Introduction
Osteochondritis dissecans (OCD) is a focal, pathological lesion involving subchondral bone and the attached overlying articular cartilage. Importantly, the focal lesion is at subsequent risk for the development of subchondral degeneration, destabilisation of the lesion from the articular surface, formation of loose bodies and accelerated joint degeneration. OCD characteristically affects two distinct populations, juvenile and adult, differentiated by the presence or absence of open physes. The risk for disease progression varies considerably between these two populations, with juvenile patients generally having a more favourable prognosis. Although OCD lesions occur in various joints within the body, they are most commonly found within the knee. Despite similarities, OCD lesions should be differentiated from other pathological articular processes, including chondral lesions, avascular necrosis, osteochondral fractures and spontaneous osteonecrosis. The goals for treatment of OCD lesions are to maintain joint congruity and reduce the risk of cartilage deterioration and subsequent development of symptomatic knee osteoarthritis. Treatment methods are largely dictated by the stability of the OCD lesion and the presence or absence of open physes.
Epidemiology
The overall annual incidence of knee OCD ranges from 6 to 11.5 per 100,000 people. However, it is well established that this incidence also demonstrates age and sex dependence. Knee OCD incidence is highest among both boys and girls aged 11 to 15 years, with the incidence peaking at 39.06 per 100,000 in boys and 16.15 per 100,000 in girls in this age range. Additionally, the overall incidence of juvenile knee OCD is 9.5 per 100,000 compared with 1.2 per 100,000 for adult knee OCD. , Despite this age dependence, boys have an overall 2- to 3.8-fold increased risk of developing knee OCD lesions in their lifetime compared with girls. , , Several large population-based epidemiological studies have described additional risk factors for the development of knee OCD. Notably, there is a higher overall incidence of knee OCD among African Americans compared with other racial or ethnic groups. In fact, African Americans were demonstrated to have double the risk for developing OCD compared with the second highest incidence group, non-Hispanic whites. Patients with knee OCD are also more likely to have a significantly higher body mass index (BMI) compared with the general population, and studies show that patients categorised as moderately obese (BMI for age in the 95th percentile or greater or BMI 30 kg/m 2 or more) are at an increased risk for the development of knee OCD. Importantly, the overall rate of knee OCD appears to be increasing, likely influenced by growing participation and specialisation in youth sporting activities, increased rates of adolescent obesity and increased rates of diagnosis related to the availability and widespread use of magnetic resonance imaging (MRI) and knee arthroscopy in adolescent age groups. , , ,
OCD lesions of the knee are classically designated by two categories: juvenile or adult. Juvenile OCD is typically associated with highly active children and adolescents, whereas adult OCD is typically attributed to incompletely healed, previously asymptomatic juvenile lesions. , Although de novo adult OCD lesions of the knee have been described, they are rare, occurring with an unknown incidence. Multifocal knee OCD, defined as patients with more than one identified lesion on the same or contralateral knee, are often seen at different stages of development, with the majority of patients presenting with bilateral involvement and two to three lesions. In a study by Cooper et al., the incidence of bilateral OCD lesions in newly diagnosed juvenile OCD patients was 29%, and 40% of the identified contralateral lesions were asymptomatic at the time of diagnosis. Significant risk factors identified for multifocal lesions were female sex and younger age at presentation.
Anatomy and Pathogenesis
Classically, knee OCD was described as occurring on the lateral aspect of the medial femoral condyle. Subsequent large population-based studies have demonstrated that knee OCD occurs on the medial femoral condyle in the majority of patients (61% to 64% of patients), followed by the lateral femoral condyle (32% to 33%), trochlea (1% to 3.4%), patella (1.5% to 3%) and tibia (0.5% to 2%). ,
Despite OCD first being described by the German Surgeon Franz Kӧnig in 1887, subsequent research has predominantly been low-level evidence studies with heterogeneous methodology that have failed to establish a clear, agreed-on cause. , Multiple hypotheses point to inflammatory, genetic, endocrine, vascular, ossification or mechanical/traumatic mechanisms as potential causes. In his original assessment, Kӧnig proposed that OCD was the result of a subchondral inflammatory process. However, this was subsequently refuted when histological studies revealed no underlying evidence of inflammation. ,
Familial occurrences of OCD, in conjunction with the common presence of bilateral or multifocal lesions, have led many researchers to consider a genetic component in the development of OCD. Several studies propose familial inheritance patterns, including some forms of autosomal dominant disease. , In addition, certain genes (including ACAN and the genes encoding type IX collagen) have been implicated because alterations in these genes can result in the OCD phenotype. , , Although these findings suggest that there could be a genetic role in the predisposition and development of OCD lesions, many authors debate whether these findings are widely applicable to the most common form of the disease.
Endocrine abnormalities are also associated with OCD, leading some experts to propose that hormonal changes may affect bone metabolism, bone remodelling and ultimately the pathogenesis of OCD. In a study by Bruns et al., 91% of consecutive patients undergoing surgical treatment of OCD lesions had measurably low levels of vitamin D. Additionally, Krause et al. performed a cross-sectional study of OCD patients in which 89% of their patients had vitamin D deficiency, and OCD histological specimens obtained from these patients demonstrated focal accumulations of bone matrix with a lack of mineralisation.
Vascular ischemic changes resulting in disruptions in subchondral blood supply have also played a prominent role in the early etiological theories of OCD. Although embolic phenomena were once thought to contribute to the disease development, this belief has largely been discredited. Rather, fracture or injury to the subchondral bone in areas of tenuous blood supply, leading to poor healing or regenerative processes, is thought to be the underlying pathogenic mechanism. , Tóth et al. supported this theory using a novel MRI technique to highlight the anatomical vascular architecture of human paediatric cadaveric femoral condyles. Areas of relative hypovascularity were found at the characteristic sites of knee OCD lesions. This approach was further validated in anatomically similar animal models.
Aberrancy in epiphyseal endochondral ossification is yet another hypothesis for the pathogenesis of OCD. It was first proposed by Ribbing in 1954 and further supported by Barrie in his histological assessment of OCD loose bodies in 1980. , Notably, this hypothesis has gained additional traction in the literature with the support of advanced imaging and animal studies. The proposed hypothesis suggests that a focal failure of endochondral ossification occurs within the developing cartilage complex. As a result, this region is unable to undergo vascular invasion or subsequently convert to bone. Ultimately this leads to focal cartilage necrosis. Adjacent normal bone then attempts to repair the focal lesion, and a surrounding calcific rim forms with areas of anchoring osseous bridging. At this stage in the pathological process, the lesion stability develops, with the ability to heal or evolve into a symptomatic lesion. Some authors have speculated that these subunits of failed endochondral ossification can serve as predilection sites for the development of OCD lesions, and subsequent exposures to stress or strain may ultimately instigate lesion instability. , ,
The last of the etiological hypotheses is a proposed mechanical or traumatic origin of OCD. Despite abundant theorised mechanisms, the underlying principle is pathology as a result of focal chronic loading and associated repetitive microtrauma. Fairbanks and Smillie popularised the notion that impingement of the medial tibial eminence on the lateral aspect of the medial femoral condyle was the attributable cause. Subsequent biomechanical analyses noted that increased shearing forces of these two bony interfaces were possible with internal rotation of the tibia and loading of the knee in flexion or with increased tibial external torsion. Studies have also noted that size of the tibial eminence, width of the intercondylar notch, morphology of the proximal tibia and alterations in mechanical alignment of the knee may serve as additional risk factors for OCD development. Furthermore, links between meniscal pathological conditions, including meniscal hypermobility or abnormal morphology (e.g., discoid meniscus), and knee OCD lesions are well documented. These links are hypothesised to alter the contact forces of the knee joint and subject areas to focal repetitive microtrauma.
Although the precise cause of knee OCD lesions remains controversial, many of the proposed hypotheses share fundamental pathophysiological principles. This could indicate that their mechanisms act either independently or compound together to increase a patient’s risk for OCD. Regardless of exact etiological mechanism, the resulting pathophysiological process leading to development of OCD is generally agreed on. This process begins with an initiating insult to the subchondral bone by means of ischemia or stress injury. Subsequently, if the damaged subchondral bone does not establish revascularisation, subchondral necrosis may occur. It should be noted that the overlying articular cartilage may remain viable because of the diffusion of nutrients from the synovial fluid. The osteochondral subunit is then at risk for further damage or instability compared with the surrounding subchondral bone. This has the potential to secondarily injure the overlying cartilage as shear stresses are placed over the site of the lesion. Should significant injury and breach of the overlying cartilage occur, the subchondral bone may be exposed to synovial fluid. This is associated with a significant impairment in healing potential and increases the risk for fragmentation or detachment of the OCD lesion.
Natural History
The natural history of untreated knee OCD is not well defined. The available literature is unable to provide insight into the incidence of symptomatic knee osteoarthritis after untreated knee OCD. However, it is generally accepted that juvenile OCD has a significantly better overall prognosis compared with adult OCD. , , The increased healing potential of juvenile OCD lesions is attributed to the robust vascular supply present with open physes and ongoing endochondral ossification. , This epiphyseal vascular supply disappears with age and physeal closure, presumably contributing to the diminished healing ability of adult OCD lesions.
A large number of juvenile OCD lesions can be treated nonoperatively. This is determined predominantly by overall lesion stability. Studies show an overall average healing rate of 61.4% for patients treated nonoperatively. Moreover, in a long-term follow-up of juvenile patients treated nonoperatively for their OCD lesion, the 35-year cumulative incidences of knee osteoarthritis and knee arthroplasty were 30% and 8%, respectively. For juvenile OCD patients who subsequently require operative management, the overall prognosis is promising, with healing rates of approximately 94% under the current standard of care.
In contrast, Linden found that only 10% of adults with OCD treated nonoperatively demonstrated signs of healing on long-term follow-up MRI studies, and these patients developed arthritis 10 years earlier than the normal population. Given these findings, the majority of adult OCD lesions are treated operatively. Modern operative treatments demonstrate good outcomes with overall high rates of healing and satisfaction at mid- to long-term follow-up. ,
Other than age and skeletal maturity, there are several other negative prognostic factors related to knee OCD lesions and potential nonoperative management. Specifically, lesions of larger size and lesions that involve weightbearing surfaces are associated with a worse prognosis. , Krause et al. reported that among juvenile patients the most predictive characteristic of healing was the presence of cystic changes around the OCD lesion; cystic changes greater than 1.3 mm were noted to have significantly reduced rates of healing. Additional negative prognostic indicators include more severe lesion stage (lesion instability), , longer symptom onset before medical consultation, discoid meniscus and swelling or mechanical symptoms. , Patients with higher body mass index (>25 kg/m 2 ) and patellar OCD lesions also have an increased risk of developing arthritis.
Patient History and Physical Findings
Patients with knee OCD lesions commonly present complaining of vague, poorly localised knee pain. The pain is often associated with swelling and mechanical symptoms such as catching, locking or clicking, which could suggest the presence of a loose body. Patients often complain that their pain is exacerbated by activity and may even lead to activity intolerance in previously highly active patients.
On physical examination, patients may have a joint effusion or quadriceps atrophy. Depending on the location of the lesion, there may be focal tenderness over the lesion site. These patients may exhibit decreased range of motion (especially in extension), patellofemoral crepitus, patellar subluxation or apprehension with examination manoeuvres. Patients also commonly walk with an antalgic gait.
After observation of patients walking with antalgic gaits with their ipsilateral foot externally rotated, Wilson hypothesised that external rotation relieved pressure of the tibial spine impinging on the medial femoral condyle in classically located lesions (lateral aspect of the medial femoral condyle). To evaluate this he developed the Wilson sign. A positive Wilson sign is defined as reproduction of a patient’s pain while the patient is placed supine with the knee flexed at 90 degrees, the tibia is internally rotated, and the knee is slowly extended. Unfortunately, although this diagnostic sign is emphasised in the early literature, its diagnostic accuracy is only 25% and is theoretically limited to patients with classically located lesions. This examination manoeuvre may have utility in patients with a positive sign who later convert to a negative sign, supporting possible lesion resolution. ,
Imaging and Diagnostic Studies
Imaging is essential in the diagnosis, management, and monitoring of OCD lesions. Radiographs are the mainstay in initial assessment of patients with knee symptoms and suspected OCD ( Fig. 23.1 ). Appropriate radiographic imaging includes standard anteroposterior (AP), lateral, tunnel (or notch view) and Merchant (or sunrise view). The tunnel view allows for better visualisation of lesions involving the posterolateral surface of the medial femoral condyle and is obtained via a posteroanterior (PA) beam direction with the patient weightbearing with 30 to 40 degrees of knee flexion. The Merchant view depicts the patellofemoral articular surface and is obtained with the patient supine and the knee flexed 30 degrees. The radiographic appearance of juvenile OCD lesions begins with subtle flattening and radiolucency of the cortical surface, which may then evolve into the characteristic well-circumscribed defects involving the subchondral bone. Care should be made to not confuse such findings with variations in developmental ossification common to younger patients. , Adult OCD lesions may appear more osteosclerotic and could have other additional signs of advanced disease such as fragmentation or loose bodies.
If OCD is suspected radiographically, MRI studies should subsequently be obtained for confirmation of the diagnosis and for better characterisation to assist with management decision making ( Fig. 23.2 ). MRI provides substantial value in demonstrating signs of bony oedema, subchondral separation, presence of damage to the overlying articular cartilage and accurate size and location of the lesion. Multiple MRI criteria have been proposed to distinguish lesion stability and staging; however, the practical reliability of current criteria remains debated. , Furthermore, MRI can be used to assess healing or disease progression in patients being treated conservatively as well as being used postoperatively.
Less common imaging modalities include computed tomography (CT) and scintigraphic examination (nuclear bone scan). Because of the utility of MRI studies and the exposure of radiation and radioactive tracers, these imaging modalities are rarely used in OCD. CT scanning is most useful for evaluating osseous integration such as after fixation of OCD fragments, and bone scans have previously been used for classification and as a prognostic indicator.
Diagnosis and Classification
In the initial assessment of knee OCD, other possible differential diagnoses should be considered, including but not limited to normal developmental variations in ossification centres, acute osteochondral fracture, loose bodies, meniscus pathological conditions, avascular necrosis and epiphyseal dysplasia. Ideally, diagnosis of knee OCD occurs early in the stage of the disease to maximise treatment benefit and prognosis. However, because of the nonspecific presentation of knee OCD, most patients are diagnosed more than 1 year from the onset of their symptoms. , The diagnostic accuracy of knee OCD with a combination of medical history, physical examination and radiographs has a reported sensitivity of 77% and a specificity of 98%. The diagnostic accuracy of MRI is similar to this multifaceted approach. , Finally, knee arthroscopy is a powerful diagnostic tool and remains the gold standard for judging the stability and classification of OCD lesions. ,
Multiple radiographic, , MRI , , , and arthroscopic classifications exist; however, there is no reliable or universally accepted classification system to standardise the diagnosis and treatment of knee OCD lesions. , Classification systems have focused on assessing lesion stability, synonymous with the mechanical integrity of the lesion. Guhl popularised this concept and is responsible for perhaps the most well-known and commonly used classification system ( Table 23.1 ).
| Type I: Intact lesion |
| Type II: Fragmentation in situ (early separation) |
| Type III: Partial detachment |
| Type IV: Complete detachment, loose body present |
Nonoperative Management
For juvenile patients with stable OCD lesions, nonoperative treatment is well established as the primary initial treatment approach because of favourable natural history and rates of healing. Most authors advocate for a total of 3 months of attempted nonoperative treatment. This treatment consists of 6 weeks of protected weightbearing, followed by 6 weeks of activity modification and repeat radiographic evaluation at the end of the 3-month period. Patients and families should be advised to stop participation in activities that cause repetitive stress or compressive forces to the affected knee. Because of common challenges with compliance, knee immobilisation or bracing may be used to increase patient adherence; however, patients should be encouraged to maintain knee range of motion to prevent stiffness and maintain cartilage health. If subsequent follow-up radiographs demonstrate adequate healing and the patient is pain free, the patient may then gradually return to activities.
When there is low clinical suspicion of OCD and imaging is negative, patients should undergo conservative care. This includes physical therapy for strengthening, a gradual return to desired activity, nonsteroidal antiinflammatory drugs (NSAIDs) and follow-up to clinic in 6 weeks.
Nonoperative management is rarely indicated in adult OCD patients as a result of their limited healing potential and strong likelihood of disease progression. Multiple sources cite trends towards poorer outcomes in adults (patients with closed physes) undergoing nonoperative treatment. , , However, patients with stable lesions and no loose bodies or mechanical symptoms may attempt nonoperative management with close surveillance for changes in lesion stability. , Subsequently there should be a low threshold to progress with operative management if adult patients develop worsening pain, mechanical symptoms, or signs suggesting lesion instability or joint incongruity on repeat imaging studies.
Operative Management
Operative management is indicated for all patients with detached or unstable OCD lesions and for juvenile patients who have failed nonoperative measures and have impending closure of the distal femoral physis. , The overall goals of operative management are to maintain joint congruity, enhance the healing potential of subchondral bone, rigidly fix unstable lesions and repair osteochondral defects to ameliorate symptoms and prevent ongoing cartilage degeneration. The initial approach is typically reparative surgical methods, such as drilling or internal fixation; however, for OCD lesions that are not amenable to or fail reparative methods, restorative treatments are indicated. Fig. 23.3 shows the algorithm we use at our institution.
Reparative Procedures
The goals of reparative procedures in OCD treatment are to enhance healing potential by reestablishing blood supply to the subchondral bone, limit lesion instability by restoring the integrity of the native subchondral interface and preserve the overlying articular cartilage.
Drilling
The aim of drilling techniques is to disrupt the sclerotic margin of the OCD lesion, allowing for establishment of channels for revascularisation of the subchondral bone, which in turn facilitates subsequent osseous bridging and healing of the lesion. Drilling procedures are primarily indicated for juvenile patients (open physis) with stable or minimally unstable lesions (no gross instability with palpation), corresponding to Guhl classification types I and II, respectively. The drilling is performed through antegrade , or retrograde , approaches; multiple retrograde techniques, including extraarticular and intercondylar techniques, have been described. Antegrade drilling is perhaps the easiest approach; however, it requires violating the articular surface. Conversely, retrograde drilling attempts to enter the lesion from behind, or indirectly, to limit iatrogenic damage to the overlying articular surface ( Fig. 23.4 ).
Postoperatively, full knee range of motion is immediately encouraged with the goal of enhancing synovial fluid nutrient diffusion and preventing loss of range of motion. The patient should remain touchdown weightbearing for 6 weeks, advancing to weightbearing and resistance exercises at 6 to 12 weeks with possible return to sport after evidence of radiographic union or 3 months postoperatively.
There are overall favourable reported outcomes with drilling techniques for juvenile OCD patients. The reported healing rate ranges from 71% to 100%, with significant improvements in various outcome measurement tools. , , The reported healing rate in adult OCD patients is less favourable (25%), with an associated high rate of poor outcomes.
Loose body excision
Loose body excision, performed as a solitary procedure, is generally not recommended because of low overall success rates and an increased risk for development of osteoarthritis. , Moreover, the cartilage viability of loose body fragments is similar to that of native cartilage, and fixation of loose bodies has been shown to be a better alternative to excision in long-term studies. Depending on the size and location of the OCD lesion, good results have been reported with solitary excision, , , especially if the lesion is in an area without significant weightbearing or contact pressure. However, many of these results remain controversial. At this time, loose body excision is reserved for symptomatic patients with low functional demands or patients who are not able to adhere to postoperative rehabilitation protocols necessary for more sophisticated procedures.
Internal fixation
Internal fixation methods are used to help reestablish joint congruity by aiding in the reduction and subsequent healing of unstable OCD fragments. This applies to hinged chondral flaps (or so-called trapdoor fragments) and loose bodies, which correspond to Guhl classification types III and IV, respectively. Although arthroscopic techniques are typically used, open techniques can also be helpful for direct fragment visualisation and optimal reduction. , Before fixation, the fragments and their native subchondral bed should be debrided of fibrous tissue. Subsequent drilling or microfracture of the subchondral bed should be performed to stimulate vascularisation and to improve healing potential. , Additionally, cancellous autograft is useful for larger fragments and can help enhance revascularisation, as well as provide a foundation to ensure the fixed osteochondral fragment sits flush with the articular surface. The cancellous autograft required for this procedure can be easily obtained from the ipsilateral extremity, proximal and medial to the tibial tubercle (above the physes) or at the Gerdy tubercle. Fixation can be achieved with metal implants (Kirschner wires, cannulated screws, variable-pitch screws), bioabsorbable devices (screws, arrows or smooth or barbed pins) ( Fig. 23.5 ) or osteochondral autologous grafts (also referred to as bone pegs or bone sticks). , The fixation is generally accomplished in an antegrade approach, although retrograde techniques are commonly used for lesions involving the patella. Importantly, for stable fixation resistant to shear stress, there must be adequate subchondral bone attached to the fragment to allow for implant purchase and compression. Furthermore, fixation should be placed at a minimum of two locations to confer rotation stability and equal, distributed compression. The fixation devices should be recessed beneath the cartilage surface and overtightening of screw fixation should be carefully avoided because it can cause fracture of the osteochondral fragment.
