Journal of the American Academy of Orthopaedic Surgeons 180 Treatment of full-thickness articu- lar surface lesions in the knee re- mains a challenge for the practicing orthopaedic surgeon. These lesions may be small and asymptomatic at the time of discovery, yet may increase in size and become painful at a later date if left untreated. De- cisions about whether and how to treat an individual lesion are prob- lematic. Untreated articular surface le- sions have little or no potential to spontaneously heal with normal hyaline surface cartilage. Curl et al 1 found a 63% incidence of chondral lesions (averaging 2.7 lesions per knee) when they reviewed more than 31,000 arthroscopic surgical procedures. Grade IV (modified Outerbridge classification system) articular cartilage lesions were noted in 20% of patients, with 5% of these occurring in patients less than age 40 years. Seventy-five percent of the patients in this group who were less than 40 years old had solitary lesions, and only 35% of them had no accompanying meniscal or liga- mentous lesion. With such a high incidence of articular surface lesions, the orthopaedic surgeon should expect to see a high percentage of symptomatic patients in his or her office. However, Messner and Mal- etius 2 reported that 22 of 28 patients with isolated chondral lesions had good or excellent clinical results without treatment 14 years after diagnosis. Although this might imply that most chondral lesions are asymptomatic, the majority of their patients had abnormal radiographic findings, suggesting that some asymptomatic lesions do go on to permanently damage the knee. Maletius and Messner 3 also re- ported on a 12- to 15-year follow-up of 42 matched patients with chon- dral damage who were treated with or without partial meniscectomy. Radiographic follow-up revealed more significant changes (P<0.03) in patients with both meniscectomy and chondral damage; however, those with chondral damage alone still had some radiographic evi- dence of joint-space narrowing. While this limited evidence sug- Dr. Browne is Clinical Associate Professor of Orthopaedic Surgery and Director of the Orthopaedic Sports Medicine Fellowship Program, University of Missouri, Kansas City. Dr. Branch is Director, University Orthopaedic Clinic, Decatur, Ga. Reprint requests: Dr. Browne, Orthopaedic and Sports Medicine Clinic, Suite 400, 6675 Holmes Road, Kansas City, MO 64131. One or more of the authors or the departments with which they are affiliated have received something of value from a commercial or other party related directly or indirectly to the sub- ject of this article. Copyright 2000 by the American Academy of Orthopaedic Surgeons. Abstract Articular cartilage injuries in the knee are common; fortunately, full-thickness articular cartilage defects constitute only a small portion of this group. These lesions may be incidentally encountered during ligament or meniscal surgery, having been silent or asymptomatic for an unknown period of time. However, when they are large and symptomatic, the surgeon may choose from a wide array of techniques available for treatment. The relatively small number of nat- ural history studies regarding full-thickness articular surface lesions compli- cates the decision-making process. Accurate evaluation and classification of the anatomic defect aids in the development of a clinical algorithm for treatment. Surgical techniques are either reparative or restorative in nature. Reparative techniques fall short of complete reestablishment of the articular cartilage; how- ever, the resultant repairs may remain quite functional for varying periods of time. Restorative techniques attempt to reestablish the native articular surface. To date, no peer-reviewed, prospective, randomized, controlled studies of opera- tive versus nonoperative treatment for full-thickness articular cartilage lesions have been published. Even though the long-term results of surgical treatment for full-thickness articular surface lesions remain unknown, the early results are encouraging. J Am Acad Orthop Surg 2000;8:180-189 Surgical Alternatives for Treatment of Articular Cartilage Lesions Jon E. Browne, MD, and Thomas P. Branch, MD Jon E. Browne, MD, and Thomas P. Branch, MD Vol 8, No 3, May/June 2000 181 gests that chondral damage in the knee predicts early development of osteoarthritis, there is a decided absence of matched controlled nat- ural history studies. It is important that arthroscopic surgeons be familiar with the cur- rent techniques available for the treatment of full-thickness articular surface lesions and the guidelines for treatment of both symptomatic and asymptomatic lesions. The techniques and guidelines dis- cussed in this review are limited to those applicable to chondral defects that are traumatic in origin and are not related to osteoarthrotic and in- flammatory arthritic conditions. Anatomy Knowledge of the microanatomy of the articular surface cartilage pro- vides a framework on which the surgeon can base selection of the appropriate surgical procedure. The goal is to reestablish the articu- lar surface to normal biomechanical and histologic integrity. The basic structural components of articular cartilage include chondrocytes, col- lagen, extracellular matrix proteo- glycans, noncollagenous proteins, and water. The distribution of each component varies within four dis- tinct histologic zones: superficial, middle, deep, and calcified (Fig. 1). The basic building block of the articular surface is the chondrocyte, which originates from undifferenti- ated mesenchymal marrow stem cells. These cells in turn propagate through the calcified cartilage zone to become chondroblasts. When the chondroblasts become isolated in lacunae, they become chondro- cytes, which receive their nutritional support from the synovial fluid. In skeletally mature articular carti- lage, chondrocytes no longer di- vide but still remain alive via the glycolytic anaerobic metabolism pathway. Skeletally immature articular cartilage chondrocytes undergo cell division and an in- crease in cell matrix volume. As chondrocytes age, they exhibit a decrease in cellular activity, espe- cially production of both collagen and proteoglycan. Although chon- drocytes constitute only 5% of the wet weight of articular cartilage, they are the major source for new synthesis and maintenance of its components. This includes the production of collagen, proteogly- cans, and noncollagenous proteo- glycans as well as enzymes. They maintain the balance of synthesis and degradation of the protein macromolecular complex. Water constitutes approximately 75% of the weight of articular carti- lage. Because of its role as a cation, water is one of the most important components of cartilage. Collagen, predominantly type II, underlies the form and tensile strength of ar- ticular cartilage. It makes up ap- proximately 10% of the weight of cartilage. Proteoglycans, with their structural subunits, glycosamino- glycans, provide the compressive strength of articular cartilage. They account for the remaining 10% of cartilage weight. Proteoglycans trap and hold water within articu- lar cartilage. Like other systems within the body, articular cartilage Figure 1 Basic structural anatomy of articular cartilage. Zones Superficial Middle Deep Flat, parallel Flatter, more rounded Random, oblique Spherical, in columns Tidemark Smaller- volume cells Cortical bone Cancellous bone Mesenchymal stem cells Calcified Chondrocyte AppearanceCollagen Orientation Articular Cartilage Lesions Journal of the American Academy of Orthopaedic Surgeons 182 contains special subunits, which interact with cytokines and growth factors. Interleukin-1, insulinlike growth factor-1, and transforming growth factor-β1 combine with articular cartilage in an anabolic, a catabolic, or a mixed fashion. The microarchitecture of articular cartilage is unique. The outermost layer, or superficial zone, which con- tains a relatively small amount of proteoglycan, is thin, noncellular, and porous. In this layer, called the lamina splendens, the fibers are arranged parallel to the joint surface. Farther down in the articular carti- lage, the collagen fibers are oriented perpendicular to the joint surface. In the middle zone, the collagen fibrils have a larger diameter com- pared with those in the superficial zone, with a higher concentration of proteoglycans and lower amounts of water and collagen. In the third layer, or deep zone, the largest- diameter collagen fibrils, the high- est concentration of proteoglycans, and the lowest concentration of water are noted. The collagen fi- brils eventually pass through the tidemark boundary and extend into the remaining area, the calcified zone that separates the noncalcified zone from the underlying subchon- dral bone. The biomechanics of articular car- tilage utilize this microanatomy to reduce the forces of friction across the joint to extremely low values. This system incorporates three ma- jor avenues to lessen the friction in the joint. First, the parallel fibers of the lamina splendens provide a flat surface for the joint to roll or slide across during motion. Second, the porous nature of the lamina splen- dens in combination with the water- attracting characteristics of the pro- teoglycans allow fluid flow through the surface of articular cartilage dur- ing compression. This fluid flow pro- duces hydrostatic pressure, which helps decrease the forces of friction across the joint. Third, the lamina splendens surface becomes coated with phospholipids, which have a hydrophobic head attracted to the collagen surface and a hydrophilic tail pointed toward the opposite articular surface. This creates an electrostatic pressure similar to that of magnets opposing one another. Recreating this complex microstruc- ture makes surgical reconstruction of articular cartilage very difficult, as all three parts of this biomechanical system must work together for opti- mal function. 4,5 Articular cartilage lacks vascu- lar, neural, and lymphatic access networks, which creates a limited environment for spontaneous re- pair. Injuries that do not penetrate into the subchondral bone show lit- tle sign of spontaneous repair, whereas those that extend into the depth of subchondral bone initiate a vascular proliferative response that produces only fibrocartilage. Current surgical procedures either incorporate penetration into the subchondral bone as part of their technique or utilize it as a bound- ary or base for surface restoration. Clinical Presentation of Articular Cartilage Lesions The most common clinical presen- tation of a full-thickness articular cartilage lesion is a loose body. It may be associated with an acute injury, with a concomitant large knee effusion, or, more likely, it may have an insidious onset with no effusion. Other patients may have a progressive onset of joint- line and/or patellofemoral pain with occasional mechanical symp- toms of locking or catching. Com- mon scenarios for the presentation of full-thickness articular surface injuries include patellar dislocation with lateral femoral condylar and medial-patella facet lesions, Òdash- boardÓ injuries in which the patella is driven into the trochlea, and liga- ment injuries, most often to the anterior cruciate ligament. The physical examination usually does not elicit a distinct consistent finding other than localized pain with or without an effusion. The presence of a loose body should be considered predictive of the occur- rence of an articular surface injury until proven otherwise. A routine complete examination should be performed to rule out other factors, such as malalignment and other meniscal, ligamentous, and extensor mechanism problems. Various sub- jective and objective criteria related to articular surface injury and repair may be used to categorize the status of the knee in both the history and the physical examination. Plain radiographs, including standing posteroanterior flexion views, may visualize compartment joint-space narrowing or an osteo- chondritis dissecansÐtype defect, with or without a loose body. With full-thickness articular cartilage lesions, plain radiography might not reveal any changes; in that set- ting, magnetic resonance (MR) imaging may be more helpful. Herzog 6 reviewed current MR imaging techniques for assessing chondral injury and concluded that proton-density imaging of thin (3- to 4-mm) sections and T2-weighted imaging with fat-saturation sequences optimize resolution of the articular chondral surface. High-resolution gradient-echo imaging has also been proposed to allow more careful evaluation of the articular surface of the patella. Defects are best ana- lyzed with three orthogonal planes. In this way, the image obtained will be perpendicular to the area of con- cern. With the known high inci- dence of subchondral bone contu- sions associated with ligamentous injuries, the identification of edema in the subchondral bone should serve as a flag to carefully review and analyze the overlying articular surface. Jon E. Browne, MD, and Thomas P. Branch, MD Vol 8, No 3, May/June 2000 183 Although MR imaging remains the benchmark for musculoskeletal soft-tissue imaging, its usefulness in consistently analyzing changes in the articular surface has been questioned. Arthroscopy is a more accurate technique for diagnosing articular surface lesions. Ochi et al 7 prospectively and retrospectively analyzed preoperative MR imaging studies of 65 patients who under- went surgical procedures and were found to have 72 articular surface defects. The overall prospective sensitivity of MR imaging for these defects was 40%, with a retrospec- tive sensitivity of 70%. The role of the bone scan remains controversial. Isolated articular sur- face defects that do not penetrate the subchondral bone might not be identified by bone scanning. Dye and Chew 8 stressed that the change in joint homeostasis occurring with any significant joint injury will be reflected in a persistent increase in scintigraphic activity. The return to the normal state is concomitant with the return of a normal scintigraphic appearance. Bone scanning has not been used to document joint homeo- stasis during the treatment of artic- ular surface lesions; however, it may ultimately provide the best tool for evaluating whether surgical intervention has restored the joint to its normal state. Documentation of Arthroscopic Findings Classifying the condition of the joint and the nature of a chondral lesion necessitates a documentation system. The grading system de- vised by Outerbridge 9 is the sim- plest working tool for describing chondral lesions (Fig. 2). Other sys- tems may be more elaborate and specific, but the clinical usefulness of the Outerbridge system in daily practice makes it still a practical working approach. This must be combined with an accurate notation of the location, size (i.e., surface area), and shape (i.e., circular, rec- tangular, or elliptical) of the articu- lar surface lesion and a description of the walls (i.e., whether they are contained, partially contained, or open). The depth of the lesionÑ designated as mild (partial thick- ness), moderate (characterized by extension to subchondral bone), or severe (extending into subchondral bone)Ñmay be the major determi- nant in the final selection of the sur- gical technique to be utilized. The appropriate treatment for the asymptomatic patient with an incidental finding of a full-thickness articular cartilage lesion is problem- atic. If such a lesion is left untreated, will it then go on to be symptomatic within a short period of time? Con- versely, if it is treated, will it be- come symptomatic as a result? Without treatment, might it have remained asymptomatic? The ab- sence of a documented natural his- tory makes these decisions difficult. Until the natural history of the sur- gically treated symptomatic lesion is confirmed, surgical treatment cannot be recommended; however, continual reevaluation and follow- up monitoring are warranted. Nonoperative Treatment The goal of nonoperative treatment is to reduce symptoms related to the articular cartilage lesion, not to restore anatomy. Physical therapy for muscle strengthening, gait training, and application of appro- priate bracing or use of an orthotic device may eliminate some of the symptoms. Use of intra-articular viscosupplementation products and oral chondroprotective agents for the treatment of osteoarthritis may also provide symptomatic relief, but to date there has been no evidence of structural improve- ment. Operative Choices The various techniques available for surgical intervention result in a tissue response that is either repar- ative or restorative (Table 1). The ultimate response to surgical inter- vention may be correlated with the numbers and kinds of cells used and how closely the surgical reconstruc- tion seeks to emulate the micro- anatomy of the articular cartilage. The chondrocytes for all of these procedures are facilitated from mesenchymal stem cells induced Figure 2 System for grading the status of the articular cartilage, as described by Outerbridge. 9 In grade I, the articular sur- face is swollen and soft and may be blis- tered. Grade II is characterized by the presence of fissures and clefts measuring less than 1 cm in diameter. Grade III is characterized by the presence of deep fis- sures extending to the subchondral bone, measuring more than 1 cm in diameter. Loose flaps and joint debris may also be noted. In grade IV, subchondral bone is exposed. Grade I Clefts Blister Subchondral bone Deep fissures Subchondral bone exposed Grade II Grade III Grade IV Articular Cartilage Lesions Journal of the American Academy of Orthopaedic Surgeons 184 from periosteum or perichondrium, harvested as autologous chondro- cytes, or transplanted as allogeneic chondrocytes. The goal of restorative surgical techniques is complete reconstruc- tion of the microarchitecture of articular cartilage, with restoration of all biomechanical and physiolog- ic functions and resultant complete relief of symptoms. In contrast, a reparative surgical technique re- constructs the defect in a manner that does not necessarily restore the articular cartilage architecture but still may relieve symptoms. Conse- quently, only some of the biome- chanical functions of the articular cartilage are restored, which com- promises the longevity of the artic- ular surface due to a higher coeffi- cient of friction. There are also some operative techniques that have no impact on the articular cartilage defect itself. For example, arthroscopic lavage and/or debridement (chondroplas- ty) may lessen symptoms, but the effects diminish with time. 10 Pa- tients with angular deformity and articular surface lesions (generally due to osteoarthritis) may show signs of clinical improvement and increased joint-space widening after osteotomy; however, biopsy specimens obtained from the ar- thritic compartment consistently show proliferation of a fibrocarti- laginous response with little hya- linelike cartilage restoration. 11 Sim- ilarly, varus or valgus bracing may offer symptomatic relief to the patient with a malaligned knee without changing the damaged articular surface structure. Truly restorative procedures for the treatment of full-thickness ar- ticular surface lesions are limited to single-plug osteochondral auto- graft transfer (i.e., with the use of plugs measuring 5 to 12 mm in diameter) and osteochondral allo- graft reconstruction. The other avail- able procedures attempt to achieve full restoration of only the articular surface and therefore should be considered merely reparative. Abrasion arthroplasty and micro- fracture rely on facilitation of local mesenchymal stem cells for articu- lar cartilage reconstruction; unfor- tunately, the repair tissue is pre- dominantly fibrocartilaginous in nature. Surgery utilizing periosteal or perichondrial tissue can achieve a biologic response that is closer to full restoration, with induction of chondroneogenic cells; neverthe- less, the result falls short of full restoration because microfracture is still a key component in the tech- nique. Mosaicplasty is a technique that involves the use of multiple donor osteochondral dowel plugs. This procedure would approach being restorative if it were not for the fibrocartilage that invariably grows between the plugs. Autol- ogous chondrocyte implantation appears to offer the best potential for restoration, involving as it does reimplantation of the patientÕs own cultured chondrocytes; however, core biopsy specimens include residual periosteum from the artic- ular surface and therefore repre- sent some fibrocartilage mixture. Table 1 Goals and Source of Chondrocytes for Surgical Treatment of Articular Cartilage Lesions Goals Source of Chondrocytes Facilitated Intra- Extra- Procedure Reparative Restorative MSC * articular articular Cultured Allogeneic Chondroplasty (debridement) ÐÐ ÐÐÐÐÐ Laser chondroplasty ÐÐ ÐÐÐÐÐ Abrasion arthroplasty + Ð + ÐÐÐ Ð Microfracture + Ð + ÐÐÐ Ð Periosteum/ perichondrium + ÐÐÐ + ÐÐ Autologous chondro- cyte implantation ++ ÐÐ ++ Ð Osteochondral auto- graft transfer Ð + Ð + ÐÐ Ð Mosaicplasty + + + + ÐÐ Ð Allograft Ð + ÐÐÐÐ + * MSC = mesenchymal marrow stem cells. This procedure has both reparative and restorative qualities, but it is predominantly restorative in nature. Jon E. Browne, MD, and Thomas P. Branch, MD Vol 8, No 3, May/June 2000 185 Surgical Procedures Arthroscopic Debridement Arthroscopic debridement (chon- droplasty) to remove loose flaps or edges that mechanically impinge on the joint will temporarily improve symptoms. On the basis of a 1-year follow-up on 15 patients, Levy et al 12 noted 100% good or excellent results from simple arthroscopic debride- ment. In their study, they limited surgical intervention to debridement of the lesion to a stable rim and removal of the calcified cartilage base. Remarkably, 33% of the lesions found in this homogeneous popula- tion of soccer players were less than 10 mm in diameter and were consid- ered to be the source of their symp- toms. Repeat biopsy specimens obtained from 4 patients revealed fibrocartilage in the lesions, suggest- ing a reparative response. Longer follow-up is necessary to decide whether this form of treatment car- ries the longevity of modern articular cartilage repair techniques. Abrasion Arthroplasty Popularized in the early 1980s by Johnson, abrasion arthroplasty is indicated in the treatment of an exposed sclerotic degenerative ar- thritic joint lesion. It involves careful intracortical superficial abrasion to create a vascular response not medi- ated by the subchondral bone mar- row elements, but rather by cells within the joint itself. At the follow- up evaluation of 10 patients 1 year after treatment, Johnson 13 found that 1 patient had 100% hyaline type II collagen formation; biopsy specimens from the remaining patients showed predominantly fibrocartilaginous re- sponses, with varying amounts of hyaline articular cartilage. Repara- tive tissue appears to be the domi- nant result of this technique. Microfracture Techniques Microfracture techniques, such as drilling of sclerotic subchondral exposed bone, 14 stimulate the for- mation of a smooth fibrocartilagi- nous surface. Steadman et al 15 ex- panded their use for the treatment of full-thickness traumatic chondral injuries. In a series of more than 200 treated patients, the authors found that 75% had an improve- ment in pain at a minimum follow- up interval of 7 years. This tech- nique involves the use of surgical awls (rather than drilling, which generates heat) to create several subchondral puncture holes 3 to 4 mm apart. Important technical adjuncts are careful debridement of the calcified cartilage layer and the use of postoperative continuous passive motion (CPM) with protected weight bearing for 6 to 8 weeks. Long-term follow-up histologic analysis is needed to allow evalua- tion of the repair tissue. Laser Chondroplasty Laser chondroplasty allows pre- cise molding and contouring of soft-tissue joint structures. How- ever, there is concern about poten- tial cellular necrosis of chondro- cytes near the directed laser beam. Therefore, care should be taken when using this technique. 16 Periosteal and Perichondrial Grafting Periosteal and perichondrial grafts have been demonstrated to effect chondroneogenesis in vitro from their cambium layer. 17 Hom- minga et al 18 implanted 30 costal perichondrial grafts in 25 knees and noted very good early func- tional rating scores. At 5 to 7 years postoperatively, 20 of 30 grafts had developed enchondral ossification. Lorentzon et al 19 reported on 26 tibia-based periosteal grafts im- planted into patellar defects that had been concurrently treated with microfracturing and debridement accompanied by an aggressive postoperative regimented CPM program. At an average follow-up interval of 42 months, 16 excellent and 9 good results were noted; only 1 patient had a poor result. Biopsy specimens obtained randomly from 5 patients revealed a hyalinelike car- tilage appearance. To date, clinical ex- periences with isolated periosteal transplants in humans remain limited. Autologous Chondrocyte Implantation Autologous chondrocyte implan- tation was first reported in 1994 by Brittberg et al. 20 They initially har- vested autologous chondrocytes from 23 patients and then expanded and manipulated these cells in cul- ture, prior to reimplantation under a periosteal flap. The mean follow- up of these procedures was 39 months. Second-look arthroscopy and biopsy was performed on 15 of 16 treated femoral lesions in 16 pa- tients. Hyalinelike tissue repair was found in 11 lesions. Fourteen patients rated their results as either good or excellent. The patellar defects fared worse, with only 2 of 7 patients rating their knees as excellent or good, and 1 having hyalinelike tissue on second-look arthroscopy and biopsy. Unfortu- nately, the results noted in in vivo animal models are conflicting. 21 Improvement was noted in rabbit models with periosteum plus chon- drocyte implantation versus perios- teum implantation only; however, these results were not replicated in a canine model. 22 The United States and European experience in 50 patients (not in- cluding the Swedish experience) with at least 2 years of postopera- tive follow-up has been reported (Cartilage Repair Registry Report, vol 4, Genzyme Tissue Repair, Cambridge, Mass, February 1998). Clinicians noted a good to excellent result in 86% of their patients, and 79% of the patients also rated their results as good to excellent. A total of 891 transplants are included in this report. There was a 12.6% com- Articular Cartilage Lesions Journal of the American Academy of Orthopaedic Surgeons 186 plication rate (112 patients), and 88 patients (9.9%) required a second operative procedure. Treatment fail- ures were noted in 18 patients (2%). The cumulative index rate of failure at 2 years was estimated at 5.8%. Autologous chondrocyte im- plantation (Fig. 3) is indicated for the younger (aged 20 to 50 years) active patient with an isolated traumatic femoral chondral defect greater than 2 to 4 cm 2 . Care should be taken to ensure that the lesion is not so deep (i.e., 3 to 6 mm into the subchondral boundary) that an initial repair of the subchondral base might be neces- sary. Accompanying ligamentous and meniscal lesions, joint malalign- ment, and patellofemoral instability must be corrected concurrently. Absence of a meniscus may preclude such treatment even with a meniscal allograft due to the persistence of residual high joint-reaction forces. 23 Bipolar lesions of the articular sur- face also militate against its use. Osteochondral Autograft Osteochondral autograft was first reported by Outerbridge et al 24 for treatment of osteochondritis dissecans defects in the femur. They used the lateral patellar facet as an autograft. As much as one third of the surface width may be removed. A follow-up study of 10 patients an average of 6.5 years after the procedure revealed satisfactory functional results with decreased symptoms. Postoperatively, the pa- tients had mild donor-site patello- femoral pain. The mosaicplasty procedure popularized by Hangody and co- workers provides treatment op- tions for much larger and deeper femoral condylar or patellar de- fects. In their series of 227 cases, 25 the follow-up interval for 57 pa- tients was more than 3 years. As evaluated with use of a modifica- tion of the Hospital for Special Surgery scoring system, 91% of these 57 patients achieved a good or excellent result. Twelve patients underwent second-look arthroscop- ic biopsy, which revealed that the transplanted cartilage remained hyaline in character and that donor- graft bonding sites were fibrocarti- laginous. The use of autografts is appeal- ing; however, there is a limited amount of donor-graft tissue avail- able for transfer and a potential risk of donor-site morbidity. Mo- saicplasty currently is dependent on surgical skill to recreate the nor- mal radius of curvature in the femoral condyle. This is particular- ly true when multiple small grafts (2.7 to 4.5 mm in diameter) must be press-fitted together to repair a large defect. The two-dimensional surface area can be covered with this technique, but it is difficult to reproduce the three-dimensional surface of the femoral condyle. Collapse of the osteochondral dow- els by migration or degradation leads to flattening in the area of the mosaicplasty. This procedure may also result in additional damage to the subchondral bone structure of the femur, resulting in a change in the osseous contour of the femoral condyle in those cases in which the original lesion affected only chon- dral tissue. Osteochondral Allografts Osteochondral allografts (Fig. 4) may be used for larger (>10 cm 2 ) full-thickness lesions after the fail- ure of one or two previous surgical procedures. Fresh allografts (i.e., obtained within 24 to 72 hours) provide the greatest likelihood of chondrocyte survivability, but also carry a higher risk of immunogenic and transmissible disease. Incuba- tion periods for infection screening may be too long to allow implanta- tion of a fresh graft within the 72- hour time limit. Use of a ÒshellÓ graft (one with <1 cm of subchon- dral bone base) reduces immuno- genicity of the graft by decreasing exposure of white cells found in can- cellous bone. A factor contributing to the fail- ure of osteochondral allografts is the host-directed tissue remodeling of the graft by Òcreeping substitu- tion.Ó The speed of this substitu- tion is reduced by ÒcorkÓ fixation of the allograft (in which a graft shaped like a tapered cone is press- fitted into the site) into the host knee but is increased when trans- graft stabilization (e.g., with screws or absorbable pins) is needed. The technical constraints of sur- gical implantation of fresh osteo- chondral allografts are extremely Figure 3 Arthroscopic images obtained 1 year after autologous chrondrocyte implantation show an intact graft (arrows) in the lateral femoral condyle of a 35-year-old man who had had a 6-cm 2 lesion. An acute lateral meniscus tear was noted during examination (A) and was subsequently resected (B). A B Jon E. Browne, MD, and Thomas P. Branch, MD Vol 8, No 3, May/June 2000 187 demanding. Fresh tissue from a young donor (<30 years old) must be available; the recipient patient must be on call; and the surgeon must be able to transplant the graft at all hours of the day and night. Gross 26 considers the best indica- tion for the procedure to be a post- traumatic or osteochondritis disse- cans defect. Any associated angular deformity must be corrected by oste- otomy, usually performed at the same time. A total of 126 proce- dures on 123 knees were reviewed at an average follow-up interval of 7.5 years. The success rate (defined on the basis of achieving good or excel- lent results) at 5 years was 95%; at 10 years, 71%; and at 20 years, 66%. Failures were most common in bipo- lar grafts and in workerÕs compensa- tion cases. This procedure is not rec- ommended for the patient with osteoarthritis. Typically, patients present with a large monopolar traumatic lesion, which, if left untreated, will result in permanent damage to the opposite side of the joint (becoming a bipolar lesion). Failed primary reconstruc- tions of intra-articular femoral condylar fractures or of complex lat- eral tibial plateau fractures are appropriate situations. Even when these patients are maintained in non-weight-bearing status on crutches, rapid deterioration of the other side of the joint occurs. In these situations, fresh allografts may be the answer. Although fresh- frozen allografts have a decreased risk of immunogenic response and viral transmission, there is concern that the viability of chondrocytes in the donor articular surface will be reduced, potentially decreasing the longevity of the osteochondral allo- graft. Rehabilitation Guidelines All of these procedures except debridement require protected weight bearing for varying time periods (a minimum of 6 weeks). Continuous passive motion may be helpful for improving surface con- tour during the postoperative peri- od. Limitation of the arc of motion may be necessary, but its value in articular surface nutrition and function has been well documented by Salter. 27 Allograft techniques usually require longer periods of protected weight bearing (3 to 4 months). Return to functional work and sports activities is possi- ble with all the procedures, but allograft transplantation necessi- tates consideration of permanent moderation of activities. Authors’ Preference Our preferred approach for treating full-thickness articular surface inju- ries assumes that six basic criteria have been satisfied: (1) age range from skeletal maturity to 50 years; (2) stable knee ligaments, with either preoperative or concurrent recon- struction of any defects; (3) stable neutral tracking extensor mecha- nism; (4) intact menisci (meniscal allograft may be necessary); (5) sin- gle or multiple full-thickness fe- moral condyle or patellar articular surface defects without bipolar defect (i.e., femoral tibial and/or patellofemoral joint-surface changes A B C D Figure 4 A, Preoperative photograph of the knee of a 50-year-old patient with a failed osteochondral autograft transfer. B, Implanted osteochondral allograft. C, Osteochondral allograft and meniscus. D, Photograph obtained during second-look arthroscopy shows restored contour of the articular surface. Articular Cartilage Lesions Journal of the American Academy of Orthopaedic Surgeons 188 greater than grade 2); and (6) a defect that is not osteoarthrotic or associated with inflammatory joint disease. The algorithm shown in Figure 5 outlines an approach to simplification of decision making in the surgical treatment of these defects. Traumatic joint injuries that have resulted in loss of surface contour involving more than half of the joint compartment are best handled with fresh allograft even as a primary surgical procedure. To date, little information regarding resurfacing of tibial defects (other than with al- lograft) has been published. There- fore, at this time, smaller lesions should be treated with debride- ment, with or without microfractur- ing. Lesions with loss of surface contour congruity should be man- aged with fresh allograft. As for the tibial articular surface, technical details of the osteochondral auto- graft transfer and the mosaicplasty procedure limit their use to the ante- rior third of each compartment. Summary Articular cartilage is a nearly fric- tionless system that can provide maintenance-free service for de- cades of activity. Unfortunately, its intrinsic reparative processes can- not cope with full-thickness injury. It is difficult to predict which full- thickness chondral lesion will progress to become symptomatic, but current reparative or restora- tive surgical procedures provide an opportunity to return the surface to its normal or nearly normal func- tional status. Obviously, many associated factors, such as accom- panying joint abnormalities, body weight, job description, and activi- ty level may influence the necessity of treating these defects. Postoperative evaluation of all of these techniques requires constant documentation of patient progress. Occasionally, the need for second- look arthroscopy will arise. Assess- ment of the defect should include inspection, instrumented indenta- tion probing to measure cartilage stiffness (compared with the oppo- site surrounding normal tissue walls), and biopsy. Surgical biopsy establishes a more dynamic picture with histologic evaluation, particu- larly when it extends to the zonal base of the calcified and noncalci- fied subchondral bone as well as the junction between normal tissue and the treated defect. Collagen typing, weight-bearing plain films, MR imaging, and possibly bone scanning may also be useful. New developments might influ- ence the ease and reproducibility of articular-surface restoration proce- dures. Growth factors, adhesives, artificial bioabsorbable scaffolding matrices, and gene therapy manip- ulation are being investigated as possible adjuncts to the current standard surgical techniques. Also being explored is the use of mar- row aspiration to obtain pluripo- tential mesenchymal marrow stem cells, which can then be injected into the defect and covered by bio- absorbable artificial matrices or scaffolding. Early results need to be carefully assessed over many years with continual monitoring and updating before clinical recommendations about the durability of results can be made. Acknowledgment: The authors would like to thank Spencer P. Browne for his techni- cal assistance in manuscript preparation. Lesions <1.5 cm 2 Lesions >1.5 cm 2 but <4 cm 2 Restorative • Single-plug OAT Reparative • Microfracture • Chondroplasty Restorative • ACI • Allograft • Mosaicplasty Reparative • Microfracture • Mosaicplasty • Microfracture plus periosteal flap Restorative • ACI • Allograft Reparative • Possibly microfracture • Possibly mosaicplasty Restorative • Allograft • Possibly ACI Reparative • None Lesions >4 cm 2 but <8 cm 2 Lesions >8 cm 2 Symptomatic full-thickness articular surface lesion (assuming that all preoperative requirements have been satisfied) Salvage procedures for all of these techniques should be limited to allograft reconstruction for lesions >1.5 cm 2 . Lesions <1.5 cm 2 may merit attempts at other surgical techniques without risking transformation to a bipolar lesion. Figure 5 Algorithm for the treatment of articular cartilage lesions. ACI = autologous chondrocyte implantation; OAT = osteochondral autograft transfer. Jon E. Browne, MD, and Thomas P. Branch, MD Vol 8, No 3, May/June 2000 189 References 1. Curl WW, Krome J, Gordon ES, Rushing J, Smith BP, Poehling GG: Cartilage injuries: A review of 31,516 knee arthroscopies. Arthroscopy 1997; 13:456-460. 2. Messner K, Maletius W: The long- term prognosis for severe damage to weight-bearing cartilage in the knee: A 14-year clinical and radiographic fol- low-up in 28 young athletes. Acta Orthop Scand 1996;67:165-168. 3. Maletius W, Messner K: The effect of partial meniscectomy on the long-term prognosis of knees with localized, severe chondral damage: A twelve- to fifteen-year follow-up. Am J Sports Med 1996;24:258-262. 4. Buckwalter JA, Mankin HJ: Articular cartilage: Part I. Tissue design and chondrocyte-matrix interactions. J Bone Joint Surg Am 1997;79:600-611. 5. Buckwalter JA, Mankin HJ: Articular cartilage: Part II. Degeneration and osteoarthrosis, repair, regeneration, and transplantation. J Bone Joint Surg Am 1997;79:612-632. 6. Herzog RJ: Radiologic imaging in reha- bilitation, in Kibler WB, Herring SA, Press JM, Lee PA (eds): Functional Rehabilitation of Sports and Musculo- skeletal Injuries. Gaithersburg, Md: Aspen Publishers, 1998, pp 20-56. 7. Ochi M, Sumen Y, Kanda T, Ikuta Y, Itoh K: The diagnostic value and limi- tation of magnetic resonance imaging on chondral lesions in the knee joint. Arthroscopy 1994;10:176-183. 8. Dye SF, Chew MH: The use of scintig- raphy to detect increased osseous metabolic activity about the knee. J Bone Joint Surg Am 1993:75:1388-1406. 9. Outerbridge RE: The etiology of chon- dromalacia patellae. J Bone Joint Surg Br 1961;43:752-757. 10. Jackson RW: Arthroscopic surgery and a new classification system. Am J Knee Surg 1998;11:51-54. 11. Odenbring S, Egund N, Lindstrand A, Lohmander LS, WillŽn H: Cartilage regeneration after proximal tibial osteotomy for medial gonarthrosis: An arthroscopic, roentgenographic, and histologic study. Clin Orthop 1992;277: 210-216. 12. Levy AS, Lohnes J, Sculley S, LeCroy M, Garrett W: Chondral delamination of the knee in soccer players. Am J Sports Med 1996;24:634-639. 13. Johnson LL: Arthroscopic abrasion arthroplasty, in McGinty JB, Caspari RB, Jackson RW, Poehling GG (eds): Oper- ative Arthroscopy, 2nd ed. Philadelphia: Lippincott-Raven, 1996, pp 427-446. 14. Pridie KH: A method of resurfacing osteoarthritic knee joints [abstract]. J Bone Joint Surg Br 1959;41:618-619. 15. Steadman JR, Rodkey WG, Singleton SB, Briggs KK: Microfracture technique for full-thickness chondral defects: Technique and clinical results. Opera- tive Techniques Orthop 1997;7:300-304. 16. Vangsness CT Jr, Ghaderi B: A litera- ture review of lasers and articular car- tilage. Orthopedics 1993;16:593-598. 17. OÕDriscoll SW, Keeley FW, Salter RB: The chondrogenic potential of free autogenous periosteal grafts for bio- logical resurfacing of major full-thick- ness defects in joint surfaces under the influence of continuous passive motion: An experimental investigation in the rabbit. J Bone Joint Surg Am 1986;68:1017-1035. 18. Homminga GN, Bulstra SK, Bouw- meester PSM, van der Linden AJ: Perichondral grafting for cartilage lesions of the knee. J Bone Joint Surg Br 1990;72:1003-1007. 19. Lorentzon R, Alfredson H, Hildings- son C: Treatment of deep cartilage defects of the patella with periosteal transplantation. Knee Surg Sports Traumatol Arthrosc 1998;6:202-208. 20. Brittberg M, Lindahl A, Nilsson A, Ohlsson C, Isaksson O, Peterson L: Treatment of deep cartilage defects in the knee with autologous chondrocyte transplantation. N Engl J Med 1994; 331:889-895. 21. Grande DA, Pitman MI, Peterson L, Menche D, Klein M: The repair of experimentally produced defects in rabbit articular cartilage by autologous chondrocyte transplantation. J Orthop Res 1989;7:208-218. 22. Breinan HA, Minas T, Hsu HP, Nehrer S, Sledge CB, Spector M: Effect of cultured autologous chondrocytes on repair of chondral defects in a canine model. J Bone Joint Surg Am 1997;79:1439-1451. 23. Paletta GA Jr, Manning T, Snell E, Parker R, Bergfeld J: The effect of allo- graft meniscal replacement on intraar- ticular contact area and pressures in the human knee: A biomechanical study. Am J Sports Med 1997;25:692-698. 24. Outerbridge HK, Outerbridge AR, Outerbridge RE: The use of a lateral patellar autologous graft for the repair of a large osteochondral defect in the knee. J Bone Joint Surg Am 1995;77:65-72. 25. Hangody L, Kish G, K‡rp‡ti Z, Udvar- helyi I, Szigeti I, BŽly M: Mosaicplasty for the treatment of articular cartilage defects: Application in clinical prac- tice. Orthopedics 1998;21:751-756. 26. Gross AE: Fresh osteochondral allo- grafts for post-traumatic knee defects: Surgical technique. Operative Tech- niques Orthop 1997;7:334-339. 27. Salter RB: The biologic concept of con- tinuous passive motion of synovial joints: The first 18 years of basic research and its clinical application. Clin Orthop 1989;242:12-25. . the joint to its normal state. Documentation of Arthroscopic Findings Classifying the condition of the joint and the nature of a chondral lesion necessitates a documentation system. The grading. concomitant with the return of a normal scintigraphic appearance. Bone scanning has not been used to document joint homeo- stasis during the treatment of artic- ular surface lesions; however, it may. symptomatic as a result? Without treatment, might it have remained asymptomatic? The ab- sence of a documented natural his- tory makes these decisions difficult. Until the natural history of the sur- gically