In Brief: OCD lesions are uncommon and poorly understood joint surface defects that are found most frequently in the knees of children and young adults. While some OCD lesions heal on their own, many require surgical treatment. Unhealed or unsuccessfully treated OCD defects compromise the "ball-bearing" function of the knee's gliding surfaces (which are normally quite smooth and almost frictionless), leading to premature joint arthritis. The goal of surgical management is to preserve or restore a normally contoured, smooth, firm joint surface that will function properly in load-bearing throughout life. This goal is by no means easy to achieve.
Osteochondritis dissecans is an unusual affliction of human joints that is not rare but also cannot be considered common. The knee is the most frequently affected joint in the body. Curiously, one particular location on the medial femoral condyle (inner aspect of the lower end of the thigh-bone) is where the majority of knee OCD lesions are found (see FIGURE 1).
The next most common site is the posterior aspect (rear portion) of the lateral (outer) femoral condyle (see FIGURE 1), followed by rarer forms in the patella (kneecap) and upper tibia ("shinbone"). OCD begins in childhood and is therefore most commonly seen in teenagers and young adults. Severe (large) OCD lesions that remain unhealed can ultimately wreak havoc on a knee joint, with long-term arthritic consequences that may require joint replacement surgery.
Osteochondritis dissecans is a truly mysterious joint disease process
that was studied by 19th century pathologists and given its name because
it was considered an inflammatory (the "itis"
in osteochondritis refers to inflammation) condition of articular (joint
surface) cartilage and the underlying (subchondral) bone. (Note: "osteo"
refers to bone and "chondro" refers to cartilage).
The disease causes a section of joint surface cartilage and the bone
beneath it to loosen and separate (by way of a gradual dissection process,
hence the "dissecans" in OCD's name) from the main or "parent"
bone structure such as a femoral condyle (see
The actual cause of OCD is unknown. There are three theories regarding the origin of OCD, the first being that it is the result of traumatic impact injuries delivered to the adolescent joint surface, either acutely (suddenly) or chronically (repetitively) over time. This view considers OCD to represent a fracture of sorts. The second theory is that the separating osteochondral (bone and cartilage) fragment starts out as a small, anomalous (aberrant or extra), independent zone of ossification (bone formation) during early adolescent skeletal growth that simply fails to fuse (merge) with the main ossification center as the bone matures. This ultimately leaves the bone tissue in that extra ossification center (the "ossific nucleus" or subchondral ossicle of OCD) isolated and without a blood supply, depriving it of oxygen and nutrients. The bone ossicle may then shrink and atrophy, thereby undermining the overlying joint surface and making the involved segment of articular cartilage (with attached ossicle) subject to gradual loosening and separation from the parent bone. The third theory is that normal bone underlying a region of articular joint cartilage somehow loses its circulation suddenly and therefore dies, similar to the way a localized area of heart muscle dies when a blood clot cuts off its circulation in the case of a myocardial infarct (heart attack). While my personal experience with many cases of OCD over the years has led me to believe that the anomalous ossification center theory is the most likely of the three to be correct, no one knows for sure.
Natural History and Diagnosis of OCD
OCD lesions may or may not cause symptoms early on. They apparently first appear in the early part of adolescence, well before skeletal maturity is reached and the child's "growth plates" fuse (close). Prior to fusion of the growth plates, if the lesion remains undisturbed (i.e., does not loosen or separate), the chance for spontaneous healing by way of natural reunification of the abnormal subchondal ossicle with the parent bone is reasonably high. In patients between 10 and 13 years of age, simple rest and avoidance of stress to the knee joint over a period of time may be enough to allow such natural healing to take place. As skeletal maturity approaches, the chances for spontaneous healing diminish, and the probability that surgical treatment of some type will be needed increases
Some undetected cases of OCD remain unhealed and clinically silent (cause no symptoms) until a patient reaches their twenties and sometimes even their thirties or later. Some patients can harbor a fairly large osteochondritis dissecans lesion in their knee without realizing it until it frankly separates and literally pops loose inside their joint! Separation may occur either gradually, or all at once during a mildly stressful event such as a routine pivoting or turning maneuver. In a teenager or young adult, the sudden onset of knee locking or internal "clunking" followed by joint swelling (without having experienced the degree of knee stress usually required to tear healthy ligament or cartilage tissues) should suggest to an examining physician the possibility of a recently loosened or separated osteochondritis dissecans lesion. OCD is notoriously misdiagnosed and underdiagnosed, particularly in adolescents whose lesions have not yet separated. These patients usually experience only vague symptoms and may have no knee swelling whatsoever. Routine x-rays ordered by pediatricians may not include the special views necessary to see an osteochondritis dissecans lesion clearly, leaving it to go untreated and thus possibly loosen and separate later on. Unless an MRI scan or the appropriate x-rays are ordered, it usually takes an orthopedic surgeon to make the diagnosis. A scene that has recurred many times in my office is a 15 or 16-year-old patient (who had voiced vague complaints to skeptical parents over several years) saying, "See, there really was something wrong with my knee!" By then, however, surgery is almost always needed.
The potential adverse consequences of osteochondritis dissecans depend very much upon the size of the lesion, its location and how amenable it is to "restorative" treatment. Restorative treatment leads to healing of the lesion, with preservation of the weight-bearing joint surface. Small lesions in non-critical regions of the tibial joint surface and patella often have a fairly benign clinical course, even when the only treatment possible is removal of the loose osteochondral fragment from the joint, which leaves a defect in the articular surface. Unsuccessfully restored femoral condylar lesions are more likely to cause arthritis over time.
Large, advanced lesions of the femoral condyles that are fragmented or otherwise not amenable to restorative treatment leave substantial, crater-like defects in the main weight-bearing surface of the knee joint, with almost uniformly poor long-term results. Such a knee behaves like a ball-bearing with a large divot in its surface, essentially wearing itself out prematurely. Some varieties of femoral osteochondritis dissecans involve a fairly extensive zone of abnormal, underlying subchondral bone. These leave a deep crater if the lesion loosens and separates. Other varieties of OCD extend only superficially into the underlying subchondral bone. In all cases that do not heal spontaneously, the patient's body forms a zone of fibrous tissue between the parent bone and the OCD fragment. This fibrous interface layer is relatively inert (inactive) tissue. The few living cells within this fibrous tissue try to absorb the dead bone in the overlying OCD fragment, which thickens the fibrous tissue layer as the adjacent bone dissolves. Because fibrous tissue is soft, this makes for a loss of structural support beneath the OCD lesion (see FIGURE 3).
This lack of underlying support allows the cartilaginous joint surface to flex repeatedly under weight-bearing stresses, much like floorboards with weak support framing beneath them. Such repetitive, abnormal micromotion causes fatigue breakdown of the articular (surface) cartilage along the lesion's perimeter (the border interface between the OCD fragment and the surrounding, structurally stable femoral surface), culminating in an open crack or fissure demarcating some or all of the lesion. As noted previously, this is the "dissection" process that gave OCD the last part of its name.
If the ossific (bone tissue) nucleus of the OCD lesion shrinks substantially in size, the entire lesion may weaken and break apart (see FIGURE 4), thus precluding any attempt at direct surgical repair and reunification (healing by way of incorporation back into the parent bone).
Surgical Treatment of Femoral OCD
In formulating a treatment plan for a patient with an OCD lesion, the knee specialist's approach is first directed at establishing whether the lesion is mechanically stable or unstable (i.e., still firm and well-fixed in place vs. already loose or frankly mobile). Sometimes, obvious clinical symptoms or the initial x-ray findings (such as a loose piece of bone in the joint) are a "give-away" in this regard. In less obvious cases, the combination of x-rays, a bone scan, and an MRI scan can predict the likelihood of stability versus instability with perhaps a 75-80% accuracy. (For further information regarding the value of such diagnostic screening tests in cases of osteochondritis dissecans, reprints of my published research in this area are available by writing or e-mailing the Knee and Shoulder Center). Taking into account the patient's age and clinical symptoms, the surgeon will interpret the results of these non-invasive diagnostic tests and determine if an attempt at non-operative treatment is reasonable, vs. recommending operative treatment.
The first step in the operative treatment of unstable lesions or stable
lesions that are deemed unlikely to heal on their own is direct arthroscopic
inspection. Even though the lesion may be obvious on an MRI scan or
x-ray, finding it within the joint can be tricky if it has not yet become
mobile. The overlying articular surface cartilage may appear entirely
normal, with perhaps the lone exception of an extremely subtle contour
irregularity at the perimeter of the lesion. This can be very difficult
to appreciate visually with an arthroscope (see
FIGURE 5). Probing the joint surface with firm
pressure administered by a blunt instrument may help the surgeon locate
the lesion by detecting micro-motion of the underlying bone fragment.
One way of treating stable and relatively stable lesions is by a drilling a number of small holes through the involved articular surface, proceeding down through the OCD ossicle and fibrous tissue interface into the normal bone of the parent femur. This is done in the hope of creating internal bone bleeding and an inflammatory healing reaction that ultimately fixes the underlying bone ossicle into its crater (the two bone surfaces will hopefully "knit" together, as would a fracture). This approach has the disadvantage of potentially compromising the otherwise healthy, articular cartilage cap over the OCD ossicle by way of the multiple, drill-hole punctures. Another, more technically difficult approach is to perform "retrograde" (coming in from behind) "bone grafting" of the lesion. Bone grafting refers to a process where healthy bone tissue is transplanted from an area where it is expendable, to an area where bone is deficient and/or unhealed. The OCD lesion is approached by drilling a small tunnel through the femur, starting at a distant location and approaching the lesion from behind. The drilling must stop just as the tunnel passes through the lesion's fibrous interface layer, reaching the necrotic (dead) OCD nucleus, or else the drill may go through the ossicle and exit out through the articular surface cartilage, thus damaging it. This procedure requires fluoroscopic (x-ray) guidance during drilling (see FIGURE 6) and direct visual confirmation that the fibrous interface has been reached and breached by the drill bit. This is accomplished by inserting the arthroscope directly down into the bone tunnel to inspect its depths.
A small amount of local femoral bone graft is then harvested from
an area near the tunnel entrance and pushed down to the bottom of the
tunnel. It is packed into the area where the fibrous tissue interface
layer previously existed. This creates a bone tissue bridge between
the ossific nucleus of the lesion and the parent femur, hopefully allowing
rigid, local bone-to-bone healing that will spread and ultimately result
in reunification of the entire ossific nucleus to the parent femur without
ever having disturbed the overlying articular cartilage cap.
For large lesions that are not completely stable but have not yet developed overt perimeter fissuring of their articular surface cartilage, the retrograde bone grafting technique can be supplemented by internal fixation and stabilization of the fragment in situ (right where it lies) by way of either removable, stainless steel compression screws (see Figure 7) or bioabsorbable (slowly dissolving) pins or compression screws. This requires that one or two, very small holes be drilled through the articular cartilage cap of the lesion for screw placement, but this is still probably less traumatic to the joint than drilling many holes through the cartilage cap, as is performed in a typical anterograde (frontal approach) drilling procedure. The screws hold the fragment rigidly in place to foster osseous union (healing or "knitting") at the bone graft site. Non-absorbable screws are typically removed after eight to ten weeks by way of the arthroscope, and at that point the lesion can be probed and checked for stability. If micromotion of the fragment is no longer present, then early healing can be assumed to be taking place and the patient can very gradually be allowed to resume weight-bearing activities. During the time the screws are in place (assuming proper screw position and grip), the patient can swim and bicycle without harm to their knee, but they should not bear weight on that leg. With absorbable screws a follow-up arthroscopy may not be necessary, but this eliminates the chance for the surgeon to check the early stability and healing of the lesion directly. Because it takes a long time (six to twelve months) for healing to be clearly recognizable by way of x-ray or an MRI scan, not having the opportunity to recheck the patient's knee arthroscopically leaves the surgeon with less information to go on when instructing the patient with respect to their return to weight-bearing activities.
In cases where the OCD lesion is clearly unstable and the overlying articular cartilage cap has begun to separate by way of circumferential perimeter fissuring, a somewhat simpler (but still difficult to perform) bone grafting technique can be employed wherein the bone graft tissue is introduced arthroscopically into the OCD defect from the front, as opposed to an indirect approach from behind the lesion. Such lesions are often found incompletely separated, hanging on by one area of articular cartilage or fibrous tissue near the intercondylar notch (the central, open concavity in the end of the femur). This acts like a hinge, allowing the lesion to be flipped open as would a trapdoor, thus exposing the base of the crater in the parent femur and the undersurface of the un-united bone ossicle or cartilage cap (see FIGURE 8).
The surgeon will carefully hold the lesion open,scrape out any fibrous tissue present (see FIGURE 9) and then insert fresh local bone graft material into the crater (see FIGURE 10). This bone graft material must be very carefully inserted so as to fill up (but not overfill) the empty space created by the removal of the fibrous tissue. The OCD cap must then be closed back down and held in place with internal fixation screws or pins before any of the bone graft material escapes (see FIGURE 11). The greater the degree of prior reabsorption of the dead ossific nucleus, the more fibrous tissue there is to remove and the greater the amount of bone graft material required to fill the void. One must be very careful not to overfill the crater with bone graft or the cap will not fully close back down, causing it to bulge out beyond the normal joint surface contour.
In my experience I have come across two unusual OCD lesions (one of which is shown above in FIGURES 8-11) where the entire ossific nucleus had been reabsorbed, leaving only a thick cap of articular cartilage. In both cases, bone grafting and internal fixation still successfully resulted in firm union of the articular cartilage cap to the parent femur and an ultimate clinical success. As of the time of this writing, I have never encountered an OCD lesion that failed to unite when bone grafting methods were employed, with or without internal screw fixation of the lesion, as needed. Once an individual with at least a partially unstable lesion has reached skeletal maturity by way of closure of their growth plates, I believe that non-operative treatment or less involved surgical approaches (such as simple drilling) are less effective than the more complex techniques utilizing arthroscopic bone grafting.
I have also not yet come across a case where bone grafting and screw fixation of a lesion could not be done arthroscopically, as opposed to opening up the knee by way of a large incision known as an arthrotomy. Such arthroscopic procedures can be lengthy and tedious, and somewhat akin to building a model ship in a bottle, but the post-operative morbidity experienced by the patient (pain, swelling, stiffness and scarring) is usually significantly less and the cosmetic results to be obtained are superior to any procedure that involves an arthrotomy (an important factor to many teenage females).
If an OCD lesion comes completely loose, creating a free fragment within the knee joint, the fragment should be carefully checked with an arthroscope to see if it still has the appropriate size and shape to fit back into its crater in the femoral condyle. If so, then bone grafting and internal fixation methods should be employed to try to replace the fragment in its bed and get it to heal. If that is not feasible, then it must be removed from the joint. The base of the empty crater can be abraded somewhat to make it bleed and thereby provide for subsequent in-filling by fibrocartilage (cartilage scar) tissue, but that almost never provides a good weight-bearing surface and ultimately degenerative joint arthritis will ensue. A possible salvage option in this situation is packing bone graft material into such an open crater to a level just slightly below the level of the surrounding surface articular cartilage, hoping that a surface cartilage layer will form over the solidifying bone graft. Another possible salvage option in such a circumstance is to perform an "OATS" (osteo-articular transplant surgery) procedure (see FIGURES 12a-12d).
During such surgery one or more plugs of healthy articular cartilage and underlying subchondral bone are transplanted from less important articular surface areas in the knee (donor zones) to the open crater left by the osteochondritis dissecans lesion, which usually occupies a more critical, weight-bearing zone of the joint surface. Some surgeons are also using one or more small OATS plug transfers as a substitute for the multiple drill-hole or retrograde bone graft procedures mentioned earlier for the treatment of stable, in situ OCD lesions.
It is worth remembering that the goal of each and every treatment for osteochondritis dissecans is to preserve or restore a normally contoured, well supported, weight-bearing articular cartilage layer so that the gliding/load-bearing function of the joint surface is not compromised. It is currently believed that success in this regard will minimize the long-term development of degenerative arthritis.