Introduction

Anterior cruciate ligament (ACL) injury is a common knee injury suffered predominantly by active individuals. Groups at highest risk for ACL tear include professional and amateur athletes, with annual incidence rates estimated to be between 0.15% and 3.67% and 0.03% and 1.62%, respectively. The treatment of ACL tears has evolved considerably over the last several decades, and in young, active patients generally involves surgical reconstruction of the ligament. Several issues related to the treatment of ACL injuries remain a matter of debate, particularly those pertaining to graft choice. Among panelists at a global summit on ACL reconstruction in 2011, the most popular graft choice among orthopaedic surgeons was hamstring tendon autograft (53.1%), followed by bone–patellar tendon–bone (BPTB) autograft (22.8%), allograft (13.5%), and quadriceps tendon autograft (10.6%).

Allograft use has increased in recent years, and its reported use ranges from 11% internationally to 22% in the United States. There are several advantages to the allograft concept, including lack of donor site morbidity (knee flexion strength deficit after hamstring harvest, anterior knee pain after BPTB harvest), decreased operative time, and lower risk of inadequate graft size or quality. Inherent with use of any allograft tissue are several disadvantages, including risk of disease transmission and cost of procurement and processing.

The purpose of this chapter is to discuss the use of allograft for ACL reconstruction. Issues pertaining to graft preparation will be explored, and the literature that focuses on basic science research as well as clinical outcomes after ACL reconstruction using allograft will be reviewed.

Graft Preparation

Despite accepting that allograft use has a role in orthopaedic surgery, many orthopaedic surgeons have expressed concerns regarding the safety of allograft tissue and the effect of sterilization procedures on mechanical and biologic properties of allograft tissue. To address these concerns, McAlister et al. composed an article that effectively discusses these issues. They emphasize that safety in using allograft tissue is contingent on both donor screening and tissue processing. Despite limitations, donor screening and testing can reduce the probability of transmitting a viral infection, such as HIV or hepatitis B or C, to extremely low levels. Currently in the United States there are various proprietary tissue-processing techniques used by tissue banks to achieve an acceptable level of sterility, and the US Food and Drug Administration (FDA) does not dictate which one should be used. Of these, no one technique has been found to be more effective than any other.

There are two methods of achieving sterility in the terminal step of the graft-cleansing process that have historically been used—ethylene oxide and gamma radiation—although some newer techniques use neither. Due to findings of ethylene oxide-sterilized allografts eliciting synovitis, the use of these grafts is not recommended. Gamma radiation has its sterilizing effect by generating free radicals, but these free radicals are thought to adversely affect the structure of collagen. Some processing techniques attempt to avoid the negative effects of free radicals by adding free radical scavengers. Despite claims about various proprietary techniques, both those using radiation and otherwise, the effects that these processes have on allograft tissue remain largely unknown. It is difficult to conclusively determine whether outcomes after ACL reconstruction using allograft are influenced by sterilization methods. However, there is growing clinical evidence to suggest that nonirradiated grafts may provide superior outcomes and lower retear rates compared with radiated grafts.

Basic Science

Following reconstruction of the ACL with any tendon graft, a process involving early graft healing, cellular proliferation, and eventually ligamentization ensues, and these phases, in both animal and human models, are thoroughly described elsewhere. In various animal experiments, allografts have been shown to have a slower rate of biologic incorporation and revascularization, and elicit a prolonged inflammatory response compared with autografts. In a goat model at 6 months after reconstruction, autografts had relatively better anteroposterior stability and strength-to-failure values. Similarly, human experiments suggest that allografts have a slower onset and rate of revascularization, which may lead to delayed graft incorporation relative to autografts.

Outcome Studies

Allografts have been compared with autografts in numerous meta-analyses, and the results have been mixed. Several meta-analyses have shown no difference between the two grafts, and several have found inferior outcomes and higher failure rates with allograft. Mascarenhas et al. performed a systematic review of meta-analyses on this topic and concluded that the highest quality meta-analyses suggest no difference in rupture rates or clinical outcomes between autograft and allograft, but that lower quality meta-analyses show lower rupture rates and better outcomes with autograft. A systematic review of five randomized controlled trials found no statistically significant differences between the two types of grafts, but the author warned that these results may not extrapolate to younger patient groups. Similarly, an earlier review found no difference between the two grafts but also found that none of the studies reviewed had controlled for or stratified results according to age.

To address this question, Wasserstein et al. formulated a systematic review that examined rates of failure after ACL reconstruction in young (<25 years old), active patients, and found that failure rates were significantly higher in the allograft groups (57 of 228 [25.0%]) compared with the autograft group (76 of 788 [9.6%]). Several prior meta-analyses have demonstrated higher failure rates in patients who undergo ACL reconstruction with allograft, regardless of age. Aside from increased failure rates, meta-analyses have also shown that allografts may be associated with increased A-P translation, as measured by a KT-1000 arthrometer.

Several authors have explored the possibility that allograft preparation techniques, especially the use of irradiation, may disrupt the mechanical properties of the graft. Park et al. performed a systematic review of 21 studies (1453 patients) that compared irradiated with nonirradiated grafts used in ACL reconstruction. Knees with nonirradiated grafts had better clinical outcomes and required fewer revision surgeries. Two earlier systematic reviews of 11 and 9 studies, respectively, that compared nonirradiated, nonchemically treated allograft ACL reconstruction with autograft ACL reconstruction found no difference in clinical outcomes or failure rates. It must be noted that these reviews consisted of studies that involved patients who were in their late 20s and older. After reviewing outcomes in young and active patients and finding significantly higher failure rates in those that received allografts, Wasserstein et al. emphasized that there is insufficient evidence to determine whether this difference persists based on the sterilization method, and strongly cautioned against using allograft prepared in any manner in young, active patients.

Large cohort studies have also been instrumental in understanding the risk of allograft failure, especially in young patients. Prospectively collected data from the Multicenter Orthopedic Outcomes Network (MOON), which was obtained from nearly 3000 ACL reconstructions, suggest that the risk of failure is 5.2 times greater for an allograft than for a BPTB autograft. Younger age and higher activity level were also found to be independent predictors of failure. Strikingly, in a small longitudinal study that reported on outcomes in division I National Collegiate Athletic Association athletes, Lenehan et al. reported a 62% failure rate after irradiated allograft ACL reconstruction compared with 0% failure rate after autograft ACL reconstruction. Another study utilizing data obtained from the MOON group found that odds of graft rupture were 4 times higher for allograft reconstruction than for autograft, and that for each 10-year decrease in age, the odds of rupture increased 2.3 times. Given these data, surgeons from the MOON group are extremely hesitant to recommend allograft reconstruction in patients who are already in a group at risk for failure based on their age.

Recent prospective data from the Multicenter ACL Revision Study cohort suggest that in revision cases allografts yield inferior outcomes when compared with autografts. Autografts were found to have improved sports function and outcomes, as well as a decreased risk of graft re-rupture in patients undergoing revision ACL reconstruction.