KEY POINTS

■ Acute lesions or those that have become symptomatic warrant expeditious surgical consideration, often within 2 to 3 weeks, as these injuries commonly progress and develop more significant cartilage deterioration.

■ Although fibrocartilage does not have the same biomechanical capabilities as hyaline cartilage to withstand stresses, stimulating the production of fibrocartilage in appropriate osteochondral defects can allow patients to perform asymptomatically for years.

■ Microfractures promote propagation of a blood clot. Marrow elements within the clot fill in the defect and form fibrocartilage mixed with hyaline cartilage.

■ Mosaicplasty requires the use of equipment that harvests a plug of articular cartilage and subchondral bone and transfers it to the site of a full-thickness osteochondral defect.

■ Autologous chondrocyte implantation is the only available procedure that attempts to regrow hyaline cartilage cells for implantation into chondral defects of the knee. What we have learned through all these studies is that younger patients with smaller lesions, low BMI, and short duration of symptoms generally have better outcomes.

The articular surface of the knee is frequently injured (Figure 1). Such injuries are often classified according to the degree of chondro­malacia, or softening of the cartilage, using a system developed by Outerbridge1 (Table 1). Because articular cartilage has no ability to heal itself, significant chondral injuries may require surgical correction. In a retrospective analysis of 31,516 knee arthroscopies, Curl and colleagues found that 63% involved lesions of the articular cartilage; of these, 41% were grade III and 19.2% were grade IV chondromalacia.2 Five percent of the arthroscopies were performed in patients younger than 40 years. The most common locations of the cartilaginous defects were the medial femoral condyle, the patella, and the lateral femoral condyle. Over the past 50 years, advancements in surgical techniques have allowed patients with these types of injuries to remain active, sometimes returning them to previous levels of activity. This article will review current surgical options for treatment of chondral injuries of the knee.

CLINICAL EVALUATION


Chondral or osteochondral damage often results from an acute injury that involves a shearing or axial load across the knee joint surfaces during an athletic event or fall or following impact to the area. Patients may present with rapidly developing hemiarthrosis. Focal pain often originates in the medial or lateral aspect of the knee or beneath the patella. There may be pain with weight bearing and concomitant soft-tissue injuries involving the cruciate ligaments, collateral ligaments, patellar ligament, and menisci. 


Physical examination may reveal joint-line tenderness and pain, joint effusion, ligamentous instability, and decreased range of motion (ROM). Initial diagnostic studies include radiographs of the patient taken during weight bearing, with the knee straight and flexed. Most revealing are lateral and patellar views, with particular attention to the appearance of the joint surfaces. Any irregularity of the joint surface—including visible pitting or shadows or indication of ligamentous disruption—necessitates an MRI, which is very sensitive and specific for chondral and osteochondral lesions. Once the clinical diagnosis is made, a referral for surgical consultation is required. Allowing these lesions to be left untreated precipitates further cartilage injury and degeneration of the joint. 


TREATMENT OPTIONS


Deciding how to treat a chondral injury can be a complex and multifaceted process. Treatment options range from nonsurgical to surgical. Miller and Cole derived an algorithm that directs a treatment strategy for chondral injuries (Figure 2).3 The algorithm categorizes the chondral defects by size (<2 cm2 and ≥2 cm2) and then considers primary and secondary treatment options, which include surgery. Typically, the patient's age, activity level, and habitus are taken into account when deciding which form of surgical treatment will be utilized. Current techniques include arthroscopic lavage/debridement with a mechanical shaver or a thermal radiofrequency device; marrow-stimulating techniques, such as microfracture; an osteochondral autologous transfer system (OATS) procedure; and autologous chondrocyte implantation (ACI). These surgical techniques require specific skills, a thorough understanding of procedure and surgical expertise in the technique, patient compliance with postoperative protocols, and appropriate patient selection. Surgical treatment based on appropriate patient selection and defect size is key to a successful outcome. 


Nonsurgical treatment for chondral injuries is usually reserved for asymptomatic lesions less than 5 mm2. Over time, these lesions may progress to a symptomatic state. At that point, the patient's profile (body habitus), activity level, and age will determine appropriate treatment. Nonsurgical treatment involves activity modification, NSAIDs, and possibly viscosupplementation or corticosteroid injections. Acute lesions or those that have become symptomatic warrant expeditious surgical consideration, often within 2 to 3 weeks, as these injuries commonly progress and develop more significant cartilage deterioration.


Arthroscopic lavage and/or debridement is reserved for defects less than 2 cm2; lesions not amenable to repair; and lesions in patients whose knees are not exposed to great stresses, that is, low-demand patients.3 This technique is used for temporary pain reduction and to increase function. Debridement can remove and minimize any articular cartilage defects causing pain, such as degenerative cartilage on femoral condyles and the tibial plateau, degenerative menisci, and articular cartilage flaps. Additionally, this procedure is used when incidental chondral lesions are discovered or when concomitant procedures are being performed in the knee. A mechanical shaver is used to debride any derangement of the articular cartilage to a homogenous appearance. The intent is to smooth out the articular surface as much as possible in order to prevent future fibrillation and fissuring of the hyaline cartilage. Orthopedists recognize the limitations of this technique, which can fail to relieve pain.


Marrow-stimulating techniques use mechanical means to promote production of fibrocartilage that will fill in chondral defects. When the subchondral bone is involved, creating a lesion that bleeds into the defect will stimulate production of fibrocartilage to fill in the hole. Although fibrocartilage does not have the same biomechanical capabilities to withstand stresses that hyaline cartilage does, stimulating the production of fibrocartilage in appropriate osteochondral defects can allow patients to perform asymptomatically for years. Defects ranging from 1 cm2 to 5 cm2 are most likely to have positive results with this approach.3


The most commonly used marrow-stimulating technique for osteochondral defects is microfracture, which was developed in the 1980s by J. Richard Steadman, MD. This technique is used in low- to high-demand patients with lesions smaller than 3 cm2 and in low-demand patients with some lesions greater than 3 cm2. Microfracture is performed by using a curette or a mechanical shaver to debride the lesion of all hyaline cartilage. A vertical wall of articular cartilage is then established around the lesion, taking care not to damage the subchondral plate, that is, the exposed bone beneath the layer of articular cartilage. Arthroscopic awls with varying degrees of angulation are used to create artificial fractures perpendicular to the subchondral bone and approximately 3 to 4 mm apart, starting from the periphery and working to the center of the lesion (Figure 3). These microfractures promote propagation of a blood clot. Marrow elements within the clot fill in the defect and form fibrocartilage mixed with hyaline cartilage. Gentle, passive range of motion is initiated immediately, followed by continuing passive motion (CPM) and enforcement of non-weight-bearing status for 6 to 8 weeks postoperatively.


If the technique is successful, fibrocartilage will fill the defect after 8 weeks. At that point, weight bearing may be initiated, and participation in sports may be allowed 4 to 6 months later. Results are encouraging. In several studies 
by Steadman and colleagues, most patients had good to excellent results.4-6 Investigators led by Blevins followed athletes who also improved but saw their improvement plateau over the next 4 to 5 years.7 All studies discovered trends demonstrating that younger patients fared better than older ones. In a 24-month prospective study of 48 patients, researchers led by Mithoefer found excellent outcomes in 67%, fair outcomes in 25%, and poor outcomes in 8%.8 The authors also correlated lower body mass index (BMI) and short duration of preoperative symptoms with higher functional improvement.8 


Osteochondral autologous transplant techniques are analogous to changing the cup on the green of a golf course. The procedure requires the use of equipment that harvests a plug of articular cartilage and subchondral bone and transfers it to the site of a full-thickness osteochondral defect, usually in the weight-bearing area of a condyle. The typical harvesting site is the lateral trochlear ridge, although the intercondylar notch can also provide the plug for transplantation. The donor site can be backfilled with the plug from the defect site to decrease tissue morbidity at the harvesting site. A full-thickness, grade IV osteochondral lesion smaller than 2 cm2 is ideal for an autograft implantation using an osteochondral autologous transfer system (OATS) procedure (Figure 4).3 Larger lesions (size range, from 2 cm2 up to a hemicondyle) are appropriate candidates for an allograft OATS procedure (Figure 5).3 OATS procedures can also be beneficial for some grade III lesions, depending on the patient's profile and activity level. 


A prospective study by Marcacci and colleagues evaluated 30 patients (mean age, 29.3 years) 7 years after they had undergone an OATS procedure and found good to excellent results in 76.7%.9 This study utilized multiple plugs for treatment of the lesion; three failed cases used three and four plugs. Following an older population (average age, 44.6 years), Chow and his team reported good to excellent outcomes in 83% of patients 2 to 5 years post OATS procedure.10 This study also used multiple plugs for some lesions. A comparative analysis found no difference in results between patients younger than 45 years and those older than 45 years.10 In a second study, Marcacci and others reported on 13 patients with a mean age of 31 years. Results in 12 patients were rated as good to excellent; all subjects reported satisfaction with their postoperative experience. Duration of follow-up was between 13 and 141 months.11 Other researchers led by Miura treated unstable in situ osteochondritis dissecans lesions in 12 patients, aged 11 to 26 years. Follow-up at 33 to 71 months postprocedure found good to excellent results in 91.7%.12 


Mosaicplasty is a form of OATS that uses multiple auto­logous osteochondral plugs of various sizes to fill in a large defect (Figure 6). The challenges with this technique lie in establishing normal joint geometry when using multiple small plugs and reducing harvest-site morbidity when using large plugs. Nevertheless, several authors report good to excellent results with mosaicplasty. Hangody and colleagues evaluated 168 patients with lesions in the knee or ankle 18 months to 5 years after undergoing mosaicplasty. An average of eight plugs (range, 1-18) were used in the knee repairs, whereas an average of two plugs (range, 1-4) were used in the ankle repairs.13 Follow-up arthroscopies demonstrated integration of the grafts as well as fibrocartilage formation around the plugs. 


In another study, researchers led by Hangody followed up on 57 patients 3 years after they underwent mosaicplasty and found good to excellent results in 91%.14 The team used an average of eight plugs (range, 3-17) in each patient. At 10-year evaluations of patients who had undergone mosaicplasty, Hangody found good to excellent results with 92% of femoral implants and 79% of patellar and/or trochlear implants; morbidity after mosaicplasty was 3%.15 In a similar study, Kish and associates studied 52 athletes, aged 17 to 50 years, who had good to excellent results, with 63% returning to full participation in their sport, 31% to a lower level in that sport, and 4% to less strenuous activities; 2% of study participants had retired from athletics.16 When the patient population was subdivided into two groups by age, the researchers found that 90% of patients younger than 30 years had returned to full activity levels, 7% were participating at reduced levels, and 3% were playing low-level sports; none had retired from sports entirely. The group older than 30 years reported less return to full sports and more participation at restricted levels; 4% had retired altogether.16 Another study on athletes found a good to excellent outcome in 78.3%, with 72% of them able to participate in sports at their preoperative level and 13% at a reduced level; another 13% were unable to resume any sports activity.17 The researchers also found that younger athlete-patients did better with mosaicplasty than did older athlete-patients.17 


An alternative approach to mosaicplasty is to use the OATS procedure to place an allograft for repair of full-thickness osteochondral defects. Typically, the use of an allograft is necessitated by a large defect that would require harvesting a similarly large cylinder for the repair.3 In that situation, the potential for morbidity at the harvest site increases because of the need to harvest a large plug or multiple plugs. Using fresh frozen allografts appears to be a viable and reasonable alternative treatment choice. In one of the first studies published on allografts, Garrett followed 24 subjects who received size-matched fresh osteochondral allografts that were implanted within 12 hours of harvesting.18 Follow-up was done at 2 to 4 years in 10 cases and at 1 to 2 years in 14 other cases. All patients reported minimal to no pain, swelling, or buckling of the knee. Eleven grafts that were evaluated arthroscopically appeared viable, with no evidence of graft collapse. Two additional studies evaluated fresh osteochondral allografts. Of the 18 patients in the first study, outcomes were reported as excellent in 11, good in 3, fair in 3, and poor in 1 patient.19 The other study found good to excellent results in greater than 70% of participants.20 Despite these positive findings, utilizing allograft tissue is no easy feat. Matching the graft according to skeletal size is challenging, obtaining a fresh graft is expensive, and there is always a small risk of disease transmission. The technique is demanding and challenging, and it requires experience for a successful outcome. 


Overall, use of an osteochondral autologous transfer systems procedure for a single plug, multiple plugs, or an allograft plug is a reliable single-stage procedure that replaces damaged articular cartilage with existing articular cartilage. Currently, no other single-stage procedure does this. Studies comparing microfracture and OATS have demonstrated the efficacy of microfracture; however, patients undergoing an OATS procedure had higher levels of function, higher rates of return to sports, fewer failures, and an overall better outcome for those who were active and athletic.21 In another study, patients who received an autograft or allograft via OATS and/or mosaicplasty underwent similar rehabilitative programs. All patients remained nonweight-bearing for 6 to 8 weeks and utilized a hinged knee brace with motion set at 0° to 90°. Stationary cycling was initiated when the patient was able, and progressive weight bearing was allowed at 6 to 8 weeks postoperatively if radiographic findings appeared satisfactory. Follow-up MRI at 6 months was used to assess healing of the graft, and a return to sports was allowed 
when the affected leg showed evidence of healing and functional strength.22 


Autologous chondrocyte implantation (ACI) is the only available procedure that attempts to regrow hyaline cartilage cells for implantation into chondral defects of the knee. The two-stage procedure is expensive. The initial procedure is to biopsy the knee joint to obtain autologous chondrocyte cells, which are placed in a culture medium at 4°C and sent to Genzyme Biosurgery (Cambridge, Massachusetts) for processing and growth of the cells. This process takes 3 to 5 weeks; cells may be stored for up to 18 months. The second surgical procedure involves arthrotomy of the joint. The lesion area is prepared, and a periosteal graft is obtained; sutured to the edges of the prepared lesion; and glued in place with fibrin glue, producing a watertight seal. The culture specimen is then injected beneath the periosteal graft into the defect, and the injection site is closed with a suture or fibrin glue (Figure 7).23 Once injected, the wound is closed. For the next 6 weeks, the patient is placed in continuous passive motion for 6 to 8 hours a day and maintains a nonweight-bearing status to protect the newly forming articular surface. Gradual loading of the knee starts at around 3 months and impact activities start after 6 months. Protecting the maturing graft for 3 to 6 months is required for the cartilage to develop resistance to compressive and shearing forces. When the tissue matures, which can take 12 to 24 months, its histologic appearance is the same as that of the surrounding articular cartilage.23


The depth and range of studies examining ACI is impressive, with most demonstrating good results in the short term. Few long-term studies are available, and histologic examinations demonstrate near hyalinelike cartilage in the defects. Mithöfer and colleagues treated 45 soccer players with ACI.24 Of the 41 players available for follow-up, 31 (72%) graded their knee function as good or excellent, and 15 (33%) were able to return to playing soccer. The average return time was 18.1 months (range, 12-24 months). Thirteen of the returning players maintained their ability to play at 52 ± 7.8 months after the procedure, and two world-class soccer players have maintained their level of play for 9 years. Two studies compared ACI to OATS. One study noted that the recovery time was slower for ACI; histologic examination found mostly fibrocartilage at the ACI site as well as a stable surface in all patients.25 One patient failed ACI, and two patients experienced only slight improvement. Seventeen patients noted substantial improvement with an OATS procedure, whereas three felt only slightly better; none failed. The second study involved ACI for 58 patients and mosaicplasty for 42 patients; 88% of ACI patients had good to excellent results, whereas only 69% of mosaicplasty patients had a similar outcome.26 One year later, arthroscopy revealed that 82% had a good to excellent repair with ACI, whereas mosaicplasty results were good to excellent in just 34%. Five patellar mosaicplasties failed. 


When Knutsen and colleagues compared ACI and micro­fracture at 2 years, they found no significant differences between the two treatment groups.27 Two failures occurred in the ACI group compared with one failure in the microfracture group. Researchers found no differences in samples collected for histologic examination. Younger patients did better than older patients in both groups. In a follow-up study, a group led by Knutsen found that 77% of ACI patients and 77% of microfracture patients reported satisfactory results after 5 years.28 Both groups had 9 failures at 5 years compared with 2 failures in ACI and 1 failure in microfracture at the 2-year mark. Patients with predominantly hyaline cartilage at the defect site after 2 years did not go on to failure. In a multicenter randomized controlled study of ACI versus microfracture, investigators utilized strict control measures and followed patients for 18 months.29 Histologic comparison at 12 months postprocedure found that statistically, ACI patients presented with more type II collagen than microfracture patients did. Short-term clinical outcome was similar between the groups, but regenerated tissue was superior with ACI compared to that with microfracture. Although short-term results of ACI are promising, no long-term randomized clinical trials are available to observe its efficacy. 


SUMMARY


The results of all the studies presented demonstrate short-term relief of pain with varying rates of success and failure. Few long-term prospective randomized trials exist for any of these surgical techniques. Microfracture, which has been utilized for almost 20 years, has the largest body of clinical research available. In most cases, it is the treatment of choice for surgeons dealing with chondral defects. Microfracture has been reported to have good success and is technically the easiest of all the available procedures. OATS procedures, including those done to place an allograft, produce good outcomes, and OATS may become the treatment of choice over microfracture as the results of this procedure continue to emerge. Mosaicplasty is seen primarily in European medical centers, because much of the reported literature comes from European orthopedists. The failure rate with mosaicplasty is approximately 30%, and donor-site morbidity is higher. These issues raise concerns about utilizing mosaicplasty for chondral defects. Autologous chondrocyte implantation is a relatively new technique with few long-term studies behind it and is still considered by most to be experimental. The currently available studies show promising results, as does the science. Long-term studies of ACI are needed to determine the efficacy of this surgical approach. What we have learned through all these studies is that younger patients with smaller lesions, low BMI, and short duration of symptoms generally have better outcomes. 


Once the articular knee cartilage is damaged, it is difficult to treat. Currently available surgical procedures are valiant attempts at returning people to active and sporting lifestyles; in many cases, the attempts are successful. Each procedure has its own advantages and disadvantages. Patients who cannot be compliant with rehabilitation protocols will not have good outcomes with microfracture, OATS, or ACI. Those with damage to the subchondral bone are not good candidates for ACI. Large defects do not do well with autograft OATS but might fare well with mosaicplasty or an allograft OATS. The latter two procedures can be technically demanding, and the chance for error increases. Understanding the available procedures, indications, and expectations of your patient can help guide selection of the most appropriate treatment option. JAAPA

Rob Powers practices at Virginia Orthopaedic in Salem, Virginia. The author has indicated no relationships to disclose relating to the content of this article.


Acknowledgment: The author wishes to thank Mark D. Miller, MD, University of Virginia Orthopaedics, Charlottesville, Virginia, and Brian J. Cole, MD, MBA, Rush University Medical Center, Department of Orthopaedics, Chicago, Illinois, for their assistance with the photographs accompanying this article.


REFERENCES


1. Outerbridge RE. The etiology of chondromalacia patellae. J Bone Joint Surg Br. 1961;43-B:752-757.


2. Curl WW, Krome J, Gordon ES, et al. Cartilage injuries: a review of 31,516 knee arthroscopies. Arthroscopy. 1997;13(4):456-460.


3. Miller MD, Cole BJ. Atlas of chondral injury treatment. Oper Tech Orthop. 2001;11(2):145-150.


4. Steadman JR, Rodkey WG, Singleton SB, Briggs KK. Microfracture technique for full-thickness chondral defects: technique and clinical results. Oper Tech Orthop. 1997;7(4):300-304.


5. Steadman JR, Briggs KK, Rodrigo JJ, et al. Outcomes of microfracture for traumatic chondral defects of the knee: average 11-year follow-up. Arthroscopy. 2003;19(5):477-484.


6. Steadman JR, Rodkey WG, Briggs KK. Microfracture to treat full-thickness chondral defects: surgical technique, rehabilitation, and outcomes. J Knee Surg. 2002;15(3):170-175.


7. Blevins FT, Steadman JR, Rodrigo JJ, Silliman J. Treatment of articular cartilage defects in athletes: an analysis of functional outcome and lesion appearance. Orthopedics. 1998;21(7):761-768.


8. Mithoefer K, Williams RJ III, Warren RF, et al. Chondral resurfacing of articular cartilage defects in the knee with the microfracture technique: surgical technique. J Bone Joint Surg Am. 2006;88(suppl 1 pt 2):294-304.


9. Marcacci M, Kon E, Delcogliano M, et al. Arthroscopic autologous osteochondral grafting for cartilage defects of the knee: prospective study results at a minimum 7-year follow-up. Am J Sports Med. 2007;35(12):2014-2021.


10. Chow JC, Hantes ME, Houle JB, Zalavras CG. Arthroscopic autogenous osteochondral transplantation for treating knee cartilage defects: a 2- to 5-year follow-up study. Arthroscopy. 2004;20(7):681-690.


11. Marcacci M, Kon E, Zaffagnini S, Visani A. Use of autologous grafts for reconstruction of osteochondral defects of the knee. Orthopedics. 1999;22(6):595-600.


12. Miura K, Ishibashi Y, Tsuda E, et al. Results of arthroscopic fixation of osteochondritis dissecans lesion of the knee with cylindrical autogenous osteochondral plugs. Am J Sports Med. 2007;35(2):216-222.


13. Hangody L, Kish G, Karpati Z, Eberhart R. Osteochondral plugs: autogenous osteochondral mosaicplasty for the treatment of focal chondral and osteochondral articular defects. Oper Tech Orthop. 1997;7(4):312-322. 


14. Hangody L, Kish G, Kárpáti Z, et al. Mosaicplasty for the treatment of articular cartilage defects: application in clinical practice. Orthopedics. 1998;21(7):751-756.


15. Hangody L, Fules P. Autologous osteochondral mosaicplasty for the treatment of full-thickness defects of the weight-bearing joints: ten years of experimental and clinical experience. J Bone Joint Surg Am. 2003;85-A(suppl 2):25-32.


16. Kish G, Módis L, Hangody L. Osteochondral mosaicplasty for the treatment of focal chondral and osteochondral lesions of the knee and talus in the athlete. Rationale, indications, techniques, and results. Clin Sports Med. 1999;18(1):45-66.


17. Marcacci M, Kon E, Zaffagnini S, et al. Multiple osteochondral arthroscopic grafting (mosaicplasty) for cartilage defects of the knee: prospective study results at 2-year follow-up. Arthroscopy. 2005;21(4):462-470.


18. Garrett JC. Treatment of osteochondral defects of the distal femur with fresh osteochondral allografts: a preliminary report. Arthroscopy. 1986;2(4):222-226.


19. Fox JA, Freedman KB, Lee SJ, Cole BJ. Fresh osteochondral allograft transplantation for articular cartilage defects. Oper Tech Sports Med. 2002;(10)3:168-173.


20. Emmerson BC, Görtz S, Jamali AA, et al. Fresh osteochondral allografting in the treatment of osteochondritis dissecans of the femoral condyle. Am J Sports Med. 2007;(35)6:907-914.


21. Gudas R, Kalesinskas RJ, Kimtys V, et al. A prospective randomized clinical study of mosaic osteochondral autologous transplantation versus microfracture for the treatment of osteochondral defects in the knee joint in young athletes. Arthroscopy. 2005;21(9):1066-1075.


22. Morelli M, Nagamori J, Miniaci A. Management of chondral injuries of the knee by osteochondral autogenous transfer (mosaicplasty). J Knee Surg. 2002;15(3):185-190.


23. Gillogly SC. Treatment of large full-thickness chondral defects of the knee with autologous chondrocyte implantation. Arthroscopy. 2003;19(suppl):147-153.


24. Mithöfer K, Peterson L, Mandelbaum BR, Minas T. Articular cartilage repair in soccer players with autologous chondrocyte transplantation: functional outcome and return to competition. Am J Sports Med. 2005;33(11):1639-1646.


25. Horas U, Pelinkovic D, Herr G, et al. Autologous chondrocyte implantation and osteochondral cylinder transplantation in cartilage repair of the knee joint: a prospective, comparative trial. 
J Bone Joint Surg Am. 2003;85-A(2):185-192.


26. Bentley G, Biant LC, Carrington RWJ, et al. A prospective, randomised comparison of autologous chondrocyte implantation versus mosaicplasty for osteochondral defects in the knee. J Bone Joint Surg Br. 2003;95(2):223-230.


27. Knutsen G, Engebretsen L, Ludvigsen TC, et al. Autologous chondrocyte implantation compared with microfracture in the knee: a randomized trial. J Bone Joint Surg Am. 2004;86-A(3):455-464.


28. Knutsen G, Drogset JO, Engebretsen L, et al. A randomized trial comparing autologous chondrocyte implantation with microfracture: findings at five years. J Bone Joint Surg Am. 2007;89(10):2105-2112.


29. Saris DB, Vanlauwe J, Victor J, et al. Characterized chondrocyte implantation results in better structural repair when treating symptomatic cartilage defects of the knee in a randomized controlled trial versus microfracture. Am J Sports Med. 2008;36(2):235-246.