|
Dunn, M.G. (1998). Anterior Cruciate Ligament Prostheses. In: Encyclopedia of Sports Medicine and Science, T.D.Fahey (Editor). Internet Society for Sport Science: http://sportsci.org. 10 March 1998.
Initial Mechanical
Properties of ACL Prostheses
Effect of
Surgical Fixation
Post-Surgical
Changes in Prosthetic Ligaments
Permanent
Prostheses
Non-Degradable
Scaffolds Designed for Tissue Ingrowth
Graft-Augmentation Devices
Future
Directions: Tissue Engineering
Resorbable
Synthetics and Tissue-Derived Biomaterials
Fibroblast-Seeded Scaffolds
Summary
References
There are several approaches to treating severe knee ligament injuries, including non-operative management, primary repair, and surgical reconstruction using biological grafts or synthetic prostheses. Collateral ligament injuries heal well with conservative treatment or surgical repair. In contrast, surgical reconstruction is recommended for cruciate ligament tears, due to their poor intrinsic healing potential. The anterior cruciate ligament (ACL) is injured more frequently and has been studied much more extensively than the posterior cruciate ligament (PCL). Although significant progress has been made toward understanding the anatomy, composition, biomechanics, and healing of the ACL, there is still no graft or prosthesis ideally suited for ACL reconstruction.
Initial Mechanical Properties of ACL Prostheses
The ACL is considered to be the primary mechanical link between the femur and the tibia, so its mechanical properties-are crucial to its function. The mechanical behavior of the ACL is complex, with variable tension among fiber bundles, variable modulus as a function of load level, viscoelastic effects (creep, relaxation, strain-rate sensitivity), etc., which are discussed elsewhere. Here, the ultimate mechanical properties (e.g., breaking load and stiffness; tensile strength and modulus) of the ACL and prostheses are considered. The ACL has considerable strength and modulus due to an aligned type I collagen network that bears great loads while undergoing little deformation. Breaking loads of 1725-2160 N have been reported for young human ACLs. The tensile strength of the human ACL is at least 38 MPa. The ultimate mechanical properties of ligaments generally increase during development, then decrease with aging.
Various synthetic materials have been used either as 'permanent' ACL prostheses, tissue ingrowth devices, or graft augmentation devices. The first generation of ACL prostheses had very poor initial mechanical properties. Today, prostheses can be fabricated with more appropriate initial mechanical properties, equaling or exceeding the properties of the normal human ACL. For example, the Gore-Tex ligament has a breaking load of 4800 N. Regardless of their initial mechanical properties, however, most permanent ligament prostheses still perform poorly in the long-term. Furthermore, the initial mechanical properties of the reconstructed bone-ligament-bone (BLB) complex depend on other factors besides the mechanical properties of the prosthesis - the method of surgical fixation to the bones is a limiting factor.
Although the mechanical properties of the prosthesis are important, the initial mechanical properties of the reconstructed BLB complex are limited by the method of surgical fixation to the femur and tibia. The freshly reconstructed BLB complex consistently fails at the osseous attachment sites, not in the midsubstance of the prosthesis. Typically, surgical bone tunnels are made for fixation at the anatomic ACL attachment sites, although some have argued that over-the-top fixation can be biomechanically equivalent or superior to anatomic tunnel fixation. Prostheses can be attached to bone using sutures, staples, screws, washers, and combinations of these methods. Simple suture or staple fixation fails at very low loads, less than 200 N. Screws with washers or plates perform better, failing at loads of about 200-250 N. A double 'locking ligament' suture fails near 400 N; when reinforced with a staple the failure load exceeds 400 N.
Some ACL prostheses have plugs on the ends, resembling bone plugs, and are fixed within the bone tunnel using interference screws, similar to patellar tendon grafts. The strength of this type of fixation is heavily dependent on the quality of bone and bone compression. The Kurosaka custom 9.0 mm screw provides improved breaking load (437 N) and stiffness (58 N/mm) compared to an AO 6.5 mm screw (breaking load 161 N; stiffness 36 N/mm). Failure loads as high as 600 N have been obtained using this larger diameter screw. High failure loads (454 N) have also been reported recently for a 'suture over screw post' method of anchorage.
In short, surgical fixation is a weak link that severely limits the initial mechanical properties of the reconstructed BLB complex. Thus, even if a prosthesis has a breaking load exceeding 2000 N, the initial breaking load of the reconstructed BLB complex may be only 25% of that value, about 450 N. We have focused on breaking load and strength in this discussion, but be aware that stiffness and modulus can be similarly reduced by surgical fixation. That is, under a given load, the deformation of the reconstructed BLB complex is increased compared to an isolated prosthesis, due to the effects of surgical fixation.
Several other surgical factors strongly influence the initial and long-term results of ACL reconstruction. For example, bone tunnel placement and initial tension on the implant must be carefully controlled to achieve consistent results. Determination of the best values for these and other surgical variables, to optimize knee biomechanics following ACL reconstruction, continues to be an important and active field of orthopedic research.
Post-Surgical Changes in Prosthetic Ligaments
Synthetic polymers evaluated clinically for ACL reconstruction include polytetrafluoroethylene (Gore-Tex), polyethylene terephthalate (Dacron; Stryker-Meadox and Leeds-Keio ligaments), carbon fibers (Integraft), and braided polypropylene (Kennedy Ligament Augmentation Device). Although several synthetic ACL prostheses are conditionally approved by the United States Food and Drug Administration (FDA) for salvage cases, or as graft augmentation devices, no prosthesis is unconditionally approved for primary ACL reconstruction. Experimental and clinical ACL reconstruction studies have generally shown poor long-term results due to persistent pain, synovitis, sterile effusions, arthritis, and mechanical breakdown of the synthetic polymers.
A permanent synthetic prosthesis that does not receive host tissue ingrowth is prone to long-term mechanical failure in the joint. Potential mechanical problems associated with permanent synthetic prostheses include creep (gradual stretching under load) and fatigue failure due to cyclic loading. Furthermore, the prosthesis may fray or rupture at the sharp bone tunnel attachment sites. Then, prosthesis fragments or wear debris particles may accelerate damage to the surrounding tissues.
The Gore-Tex prosthetic ligament is composed of braided bundles of polytetrafluoroethylene (PTFE), an inert polyethylene with fluorine substituted for hydrogen side groups. The failure load of the prosthesis is 4830 N and the stiffness is 322 N/mm. The Gore-Tex device performed well as a permanent ACL prosthesis in 15 month clinical trials, improving knee stability in 129 out of 130 patients. In 1986, the FDA conditionally approved the Gore-Tex ligament prosthesis to salvage failed autografts, sparing the patient from a total knee replacement procedure. Clinical trials using the device for ACL reconstruction continue to be reported, and the results seem to deteriorate with time.
In two-year clinical trials, 87% of the patients had satisfactory results, but 4 of 39 prostheses had ruptured. In a 3 year study, the Gore-Tex prosthesis performed poorly due to sterile effusions and persistent pain in 6 out of 18 patients. The probability of survival of the prosthesis decreased sharply from 2 to 5 years post-implantation, and the overall failure rate of the Gore-Tex prosthesis was 33%. In 268 patients with average follow-up of 4 years, unacceptable results were reported in 56% of the patients; effusions occurred in 34% of the patients, and the device ruptured in 12% of the patients.
Permanent ACL prostheses have also been made from Dacron (polyethylene terephthalate), a polyester with a rigid aromatic ring incorporated into the backbone chain. The Stryker-Meadox Dacron ACL prosthesis was not approved by the FDA for ACL reconstruction. Using a modified Macintosh procedure and augmentation of the Dacron prosthesis with iliotibial band, a 60% failure rate was found at a mean follow-up time of almost 4 years. Five of 40 Dacron prostheses (12.5%) had ruptured by this time. At 5 years follow-up, 23% of unaugmented Dacron prostheses had ruptured. Similarly, 28% of Dacron prostheses had ruptured at an average follow up of 4.4 years in another study. This study was particularly disturbing because degenerative arthritis was found in the patients with the ruptured prostheses, perhaps due to wear debris. Others reported overall failure rates of 20% at 2 years increasing to 35.7% at 5 years follow-up for permanent Dacron ACL prostheses.
The results of ACL reconstruction deteriorate with time for both Gore-Tex and Dacron permanent prostheses. Permanent ACL prostheses that do not induce supportive tissue ingrowth will likely fail in the long-term, due to synovitis, effusions, arthritis, or mechanical deterioration of the prosthesis.
Non-Degradable Scaffolds Designed for Tissue Ingrowth
Tissue ingrowth or scaffolding devices for ACL reconstruction have been developed from several types of polymers, including Dacron and carbon fibers. These devices are designed with a porous or filamentous structure to promote host neoligament tissue ingrowth. The amount of host tissue ingrowth depends more on the mechanical properties and surface morphology than the chemical composition of the implant. The implanted scaffold should provide mechanical support and promote ingrowth of host neoligament tissue. Mechanical loads should be transferred gradually from the implanted scaffold to the neoligament tissue.
The Leeds-Keio synthetic ligament, composed of a Dacron mesh, was designed as a scaffolding structure to support host tissue ingrowth. The inventors of this device have reported successful clinical results, with arthroscopic observations documenting neoligament tissue development within the implanted Dacron scaffold. Other investigators, however, reported nonaligned fibrous tissue, not true neoligament tissue, ingrown within the device after implantation in a sheep model. They suggested that the Leeds-Keio ligament did not serve as a true scaffolding device, but instead behaved as a permanent load-bearing prosthesis, subject to long-term failure in the joint.
Filamentous carbon has also been used to develop a scaffolding device to induce neotendon and neoligament tissue ingrowth. Implants consist of approximately 10,000 individual fibers, each with a diameter of about l0 mm. The individual fibers have very high tensile strength (2100-2350 MPa), but are brittle, resulting in fiber fracture and the release of potentially harmful debris. A resorbable polymeric coating was somewhat successful in preventing carbon fiber breakage and localizing debris. However, due to poor performance and permanent wear debris in the joint, the Integraft carbon fiber device was not approved by the FDA for ACL reconstruction.
In theory, a tissue ingrown device, unlike a permanent prosthesis, should remodel in response to long-term mechanical loads. Although the concept is interesting, there have been problems in implementing this design. Tissue ingrowth may be delayed, or may occur only in limited amounts. Even if ingrowth is substantial, the tissue can be disorganized and weak, resembling scar tissue instead of neoligament tissue. If the stiffness of the implant greatly exceeds that of the ingrown host tissue, most of the mechanical load will be borne by the implant, and the load-deprived host tissue will not remodel or mature. This phenomenon of stress-shielding is a major concern for both tissue ingrowth devices and graft augmentation devices.
Graft augmentation devices were developed to protect biological grafts from high loads in the early postoperative period of graft weakness. These devices are normally implanted in parallel with a biological graft (autograft or allograft) to share the mechanical loads. The amount of load borne by each component is proportional to its stiffness. Therefore, the device stiffness should approximate that of the biological graft to avoid stress-shielding.
Dacron has been used as a graft augmentation device in addition to being used as a permanent prosthesis and a tissue ingrowth device. The Dacron device reportedly shielded autogenous grafts from bearing mechanical loads, resulting in poor long-term neoligament development in augmented grafts compared to unaugmented grafts. Stress-shielding continues to be a major concern for graft augmentation devices, and tissue ingrowth devices as well.
To address the problem of stress-shielding by stiff synthetic polymers, investigators augmented patellar tendon autografts in primates using the Gore-Tex ACL prosthesis placed in an over-the-top position (nonparallel, nonisometric). The device was implanted this way to provide increased protection to the graft as the knee was extended. Knee laxity was decreased by this augmentation procedure, and autograft incorporation and remodeling occurred at similar rates for augmented and control groups.
The Kennedy LAD (ligament augmentation device) is a braided polypropylene yarn with an initial breaking load of about 1500 N and a stiffness of 36 N/mm. The stiffness of the Kennedy LAD is low, and the device is attached to bone on only one end of the autogenous repair. Due to the low stiffness of the device, stress-shielding of the graft is reduced, so normal neoligament remodeling should occur. Patellar tendon allografts in sheep had improved initial strength and stability in LAD-augmented knees. In the long-term, the biomechanical properties of augmented and unaugmented allografts were similar, suggesting that normal tissue remodeling occurred in the presence of the augmentation device. Clinical results using this LAD were promising, and the FDA conditionally approved the device only for augmentation of a Marshall Macintosh ACL reconstruction procedure. However, investigators recently reported no difference in the clinical outcome for augmented and unaugmented semitendinosus tendon autografts at about 2 years follow-up.
An augmentation device is needed only temporarily, to share mechanical loads with the biological graft, so long-term maintenance of the mechanical properties of the device is not necessary or even desirable. Ideally, graft augmentation devices should be resorbable, gradually transferring mechanical loads completely to the biological graft.
Future Directions: Tissue Engineering
The 'tissue engineering' approach to ACL reconstruction uses resorbable scaffolds consisting of tissue-derived and/or synthetic materials to induce neoligament formation. This concept was originally developed for repair of other connective tissues including skin, bone, and cartilage. In contrast to permanent synthetic prostheses that lose strength with time, the mechanical behavior of these implants should improve with time due to neoligament tissue development and remodeling. Fibroblast-seeding of resorbable scaffolding devices is another promising concept that is currently a focus of ACL reconstruction research in our laboratories at Robert Wood Johnson Medical School.
Resorbable Synthetics and Tissue-Derived Biomaterials
Although the use of resorbable synthetic polymers for tendon/ligament reconstruction is an interesting idea, few reports are found. A resorbable device composed of braided polyglycolic acid suture (Dexon) was successfully used to reinforce ACL repair in dogs. Others reconstructed the rabbit Achilles tendon with resorbable fibers composed of dimethyltrimethylene carbonate and trimethylene carbonate. At 26 weeks, the implanted fibers were still intact, and fibrous tissue in the center of the prosthesis was disorganized. A more rapid fiber degradation rate is needed to promote earlier tensile loading and organization of the host ingrowth tissue.
In addition to synthetic polymers, tissue-derived materials also show promise for development of novel ACL reconstruction devices. Type I collagen is the major structural component of tendons and ligaments. It can be extracted from tissues and processed into thin, high-strength fibers to make resorbable scaffolds with in vivo mechanical behavior similar to autografts. The resorption rate of collagen can be controlled by the extent of crosslinking, and no permanent wear debris is generated during its resorption. Collagen is not highly antigenic, and is chemotactic for fibroblasts and other cells involved in tissue repair. Reconstituted collagen has been used extensively as a scaffold to accelerate dermal repair. Growth factors and other extracellular matrix components can be added to enhance the biological activity of collagen.
We have surgically reconstructed both the rabbit Achilles tendon and ACL using collagen fiber-based scaffolds. These degradable implants induce deposition of neotendon and neoligament tissue. The collagen scaffolds experience early strength loss, followed by tissue ingrowth and strength regain, similar to autogenous grafts. The rate of strength regain in neotendon and neoligament tissue is proportional to the degradation rate of the implanted collagen scaffold. In surgical bone tunnels, the collagen scaffolds resorb rapidly, and induce rapid ingrowth of fibrous tissue and bone. These results suggest that tissue ingrowth in the bone tunnels might provide biological fixation for collagen prostheses used for ACL reconstruction.
Although our preliminary results of Achilles tendon and ACL reconstruction studies have been encouraging, we need to optimize the initial strength and resorption rate of these collagen based devices. We found that by minimizing the diameter, fiber strength can be increased without prolonging fiber resorption rate. Recently, we used a polylactic acid polymer matrix to further improve the initial strength of the collagen fiber scaffold without severely prolonging its resorption rate. In addition, we are seeding the collagen scaffolds with fibroblasts in an attempt to improve neoligament formation.
In vitro 'seeding' of viable fibroblasts within the scaffold implant may stimulate early healing and improve neoligament formation. Fibroblasts play a major role in collagen turnover during healing and tissue remodeling. Furthermore, fibroblast function depends on the source or type of fibroblast. For example, fibroblasts from the ACL and medial collateral ligament have different intrinsic capacities for migration and proliferation. We recently showed that 'ligament analogs' consisting of collagen scaffolds seeded with extra- or intra- articular fibroblasts have different healing potentials in vitro depending on the fibroblast source. These fibroblast-seeded collagen scaffolds should be a useful tool to explore the role of different fibroblast types in neoligament formation. The clinical goal is to promote early healing and improve long-term remodeling and biomechanical performance in the reconstructed ACL by using ligament analogs (fibroblast-seeded scaffolds) instead of biological grafts or synthetic prostheses.
The long-term biomechanical performance of ACL prostheses is unsatisfactory. Permanent ACL prostheses are prone to creep, fatigue, and mechanical failure within several years after implantation. Tissue ingrowth scaffolds and ligament augmentation devices require further refinement to provide effective mechanical support while avoiding stress-shielding of the host tissue. Perhaps development of resorbable, tissue-inducing biomaterials and fibroblast-seeded biomaterials will improve the long term biomechanical performance of the reconstructed femur ACL-tibia complex. Carefully planned and executed research studies, in experimental animals and in the clinic, are required to reach this difficult goal.
This essay was condensed and adapted from Dunn, M.G.: Biomechanics of Ligament Reconstruction, in: A.J. Tria, Jr., (ed.). Ligaments of the Knee, New York: Churchill Livingstone, 1995, with permission.
Daniel, D. , W. Akeson, and J. O'Connor (Eds.). Knee Ligaments: Structure, Function, Injury, and Repair. New York: Raven Press, 1990.
Dunn, M.G., A.J. Tria, J.R. Bechler, R.S. Ochner, J.P. Zawadsky, Y.P. Kato, and F.H. Silver. Anterior cruciate ligament reconstruction using a composite collagenous prosthesis. A biomechanical and histologic study in rabbits. Am. J. Sports Med. 20: 507, 1992.
Dunn, M.G., J.B. Liesch, S.H. Maxian, M.L. Tiku, and J.P. Zawadsky JP: The tissue engineering approach to ligament reconstruction. Proc. Material Res. Soc. 33: 13, 1994.
Friedman, M.J. and R.D. Ferkel (Eds.). Prosthetic Ligament Reconstruction of the Knee. Philadelphia: W.B. Saunders, 1988.
Fu, F.H., C.D. Harner, D.L. Johnson, M.D. Miller, and S.L.Y. Woo. Biomechanics of knee ligaments: Basic concepts and clinical application. J. Bone Joint Surg. 75A: 1716, 1993.