechanical strength, leading to tears and dysfunction of the allograft. More recently, tissue-engineered meniscus implants have been developed as a viable alternative to allografts, as reviewed extensively by Brophy and Matava and van Tienen et al. The tissue-engineered graft serves as a scaffold structure at the defect site, which allows for cell infiltration from the surrounding native tissue into the structure. The cells are then able to regenerate new tissue while the scaffold structure sustains the mechanical loading and degrades away for eventual total replacement with regenerated, functional tissue filling the defect site. This section will focus on two main types of scaffolds: the collagen meniscus implant and polyurethane-based implants Actifit from Orteq and NUsurface from Active Implants ( Table 4-3 ). These implants have all undergone extensive in vitro and in vivo testing for biocompatibility, cell response, biodegradability, and mechanical properties. Moreover, all three implants have received the Conformité Européenne (CE) marking in Europe.

The collagen meniscus implant (CMI) is a purified type I collagen scaffold derived from bovine Achilles tendon, and glycosaminoglycans are added to the collagen, after which the structure is cast in a mold, lyophilized, and cross-linked in formaldehyde. The implant requires a meniscal rim and intact anterior and posterior meniscal horns for attachment during surgery and is used to treat medial and lateral meniscus injuries. Multiple clinical studies were conducted in which the CMI scaffold was implanted in patients with meniscal tears to evaluate the feasibility of the scaffold. In terms of the chondroprotective capabilities of the CMI, several studies have shown that the joint space and chondral surfaces were preserved after implantation. MRI revealed that at 5 to 10 years after surgery, new meniscal tissue was formed and integrated well with host tissue. However, the neotissue differed in MRI signal from the surrounding native tissue, suggesting that the regenerated tissue did not completely match the native tissue in structure and composition. In most cases, the newly formed meniscus tissue was reduced in size compared with the host tissue, but based on patient scoring, these differences were not found to be clinically significant. In addition, the defect filling was estimated to reach 70% in 1 year. Mixed results on implant resorption have been reported, spanning from no observable resorption to complete implant degradation in 5 years; as such, in all cases, no inflammation or negative effects have been reported. To assess the efficacy of the CMI compared with a control group that underwent partial meniscectomy without use of a scaffold, Rodkey et al. followed up on 311 patients (aged 18 to 60 years) who underwent either CMI implantation or a partial meniscectomy. Results showed that patients treated with CMI had an improvement in activity levels, as evaluated at 5 years using the Tegner Index for activity, whereas Lysholm scores for pain were found to be the same for both procedures. Moreover, the number of revision surgeries required was reduced by 50% with the CMI. A 10-year study was performed by Zaffagnini et al. with 36 male patients (aged 24 to 60 years). Regardless of CMI implantation, two patients per group required revision surgery during the 10-year period. The improved activity level compared with the control group at 5 years as evaluated by the Tegner index was also observed here, with a similar activity level being maintained for over 10 years. In addition, patients implanted with the CMI showed either significant pain improvement at 10 years (when evaluated using a visual analog pain scale), or a similar pain score compared with the control group (when evaluated using the Lysholm score, as was found previously by Steadman and Rodkey ).

Another class of meniscal implants with CE approval is based on polymeric materials. Actifit is an aliphatic polyurethane implant developed for medial and lateral partial meniscus tears. The graft is composed of two types of degradable polymers with distinct mechanical properties: 80% of a mechanically soft polymer poly(ε-caprolactone) and 20% of a mechanically stiffer segment polyurethane. This polymer blend was optimized through in vitro and in vivo testing for mechanical properties, degradation properties, and biocompatibility. The scaffold exhibits a relatively slow degradation rate (it takes up to 5 years to fully degrade). During clinical testing by Verdonk et al., of the 52 patients with partial meniscus defects, patients receiving the Actifit implant were compared with patients undergoing a standard partial meniscectomy. For one group treated with Actifit, tissue ingrowths of 85.7% were observed at 3 months, with 12-month biopsies showing cells with meniscus-like differentiation potential, determined through CD34 immunostain for vessels, Sirius red for collagen, and hematoxylin and eosin for cellular location and morphology. All patients receiving the Actifit implant measured significant improvements in terms of International Knee Documentation Committee score, Knee Injury and Osteoarthritis Outcome score, and Lysholm knee scale, showing that in 2 years, the scaffold is capable of restoring a certain level of knee functionality.

Another meniscal implant of interest is the NUsurface meniscus implant, a polycarbonate-urethane implant reinforced circumferentially with Kevlar fibers to mimic the functional properties of the meniscus. These fibers are based on the native menisci’s oriented collagen fiber network. Preliminary studies in a sheep model showed that the implant resisted wear and mild cartilage degeneration, although the total osteoarthritis score was not affected.

Other promising technologies are being investigated, although many have not reached the clinical testing stage. For example, biodegradable polycaprolactone nanofibers have been used to mimic the native fiber alignment of collagen found in the meniscus with promising results and potentially superior long-term biocompatibility. In vivo testing of foam polycaprolactone scaffolds revealed the formation of meniscus-like tissue with initial mechanical properties approaching those of the native meniscus. Hydrogel scaffolds (e.g., polyvinyl alcohol) for meniscus repair have also been evaluated in vivo and have been shown to be chondroprotective in addition to promoting meniscus-like tissue regeneration. Although these tissue-engineered implants appear to be promising, clinical applications are pending based on preclinical and clinical outcomes.

At present, two meniscal implants (CMI and Actifit) have reached the commercial stage in Europe, with NUsurface undergoing a large multicenter clinical test before being made available to the market. In the United States, CMI received market approval from the FDA in December 2009, but this approval was rescinded in October 2010 because of its intended use being “for different purposes [and being] technologically dissimilar from devices already on the market, called predicate devices.” Both the Actifit and NUsurface implants require larger randomized clinical trials for consideration by the FDA, both of which are in progress as of the writing of this chapter. In addition, several promising meniscal implants are currently in the research phase, with signs of advantages over traditional allografts. The outlook on meniscus implants is promising, and current grafts show a restoration in knee function and pose no safety issues. Further improvements can be made with new advances in scaffold design to demonstrate efficacy of these implants over current meniscal repair techniques.

Implants for Cartilage and Osteochondral Repair

Articular cartilage lines the surfaces of joints and enables near frictionless motion and load bearing. Composed of both liquid and solid phases, cartilage is a highly specialized tissue with complex structure-function relationships. It is largely avascular and aneural, and consequently, it has a limited capacity for self-repair. Clinical treatments of cartilage defects include joint lavage, subchondral drilling, microfracture, and osteochondral transplantation (with autografts or allografts). However, poor long-term outcomes are associated with many of these techniques because of unwanted fibrocartilage formation and inadequate graft-to-bone integration. Furthermore, although osteochondral allografting has demonstrated positive results, risk of disease transmission and issues with graft availability, preservation, and storage remain (see the review by Redman et al. ).

Targeted at addressing the limitations of these treatment options, Brittberg et al. published the first clinical report demonstrating the promise of autologous chondrocyte implantation (ACI), a two-step procedure in which autologous chondrocytes are arthroscopically harvested from a healthy cartilage donor site (typically the intercondylar notch of the knee), expanded in vitro, and then reintroduced to the defect site through a second open procedure. In addition, an autologous periosteum flap is harvested from the patient and used to contain the cells within the defect site. This procedure was the first FDA-approved cell-based product for cartilage repair and served to bridge the gap between previously available techniques and total joint replacement, because this procedure eliminated issues of disease transmission, as well as graft availability, preservation, and storage. Although ACI results in satisfactory clinical outcomes, several limitations are inherent to the procedure. Of note, the necessity of two surgical procedures increases cost and leads to extended recovery times. Furthermore, cases of hypertrophy have been reported. Finally, it is unclear if the cells recover completely from long-term monolayer culture, if they are homogeneously distributed within the repair tissue, and how many cells are actually retained within the defect site.

To tackle the shortcomings of the ACI procedure, second- and third-generation techniques have been developed that use matrices that eliminate the need for periosteal harvest, support and retain the chondrocytes in a three-dimensional matrix, and enable a homogeneous distribution of the cells within the defect. The matrices that have been developed are composed of a wide variety of materials, both natural and synthetic, and have shown great promise in initial trials. Natural matrices are attractive because they can be designed to closely mimic the native cartilage matrix. Matrix-assisted chondrocyte implantation (MACI) is a second-generation ACI technique in which a porcine-derived collagen I/III bilayer is seeded with autologous expanded chondrocytes. The use of a scaffold effectively eliminates the need for periosteal harvest. Bartlett et al. conducted a randomized comparison of MACI and ACI-C (ACI with a collagen I/III flap used in place of periosteum) for treatment of chondral knee defects in 91 patients and found that the two treatments resulted in clinically comparable outcomes ( Table 4-4 ). More recently, MACI has been compared with microfracture by Basad et al. in a study of 60 patients with isolated cartilage defects, and it was found that MACI was significantly more effective over time than microfracture according to three different scoring systems (the Tegner index, International Cartilage Repair Society–patient, and International Cartilage Repair Society–surgeon). Zheng et al. performed histologic analysis on a cohort of 56 MACI-treated patients and found that this technique supported chondrocyte phenotype maintenance, which was determined by aggrecan, type II collagen, and S-100 expression. After 6 months, 75% hyaline cartilage regeneration was reported. Behrens et al. and Ebert et al. have both performed 5-year follow-up studies for 11 and 41 patients, respectively, and found high patient satisfaction and low failure rates.

The Cartilage Regeneration System (CaReS) also uses a collagen matrix for cell-based cartilage regeneration; however, CaReS is made from a rat-derived collagen type I matrix that is seeded with primary cells that have not been expanded in monolayer culture. This technique is based on the assumption that cells that have not been exposed to monolayer expansion are more effective for regenerating cartilage tissue. Flohe et al. compared the CaReS system with the MACI procedure for repair of cartilage defects in the knees of 20 patients and found that both treatments resulted in improved clinical outcomes after 1 year, with no significant differences detected between treatments. In a multicenter clinical trial, Schneider et al. followed up with 116 patients who received the CaReS implant between 2003 and 2008. The overall treatment satisfaction was judged as good or very good in 88% of the cases by the surgeon and in 80% of the cases by the patient. These observations are highly promising, but longer term follow-up will be necessary to determine if the CaReS system has distinct advantages.

Monolayer expansion of autologous chondrocytes is also avoided in the NeoCart system in which chondrocytes are cultured on scaffolds made of bovine collagen type I in custom bioreactors that mimic the conditions of the knee through varying pressure and low oxygen tension. Crawford et al. demonstrated the clinical safety of this system in a small trial in which eight patients received the NeoCart treatment. In this study, pain scores were significantly reduced after treatment, and none of the patients experienced hypertrophy or arthrofibrosis. More recently, the NeoCart implant was compared with microfracture in a randomized trial of 30 patients, and it was found that significantly more patients treated with NeoCart responded positively to the treatment at both 6 and 12 months, with the trend continuing at the 2-year follow-up.

A more complex implant, ChondroMimetic, is composed of three natural materials and is designed to closely mimic the natural cartilage environment. This implant is a dual-layer porous plug composed of collagen, glycosaminoglycans, and calcium phosphate. The scaffold can be prehydrated with sterile fluids and autologous blood. Although ChondroMimetic is an acellular, off-the-shelf product, it can be used in conjunction with ChondroCelect, which is a cell-based technology that is offered by the same company, TiGenix. ChondroMimetic was launched in October 2010 in Europe; however, clinical study results have not yet been published.

In addition to collagenous implants, several hyaluronan-based matrices have been developed. Hyalograft C is a hyaluronic acid scaffold (HYAFF) that is combined with autologous chondrocytes. Hyalograft C can be used in both arthroscopic and open procedures, and satisfactory clinical outcomes have been reported after 7 years. Improved clinical outcomes were reported for young patients in a prospective study of 70 patients with 3- and 4-year follow-up and in a study of 36 patients with 2- and 3-year follow-up analysis. A study of 62 patients with 7-year follow-up found that, when compared with female patients, young active men had the best clinical outcomes when treated with Hyalograft C. Nehrer et al. reported that although Hyalograft C results in satisfactory repair for patients with a primary indication (e.g., young patients with a stable and healthy knee joint with an isolated chondral defect), it is a poor option for salvage procedures or for patients with osteoarthritis. Kon et al. compared Hyalograft C with microfracture in 41 professional or semiprofessional soccer players and found that although microfracture allowed players to return to competition more quickly, repair with Hyalograft C may offer more durable clinical results. In addition, Hyalograft C was compared with MACI by Kon et al. in a trial of 61 patients who were older than 40 years. A faster improvement in the International Knee Documentation Committee subjective score was reported for the patients treated with Hyalograft C, whereas similar scores were reported at the 2-year follow-up.

A more recent product, BioCart II, combines recombinant hyaluronan with homologous human fibrin to form a macroporous sponge that is seeded with autologous chondrocytes that have been primed with a recombinant fibroblast growth factor 2 variant. A preliminary study by Nehrer et al. reported good defect filling with the BioCart II system in a study of eight patients. Significant improvement in defect healing over time was subsequently reported by Eshed et al. in a study that evaluated 31 patients at time points ranging from 6 to 49 months after BioCart II implantation.

Another scaffold based on a polymer derived from nature is BST-CarGel, which is an injectable chitosan-based scaffold that is used in conjunction with bone marrow stimulation to form a volume-stable clot that drives cartilage regeneration. BST-CarGel is injected into the defect in a single-step procedure and cross-linked in situ. Shive et al. followed up with 33 patients treated with BST-CarGel and reported preliminary evidence suggesting that BST-CarGel has the potential for treatment of focal cartilage defects with varying etiology. Additionally, alginate and agarose-based scaffolds such as the Cartipatch system and a bead system have been investigated; however, few clinical reports of these approaches have been published.

In addition to naturally derived products, several synthetic polymer-based scaffolds are currently on the market in Europe. BioSeed-C is a polyglycolic/polylactic acid- and polydioxane-based material that is combined with culture-expanded autologous chondrocytes that are suspended in fibrin. Kreuz et al. followed up on 19 patients with osteoarthritis who had received BioSeed-C treatment and reported good clinical outcomes 1 year after implantation. Moreover, BioSeed-C remained stable over the course of a period of 4 years, suggesting that it may be a promising treatment option for the repair of focal degenerative cartilage defects in the knee. In a larger study of 52 patients with full-thickness defects by Kreuz et al., BioSeed-C treatment resulted in good clinical outcomes after 4 years despite a persisting deficit in mechanical strength. The authors suggest that this deficit may be addressed with a focus on muscular strength during rehabilitation.

The Cartilage Autograft Implantation System (CAIS) is another polymer-based approach that consists of absorbable copolymer foam of 35% polycaprolactone and 65% PGA, reinforced with a polydioxanone (PDO) mesh (Advanced Technologies and Regenerative Medicine). In a one-step procedure, autologous cartilage is harvested, minced, and uniformly distributed within the scaffold using a fibrin sealant. The polymer foam is designed to keep the tissue fragments in place while the PDO mesh enables the foam to have adequate mechanical strength during implant handling. Cole et al. compared CAIS treatment with microfracture in a randomized study of 29 patients at 1 and 2 weeks and periodically up to 2 years after surgery. It was found that CAIS resulted in significant increases in select subdomains in the Knee Injury and Osteoarthritis Outcome Score assessment tool, and it was concluded that CAIS is a safe, feasible, and effective method for treating patients with focal chondral defects that may improve long-term clinical outcomes.

In addition to chondral grafts, several implant systems have been developed to address both chondral and osteochondral defects. Cartiva is a polyvinyl alcohol cryogel that has been used in patients since 2002 and consists of cylindrical gels that can be used to replace osteochondral grafts (either autografts or allografts). Falez and Sciarretta performed a preliminary clinical study and concluded that the use of this type of treatment should be limited to precise indications: grade 3 and 4 chondral or osteochondral symptomatic defects, focal unicompartmental defects with 15-mm maximum extent, limitation of the patient’s age to the fourth to seventh decade of life, or absence of angular deformities or articular instabilities. Although cases of failure and dislocation have been reported, the synthetic cartilage resurfacing technique has the advantages of no donor defect, one short-step surgical procedure, and immediate weight-bearing ability.

The TruFit CB plug is also a synthetic osteochondral implant that consists of a porous PLGA scaffold that is reinforced with polyglycolic acid fibers and calcium sulfate mineral. Dhollander et al. investigated the TruFit CB for osteochondral repair and observed modest outcomes. In a more recent study, Joshi et al. reported that whereas the TruFit CB system led to initial symptom relief in 10 patients with a median age of 33.3 years, a failure to regenerate subchondral bone over a 2-year period was observed, which in the long term could lead to implant failure and a repeat operation.

Overall, a plethora of implants based on both natural and synthetic materials for cartilage and osteochondral repair are commercially available. Implants offer advantages over the ACI procedure because they circumvent the use of periosteal tissue, they address the challenge of homogenous cell distribution and retention within the defect site, and they can provide a three-dimensional matrix that supports chondrocyte phenotype maintenance. Although most implants are engineered to regenerate cartilage tissue, osteochondral implants provide support for both bone and cartilage tissue regeneration. Currently no FDA-approved implants are available in the United States; however, more clinical data are expected to elucidate the long-term stability and advantages of synthetic graft systems. As the field matures, a focus on interface regeneration and consistent bone-cartilage integration would further enhance the utility of cartilage and osteochondral grafts.