Graft Options

ACL reconstruction using autograft patellar tendon was first described by Jones in 1963.⁴⁰ Now, patellar tendon autograft is considered the gold standard for ACL reconstruction.⁷⁷ Patellar tendon autograft offers high initial strength, earlier bone-to bone healing, and proven success by a multitude of outcome studies.^1,6,16,20,53 However, despite being the gold standard graft choice, there are several disadvantages to autogenous patellar tendon harvest. These include anterior knee pain, patellar shortening, decreased range of motion, longer surgical time, infrapatellar fat pad fibrosis, patellar fracture, and patellar chondrosis.* Dynamic disadvantages of autograft patellar tendon harvest include decreased quadriceps strength, quadriceps inhibition, altered patellar alignment, and decreased active range of motion.^36,44,51

Quadrupled hamstring autografts (semitendinosus and gracilis tendon autograft) have been shown to have the highest tensile strength and excellent clinical results. However, disadvantages include decreased knee flexion and hip extension strength, which can be detrimental to athletes who rely on knee flexion strength beyond 90 degrees of flexion (e.g., sprinters, wrestlers, gymnasts, martial arts practitioners).^36,74,79 Furthermore, hamstring strength has been shown to be protective of ACL reconstruction by way of the ACL-hamstring reflex arc. Taking the semitendinosus and gracilis tendons results in disruption of this arc and a decrease in the protective effect of the hamstrings on the ACL graft.⁹

Quadriceps tendon autograft is also being used in some centers for ACL reconstruction. Limited outcome data are available but early studies have suggested equivalent results between quadriceps tendon (with patellar bone plug) autograft and patellar tendon autograft. After a 2-year follow-up, a recent study has found that kneeling pain is significantly less than with a bone-patellar tendon-bone (BPTB) autograft but anterior knee pain in the two groups is similar.⁴⁵

Because of the potential drawbacks of autograft ACL reconstruction with patellar tendon, hamstring, or quadriceps tendon, allograft ACL reconstruction is increasing in the United States.^5,77 Advantages of an allograft include elimination of donor site morbidity, decreased surgical time, smaller incisions, lower incidence of arthrofibrosis, decreased postoperative pain, no loss of donor graft function, faster subjective recovery time, and predictable graft sizes.^19,27,41 Allograft options include patellar tendon, hamstring, anterior tibialis, posterior tibialis, Achilles, and fascia lata. However, allograft drawbacks also exist, such as disease transmission, slower incorporation, possible immunologic reaction, finite supply, tunnel expansion, inferior biomechanical strength, possibly increased failure rate, and cost.^{31,33,65,71,73}

Graft Procurement and Sterilization

There are over 150 individual tissue banks that provide allograft for surgical use. Most of these tissue banks are members of the American Association of Tissue Banks (AATB). Beginning in May 2005, federal legislation was passed mandating that all United States tissue banks be subject to the U.S. Food and Drug Association (FDA) “Good Tissue Practice” guidelines. These guidelines specify a minimum standard for tissue procurement, as well as testing and processing, require periodic inspections of tissue bank facilities, and necessitate reporting adverse events to the FDA.^19,77

Graft procurement begins with donor screening to eliminate donors who are at high risk for transmitting communicable diseases. This involves screening prior to death and evaluating cause of death. High-risk donors are eliminated; these include those who have a known communicable disease, show active signs of infection despite not having a diagnosis, or whose lifestyle places them at high risk for infection. Physical examination is the next step to assess the potential donor for high-risk behavior (e.g., needle marks, skin findings). Third, blood tests and tissue cultures are taken from the potential donor to rule out an otherwise undetermined communicable disease. Required serologic tests include human immunodeficiency virus (HIV) types 1 and 2, hepatitis B, hepatitis C, and syphilis. The AATB-affiliated tissue banks also test donors for human T-lymphotropic virus (HTLV) types 1 and 2. Certain tissue banks may accept positive bacterial cultures, relying on the sterilization process to eliminate the infection. Tissue banks that are part of the AATB require destruction of all tissue that demonstrates positive cultures for Clostridia or group A streptococcus.¹⁹ Once these criteria are met and passed, the graft is suitable for procurement in the standard sterile surgical fashion. Each tissue bank has a specific time window for acceptable retrieval of graft tissue. Once obtained, the graft is placed in an antimicrobial solution and taken to the sterilization plant.

Sterilization is performed via two techniques, tissue irradiation and chemical processing. It has been found that high levels of irradiation (>25 kGy) can reliably inactivate HIV and spores. However, these same high levels of irradiation can be detrimental to the biomechanical properties of the graft itself.^43,71,77 Thus, most tissue banks use lower dose irradiation, which inactivates most organisms but does not alter the strength of the graft. Low-dose radiation may not inactivate viruses such as HIV.⁷⁶ Various tissue banks will use different radiation dosages and some banks have abandoned irradiating as a sterilization technique because of adverse effects on graft biomechanics. Chemical processing involves a succession of cleansing, disinfection, and rinsing of the tissue to remove viable cells, lipids, and microorganisms. Penetration into the deepest parts of the tissue is the primary challenge in chemical sterilization. Various tissue banks have their proprietary chemical cleansing solutions. Tissue “sterility” (sterility assurance level) is defined by the tissue banking industry as P < .000001 that a viable microbe is present in tissue after having undergone the sterilization process. Once the sterilization process is complete, the graft is then frozen at −70° C to −80° C until surgery. Deep freezing has been shown to decrease the risk of graft rejection by causing cell necrosis and loss of immunogenicity.²⁷ Freezing of allograft tissue has led to variable results in biomechanical properties, with some studies suggesting no difference, whereas others demonstrating that freezing is detrimental to graft strength.^54,60 Deep freezing does not destroy HIV or the hepatitis C virus. Tissue irradiation, chemical processing, and deep freezing have been found in some studies to result in decreased graft biomechanical strength. Because the procurement and sterilization protocols of tissue banks can vary, the surgeon and hospital should be well aware of the quality and techniques of tissue processing of the company that they choose to use as their source of graft tissue.

Infection

The risk of disease transmission is an important factor when weighing the options of allograft versus autograft ACL reconstruction. Possible infectious agents include HIV, hepatitis B, hepatitis C, HTLV, syphilis, aerobic bacteria, and anaerobic bacteria. With tissue banks adopting varying procedures on graft sterilization, the risk of disease transmission is also variable. The risk of HIV infection has ranged in the literature from 1 to 400,000 to 8,000,000, with a commonly quoted figure of 1 to 1,600,000.^5,11,12,19 The risk of hepatitis B and C virus infection has consistently been shown to be higher than that of HIV transmission. Bacterial infection is also a concern with regard to disease transmission with Clostridium spp. (a spore-forming anaerobe) being a common pathogen. In a Centers for Disease Control and Prevention (CDC) study performed in 2002, 26 cases (18 used for ACL reconstruction) of allograft-associated bacterial infections were identified in approximately 1 million transplanted allografts. Half of these were found to be caused by Clostridium spp., with one death.¹³ In another CDC study,⁴² published in 2004, 70 cases of allograft-associated infection were reported; it was found that since 1995, 6 were caused by hepatitis C and none by HIV. Of these patients, 14 were found to have Clostridium infections.

Routine culture of allografts prior to implantation has yielded 4.8% to 9.7% positive cultures for bacterial organisms, but no clinical infection was correlated with these positive culture results.^14,30 Thus far, routine preimplantation cultures have not been recommended. Because tissue banks differ in their methods of sterilization, the surgeon should be familiar and comfortable with the tissue processing of the bank selected and be able to discuss possible infection transmission risks with patients as part of the informed consent.

Graft Biology

Allograft incorporation and healing is critical to the success of ACL reconstruction. One commonly cited disadvantage of allograft usage for ACL reconstruction is slower and less extensive incorporation compared with autograft ACL reconstruction.²⁴ Allograft versus autograft ACL reconstruction has been compared from histologic, biomechanical, and radiographic perspectives. A recent sheep study⁷² comparing native ACL, reconstructed soft tissue autograft ACL, and reconstructed soft tissue allograft ACL has suggested that allograft demonstrated delayed remodeling histologically at 6 and 12 weeks postoperatively. However, at 52 weeks, the differences were less apparent. Biomechanically, allografts at 52 weeks demonstrated statistically significant anteroposterior laxity compared with autografts; this was not present at 6 and 12 weeks. Allograft healing was improved in a rabbit histologic study when the fresh-frozen Achilles allografts were coated with mesenchymal stem cells compared with controls (Achilles allograft only).⁷⁵ Radiographically, BPTB allografts were found to have less revascularization by contrast-enhanced magnetic resonance imaging (MRI) at 1, 4, 6, and 12 months after surgery, but were found to equalize at 18 months postoperatively. The authors suggested that revascularization is slower in BPTB allografts compared with BPTB autografts.⁵⁸ However, computed tomography (CT) imaging of BPTB bone plugs at 1 week, 2 months, and 5 months did not show a significant difference in bony incorporation of BPTB allograft versus BPTB autograft.⁵⁰

For allograft ACL reconstruction to be successful, the graft must heal adequately in the bone tunnel. The intra-articular portion of the allograft must undergo the process of ligamentization in which the graft remodels to resemble the histology of a native ligament more closely.³¹

Soft tissue allografts must undergo tendon to bone tunnel healing. The native ACL insertion site is an example of direct insertion from tendon to bone. Four distinct histologically appreciable zones comprise this insertion site—tendon, unmineralized fibrocartilage, mineralized fibrocartilage, and finally bone. With soft tissue ACL reconstruction, this direct insertion site is not replicated. Rather, indirect insertion is relied on for soft tissue graft healing within the bone tunnel. Indirect insertion is naturally exemplified by other ligaments, such as the medial collateral ligament (MCL), which broadly inserts along the surface of the bone through fibers that travel obliquely from the long axis of the ligament to the long axis of the bone. These fibers, known as Sharpey’s fibers, are also seen histologically in ACL reconstruction with tendon grafts and correlate with the biomechanical properties of the graft insertion site.^28,31,70

A number of strategies have been used to improve the strength and healing of the soft tissue graft within the bone tunnel. Creating a longer tunnel has been shown to increase the strength of the graft-tunnel interface, presumably by increasing the amount of contact between the graft and the bone.⁸¹ Impregnating the graft with mesenchymal stem cells, as suggested earlier, has shown superior healing as well as more normal tendon to bone insertion histology.^49,62,75 The application of bone morphogenic proteins to the graft has also been shown to increase bone formation around the graft as well as graft pull-out strength.⁵⁵ Inhibition of osteoclast activity has been investigated with the use of osteoprotegerin (OPG). OPG-treated grafts in rabbits demonstrated greater bone formation and smaller sized bone tunnels around the graft.²⁵ Inhibiting degrading matrix metalloproteinases (MMPs) in rabbit allograft ACL reconstruction has shown more Sharpey’s fibers and a stronger load to failure.²¹ The primary healing response after graft implantation involves the arrival of inflammatory cells facilitated by cyclooxygenase-2 (COX-2). Studies have shown that avoiding anti-inflammatory medications such as COX-2 inhibitors allows the primary healing phase to proceed without chemical interruption. Some have cautioned against the use of nonsteroidal anti-inflammatory drugs (NSAIDs) to prevent inhibition of the primary healing response.^18,31,61

Compared with soft tissue grafts, which require soft tissue to bone healing within the osseous tunnel, BPTB grafts require bone to bone healing. This bone to bone healing is widely accepted as the strongest form of healing for ACL reconstruction. Histologic studies have demonstrated that the implanted bone plug demonstrates initial osteonecrosis and hypocellularity followed by revascularization, fibroblast invasion, and collagen synthesis, with subsequent rapid incorporation of the plug by surrounding bone.^22,39,64,80 These changes have been found to be similar in autograft versus allograft BPTB healing. Within 3 weeks, histologic incorporation is visible, but still fragile. At 3 weeks after implantation, the weakest point of a BPTB autograft remains the graft-tunnel interface. However, at 6 weeks, this interface has healed so that the site of graft failure is at the patellar tendon insertion into the bone plug.³¹ Although some studies evaluating timing of incorporation have suggested no significant difference between autograft and allograft bone plugs, others have shown that allograft bone plugs require a longer time to incorporate and central portions of the plug may not incorporate at long-term evaluation.^37,38,52 Healing of the bone plug likely occurs from the end of the tunnel furthest from the influences of degradative enzymes in the synovial fluid.⁸

Ligamentization of the graft is the phase of graft incorporation in which the intra-articular portion of the graft undergoes changes that more closely resemble the histologic properties of a native ligament. Whether autograft or allograft, this process takes several months and undergoes a series of steps.^31,38 Similar to the initial events of bone plug incorporation, the ligamentous portion also undergoes an initial phase of avascular necrosis and acellularity. Despite being devoid of cells, the graft maintains its collagenous structure, which serves as a scaffold for subsequent steps—cellular repopulation, revascularization, and ligament maturation.^3,4 The timing of these steps, outlined in a rabbit model, has demonstrated that at 2 weeks, necrosis of the graft begins. At 4 weeks, the graft is completely devoid of cellularity but the collagen scaffold remains intact. At 12 weeks, vascular proliferation and cellular repopulation are appreciated. At 6 months, the cellularity of the graft is similar to that of a native ligament. Finally, at 9 months, the graft is mature and histologically similar to a native ACL.⁶¹ In humans, surface blood flow studies have suggested that after an initial period of increased flow, ACL allografts demonstrate normal surface blood flow at 18 months, implying the end of graft remodeling.⁷¹