Osteochondral Lesions of the Talar Dome: Cartilage Replacement Using Osteochondral Autogenous Transplantation and Mosaicplasty

Published on 10/03/2015 by admin

Filed under Orthopaedics

Last modified 22/04/2025

Print this page

This article have been viewed 1777 times

CHAPTER 13 Osteochondral Lesions of the Talar Dome

Cartilage Replacement Using Osteochondral Autogenous Transplantation and Mosaicplasty

Mark Glazebrook, Johnny Tak-Choy Lau, Jean-Pascal Allard

TERMINOLOGY

The term osteochondritis dissecans was first used in 1887, when Konig ¹ described an inflammatory lesion of cartilage and underlying bone in the knee joint that resulted in loss of structural integrity, causing pain and dysfunction. Since then, it has been recognized that lesions of the bone and cartilage may appear spontaneously or may be caused by a traumatic event to the joint surface.² The general term osteochondral lesion (OCL) has gained more widespread use than osteochondritis dissecans.

ETIOLOGY, EPIDEMIOLOGY, AND CLASSIFICATION

In the ankle joint, an OCL can occur spontaneously or as the result of trauma, and there appears to be a relationship between the location of the lesion and the cause.³^,⁴ In a comprehensive review of the early literature that included more than 500 cases of OCL, Flick and Gould⁵ found that 98% of lateral lesions and only 70% of medial lesions were associated with a traumatic event. This and other reports²^,⁶ led to the associations of traumatic OCLs with lateral lesions and spontaneous OCLs with medial lesions.

The incidence of OCL depends on the etiologic factors involved. In an early study, Bosien and colleagues ⁷ reported the incidence of OCL to be 6.5% with ankle sprains, and Hinterman and coworkers⁸ and Leontaritis and associates⁹ reported the incidence of OCL as 69% and 74%, respectively.

The most recognized classification system is that of Berndt and Harty,² which is useful for research purposes and preoperative planning. It is based on radiographic imaging or intraoperative findings. Stage I is a small, subchondral compression; stage II is partial fragment detachment; stage III is complete fragment detachment with no displacement; and stage IV is complete fragment detachment with displacement.

PATIENT EVALUATION

History and Physical Examination

Patients with an OCL usually present with the complaint of persistent pain in the ankle that is frequently localized to the side of the lesion. In acute cases, the pain is accompanied by swelling, ecchymosis, and decreased range of motion. In chronic cases, the pain is usually a dull ache in the ankle joint and is associated with stiffness. Crepitus, discomfort with weight bearing, decreased range of motion, mechanical clicking, and recurrent swelling should raise the suspicion of an OCL. If a thorough history does not elicit a traumatic event or instability, the examiner should seek factors predisposing to vascular insufficiency, including endocrine, vascular, hematologic, and coagulation disorders and any relevant family history. Patients also should be questioned about symptoms in the contralateral ankle.

Physical examination should begin with assessment of the standing alignment to identify an associated coronal plane deformity. Inspection of the ankle with an acute injury usually reveals swelling and ecchymosis, but the ankle sometimes appears completely normal, as is often the case with delayed presentations. An attempt to elicit tenderness with palpation should be made by focusing on the common OCL sites. Posteromedial lesions may be tender in response to palpation when the ankle is dorsiflexed and the region posterior to the medial malleolus is palpated. Anterolateral lesions may be tender when the joint is palpated laterally with the ankle in plantar flexion. These findings are of ten nonspecific because the tenderness is likely related to joint synovitis rather than the OCL itself. The joint should be taken through a range of motion to assess stiffness and to feel for crepitus and mechanical signs of clicking or locking. Ligamentous stability should be assessed to ascertain evidence of associated laxity, particularly of the lateral ligament complex. The examination is completed with a thorough neurovascular assessment and examination of the contralateral ankle for comparison.

Diagnostic Imaging

Plain radiographs are the initial investigation of a suspected OCL. Anteroposterior, lateral, and mortise views should be obtained on presentation. Radiographs in various degrees of plantar flexion and dorsiflexion may help to demonstrate posteromedial and anterolateral lesions, respectively. Plain radiographs of the opposite ankle should be obtained to exclude bilateral lesions. Radiographs are often limited in their ability to demonstrate an OCL, and with continued clinical suspicion, further diagnostic imaging may be necessary.

Bone scintigraphic results correlate with osteochondral injury, and scans can be useful for discerning which joint is causing clinical discomfort if other imaging modalities have not located the lesion. More advanced imaging has largely replaced scintigraphy, but it still provides a valuable tool in the workup of persistent ankle pain if immediate access to computed tomography (CT) or magnetic resonance imaging (MRI) is not available to the evaluating physician.

CT provides best information regarding bony architecture of the lesion, and it assists in accurately determining its dimensions. Scans in the coronal, sagittal, and axial planes should be obtained, especially when initial radiographs appear normal and the result of bone scintigraphy is positive.

MRI affords an accurate assessment of the ankle’s soft tissue structures, including the articular cartilage and supporting ligaments. Multiplanar evaluation is possible with no radiation exposure to the patient. Sequences can identify the size and location, and they provide information about the stability of the lesion. DeSmet and colleagues ¹⁰ prospectively showed an excellent correlation between staging with MRI and arthroscopic results, correctly predicting stable and unstable lesions in 92% of cases. In some circumstances, MRI may overestimate the size of the OCL due to associated edema, and CT may be better suited for defining the dimensions of the defect.

TREATMENT

The use of non–weight-bearing articular surfaces as a donor site for osteochondral autogenous transplantation (OAT) grafts for the treatment of OCL began in the 1980s. Yamashita and coworkers ¹¹ published a case report on two patients who had an OAT graft harvested from the non–weight-bearing portion of the medial femoral condyle and transplanted to an OCL of the same knee.

In 1993, Matsusue and associates ¹² reported their 3-year follow-up of cases treated with a new technique that involved transplantation of multiple osteochondral plugs harvested from the same knee. They used the press-fit method of cylindrical osteochondral plugs described by Garrett¹³ for allografts. They harvested two osteochondral autogenous graft (OAG) plugs from the lateral femoral patellar groove and one from the anterolateral side of the intercondylar notch.

In 1996, Bobic ¹⁴ was the first to publish a case series (12 patients) describing multiple OAT grafts for the treatment of knee OCLs associated with anterior cruciate ligament reconstruction. In 1997, Hangody and colleagues¹⁵ published their first report of their prospective case series. They described 44 patients with 1 to 5 years of follow-up after multiple OAG plugs were used to treat OCLs of the knee. Most patients had good or excellent outcomes. They were the first to call the procedure mosaicplasty. We now commonly use this term for the technique of using multiple OAT graft plugs to treat an OCL. Some surgeons make distinction between OAT and mosaicplasty, using the former to describe the use of only one graft plug and the latter to describe the use of multiple graft plugs. These terms can be used interchangeably, but for clarification, we use the term mosaicplasty osteochondral autogenous transplantation (MOAT) when referring to transplantation with multiple graft plugs and OAT when referring to transplantation of a single graft plug.

Evidence-Based Studies of Osteochondral Autogenous Transplantation and Mosaicplasty Osteochondral Autogenous Transplantation

Given the success of OAT and mosaicplasty in the knee, there was a logical extension of this technique to the ankle. In reviewing the literature on OAT and MOAT in the ankle before February 2009, 15 studies were identified. All articles were reviewed and assigned a classification (i.e., I through IV) using the Journal of Bone and Joint Surgery levels of evidence for primary research questions.¹⁶ The 15 studies included 1 level II study, 1 level III study, and 13 level IV studies (Table 13-1).

TABLE 13-1 Clinical Outcome Studies of Mosaicplasty Osteochondral Autogenous Transplantation for Talar Osteochondral Lesions

Level IV Evidence

In 1997, Hangody and coworkers ¹⁷ were the first to report on use of MOAT for treatment of OCL of the talus. In a preliminary report of a prospective case series (level IV), MOAT was done on 11 patients with OCLs larger than 10 mm diameter. The outcome results after at least 1 year of follow-up were excellent and promising. This study included second-look arthroscopies in three patients and one cartilage biopsy. Arthroscopy showed good incorporation of the grafts, with articular surface congruency and normal cartilage appearance. Biopsy showed normal articular hyaline cartilage (i.e., collagen type II) and proteoglycans and fibrocartilage bonding at the graft-recipient junction. Donor-site morbidity was a concern, because 3 of 11 patients had occasional knee pain and 2 of 11 had knee pain with activities (one with a 10-degree extension lag).

In 2001,¹⁸ 2003,¹⁹ and 2008,²⁰ Hangody and coworkers expanded their prospective case series report (level IV) after 2 to 9 years of follow-up after MOAT for OCL of the talus. The report included 98 patients with 92% to 94% good or excellent clinical results and less than 3% had knee pain after 1 year. Further assessment included 28 MRI scans, 19 CT scans, 8 second-look arthroscopies, and 4 biopsies. CT, MRI, and arthroscopy showed good incorporation of the grafts, with articular surface congruency and normal cartilage appearance. Biopsies showed normal articular cartilage collagen (type II) and proteoglycans. Three postoperative knee arthroscopies showed a donor-site defect covered by fibrocartilage congruent with the surrounding hyaline cartilage.

In smaller case series (level IV), Gauthier and colleagues ²¹ and Baltzer and coworkers²² described 11 (retrospective) and 43 (prospective) cases, respectively, that had MOAT for OCL of the talus. More than 90% of their patients were satisfied with the results. In another retrospective case series (level IV), Al-Shaihk and associates²³ performed MOAT on 19 patients with larger OCLs (mean area of 120 mm²). Eighty-three percent of patients were satisfied with the results, and two patients (10.5%) reported mild knee pain. They found no difference in outcomes between patients with lesions larger than 1 cm² and those with lesions smaller than 1 cm.²

In 2001 ²⁴ and 2006,²⁵ Scranton and colleagues confirmed earlier promising results with a retrospective case series (level IV) that evaluated 50 patients. Their technique included the transplantation of only one cylindrical plug in patients having cystic lesions that were 4 to 20 mm of diameter. The results were equally impressive, and the investigators advocated arthroscopic harvesting and the use of larger and fewer plugs to minimize peripheral graft chondrocytes death and to simplify the technique. In a retrospective case study (level IV), Assenmacher and coworkers ²⁶ reported the use of single-plug OAT in 8 of 9 patients, who had an average decrease in the visual analog scale (VAS) score of 7.7 to 3.1 after the procedure. They also described the use of an anterior tibial wedge osteotomy (i.e., plafondplasty) in five patients for lesions in the central third of the talar dome to avoid a malleolar osteotomy. Preoperative MRI showed an excellent correlation for OCL grading, and postoperative MRI showed integration of the grafts.

In a prospective case series (level IV) reported by Lee and associates,²⁷ 18 patients had 100% good or excellent outcomes. Two of the 18 patients (11%) had mild knee pain, and 14 (87.5%) of 16 second-look arthroscopies showed cartilage continuity with the native cartilage.

To minimize the risk of knee pain associated with violating an asymptomatic joint to harvest the OAG, Sammarco and colleagues ²⁸ performed a local osteochondral graft for an OCL with a mean area of 62 mm,² and they typically used a one-plug graft. They harvested the grafts from the medial or lateral talar facet on the same side as the lesion they were treating. In the prospective case series (level IV), the investigators reported that all 12 patients were satisfied with the results and that they had no talar collapse.

In a prospective case series (level IV), Kreuz and coworkers ²⁹ reported 91% good or excellent results using local OAT grafts harvested from the medial or lateral talar facet. They subdivided their series into four subgroups: no osteotomy, medial malleolar osteotomy, tibial wedge plafond osteotomy, and posterolateral approach. They found better outcomes in the two groups not undergoing osteotomies compared with the others and found better outcomes for the tibial wedge osteotomy group compared with the medial malleolar osteotomy group.

Level III Evidence

In a small, retrospective cohort study (level III) of 14 patients, Haasper and coworkers ³⁰ did not find differences in outcomes between previously treated and previously untreated OCL patients receiving MOAT. The lesions were small (68.9 ± 32.2 mm²) and a surprisingly large number of patients reported knee pain at a minimum of 1-year of follow-up.

Level II Evidence

In a randomized, controlled trial (level II), Gobbi and colleagues ³¹ compared chondroplasty, microfracture, and MOAT in 33 ankles. The American Orthopaedic Foot and Ankle Society (AOFAS) Hindfoot Score, the Ankle-Hindfoot Scale, and the Subjective Assessment Numeric Evaluation improved in all groups, and the investigators found no significant difference in outcomes among the three groups. There was an inverse correlation between size of the defect and outcome in the microfracture and OAT groups. To have comparable techniques, the central and posterior lesions that needed a malleolar osteotomy were excluded from the study.

Technique Advantages and Disadvantages

The advantages of using MOAT over other techniques to treat OCL are a one-step procedure with relatively fast recovery and hyaline cartilage covering more than 80% of the treated lesion, and consistently more than 80% to 90% good clinical results. Multiple, small autografts have the advantage of reduced complications at the donor site, better fit to the shape of the defect, and avoidance of screws or wires for fixation with the press-fit implantation technique.³²

The disadvantages of MOAT are related to donor-site morbidity, fibrocartilage filling the dead space between OAT grafts, and the technical challenge of obtaining optimal success.

Indications and Contraindications

Débridement with subchondral bone penetration produced consistently good or excellent outcome results.³¹ In a prospective case series that used the microfracture technique, Chuckpaiwong and associates³³ had 100% good or excellent results in patients with lesions less than 15 mm in diameter but only 3% good or excellent results in patients with lesion larger than 15 mm in diameter. The only randomized, controlled trial (level II) comparing MOAT with a form of subchondral bone penetration technique showed no difference in outcomes.³¹ The only level III study of MOAT showed no difference between patients with OCLs previously treated by subchondral bone penetration and untreated patients.³⁰

Given a lack of literature to support superior outcomes for OAT and mosaicplasty for the treatment of OCL and because of the procedure’s inherently more invasive nature and the potential morbidity at the donor site, we feel that OAT and mosaicplasty should be reserved for failure of a prior attempt of arthroscopic débridement associated with any technique of subchondral bone penetration (i.e., curettage, drilling, or microfracture), at least for OCLs of the talus smaller than 15 mm in diameter. However, the literature does not provide direction about the primary treatment of OCLs larger than 15 mm in diameter. There may be a role for primary MOAT technique, especially for medial, deep, and cystic OCLs, which have worse results than lateral, more shallow, and wafer OCLs.²⁵^,³⁴^–³⁶ Even for failed prior débridement and subchondral bone penetration, there is at least one study showing 92% good results with revision using the same procedure by the same surgeon.³⁷

When deciding where to harvest the OAT grafts, the surgeon should consider the fact that patellofemoral peripheries allow up to 3 to 4 cm ² of graft plugs for harvest. That is sufficient for grafting most large, contained talar OCLs treated by MOAT.²⁰ However, for smaller lesions needing only one or two plugs, some surgeons advocate local graft harvest from medial or lateral non–weight-bearing talar facets. This prevents violation of a normal knee joint and avoids chronic knee pain.

Other than obvious contraindications such as infection, tumors, rheumatoid arthritis, diabetes with neuropathy, or sympathetic dystrophy, OAT and mosaicplasty should be use with caution in patients with signs of osteoarthritis other than mild osteophytes that can be removed during the same surgical procedure. Ankle instability and alignment deformity should be addressed before or at the same time to prevent failure due to higher than normal contact stress on the transplanted grafts. Hangody and colleagues ²⁰ advise caution in treating patients older than 50 years because of their decreased repair capacity.

Surgical Technique

The procedure can be done under spinal or general anesthesia. The patient is positioned supine. A thigh tourniquet is used in most cases. Ankle arthroscopy may be carried out first, especially if it is a primary procedure, to confirm the location and grade of the OCL and to assess other cartilage and ankle pathology (Fig. 13-1). If the lesion is small and anterior, an all-arthroscopic technique can be used. Otherwise, an open technique is needed.

FIGURE 13-1 An arthroscopic view of an osteochondral lesion of the talar dome.

For medial OCLs, a medial approach with a medial osteotomy (Fig. 13-2) is usually needed because most of those OCLs are more posterior. The osteotomy may be a step cut or an oblique osteotomy to provide a good perpendicular view of the talar dome. To achieve this view, the talus must be rotated into valgus, and a pin can be inserted into the talar body¹⁹ or a noninvasive ankle distractor can be used.³⁸ Some surgeons advocate an anterior tibial wedge osteotomy for smaller and middle-third lesions instead of a malleolar osteotomy,²⁶^,²⁸^,²⁹ but we cannot confirm this approach because we have no experience with the technique.

FIGURE 13-2 An open approach to the talar dome using a medial malleolar osteotomy.

For lateral OCLs, an anterolateral approach is usually needed. Because these OCLs often have a more anterior location, extreme plantar flexion usually can expose the lesions. If a large, posterior lesion needs more exposure, a lateral malleolar osteotomy ²¹ (Fig. 13-3) or a release of the anterior talofibular ligament and calcaneofibular ligament by a periosteal flap at their origin on the distal fibula may be used.³⁹ A good perpendicular view (see Fig. 13-3) of OCL of the talar surface is mandatory to allow perpendicular insertion of the graft plugs.

FIGURE 13-3 An open approach to the talar dome using a lateral malleolar osteotomy.

After good exposure, débridement and removal of all diseased cartilage and bone debris with a curette or knife is done until a sharply defined, viable cartilage rim and a clean bony bed of the lesion are obtained. Using MOAT-specific instruments (various companies offer complete sets of instruments), the defect is exactly sized, and the number, diameter, depth, and location of the graft plugs and recipient holes are determined (Fig. 13-4). The intended location of the planned mosaic is marked on the recipient bed with the appropriate sizer-tamp or cutting chisels.

FIGURE 13-4 An example of industry-specific instruments used for osteochondral allograft transplantation.

(Courtesy of Smith & Nephew, Andover, MA.)

Larger graft plugs typically are used first to fill most of the defect, and smaller grafts are used to fill the remaining dead space not covered by the primary plugs (Fig. 13-5). Analysis of a cadaveric contact model study showed that four 4-mm-diameter graft plugs better restored joint contact area and pressure to normal levels than two 6-mm-diameter graft plugs in a talar OCL made in the center of the contact zone.⁴⁰ More grafts may also allow for better contouring to fit the original surface anatomy and minimize the unfilled space. However, the use of more grafts in transplantation is more technically demanding and may signify a larger area of nonviable hyaline cartilage because of the rate of possible peripheral graft chondrocytes death of up to 24%.⁴¹ This is why some arthroscopists advocate using fewer larger graft plugs.²⁴^,²⁵

FIGURE 13-5 Intraoperative picture showing the harvest site for osteochondral allograft transplantation.

OAT graft plugs may be harvested from the ipsilateral knee (see Fig. 13-5). This can be done by arthroscopy or by miniarthrotomy, depending on the number and size of grafts needed and the ability of the surgeon. We prefer the open technique. The medial, proximal, non–weight-bearing surface of the medial femoral condylar edge is our preferred donor site. Although Marymont and coworkers⁴² showed with an MRI study that the best donor site for filling a medial talar lesion is the superolateral femur and although we found no literature to support our approach, we think there are more risks of patellofemoral symptoms from lateral harvesting.

For talar defects, Demirci and colleagues ⁴³ showed that the grafts should be taken from condylar edges, where the chondral thickness is decreased. Flexing the knee allows up to four graft plugs to be harvested. For more graft plugs, the lateral femoral condylar edge can be used. The graft plugs are harvested with the appropriately sized, tubular, cutting harvester chisels. It is important to be perpendicular to the articular surface in both planes when harvesting the graft. Each graft length is recorded. Expect up to 0.2 mm of normal expansion of the cartilage plugs after removing them from the harvester.¹⁹

Next, the recipient bed is cleaned again. Sequential drilling, dilation, and delivery are performed for each graft using the appropriate drill guide–delivery cylinder (see Fig. 13-4). Remaining perpendicular to the articular surface is mandatory. Drilling should be 3 to 4 mm deeper than the graft plug length.¹⁹ When delivering the grafts, care should be taken to set them as flush as possible with the surrounding articular surface of the native cartilage (Fig. 13-6). Huang and associates⁴⁴ showed that small incongruities could remodel, provided they did not exceed 1 mm in either direction. Pearce and coworkers⁴⁵ showed that grafts left 2 mm proud and repositioned by weight bearing caused perigraft fissuring, fibroplasia, and subchondral cavitations. They concluded these complications were caused by excessive motion between the graft and recipient site in the proud grafts. However, the graft plugs also must not be sunk too deeply. After all grafts are transplanted, the ankle is reduced and moved through its entire range of motion with slight axial compression. This ensures congruency of the MOAT.

FIGURE 13-6 Intraoperative picture showing the recipient site after osteochondral allograft transplantation.

If an osteotomy was used, it is reduced and fixed with two screws inserted through predrilled holes. If lateral ligament release was used, repair is made as in a classic Broström ligament repair.

Wound closure is made in layers, and a knee drain can be use. A well-padded Jones compression dressing with a below-knee splint is applied, and the foot is kept elevated for 1 or 2 days during the hospital stay. Patients receive standard prophylactic antibiotics and pain management.

PEARLS& PITFALLS

• Lateral OCLs are more likely to be associated with trauma, and medial OCLs are more likely to be spontaneous.

• The routine imaging workup of an OCL consists of plain radiography, CT, and MRI. If CT or MRI is not available, a bone scan is useful to identify an OCL.

• It is not clear what to conclude from the published literature about the primary treatment of OCLs larger than 15 mm in diameter. There may be a role for primary MOAT technique, especially for medial, deep, and cystic OCLs, which have worse results than lateral, more shallow, and wafer OCLs.

• For an OCL of the talus that is less than 15 mm in diameter, OAT and mosaicplasty should be reserved for failure of a prior attempt of arthroscopic débridement associated with any technique of subchondral bone penetration (i.e., curettage, drilling, or microfracture).

• Arthroscopic débridement is possible for lateral OCLs, but medial OCLs may require a medial malleolar osteotomy.

• For a medial OCL, a medial malleolar osteotomy should be predrilled to facilitate open reduction and internal fixation, and it should be performed high enough to obtain access to the OCL.

• For a lateral OCL, an anterolateral approach should be combined with plantar flexion of the ankle to expose the lesion.

• For MOAT, specific equipment is used to harvest the graft. Larger graft plugs are used first to fill most of the defect, and smaller grafts are then used to fill the remaining dead space not covered by the primary plugs.

Postoperative Management and Rehabilitation

Some surgeons advocate immediate ankle range of motion,¹⁸^,¹⁹ and others advocate no motion for 3 weeks.²⁴^,²⁵ We usually start controlled range of motion after stitches are removed at 10 to 14 days if the wound is clean. The patient is kept in a removable cast boot to prevent weight bearing for 6 weeks. We then allow progressive weight bearing over a 4-week period. Formal physiotherapy usually starts at 6 weeks. Unprotected weight bearing is allowed at 10 weeks, and athletic activities should be postponed until 6 months.