 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1997 The Journal of Bone and Joint Surgery, Inc.
Current Concepts Review - The Science of Reconstruction of the Anterior Cruciate Ligament* , CALGARY, ALBERTA, CANADA , LONG BEACH, CALIFORNIA*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
Very few subjects in contemporary orthopaedic surgery have evoked as much controversy, thought, and opinion as that of when and how to optimally reconstruct the anterior cruciate ligament of the knee. Fueled by an annual incidence of acute rupture of the anterior cruciate ligament that has been estimated to be one in 3000 in the American population—approximately 95,000 new injuries each year196—and combined with the fact that more than 50,000 knees that have such an injury are reconstructed annually in the United States alone149, it is not surprising that there has been a virtual explosion of literature on this topic. During the last twenty years, more than 2000 scientific articles on the anterior cruciate ligament have been published, including numerous excellent reviews47,149,151,255,282,294 and textbooks72,89,145. It is a challenging task to attempt to synthesize these sources and to try to resolve the conflicts that are inherent in such a massive body of information. Although many relevant issues will continue to be debated until better basic-science and clinical information becomes available, a number of scientifically supported concepts can be identified within the existing knowledge base. This knowledge can help orthopaedic surgeons to understand the reasons for previous and current successes and failures of reconstruction of the anterior cruciate ligament, and it can help them to plan the care of patients who have an injury of the ligament. In the present review, the current scientific understanding of reconstruction of the anterior cruciate ligament is discussed in a sequence based on the order in which clinical decisions are made. An overview of the normal function of the knee joint and the anterior cruciate ligament is followed by a discussion of decisions that must be made by the surgeon in the clinic and then in the operating room. Scientific knowledge pertaining to rehabilitation after reconstruction of the anterior cruciate ligament is then presented. Each section contains references to recent textbook chapters, reviews, and papers, which themselves contain many additional key references to the numerous authors who have contributed to this field, and each section provides statements that are intended to summarize what is currently known and to provoke further thought concerning what additional knowledge needs to be gained.
The knee joint functions within the extremely complex and interactive neuromusculoskeletal system148,306. Conscious decisions clearly influence how frequently an individual exerts substantial stress on the knee. Neuromuscular coordination also influences the forces on the knee and its component parts, including the anterior cruciate ligament. The anterior cruciate ligament therefore functions as one component within an extensive and dynamic system that can be injured and treated at many levels. In the context of the knee itself, the anterior cruciate ligament functions in concert with all other anatomical structures in and around the joint to control and limit motion and to maintain both static and dynamic equilibrium5,120,190,264,281,304. Because the anterior cruciate ligament shares tensile load-carrying functions with musculotendinous units and other ligaments, these structures can be considered to complement each other's functions directly. Some structures in the knee clearly complement the functions of the anterior cruciate ligament more than others; these structures are known as secondary restraints. For example, like the anterior cruciate ligament, the structures in the posterolateral corner of the knee190,297,312 serve to control anterior tibial translation and anterolateral tibial rotation relative to the femur. The main functions of other structures, such as the medial collateral ligament and the posteromedial aspect of the capsule, differ from those of the anterior cruciate ligament but complement them secondarily by controlling anterior translation on the medial side of the joint. Similarly, the hamstring muscles, in addition to their primary role as flexors of the knee, augment the role of the anterior cruciate ligament in restraining anterior tibial translation. These examples show how several normal structures in and around the knee have the ability to help the joint to compensate for an injury of the anterior cruciate ligament. Furthermore, these structures may be able to adapt themselves to assume even more of the functions of the anterior cruciate ligament in some circumstances. These compensatory mechanisms may account for why some patients are able to have nearly normal function of the knee in the presence of a torn anterior cruciate ligament71. Unfortunately, in a substantial number of patients who have symptoms of a torn anterior cruciate ligament, either the compensatory structures are damaged themselves297 or the complementary mechanisms are inadequate to compensate for the loss of the anterior cruciate ligament. The ability to predict which of these compensatory structures or mechanisms could provide stability in a knee that has a torn anterior cruciate ligament would help the surgeon to decide which patients might benefit from reconstruction of the ligament. Making this decision can be difficult, given the current lack of clear understanding of all of the compensatory mechanisms in the knee and how they are controlled.
It has been suggested that the anterior cruciate ligament has two complementary roles: one proprioceptive and the other mechanical. Evidence of its proprioceptive function comes mainly from extensive histological observations demonstrating that the anterior cruciate ligament appears to contain proprioceptive nerve endings34,122,147,314. Although there is some evidence to the contrary114, many physiological and clinical studies also have suggested that ligaments, including the anterior cruciate ligament, probably have a role in proprioceptive feedback to the knee joint29,63,80,168,187,197,244,306. The proportions and magnitudes of these proprioceptive contributions by the anterior cruciate ligament are currently under intense scientific investigation. Although the specific proprioceptive role of the anterior cruciate ligament remains to be quantified, its mechanical role as a tensile load-carrying element has been characterized with considerable detail and sophistication47,282. Improved knowledge of the functional anatomy of the anterior cruciate ligament constitutes what Dodds and Arnoczky have called "an essential blueprint on which to base techniques of repair and reconstruction."81 From this database, four important features of the structure and function of the anterior cruciate ligament in humans have been defined. First, the normal anterior cruciate ligament has been shown to carry loads throughout the entire range of flexion and extension of the knee35,81,107,132,313, resisting forces that would cause the tibia to translate anteriorly relative to the femur47,228 and, to a lesser degree, resisting forces and moments that would cause tibial rotation and abduction during flexion of the knee5,125,190. This is accomplished by the recruitment of different fibers within the gross structure of the anterior cruciate ligament as the knee joint moves. In fact, like the fibers in all ligaments, those in the anterior cruciate ligament are recruited differently on the basis of every subtle three-dimensional change in the position of the joint (the tibia relative to the femur in the case of the anterior cruciate ligament)48. This simple observation has clinical importance in a number of ways. To begin with, it helps to define the mechanisms of injury, which must originate in the part of the anterior cruciate ligament (often the posterolateral portion, which is tight when the knee is in slight flexion) that is carrying load at the time of the injury. In addition, because fiber bundles are recruited in different patterns, it is clear that the anterior cruciate ligament can fail very differently at different loads, depending on the position of the bones and the direction in which the loads are applied at the time of the injury307,310. In other words, the maximum strength of the anterior cruciate ligament should not be assumed to have one fixed value. Also, fiber recruitment is relevant to how diagnostic tests are performed. When the anterior cruciate ligament is tested for torn fibers, the knee joint should be assessed for stability in various positions, but most importantly it should be tested in slight flexion. As noted in recent reviews31, the Lachman test, for example, is based on testing recruitment of the anterior cruciate fibers with the knee in slight flexion—the position in which secondary restraints are most lax and that most effectively reveals the failure of the damaged ligament to recruit fibers to resist that force. Finally, the principle of fiber recruitment has key importance to operative reconstruction, which should be aimed at replacing the damaged portions of the anterior cruciate ligament at the relevant angles of the joint and, ideally, at restoring the ability of the healed structure to recruit fibers throughout the full range of motion of the knee. The second important feature of the structure and function of the anterior cruciate ligament is that evidence in both humans and animals has shown that the anterior cruciate ligament, like other ligaments, carries only small loads during normal daily function31,130,176,229. Although estimates have varied, most data suggest that normal daily loads on the anterior cruciate ligament are, at most, only about 20 per cent of its failure capacity33,130. The anterior cruciate ligament probably carries loads approaching its failure capacity (which is approximately 2500 newtons in young adults158,307) only during relatively unusual combinations of loading of the knee by external forces or muscles. The third feature of the structure and function of the anterior cruciate ligament is that extensive in vitro and in vivo evidence31,176 has shown that the circumstances that cause the highest loads and strains on the anterior cruciate ligament during daily function are quadriceps-powered extension of the knee, moving it from approximately 40 degrees of flexion to full extension; hyperextension of the knee; excessive internal tibial rotation; or excessive varus or valgus stress on the tibia if a collateral ligament is torn. Furthermore, both estimates228,229,245 and measurements32,33 have shown that the maximum measured strain differential in the anterior cruciate ligament is approximately 5 per cent during any rehabilitation exercise30. This strain represents only about one-quarter of the failure strain of the normal anterior cruciate ligament, suggesting that these exercises load the normal anterior cruciate ligament to only a small fraction of its failure capacity48,123. The importance of this observation will be discussed in several subsequent sections in this review. The fourth and final noteworthy aspect of the structure and function of the anterior cruciate ligament is that the normal ligament has been shown to have complex biomechanical behaviors well surpassing those of a simple collection of fibers. Like all tendons and ligaments, the anterior cruciate ligament behaves as a viscoelastic structure, allowing it to dissipate energy and to adjust its lengths and internal load distribution as a function of load history123,171. In clinical terms, this means that the normal anterior cruciate ligament is capable of microscopic adjustments to internal stresses over time, thus influencing the laxity, stresses, and kinematics of the joint in subtle but potentially important ways. Analogous properties of graft materials150 and healing grafts have begun to receive attention161,211 because, in addition to strength, they define potentially important functional capacities of tissues to resist elongation. For example, viscoelastic abnormalities may define the predisposition of some grafts to stretch out (creep), a potential mechanism of failure that may not cause rupture but may lead to increasing laxity of the graft and the joint over time119,214. This possibility is beginning to receive scientific attention.
As noted earlier, the anterior cruciate ligament contains many fascicular subunits81,207,293 within larger functional bands12,35,47,118 that are selectively recruited during tensile loading295. One anatomical factor that contributes to this fiber recruitment is the specific location of the insertions of the anterior cruciate ligament on the femur and the tibia22. Different components of the anterior cruciate ligament appear to attach to different locations within the insertional area on each bone, contributing to the selective recruitment of fibers as the femur and tibia move relative to each other during motion of the knee. The femoral attachment, in particular, appears to be critical to this recruitment22. A second anatomical factor that influences how fibers of the anterior cruciate ligament are recruited is related to the internal architecture of the ligament. At a gross anatomical level, individual fibers within the anterior cruciate ligament do not appear to change length during movement of the joint. At a histological level, however, fibers do change length as they are recruited into tension, and they must do so as the joint moves12,30. Fibers do this by straightening of their crimp and perhaps by some as yet unknown interactions with each other. From an operative point of view, these facts imply that restoration of the normal locations, shapes, and geometries of the fibers and insertions of the anterior cruciate ligament should optimize the chances of restoring normal fiber recruitment and fiber strains of the ligament12,22. At a histological level, the normal anterior cruciate ligament is also very complex123. Its solid structure is primarily tension-carrying fibrous collagen; however, it contains many other important viscoelastic elements, including water7,185. The normal anterior cruciate ligament also contains blood vessels22, nerves, and unique populations of fibroblasts, all of which are still in the early stages of being characterized10,179,206.
The fundamental rationale for reconstruction of the anterior cruciate ligament is that the natural history of untreated complete injuries of the ligament has been suggested to be the progression of symptomatic instability leading to recurrent injury, damage to the menisci and the articular cartilage, and osteoarthrosis17,23,90,92,184,222. According to other reviews151, this progression probably does not occur in every patient who has an injury of the anterior cruciate ligament. Most patients who have a complete tear of the anterior cruciate ligament have increased tibial translation on instrumented testing75. It is not known, however, exactly how many of these patients will have giving-way of the knee or how many knees will have overt or latent damage of the cartilage within a few years139,141. Serious functional instability, for example, appears to occur unpredictably, ranging from only 16 per cent (sixteen of 100 patients) who had a torn anterior cruciate ligament in one series101 to nearly every patient within a few years in other studies88,124. The reasons for such discrepancies are unknown, but they may be due to different definitions of instability, to varying degrees of damage of the anterior cruciate ligament246,287, to different combinations of injuries297, to different mechanisms of compensation for the loss of the anterior cruciate ligament, to differences in rehabilitation, or simply to the diverse physical demands and expectations of different populations. The progression to radiographically detectable osteoarthrosis in patients who have a torn anterior cruciate ligament appears to be variable. For example, after durations of follow-up ranging from two to ten years after the injury, mild osteoarthrosis was noted in only four (13 per cent) of thirty low-risk patients in one study57 and in eight (17 per cent) of forty-six patients in another series280. However, Fairbank grade-I or II osteoarthrosis was noted in thirteen (65 per cent) of twenty patients who were followed for four to ten years237 and in twenty-six (68 per cent) of thirty-eight patients who were followed for nine to sixteen years after the injury286. Although these data may suggest that osteoarthrotic changes that are detectable at least radiographically will develop in most patients who have a torn anterior cruciate ligament given enough time208, they also may support the concept that the natural history of a torn anterior cruciate ligament is variable219. Some of the reasons for this variability have been elucidated, but there is reason to believe that we have only begun to define all of the genetic, neuromuscular, anatomical, biomechanical, and biological causes.
Because the bulk of current evidence indicates that not all patients who have an injury of the anterior cruciate ligament will have symptoms or progression to osteoarthrosis, it is generally accepted that not all such injuries warrant operative reconstruction54,151,219. Both retrospective18,71,223 and prospective studies17,58,255, however, have shown that certain characteristics indicate that a patient is at high risk for symptomatic giving-way of the knee after an injury of the anterior cruciate ligament and would therefore most likely benefit from a reconstruction151,208,209. Such a patient is young and highly competitive, participating annually in hundreds of hours of high-speed pivoting sports253, and has a complete tear of the anterior cruciate ligament either alone43,102,133,225 or in combination with other ligamentous injuries44,309. This high-risk group also may include individuals in certain occupations (for example, fire fighting and manual labor) in which extensive knee-loading is required on a daily basis. What is also becoming clear is that a torn anterior cruciate ligament that is associated with a meniscal injury needs particular attention, as the menisci contribute to stability of the knee126,175 and their loss appears to predispose patients who have a torn anterior cruciate ligament to osteoarthrosis76,91,127,280,286. Furthermore, as nearly half of such meniscal tears may be repaired by suturing77 and may heal more effectively if the suturing is performed in combination with a reconstruction of the anterior cruciate ligament45,53,303, many authors have advocated reconstruction of the ligament specifically to save the menisci160,285. Although this approach has been suggested in order to be able to preserve the torn menisci for a few years51,53,285 and thus potentially to decrease the development of osteoarthrosis in knees that have an injury of the anterior cruciate ligament284, scientific proof of its efficacy is still needed. Beyond these observations, the criteria for the selection of patients become more controversial. Most surgeons recommend reconstruction for appropriately selected individuals who are at moderate risk for symptoms—that is, for those who participate in moderate-risk sports sporadically or less intensively. However, the definition of such risk is debatable. Some authors have recommended reconstruction for motivated older athletic individuals57, adolescents20,177, and even people who already have some established osteoarthrotic changes267. Other than the standard medical contraindications, there are currently no scientifically validated boundaries for these indications and certainly no absolute criteria for not performing a reconstruction of the anterior cruciate ligament. The literature suggests that certain factors—that is, those that may have predisposed the individual to the injury of the anterior cruciate ligament in the first place—may eventually emerge as contraindications to a reconstruction of the anterior cruciate ligament. Although none of the factors have been proved to cause failure of a graft, it can be argued that some may contribute to such failure. These factors include a narrow intercondylar notch in the femur173,178,288, excessive physiological laxity, tibial rotation, the alignment of the foot and the width of the pelvis in the female athlete138, convexity of the lateral tibial plateau169, and sphericity of the femoral condyles101. The influence of these and other factors still must be tested scientifically.
Even after a patient has been selected for a reconstruction, exactly when the procedure should be carried out also remains controversial. Meniscal repairs are reportedly best performed early127, suggesting that if meniscal damage is identified (particularly in association with a displaced tear) reconstruction should be carried out without delay. In the absence of repairable menisci, the optimum timing of the reconstruction of the anterior cruciate ligament can be debated. It can be argued intuitively that, to prevent additional injury and to minimize the duration of rehabilitation, the reconstruction should be carried out immediately after the patient is seen with the injury. Recent reviews, however, have shown that such operations performed within the first month after the injury are associated with higher risks of arthrofibrosis and loss of motion of the joint199. It has therefore been suggested that, if operative treatment is not performed within the first few days after the injury, it should be delayed for at least one week116 and perhaps for as long as two to six weeks6,64,199 while motion is restored. Others have maintained that intensive rehabilitation training with an emphasis on regaining extension can decrease the risk of arthrofibrosis sufficiently to make any delay in reconstruction unnecessary269. At this point in time, because of the lack of a prospective comparative study with validated outcomes, both approaches can be justified equally. From a basic-science perspective, scar tissue formed in wounds that are opened for a second time has long been known to gain strength more quickly than the scar that forms in primary wounds, presumably because of local inflammatory factors1. Operative treatment of a recently injured knee could have the equivalent effect of provoking increased scarring or the formation of stronger scar tissue in areas where the joint had already started healing. Such a scarring process could help to stabilize the reconstructed knee joint; however, it might also interfere with motion of the knee (a process known as arthrofibrosis). Given the inability to predict the scar response in individual patients, optimization of these healing responses to maximize stability without causing a loss of motion of the knee remains a challenge.
Preoperatively, the surgeon must decide which reconstructive option will be pursued: primary repair (that is, suturing) of the anterior cruciate ligament alone, primary repair of the anterior cruciate ligament as well as repair of other ligamentous structures, primary repair combined with augmentation of the repair with use of some other structure (intra-articularly or extra-articularly, or both), or removal of the torn anterior cruciate ligament and replacement with another structure (a biological or synthetic structure, or both). Each of these options has been the focus of considerable thought and scientific attention84,99,106,115,142,182,298. The highlights of recent reviews, not comprehensive analyses, are presented.
Primary Repair Alone
Primary Repair with Augmentation The second way to augment a primarily sutured anterior cruciate ligament is with an artificial strut. An artificial ligament-augmentation device placed through the joint and over the top of the lateral femoral condyle may stabilize some joints when it is combined with primary suturing of the anterior cruciate ligament159. However, in a prospective comparative study, such augmentation did not appear to alter the results achieved with primary repair alone119. The third way to mechanically augment a suture repair of the anterior cruciate ligament is with a lateral extra-articular procedure. After short-term follow-up, such augmented repairs appeared to result in a higher prevalence of functionally stable knees than did non-augmented repairs16,290, but the differences were modest. Interestingly, evidence suggests that lateral extra-articular procedures alone also may alter the natural history of symptoms caused by chronic laxity of the anterior cruciate ligament40,100, with the MacIntosh11,100,300 and Losee procedures having better rates of success than the Ellison procedure248. The MacIntosh procedure involves the rerouting of a long strip isolated from the iliotibial band, which remains attached to the tibia, under the lateral collateral ligament, through a subperiosteal tunnel on the lateral femoral condyle, then through the intermuscular septum, and then back on itself distally. With the Losee procedure, a similarly isolated strip of the iliotibial band is passed through a drill-hole in the lateral femoral condyle and the posterolateral structures before being looped anteriorly under the lateral collateral ligament and back to its origin on the tibia. The Ellison procedure involves the rerouting of a strip of the iliotibial band deep to the lateral collateral ligament, replacing it anteriorly into a tibial bone trough, and plicating the capsular ligament. These procedures are appealing because they are purely extra-articular; however, when used in isolation they have been associated with high rates of failure151. The Ellison procedure reportedly led to an unsatisfactory result in sixteen (76 per cent) of twenty-one patients248, and the MacIntosh procedure yielded an unsatisfactory result in thirteen (48 per cent) of twenty-seven patients11 at an average of eleven years postoperatively. These results may not be surprising, as the procedures in question involve placement of the graft in positions that do not recreate the normal anatomy170 or function13 of the anterior cruciate ligament and as the tissues that are used have distinct mechanical weaknesses compared with the anterior cruciate ligament131,233,301. The final means of augmenting a primary repair of the anterior cruciate ligament, and the one used most commonly today, is to place a tendon or fascial graft alongside of or through the repaired anterior cruciate ligament, thus bridging the gap between the torn ends and providing reattachment to the bones111. The different types of grafts that may be chosen for this approach are the same as those used for repair of a chronically lax anterior cruciate ligament (to be discussed later). However, it was noted that no study has supported the view that a primary repair augmented with a graft is better than use of a graft alone119. Preservation of the residual anterior cruciate ligament as part of such a reconstruction merits additional study.
Prosthetic Replacement
A series of decisions must be made by the surgeon in the operating room during a reconstruction of the anterior cruciate ligament. Most of these decisions have received scientific attention, and many have been evaluated in terms of their potential to cause failure of the graft302. Each decision has potential impact on the results of the operation. These decisions are discussed here in the order in which they must be made during a typical reconstruction of the anterior cruciate ligament. A synthesis of the available data on each topic is also presented.
Choice of Grafts Beynnon et al.30, in an extensive review, found that strength has been a major consideration in the choice of grafts. For example, fourteen-millimeter-wide bone-patellar ligament-bone grafts have been shown to be stronger and stiffer than single-hamstring-tendon grafts47. This advantage is diminished when the more commonly used nine or ten-millimeter-wide bone-patellar ligament-bone grafts are compared with multiply looped hamstring grafts41,235,291,298. Regardless of which of these two options is chosen, mechanical testing studies, as reviewed by Beynnon et al.30, have suggested that large bone-patellar ligament-bone grafts or multiply looped hamstring grafts must be used if the surgeon wants the graft to approach the strength of a normal anterior cruciate ligament at the time of the operation. Three additional factors should be considered when a graft is chosen on the basis of its strength. First, it is known that, for at least the first one to two months after implantation, the main factor affecting the structural strength of either patellar ligament or hamstring grafts is not the graft itself but rather the point of fixation of the graft to the bone251. Studies of cadavera have shown that the best methods of fixation, such as the placement of large interference-fit screws parallel to the osseous part of the graft in drill-holes or the use of multiple sutures, achieve a pull-out strength that is at most 50 per cent of the strength of the graft153,166,193. After some healing of the graft to the bone has taken place, between four and twelve weeks postoperatively251, the attachments to bone are probably no longer the weakest point in the complex. Although bone-to-bone healing of bone-patellar ligament-bone grafts may be faster than the healing of hamstring grafts without bone during this interval, this has not been proved. Only a small amount of clinical data, and a modest amount of histological and biomechanical data based on limited animal models, are available for speculation on this point. Second, there is considerable scientific evidence suggesting that all tendon tissue probably loses a considerable amount of its initial strength during the early healing period47,106,186. In none of the animal models that have been studied thus far have the grafts ever regained their original strength or achieved the properties of a normal anterior cruciate ligament; only 10 to 50 per cent of normal stiffness and strength24,30 were restored, after three years of follow-up210. It is not known for certain whether or not the same weakening occurs in humans. Third, the effects of the initial strength, size, surface area, and origin of the graft on its potential for weakening during healing also remain unknown. While intuition suggests that grafts that have a larger total cross-sectional area when they are implanted will remain stronger over time, this has not been demonstrated. The reasons why grafts weaken remain obscure, but obviously an understanding of these reasons has great potential importance for optimization of the strength of the graft. Biopsies have shown that similar biological processes occur in ligament grafts in humans and animals106,163,194,195,226,272-274,277,278. In several studies, all grafts that were biopsied demonstrated some early loss of fibroblasts, revascularization from the surface within one to three months after implantation, and repopulation by cells of extrinsic origin8,144,164. These processes are accompanied by what appears to be some partial breakdown and replacement of the graft by scar-like tissue226. Although some morphological and biochemical properties of this new composite are similar to those of normal ligament tissue8,9,172, others are not24,226,277. As noted earlier, the mechanical properties of grafts appear to remain abnormal for long periods24,146,210. The current focus of experimental work in this area is thus the identification of grafts that may not undergo this remodeling, prevention of the negative consequences of remodeling, and stimulation of cells in grafts to produce a more normal anterior cruciate ligament. As so many similarities have been noted between different types of grafts in both humans and animals, it may not be surprising that different grafts appear to produce similar results in clinical practice. Both direct and indirect clinical comparisons have shown that bone-patellar ligament-bone grafts and hamstring grafts have similar rates of effectiveness in adults, with only minor variations in knee stability and muscle strength at an average of two years4,189, three years129,157,236, and five years15,76,91 after implantation. It has been theorized that hamstring and fascial grafts have an advantage compared with grafts containing bone because they may not cause growth-plate bridges in reconstruction of the anterior cruciate ligament in adolescents212,289; however, this advantage has not been demonstrated in clinical practice. These grafts also have a theoretical advantage of recreating a multibundled structure, a concept that is currently only beginning to receive scientific attention. Comparison of the rates of complications associated with patellar ligament and hamstring autogenous grafts also has failed to identify one type of graft as being clearly superior. The removal of tissue for a graft has been shown to induce weakness of the hamstrings in some patients189. The removal of tissue for bone-patellar ligament-bone grafts has induced weakness of the quadriceps and pain in the anterior aspect of the knee postoperatively257. The prevalence of these symptoms was 12 to 45 per cent in series ranging from fifty to 126 patients who were followed for a maximum of two years59,152,227,259, and the symptoms lasted for three to seven years in 5 to 9 per cent of more than 200 patients in another study2,3. Long-term tenderness at the donor site after removal of tissue for a patellar ligament graft also has been reported165. In addition, concerns have been expressed about the high rate of patellofemoral crepitation after the use of bone-patellar ligament-bone grafts85. Moreover, in one study, approximately 50 per cent of 187 patients had asymptomatic but arthroscopically proved deterioration of the articular cartilage within a few years276. Importantly, however, it has also been shown that a control population of patients who had a torn anterior cruciate ligament that was treated non-operatively, as well as those who had reconstruction with use of grafts other than autogenous patellar ligament grafts, also had patellofemoral pain189 and signs of patellofemoral disease85. Some patellofemoral problems therefore probably are not related to the type of graft but rather to other causes, such as previous patellofemoral disease, patellofemoral injury related to the injury of the anterior cruciate ligament, rehabilitation exercises, flexion contracture, and arthrofibrosis238,240,257,259,270. Clearer definition of which problems are truly due to the treatment as opposed to the condition is still required. The type of graft does not appear to influence the development of osteoarthrosis in patients who have an injury of the anterior cruciate ligament. Regardless of the type of graft, there appears to be a high prevalence of osteoarthrosis in reconstructed knees, particularly those treated with a concomitant or subsequent meniscectomy2,76,91,133. The data therefore do not support the concept that a particular type of graft is more effective than any other in preventing osteoarthrosis. In a recent review106, it was concluded that, in general, allografts are a valid alternative to autogenous grafts for the replacement of the anterior cruciate ligament, provided that there is careful screening for viral disease, appropriate pretreatment (freezing or freeze-drying) of the graft, use of sterilization techniques that do not weaken the graft93,261 or cause it to retain toxic agents109, and use of operative techniques comparable with those employed for the implantation of autogenous grafts. Both animal and clinical data generally support the concept that, like autogenous grafts, allografts become revascularized and viable after implantation21,146,278,279, but their rates of incorporation and remodeling are slower than those of autogenous grafts106,146. The mechanical properties of allograft tissues, which also are similar to those of autogenous grafts, deteriorate after implantation in animals82,146,279. Nonetheless, at five19,273,275 and seven years216 postoperatively, the clinical outcomes of replacement of acutely injured anterior cruciate ligaments with allografts have been similar to those of replacement with autogenous grafts. As with the use of autogenous grafts, allografting has led to better functional stability when it has been performed to treat isolated injuries of the anterior cruciate ligament as opposed to more complex instability215. Allografting yielded a reconstructed ligament that was functional in only thirty (53 per cent) of fifty-seven patients who had had a chronically unstable knee or one in which other grafts had already failed218. Longer-term follow-up of these patients, as well as additional controlled studies for assessment of the clinical results, the rates of complications, and the ultimate cost-effectiveness of allografts compared with autogenous grafts, are needed.
Open Compared with Arthroscopically Assisted and Endoscopic Techniques
Use of a Tourniquet or Fluid Pressure
Placement of the Tunnels Various tools have been developed to help the surgeon to confirm the location of the tunnel intraoperatively. These include devices in which the key point of reference is the over-the-top position142, the roof of the intercondylar notch136, or the anterior surface of the posterior cruciate ligament200. The use of isometers also has been proposed as a means of confirming the position of the holes. Although helpful in locating anatomical areas for insertion, isometers have not proved useful for accurate prediction of the relationship between the tension and the length of the graft62,96,97,112.
Notchplasty and Roofplasty
Preconditioning, Application of Tension, or Twisting of a Graft It was recently suggested, on the basis of findings in a canine model, that twisting of a graft can influence its mechanical properties203 and theoretically influence strains37 in the graft at least at the time of the operation. The effects of twisting on the graft or the function of the joint over time in vivo are not known.
Fixation of the Graft Abundant research has shown that, for bone-patellar ligament-bone grafts with interference-fit screw fixation, the tibial fixation is generally the weak point and larger screws are slightly better than smaller ones. If a screw diverges more than 15 degrees from the orientation of the graft plug, the strength of that construct may be weakened substantially42,49,153,243,268. Lemos et al. found that, even if the surgeon is experienced, it is easier to inadvertently cause screws to diverge with use of the endoscopic approach174. Furthermore, it has been suggested that strength is not affected if screws have to be replaced166. Double-hamstring grafts fixed with soft-tissue washers probably are as strong as bone-patellar ligament-bone grafts291, but they are not as stiff. Woo et al., in unpublished work, recently found that differences in structural stiffness are due partly to the fact that the points of fixation of patellar ligament grafts differ from those of hamstring grafts308. Grafts fixed to bone close to articular surfaces (for example, those fixed inside of drill-holes) should undergo less strain and be stiffer than grafts that are fixed outside of drill-holes. At comparable loads, a longer graft will undergo more strain over-all. Fixation of the graft closer to the surface of the joint may therefore offer theoretically superior stability of the joint, at least until the drill-holes are filled in with bone. Interestingly, on the basis of the clinical results that have been reported for a number of techniques of fixation of the graft30,41,235,291,298, including some constructs that are mechanically very weak (for example, sutures over buttons), it appears that the fixation is only rarely a problem. The prevalence and functional importance of slippage associated with any method of attachment of a graft to bone is unknown.
Closure of Defects of the Patellar Ligament
Ice and Continuous Passive Motion
Rehabilitation The limits of stress, strain, frequency, and duration beyond which an intensive rehabilitation protocol can induce damage to a healing joint by putting too much mechanical stress on its structures remain controversial; it has been found that healing graft complexes are probably weak and compliant for many months and that grafts remodel over many years256. Certain exercise protocols have been particularly worrisome for surgeons and therapists, as there is abundant information that resisted quadriceps exercises put some strain on the anterior cruciate ligament33, particularly in the last few degrees of extension of the knee if the limb is not bearing weight. These extension exercises, with or without the use of ankle weights, have been called open-chain exercises. In an effort to protect the anterior cruciate graft during quadriceps exercises, it has been suggested that, instead, the patient should stand, thus loading the knee joint axially during motion, as this may protect the graft through use of the contours of the joint to stabilize the knee. These exercises have been called closed-chain exercises. A clinical test of one closed-chain exercise (a two-legged squat), however, was not found experimentally to strain the anterior cruciate ligament any differently than did an open-chain limb-extension exercise30. It is not known whether strains induced by rehabilitation exercises always have negative effects on grafts by causing them to stretch or if some strain is actually necessary to stimulate remodeling of the graft. A thorough scientific basis for exercise protocols is still lacking. Proprioceptive training during rehabilitation after reconstruction of the anterior cruciate ligament has also been advocated, mainly on the basis of theory and empirical evidence25,28,60. Unfortunately, clinical studies of rehabilitation of the anterior cruciate ligament have had too many variables and thus have continued to lack the statistical power required to define the effects of such techniques over time50,83,110,181,188,260. Until more powerful experimental designs can be developed, the optimum protocol for rehabilitation will remain controversial.
|
Top
Introduction
Normal Function of the...