ANATOMY

Hyaline cartilage is has no nerve fibers and is avascular, deriving nutrition from its surrounding synovial fluid. Additionally, hyaline cartilage is composed mostly of matrix, with a relative paucity of live, active cellular material. Because of this, full-thickness articular cartilage defects have limited healing potential. When native healing of such defects occurs, fibrocartilage is formed in the defect. This is the goal of marrow-stimulating techniques, in which osseous bleeding at the defect initiates a cascade of fibrous scar healing. However, fibrocartilage is not as mechanically sound as hyaline cartilage under compressive or shear loads.

In adults, hyaline cartilage normally has a thickness of roughly 8 to 9 mm. As a person ages, the articular cartilage thins. The presence of pathology can accelerate this eroding process. To produce symptoms, articular cartilage must be thinned by more than 75% before subchondral nerve fibers can detect the change in contact surface pressure. However, overall size of the lesion also plays a role in this. A small chondral defect will not produce symptoms because the stress is distributed to the surrounding joint surface. Only when the cartilage defect becomes large enough, and thin enough will the patient become symptomatic.

To illustrate this point, Jason and Koh⁵ have examined the contact pressures across an articular surface with respect to a focal defect and a grafted defect. Various graft height mismatches were modeled (Fig. 15-1), with the results shown in Table 15-1. Not surprisingly, plugs that were flush to their articular surfaces most closely resembled the contact pressure of the normal intact articular surface. However, plugs countersunk by 0.5 to 1.0 mm also closely mimicked the native pressures. Thus, filling a defect to a near-congruent articular surface can reproduce native articular-surfaced pressures. This is the foundation of autogenous osteochondral transplantation.

FIGURE 15-1 The effect of graft height mismatch on contact pressures.

TABLE 15-1 Effect of Graft Height Mismatch on Contact Pressures

Autogenous osteochondral transplantation, or mosaicplasty, is a technique in which multiple smaller grafts are harvested from a less weight-bearing portion of the knee and transplanted to the clinically relevant defect. The use of several small grafts has several advantages: (1) it allows the surgeon to reconstruct the natural anatomic contour lost by the defect; (2) it provides some hyaline surface structure to the reconstruction; (3) it limits fibrous cartilage growth to the gaps between the grafts and defect border; (4) it reduces donor site morbidity; (5) it is relatively inexpensive; and (6) the entire reconstruction can be done in a single surgery.

Because small chondral defects will not produce symptoms, the goal of osteochondral transplantation is to decrease the size of large defects functionally to become less clinically significant.

PATIENT EVALUATION

TREATMENT

Treatment should be tailored to various presenting factors. These include location and size of the lesion, age and weight of the patient, occupation, and type of sports involvement.

Arthroscopic Technique

Setup and Instrumentation

A spinal needle is used to determine portal placement to approximate an almost perpendicular approach to harvest and defect site. Of specific note, the senior author (JYCC) believes this to be a forgiving technique. A 10- to 20-degree arc off perpendicular can be tolerated well without clinical consequences. Excess prominences may be later trimmed with a bassinet. The number of bone grafts necessary for the grafting can be determined by using the harvester to score the site of the defect lightly.

Define the recipient sites, starting along the margin of the defect, to ensure a flush repair of the cartilage surface. Graft size should be limited to 6-mm-diameter grafts that are 8 mm deep. Smaller grafts are too fragile, whereas larger grafts may be too difficult to transplant arthroscopically and may increase donor site morbidity. A series of grafts can be arranged to optimize resurfacing of the defect. The original contour of the condylar surface can be easily re-created simply with this cluster of 6-mm grafts. If grafts longer than 8 mm are required to fill the defect, osteochondral transplantation is not indicated because of the lack of adequate supporting bone.

Preparation of the Defect Site

Débride loose fragments away from the cancellous bone bed with an oscillating shaver while avoiding generalized bone bleeding. Keeping in mind the number of grafts to be transplanted, position the Innovasive COR drill near a margin of the defect, as initially scored. Under arthroscopic vision, drill almost perpendicularly until resistance is felt. Observe that the 8-mm drill stop is resting on the stable bone surrounding the repair site.

All holes may be drilled at once or osteochondral grafts may be implanted after each hole is drilled, maintaining a tight press fit. Care must be taken to ensure that a 1- to 2-mm bone bridge exists between recipient sites.

Harvesting COR Grafts

Insert the harvester (T-handled instrument) into the disposable cutter, screwing them together (Fig. 15-2). Advance the plunger through the lumen of the harvester, where it acts like an obturator to minimize soft tissue capture when passing the instruments into the joint. Once the assembly is properly positioned within the joint, replace the plunger with the anvil to minimize loss of fluid while harvesting grafts.

FIGURE 15-2 COR graft harvester tool.

Position the cutter-harvester assembly on the non–weight-bearing donor site selected to provide the graft (Fig. 15-3). Ensure that the end of the cutter is almost perpendicular to the surface prior to taking the donor graft. In the knee, the superior and lateral aspects of the intercondylar notch may provide easiest arthroscopic access.

FIGURE 15-3 The graft is harvested from the non–weight-bearing ACL notch site.

With a mallet, tap the anvil portion of this assembly into the bone 8 mm until it is fully seated at the cutter’s depth stop. Rotate the T handle of the harvester clockwise a minimum of two complete revolutions. The cutter tooth undercuts the distal bone, scoring it to ensure precise harvesting of a donor plug. Supportive pressure must be maintained on the T handle during rotation to ensure control of plug depth. Remove the cartilage graft by gently twisting the T handle while withdrawing it from the joint.

Unscrew the harvester from the cutter. The bone graft will remain protected within the harvester tube until it is ready for transplantation into the defect site. If additional bone grafts are required to repair the defect, the cutter can be assembled onto another harvester and the process repeated until the appropriate number of grafts has been taken.

Cartilage Graft Implantation

Screw the delivery guide onto the harvester, which contains the first graft to be delivered. Place the tip of the plunger through the spacer ring into the proximal end of the harvester. Gently tap the plunger, advancing the bone graft beyond the tip of the deliver guide, as limited by the spacer ring. This 1- to 2-mm lead of bone assists in aligning the graft with the drill hole as it is inserted into the recipient site.

Position the graft over the drill hole. Withdraw the plunger from the harvester, removing the spacer ring. Gently reinsert the plunger, carefully pressing the plug to fit in the undersized drill hole, flush with the surrounding bone (Fig. 15-4). Excess prominences may be trimmed with a basinet. Avoid impacting the graft to round off the corners, because this will mushroom the graft and damage the graft’s integrity.

FIGURE 15-4 Defect filled with three osteochondral grafts.

If the defect is deep or the decision is made to place a solitary graft (or grafts) away from the margin of the defect, the delivery guide may be partially unscrewed to reduce the insertion depth of the graft. This allows for a supportive press fit between the sides of the plug and drill hole while maintaining an elevated face on the cartilage graft.

PEARLS& PITFALLS

PEARLS

Recipient Site

1. Trim loose cartilage to a stable base at the periphery of the defect using the graft cutter tool.

2. Plan for the number and position of the grafts to be transplanted by using the harvester to score the defect.

3. There should be 1- to 2-mm osteochondral bridges remaining between recipient holes. To preserve the integrity of these bridges, holes should be drilled in an alternating fashion to sites of insertion of the donor plugs.

4. Prepare the recipient site prior to obtaining donor osteochondral grafts. This can allow immediate transfer of the donor graft to the defect. Donor graft size can also be adjusted if difficulty is encountered during the recipient site preparation.

5. Take note of the angle of flexion of the knee when drilling the recipient site graft holes. By using the same angle of knee flexion when inserting the grafts, it is easier to match the hole-graft trajectory.

6. The clear transfer tube is provided to allow complete visualization so that the graft can be inserted completely initially. Attempting to reseat the graft after a partial initial insertion without the supporting tube can increase the chances of mushrooming and graft damage.

7. When flexion of the knee is required to access a defect of the distal femoral condyle, the capsule and soft tissue will press against the defect under tension; this makes visualization with an arthroscope almost impossible. To alleviate this, an accessory portal can be used to introduce a lever arm tool (e.g., a joker) to retract the tissue away from the defect. An alternate method is to introduce Merseline tape (Ethicon; Somerville, NJ) around the tissues arthroscopically. Pulling directly on these tape loops creates a soft tissue intra-articular “tent” to aid in visualization.

Donor Site

1. The cutter tool must be turned 720 degrees (two full turns) to ensure complete shaping of the graft. Simple twisting of the cutter will free the graft from its donor site. Rocking is not necessary in the COR technique, and may damage the surrounding bone in the donor site.

PITFALLS

1. It is possible to have a graft-hole diameter mismatch when using holes and grafts of different diameters. Use of a single diameter for all grafts and holes can streamline the procedure and eliminate the possibility of this error.

2. To avoid damage to the adjacent subchondral bone during harvest, a maximum depth of 8 mm should be used during graft harvesting.

3. It is difficult to match the articular contour if the graft or recipient hole is made at an oblique angle. This may have an adverse effect on surrounding stresses at the defect, durability of the graft, and clinical outcomes. Both grafts and recipient holes must be harvested and inserted at a 90-degree angle to the articular surface.

4. If too much of the graft is presented outside the cannula prior to insertion, the graft will not be adequately supported and insertion forces may ruin the graft. To avoid this, we recommend leaving the graft only 1 mm proud from the cannula, thus aiding in graft alignment while still protecting the integrity of the construct during press fit.

5. If toggling of the drill occurs while creating the recipient hole, an oversized hole can be made. This situation can be salvaged by redrilling the hole with a larger drill and using the next size larger donor plug (e.g., step up from 6 to 8 mm).

OUTCOMES

Patient selection is paramount to success of this technique. Because of the relatively narrow indications listed above, the senior author’s experience (JCYC) includes only 40 cases from 1996 to May 2009. This includes 23 women and 17 men, ranging in age from 20 to 75 years (mean, 50.5 years). The defects ranged from 1.5 to 3 cm in diameter, with the exception of one case.

Patient follow-up results have been monitored. In 1999, 3-year results showed 86% with good to excellent results. In 2001, the 5-year results showed 76% with good to excellent results. In 2008, at 13 years, 74% of the patients still had good to excellent results. Repeat x-rays and MRI scans were obtained for asymptomatic patients who agreed to participate in this study (Fig. 15-5). In only one MRI was there a persistent bone graft margin, indicating the possibility of delayed incorporation, at 4 years and 7 months postoperatively. The patient was asymptomatic with a congruent articular surface (Fig. 15-6). The significance of this finding is uncertain, but it may not be related to clinical sequelae.

FIGURE 15-5 A, Sample radiographs at preoperative evaluation, 2 months postoperatively, and 5 years postoperatively. Maintenance of joint line congruity can be seen in these standard weight-bearing AP views of the knee in identical positions. B, MRI scan of left knee 9 years after medial femoral condyle osteochondral transplantation. Joint congruity and chondral surface remain intact.

FIGURE 15-6 Persistent bone graft margin (black arrow) with congruency at the articular surface (white arrow).

One patient had a 4-cm defect (outside of our recommended defect size), necessitating the transplantation of four graft plugs (Fig. 15-7). In this particular case, the patient was young and wished to undergo osteochondral transplantation as a final effort to temporize the necessity for a knee replacement. The patient was well-informed and accepted a potentially high risk of failure prior to surgery. Ultimately, this patient returned pain-free to playing baseball with his children (see Fig. 15-7C).

FIGURE 15-7 A, B, Large 4-cm osteochondral defect requiring four grafts. C, Clinical presentation 4 weeks postoperatively. C1, full extension; C2, full flexion; C3, full squat, frontal view; C4, full squat, side view.

Second-look arthroscopy was performed in one case. This patient had a medial femoral osteochondral transplantation graft in August 1997. Two years later, the patient returned with persistent pain and repeat arthroscopy was performed (Fig. 15-8). On second look, the original graft site showed excellent gross congruity, incorporation, and no signs of failure. A new, full-thickness defect was present in the tibia of the opposite compartment. This patient ultimately had a total knee arthroplasty in 1999.

FIGURE 15-8 Second-look arthroscopy, 2 years post–osteochondral transplantation. A, Full congruent incorporation of the graft on the medial femoral condyle. A new lateral tibial defect is present. B, Defect on lateral tibial joint surface not seen during placement of original graft. C, Loose cartilage fragment from lateral femoral condyle. D, View of loose fragment from the defect of the lateral tibial joint surface with the scope placed anteromedially.

Although difficult to prove, it is possible that osteochondral transplantation may slow the development of arthritis in the absence of mechanical or systemic pathology. Anecdotally, an 85-year -old female patient presented for left total knee arthroplasty. She had received an osteochondral transplant in her right knee 10 years earlier. The comparison x-ray of both knees shown in Figure 15-9 reveals a remarkable difference.

FIGURE 15-9 Bilateral view 10 years post–osteochondral transplantation. The joint line is maintained in the operative knee and terminal arthritis is present in the contralateral knee.

CONCLUSIONS

Osteochondral defects treated by marrow stimulation techniques have resulted in inconsistent long-term results. This has perpetuated the interest in finding alternative methods of treating full-thickness chondral lesions. These include autologous chondrocyte implantation, autologous osteochondral transplantation, fresh osteochondral allografting and bulk allograft, and a number of synthetics, scaffolds and first-, second-, and third-generation cell-based technologies.

The goal of osteochondral transplantation is twofold: (1) to remove the pathology of the osteochondral defect’s cartilage, subchondral bone, and interface; and (2) to replace the defect new graft to decrease the defects surrounding forces. Although there is no nerve fiber or blood supply in the bone plug graft, there is an assumption that the existing joint fluid will allow the cartilage cells to survive. This is further supported by the fact that the plug is harvested and inserted within a short period of time. The bone plug fills in the defect and allows the live hyaline cartilage cells to help fill the graft. If the graft is placed correctly, it will decrease the size of the lesion and direct pressure to the contact surface of the subchondral bone, thus relieving the symptoms. Also, because the bone plug lacks nerve fibers and blood supply, there is no sensitivity at the graft-harvested site.

Full-thickness defects smaller than 1 cm in diameter are not thought to be appropriate for treatment with this technique. They do not seem to be clinically troublesome and do well when left alone. It follows that a smaller defect—only 6 mm in diameter (COR technique donor site)—in an area that is not load bearing and does not affect the patellofemoral articulation should also be well tolerated. These defects fill in without additional bone grafting and are covered with fibrocartilaginous scar, much like the area between donor plugs and the adjacent intact hyaline cartilage.

Overall, experience indicates that the primary goal of osteochondral transplantation is not to reproduce a normal anatomic, physiologic, or biomechanical knee joint surface. The goal of this procedure is simply to decrease the patient’s clinical sequelae regarding the defect being treated. This may be related to decreasing the effective size of the defect, and thus decreasing the biomechanical shear and compressive forces surrounding these defects. This is illustrated by Jason and Koh’s research on force transmissions surrounding peridefect- and periplug-related contact pressures (see Fig. 15-1 and Table 1).⁵ Because of this phenomenon, this is a forgiving technique and the angle of graft insertion can be 10 to 20 degrees off of perpendicular without clinical consequences. The senior author speculates that the remodeling potential of the narrow fibrocartilage in the inter-graft space may make this possible. This is supported with our second-look arthroscopy findings (Fig. 15-8), and the good to excellent clinical results may be highly related to patient selection.

Arthroscopic osteochondral transplantation is an acceptable option for treating full-thickness osteochondral defects in appropriately selected patients. The arthroscopic technique is straightforward, with a low incidence of complications. The intergraft areas fill in, creating a homogenous appearance. The ability to manage these conditions arthroscopically at the time of discovery offers the advantage of convenience, a single procedure, and cost savings.