OCD of the Knee in Adolescents

Abstract

Osteochondritis dissecans (OCD) of the knee is a pathological condition of subchondral bone resembling focal osteolysis with subsequent bone resorption, which may lead to osteochondral fragment separation. Several etiological concepts reported for OCD development. The multifactorial theory is commonly adopted for days. Different investigators report OCD lesion healing while using conservative treatment or even “waitful watching” with a healing rate of up to 67%. In spite of these results, there are not any commonly adopted guidelines for conservative treatment. The last stage of OCD is a separation of osteochondral fragment leaving a full-thickness osteochondral defect, which is usually filled with low-quality fibrocartilaginous tissue. This tissue provides a lesser extent of resistance to peak loading forces, which poses at risk subchondral bone for further destruction and early osteoarthritis development. Appropriate treatment method should be chosen for each OCD stage in order to prevent early osteoarthritis development, increase return-to-sport rate, and decrease healing time for OCD lesions. This chapter provides short but comprehensive to date knowledge about OCD on the knee of adolescents and young adults.

1. Introduction

Osteochondritis dissecans (OCD) of the knee is a pathological condition of subchondral bone resembling focal osteolysis with subsequent bone resorption, which may lead to osteochondral fragment separation. Several etiological concepts are presented for OCD development with the leading question of what was the first—bone or cartilage? This question was raised a couple of decades ago and was similar to that one about chicken and egg. The multifactorial theory that was presented accompanying different pathological pathways leads to the main pathology.

While plenty of studies existed describing the surgical treatment of OCD, different investigators presented their data about healing OCD lesions using conservative treatment or even “waitful watching” management with a healing rate of up to 67%. Despite such a good result, there are not any common practice guidelines for conservative treatment accepted by practitioners. OCD of the knee usually goes through several stages. The last one is the separation of osteochondral fragment, leaving a full-thickness osteochondral defect, which is usually filled with low-quality fibrocartilaginous tissue. This tissue provides a lesser extent of resistance to peak loading forces, which poses at risk subchondral bone, for further destruction and osteoarthritis development. This fact raises some questions about indications for surgical treatment, its timing, and best option for surgical management of OCD in a particular stage.

Knowing the fact that adolescents involved in competitive sports are usually predisposed to knee OCD—what management strategy would you apply to such patients? How would you choose the appropriate surgical method? Will you take to attention the stage and age of patient? When will you advise this patient to start sports activities? Which criteria would you use to decide about lesion healing? Is complete lesion healing possible after OCD? All these questions need to be summarized. This chapter provides short but comprehensive to date knowledge about OCD of the knee of adolescents and young adults.

2. Epidemiology

Two types of OCD in adolescents are established—juvenile and adult—depending on whether a growth plate of the distal femur is open or closed. Juvenile OCD of the knee has an incidence from 9.5 to 29 of 100,000 knees [1, 2]. There is a three times increased incidence in children after 12 years old in comparison with children 6–11 years old, and boys are affected around four times more frequently than girls [2]. Adolescents involved in professional sports represent a large cohort of patients affected by OCD [3]. Femoral condyles are commonly affected, and posterolateral aspect of medial condyle is the typical region of OCD location (77%) [4]. Less common regions are lateral condyle (17%), patella (7%), and tibial plateau (0.2%) [4]. Lesions are bilateral in 14–30% [5].

3. Etiology and pathology

Osteochondritis dissecans of the knee was first mentioned by König in 1888, who described focus of an inflammation of the bone cartilage interface. König supposed that direct and minimal trauma can result in isolated subchondral damage. Patients that did not have any trauma prior to symptoms development were assigned to “osteochondritis dissecans.”

Many theories about OCD have been presented. It is well known that etiology-targeted treatment strategy is always the best. Thus, many investigators focused on looking for the main cause of the pathology. The pathology of OCD needs to be known to guide our way through its etiology.

According to pathological changes, OCD of the knee is usually divided into four stages, which may be seen in subchondral bone and adjoined articular cartilage [6]. Stage 1 is described as early subchondral osteopenia and usually is not detectable on X-ray, which remains a common way to visualize OCD lesions. Stage 2 represents bone marrow edema particularly located in subchondral bone. The bone is still viable in OCD lesions. Stage 3 is the most well-known stage by healthcare practitioners due to the possibility of X-ray to reveal radiolucency in subchondral bone representing its necrosis demarcated from healthy bone by sclerotic rim. Green and Banks were the first who discovered such necrosis and underlined intact articular cartilage adjoining OCD lesion (true stage 3) [7]. A variety of pathological changes was found by different investigators in OCD lesions on stage 3. Uozumi et al. found three types of bone morphology in OCD lesions—necrotic bone, viable ossifying trabecular bone, and absence of any bone. As the first type is inconsistent with the earliest study [7], other types rise the question of whether there are multiple morphological types of bone damage in OCD lesions or one process with consequent changing stages. For instance, Wagner described “malicious variant” with OCD lesions full of multiple bone fragments [8]. Other studies revealed the absence of any degenerative bone changes in stable OCD of the knee but the presence of fibrocartilaginous tissue on the periphery of the lesion between parents resembling that found between bone fragments in nonunion after bone fracture [9]—Figure 1.

Figure 1.

Osteochondritis dissecans histology. Black arrow points to the fissure between maternal bone and osteochondral fragment. Note the presence of a sclerotic rim on maternal bone adjacent to the fissure and the presence of fibrous tissue precluding adequate union. From Bruns et al. [6].

Stage 4 usually occurs under continuing mechanical loading resulting in the loosening of the bone and cartilage followed by osteochondral fragment separation.

While most of the existing studies state that subchondral bone is to be damaged first, several pathways are described. According to the systematic review of Luca Andriolo et al., 28 articles present subchondral bone fracture as the starting point of pathology cascade, and eight articles state that ischemia is the first event followed by subchondral bone degeneration [10]. Thus, there are two main general theories: mechanical and vascular, followed by special biological, genetic, and endocrine factors, which may play role in OCD development.

The mainstay of the mechanical theory is collected data about relationship between chronic repetitive microtrauma, traumatic incidents, and consequent OCD development [11]. This theory has been supported by an epidemiological study, which revealed a relatively high proportion of adolescents involved in competitive sports activities [2]. Both acute and chronic repetitive trauma may result in subchondral fracture by themselves or in combination with other mechanical factors reported—discoid meniscus [12, 13, 14], tibial spine impingement [15], hypermobile anterior horn of meniscus [16], and joint instability and genu recurvatum [17]. Several metabolic conditions affect bone quality posing subchondral bone at risk for fracture—Wilson disease [18], hyper-IgE syndrome [19], low vitamin D3 level [20], and high human growth hormone level [21].

Other well-known theory is the vascular one. Adherents of this theory suppose that ischemia is the first event in subchondral bone followed by partial bone necrosis and subsequent fracture. The first histological confirmation of this theory is related with the study of Green & Banks, who found necrotic foci in the subchondral bone of femoral condyle [7]. Campbell and Ranawat later defined OCD as a focus of aseptic necrosis particularly developing in “locus minoris resistentiae” [22] of ossifying bone of secondary ossification center in condylar epiphysis after repetitive microtrauma [23]. The ischemic nature of OCD lesions was advocated by Jans et al. Authors stated that terminal epiphyseal arteries do not have adequate collateral branches, resulting in abruption of subchondral bone vascular supply after minimal acute or chronic repetitive trauma [24]. The closed relationship between early epiphyseal vascular regression (up to 5 years old) and late enchondral ossification (up to 10 years old) of epiphyseal secondary ossification zones was described thoroughly by Ellerman et al. [25].

To date, pathophysiological pathway for OCD lesion development seems to be multifactorial: anatomy of epiphyseal vascular branches in zones of secondary enchondral ossification predisposes to subchondral bone fracture after minimal traumatic event, leading to bone necrosis, lack of cartilage supply, fissures emerging, and subsequent osteochondral fragment detachment.

4. History

History of OCD of the knee usually starts with pain, which can relate to previous traumatic episodes or not. The pain usually starts with prolonged physical activities, and worsens with deep squatting and load distribution to an affected leg [5]. As many patients with knee OCD are sportsmen, the main complaint is the inability to fully participate in competitions and even regular training activities. A separate type of history usually has patients sustained acute trauma after a period of prolonged knee pain. If such pain is long-lasting, it can be related with ongoing OCD of the knee, and the traumatic episode can lead to osteochondral fragment detachment. Such a patient may suffer from knee-locking episodes, popping, or knee catching. Cases of unstable OCD lesions are frequently accompanied by synovitis developing after osteochondral fragment partial detachment.

5. Physical examination

OCD of the knee in adolescents has poor clinical manifestation. In the case of stable OCD, there are not usually any positive provocative maneuvers. There is only one specific test described by Wilson—pain appears while tibia is internally rotated during knee extension from 30 to 90 degrees [26]. Pain alleviation, while tibia is externally rotated, is highly suspicious for OCD of femoral condyle in a typical location (posterolateral aspect of medial femoral condyle). In spite of relatively high specificity, this symptom has low clinical diagnostic value. Unstable OCD cases may be accompanied by knee effusion associated with smooth contours of the knee, positive patellar tap test, and restricted range of motion.

6. Imaging

Standing X-ray in three projections is a standard clinical protocol for OCD lesion evaluation. Anteroposterior, lateral, and notch-view usually are sufficient for diagnosis confirmation [27]. OCD lesion is usually described as either a crescent-shaped radiolucent area in subchondral bone or radiolucent line between bone fragment and parent bone with an area of sclerosis adjacent to radiolucent zone. In the case of unstable OCD with a detached osteochondral fragment, uneven contour of femoral condyle can be seen with an area of adjacent bone sclerosis. Despite the usefulness of radiography for diagnosis making, it is usually not possible to differentiate stable OCD lesions from unstable ones. So, radiography is of importance only as a first-contact diagnostic method.

MRI is the most informative method aiding in the visualization of all components of OCD lesions, including cartilage. T1 sequence is usually used for size measurement. T2 fat-saturated and PDFS sequences are used for bone marrow edema assessment, cartilage visualization, and most important for defining the type of OCD lesion according to its stability. De Smet et al. defined four MRI criteria of stability [28]:

1. A thin line of high-signal-intensity ≥5 mm in length at the interface between the lesion and underlying bone (fibrovascular granulation tissue);

2. A round area of homogeneous high-signal-intensity ≥5 mm in diameter beneath the lesion (cysts);

3. A focal defect with a width of ≥5 mm in the articular surface of the lesion (displacement of the lesion into the joint);

4. A high-signal-intensity line traversing articular cartilage into the lesion (articular fracture).

Despite relatively high sensitivity and specificity in adults, these criteria represented only 11% specificity in children [29]. After considering additional criteria by Kijowski et al., a specificity of 100% for children population achieved:

1. A high T2-signal-intensity rim or cysts surrounding an adult OCD lesion are unequivocal signs of instability

2. A high T2-signal-intensity rim surrounding a juvenile OCD lesion indicates instability only if it has the same signal intensity as adjacent joint fluid, is surrounded by a second outer rim of low T2-signal-intensity, or is accompanied by multiple breaks in the subchondral bone plate on T2-weighted MRI

3. Cysts surrounding a juvenile OCD lesion indicate instability only if they are multiple in number or large.

General visual characteristics of unstable and stable lesions described by Kijowski et al. are presented in Figures 2 and 3.

Figure 2.

Sagittal 1,5T fat suppressed T2w FSE MRI of the knee in a 15-year-old adolescent with confirmed unstable OCD lesion of the medial femoral condyle. OCD lesion consists of heterogeneous high- and low-signal-intensity (large arrow) areas and is surrounded by a single cyst of 6 mm in diameter (large arrowhead) and extensive bone marrow edema (small arrows). There is no disruption of the low-signal-intensity subchondral bone plate at edges of OCD lesion (small arrowheads). From: Kijowski et al. [28].

Figure 3.

Sagittal 1,5T fat suppressed T2w FSE MRI of the knee in a 12-year-old adolescent with confirmed stable OCD lesion of the medial femoral condyle. OCD lesion has uniform low-signal-intensity (large arrow) area that is surrounded by a single cyst 4 mm in diameter (small arrow). There is no disruption of low-signal-intensity subchondral bone plate at the edges of OCD lesion (arrowheads). From: Kijowski et al. [29].

Defining the stability of OCD lesions is of utmost importance for guiding management. CT scan also takes an important part in OCD lesion visualization. It provides data about size and subchondral bone structure in detail and can aid in finding free osseous fragments in the knee joint [30]. CT scan has the possibility of fast and precise lesion assessment in both knees in case of two-sided knee OCD and serves as a good tool for finding accidental OCD lesions in the contralateral unaffected knee. Limitations include radiation exposure and the inability to assess cartilage status in OCD lesions. CT scan cannot define if the lesion is unstable or not.

Ultrasonography is rarely used in knee OCD imaging mainly because of its operator dependence and relatively low diagnostic accuracy [31].

Bone scintigraphy is a method that is possible to provide data about bone perfusion of an OCD lesion. Almost absolutely lacking specificity in defining different stages of OCD, this imaging modality is rather a historical one. Nowadays 3-T MRI method of “arterial spin labeling” was developed to assess perfusion in femoral condyles similarly to scintigraphy [32].

7. Classifications

There are plenty of classifications described nowadays for knee OCD. A few of them are useful in treatment strategy choosing. As radiography and CT scan cannot accurately describe lesions' stability, most investigators attempted to find the best MRI classification either for treatment strategy defining or for preoperative planning. The most important among MRI classifications are the aforementioned classification by De Smet et al.’s and Kijowski et al.’s modifications for lesion stability assessment [28, 29, 33].

Arthroscopy remains to be a gold standard for the assessment of OCD lesion stability [1]. Various classifications are described. Most of them consist of four stages [34, 35, 36], where third stage is an unstable OCD lesion with a partially detached osteochondral fragment and fourth stage is a completely detached fragment with a free intraarticular body. To days, a new classification was developed by the ROCK study group consisting of six stages based on arthroscopic lesion appearance (Figure 4).

Figure 4.

ROCK study group arthroscopic OCD lesion classification. Six stages were described with 1–3 being stable and 4–6 unstable.

8. Treatment strategy

There are two main options—conservative and surgical treatments. As OCD of the knee is progressing disease, the main goal of treatment of unstable lesions is to prevent their destabilization and achieve full lesion healing.

Healing of some OCD lesions reaches 50–67% in 6–12 months [5, 37, 38]. Despite such good results, the current literature volume accumulated has scarce data about different conservative treatment methods with moderate quality studies mostly due to the inability to correct allocation of patients and non-randomized study designs [11].

In a recent systematic review, Andriolo et al. found 24 case-series studies and three case reports for conservative treatment of OCD lesions. Only 12 studies had one treatment method such as immobilization, physical instrumental therapy, or physical activity restriction. Fourteen studies describe combinations of conservative methods of treatment. Overall healing was 61.4% ranging from 10.4% to 95.8% [11]. Despite such heterogeneous data, negative prognostic values for healing were also described:

1. Lesion size (large size correlated with low overall healing rate)

2. Severe lesion stages based on different classifications

3. Older age

4. Combination with discoid meniscus

5. Long period from onset to the first consultation

6. Mechanical symptoms

7. Atypical lesion location (lateral femoral condyle, patella, and tibial condyles)

While some studies report 80–90% healing rate of stable OCD lesions, using physical activity restriction and quadriceps muscle strengthening [39], others report only 30% healing rate and 67% as “no changes” [40]. Such differences may have been appeared due to maturity status or other abovementioned factors confounding. To minimize influence of confounding factors on the healing rate, Krause et al. developed a nomogram for probability of healing assessment [38]. He used multivariate analysis of independent factors on 37 OCD lesions and found three independent factors: age, cyst-like lesion normalized width, and overall lesion normalized width—0–10 points, for each of assessing factors. The overall score is linked with probability scale (Figure 5—the example).

Figure 5.

Prediction of healing probability after 6 months of nonoperative treatment based on normalized lesion width, cyst-like lesion (CLL) size, and patient age. The case is a 13-year-old patient, who showed a tendency toward healing after 6 months and showed complete healing in 12 months. From: Krause et al. [38].

Generally, stable OCD lesions followed every 3 months with different authors suggesting either 3 or 6 months of conservative treatment prior to surgery referral decision-making [5, 11, 41].

Despite good short-term results of conservative treatment for juvenile stable OCD lesions, a recent study by Sanders et al. described bothering long-term results—in up to 30% of patients treated nonoperatively osteoarthritis was developed at 35 years [42]. Another long-term study by the same author reports osteoarthritis development cumulative incidence of 70% at 30 years after osteochondral fragment excision in patients with the last stage of knee OCD and 51% incidence in patients at 30 years after osteochondral fragment preservation [43]. This relatively high incidence determines the need for thorough timing of operative treatment.

As large part of patients with OCD of the knee is represented by professional sportsmen, the time of OCD lesion healing really matters. Faster lesion healed—faster weight-bearing activities may be achieved, and shorter return-to-sport can be reached.

In summary, indications for operative treatment of the knee OCD in adolescents are as follows:

1. Signs of instability defined by Kijowski et al. on MRI. Elective surgery may be planned but patient must observe a special regimen of non-weight bearing and activity restriction.

2. Mechanical symptoms emergence in addition to preoperative imaging data about osteochondral fragment separation. Surgery needs to be provided as fast as possible to capture a chance of osteochondral fragment refixation.

3. More than 3 months of conservative treatment without any healing signs.

9. Operative treatment

The type of surgical intervention in adolescent knee OCD depends on several factors: stability and size of the lesion and presence or absence of mechanical symptoms in patient.

Surgical treatment usually starts from arthroscopy of the knee. Infrapatellar plica, medial plica, and partial Hoffa pad resection are necessary for better visualization of condylar cartilage. The main goal is to correctly assess if the OCD lesion is stable or not. Visually stable JOCD may be represented by fully intact cartilage either without any color changes or may have dim color compared to neighboring regions of intact cartilage. Cartilage above the lesion may be depressed or fissured on the periphery of the lesion or can be partially detached. Villous overlays may be seen in patients with partially fissured cartilage.

9.1 Stable OCD lesions

When the lesion is stable, there is usually fibrocartilaginous layer between osteochondral fragment and maternal bone, resembling tissue that is usually can be found in pseudarthroses between bone fragments [9]. This finding supports data about failed ossification as the main barrier for OCD lesion to heal [44]. So far, local stimulation of osteogenesis is the main goal of OCD lesion treatment. It can be either mechanical stimulation or biological.

Common mechanical stimulation of bone regeneration in stable OCD lesions is drilling [1, 5]. There are two types of drilling: antegrade (retroarticular) and retrograde (transarticular or transchondral). Retroarticular drilling has the advantage of cartilage preservation but method is not universal and depends on OCD lesion location [45]. For instance, if OCD lesion is located posteriorly in condyle or in the center of the weight-bearing surface of cartilage, it is possible to reach this zone from the lateral subphyseal area of condyle, used as starting point for drilling. The use of fluoroscopy guidance during retroarticular drilling is associated with radiation exposure. Anteriorly located lesions are difficult to be drilled from this position. Transarticular drilling is faster, and there is no need for fluoroscopy guidance because drilling is performed under direct visual control by an arthroscope. Nonetheless, transarticular drilling is associated with cartilage damage. Cartilage appearance may have changed for years after transarticular drilling as advocated [46]. Otherwise, long-term studies with well-done designs are needed to advocate if transarticular drilling has a negative effect on knee functioning or early osteoarthritis development.

Gunton et al. published a thorough systematic review of studies that investigated either transarticular or retrograde drilling of OCD lesions [45]. Despite meta-analysis was not possible to carry out and the high heterogeneity of studies, the authors found no significant differences according to lesion healing effectiveness. There was 86% healing rate for retroarticular drilling and 91% for transarticular. Patient-reported outcomes were also comparable. Time to heal was lower in lesions after transarticular drilling (4–5 months) compared to retroarticular (5–6 months). Despite that facts, 36.9% lateral condyles were drilled in retroarticular group compared with 5.3% in transarticular, for healing rate comparison,48% and 6.3%—for time-to-heal comparing, respectively. Knowing that many factors affect the healing rate in OCD lesions, including lesion location, the results must be interpreted with caution.

The authors of this chapter carried out their own systematic review for OCD lesion drilling in children aged before 18 years old, only nine full-text studies were included in the analysis. We found 95.3% cumulative healing rate of OCD lesions after retrograde drilling and 76.8% after transarticular drilling [47]. Because of such heterogeneous information represented across the literature, there is no evidence of which type of drilling is preferable currently considering either healing rate or time-to-heal.

9.2 Unstable OCD lesions

Unstable lesions may be represented as partially or completely detached osteochondral fragments. Osteochondral fragment detached from their place and becomes incongruent with time, so there is a need for urgent treatment of patients with unstable OCD lesions suspected in MRI. If we have a deal with unstable but partially detached fragment, fixation of the fragment is the most valuable option. It can be carried out by metal screws, absorbable pins, Kirshner wires, or even autologous bone sticks [48]. A recent systematic review carried out by Leland et al. reported 67–100% healing rate after osteochondral fragment fixation in skeletally matured knees. No significant differences were found in context with healing rate between types of fixator, and complications included reoperations for cartilage resurfacing, loose body removal, or unplanned hardware removal [48]. Kocher et al. found 84,6% radiographic healing rate in adolescent OCD using fragment fixation independently of device used for fixation [49]. Adachi et al. reported 77% healing rate after osteochondral fragment fixation using bioabsorbable pins [50]. Wu et al. found 76% healing rate in both skeletally immature and mature knees in their multicenter study. They reported no differences between mature and immature knees in the context of healing rate [51].

9.3 Full-thickness cartilage lesions in the last stage of OCD

There is a high rate of osteoarthritis development after osteochondral fragment detachment—a cumulative incidence of 70% at 30 years was reported recently. Moreover, the cumulative incidence of arthroplasty is 32% at 30 years [43]. To avoid these complications and to achieve a good quality of life for patients with osteochondral defects different methods exist.

For lesions less than 2 cm², microfracture procedure proved to be the first-line treatment method [52]. Damaging subchondral bone with special awl results in multipotent stromal cell effluxes followed by fibrocartilage filling the defect. It provides good short-term results in context with patient-related outcomes [53]. Despite relatively good short-term results after microfracture procedure, there is an incidence of 45% for knee replacement at median of 12 years after this procedure [54]. Hyaline cartilage predominantly contains collagen type 2 while fibrocartilage, which fills osteochondral defects after microfracture, mainly consists of collagen type 1 [55]. Therefore, fibrocartilage does not have properties of normal hyaline articular cartilage and cannot withstand weight-bearing loads as well as intact articular hyaline cartilage [56]. Despite being the first-line treatment for osteochondral defects smaller than 2 cm², recent systematic review reported increased fill of the defect while using deep drilling comparative with microfracturing. Free access to bone marrow space can be achieved using deep drilling, while fractured bone with osteocyte necrosis was found after microfracture procedure [57]. Two studies reported superior cartilage restoration while implementing deep drilling (6 mm depth) instead of microfracture procedure [58, 59].

Osteochondral autologous transplantation (OAT) or mosaicplasty is another option to fill the full-thickness osteochondral defect. Osteochondral plugs from non-weight bearing region of femoral condyles are harvested and transferred to osteochondral defect in this method. Mosaicplasty has good mid-term and long-term results in the context of activity, and patient outcome scores risk of failure in patients with osteochondral lesions sized more than 3 cm² [60, 61]. Pareek et al. recognized in their meta-analysis that microfracture has 2.4 times more risk of failure compared with mosaicplasty [60]. Gudas et al. found a 21-times higher risk of failure after microfracture procedure compared with OAT in patients with full-thickness lesions of more than 3 cm² only after OCD [62]. Solheim et al. also reported noticeably increased patient-reported outcomes after mosaicplasty compared with microfracture procedure at long-term follow-up [63]. OAT has several complications, including cartilage hypertrophy, at the periphery of osteochondral defect and donor site morbidity while using more than two osteochondral plugs [64]. That limits the implementation of OAT in adolescents involved in professional sports activities. Limitations can be partially avoided using allograft-OAT called in literature OCA (osteochondral allograft transplantation). This technique allows for more thorough graft matching with the defect and, therefore, may be used in cartilage defects more than 4 cm² [52].

Autologous chondrocyte implantation (ACI) and its 3rd generation—matrix-induced autologous chondrocyte implantation (MACI)—are promising techniques for cartilage restoration in patients with osteochondral defects more than 2 cm² [52]. The main feature associated with MACI is the need for two operations to be performed. This fact significantly affects the rehabilitation period and return to sport velocity. The main goal of first operation is to harvest cartilage from non-weight bearing zones, usually intercondylar notch. After that, 6-week production period is needed for colony of MSC to grow and colony formation. Seeded on a special porcine collagen membrane, MSC is attached to osteochondral defect. MACI is superior to microfracture procedure in patients that have more than 2 cm² cartilage lesions at 2-year [65] and 5-year follow-up [66], according to results of patient-reported outcome measures.

Autologous induced matrix chondrogenesis (AMIC) is another technology, the main peculiarity of which is absence of second operation necessity. After lesion site debridement and subchondral bone exposure drilling perform followed by the application of ready-to-use collagen I/III membrane supported by fibrin glue [67]. Efflux of MSC from bone marrow of condyle and its proliferation and differentiation in special microenvironment provide cartilage restoration and hyaline-like tissue formation [68]. Several studies support the supremacy of AMIC over microfracture procedures in osteochondral defects more than 2 cm² on context with both patient-reported outcomes and MRI data achieved [67, 68, 69].

Another promising method for one-stage cartilage restoration is minced cartilage preparation and usage. Basic science for this method was represented by several authors [70, 71, 72, 73] and a few clinical studies conducted to date [74]. Salzmann et al. figured out that chondrocytes in live cartilage after slicing induce their proliferation and differentiation, produce intercellular matrix, and have the ability to fill cartilage defects [70]. It was also found that the smaller cartilage particles are—the bigger amount of ECM is going to be produced after cartilage slicing [73] and more efficient chondrocytes outgrowth is Ref. [75]. Firstly, reported by Albrecht et al., particulate autologous cartilage implantation (or minced cartilage implantation) procedure proved the clinical efficacy of cartilage autograft implantation system (CAIS) in the study of cole BJ et al., who reported marked improvement with statistically significant differences for KOOS and IKDC scores in favor of CAIS compared with microfracture procedure [76]. The surgical technique includes harvesting cartilage from non-weight bearing part of condylar cartilage (usually in the region of intercondylar notch), processing of cartilage (cartilage fragmentation with a shaver or other devices), cartilage defect preparation (defect walls debridement and removing of calcified layer), minced cartilage paste preparation using either fibrin glue or different orthobiologic substances—BMAC, PRP, and PRF [74, 77].

A few clinical studies confirmed positive results of minced cartilage implantation procedures. Christensen et al. reported markedly improved MOCART-MRI score, patient-reported outcomes, and more than 80% bone defect filling at 1 year after operation in eight patients using a combination of autologous bone graft and autologous cartilage chips technique supported by fibrin glue. All patients had osteochondral defects developed as the last stage of OCD [78]. Massen et al. revealed statistically significant improvement in functional knee score and MOCART score at 2 years, after minced cartilage implantation in 27 patients, and concluded that minced cartilage technique can be a good alternative to currently existing ACI/MACI techniques [79]. Cugat et al. reported excellent clinical, MRI, and patient-reported functional outcomes in 15 patients at 15 months after minced cartilage technique implementation with platelet-rich and platelet-poor 50/50 plasma addition [80].

Particulated juvenile allograft cartilage (PJAC) is the technology resembling minced cartilage procedure. The main difference is that PJAC represented by juvenile hyaline cartilage consisted of immature chondrocytes that provide increased proliferative and metabolic activity, obtained from donors, and has 45 days to be viable after the cells package [81, 82], while there are no clinical studies for use of PJAC currently available [52]. However, Ao et al. found more hyaline-like cartilage content at 1 month and 3 months after cartilage chips implantation in minipigs compared with PJAC implantation with no significant differences at 6 months [83].

10. Biological stimulation in OCD treatment

Biological stimulation is another way to improve bone regeneration locally. Orthobiological products, such as platelet-rich plasma (PRP) and bone marrow aspirate concentrate (BMAC), contain different growth factors, improving cell proliferation, migration, and differentiation [84, 85, 86]. BMAC also has multipotent stem cells, which can differentiate into different lineages themselves taking part in tissue restoration [87]. There is a literature gap about biological preparations implementation in complex surgical treatment of knee OCD.

Sharma et al. reported six adult OCD unstable ICRS and three lesions completely healed in 4 months using fragment fixation with metal compression screws and adjuvant intraarticular PRP injection treatment. The authors’ technique was represented by three intraarticular injections of PRP every 3 weeks starting on 12 postoperative days, and 8–10 ml PRP was used for each injection [88]. Davidson et al. shared the results of 52 stable juvenile OCD lesions treatment with an average size of 4.07 cm² by retroarticular drilling and BMAC augmentation back-filling through pin channel. 76.9% of lesions were healed at a mean of 10.6 months [89]. Later Andelman et al. described a technique for retroarticular core decompression of stable OCD lesions by retroarticular over drilling of femoral condyle using a drill bit with subsequent curette OCD lesion decompression and implementing BMAC-DBM paste [90]. No results have yet been published to analyze this treatment method. Baldassarri et al. recently reported good results using bone marrow-derived cell transplantation technique for restoration of osteochondral defects [91]. The technique implies a collagen scaffold, embedded with concentrated bone marrow aspirate and platelet-rich gel. Authors found significantly better Tegner and IKDC scores at 1 year after the operation and 75–100% lesion filling on MRI-MOCART score in 13 out of 18 patients. De Girolamo et al. noted lower VAS scores in patients undergone AMIC procedure with BMAC augmentation compared with AMIC procedure alone for full-thickness osteochondral defects, with significant difference in Lysholm score at 6 and 12 months after operation in favor of AMIC with BMAC procedure [68]. This finding is important in the context of the rehabilitation period shortening and faster return to sport.

Retroarticular drilling with PRP intralesional injections technique is currently used in N.F. Filatov Children's City Hospital of Moscow Healthcare Ministry by prof. Vybornov D. Yu. et al. This method was developed for stable OCD lesions and implies retroarticular drilling of the lesion with combined fluoroscopy and arthroscopy assistance—3 or 4 drill holes made—with subsequent slow (2 minutes long) PRP injection through the drill channel using 20G long needle and 3-minute exposition before needle withdrawal (Figure 6). PRP is derived from peripheral blood after 2-stage centrifugation according to the platelet-safe. Bausset et al. technique previously described maximizing live platelet count in the final product [92].

Figure 6

(A–D). Retroarticular drilling with PRP intralesional injection technique. OCD lesion is over drilled from starting point of approximately 0,5 cm under the growth plate under fluoroscopy control (A). PRP is prepared 30 minutes before operation with Bausset et al.’s protocol (B). PRP is slowly injected inside the lesion (C). Fluoroscopy control of needle position is mandatory before PRP injection (D).

A 13-year-old adolescent was admitted to orthopedics and traumatology department of N.F. Filatov Children's City Hospital of Moscow Healthcare Ministry with 9-month long complaints of knee pain that were worsening after physical activities and deep squatting. Knee provocative tests were negative. MRI and CT scan revealed an OCD lesion in medial femoral condyle without instability signs (Figure 7A and D). The patient undergone retroarticular drilling and PRP intralesional injection procedure. Touch-down weight-bearing started 1 month after the operation. A 2-month follow-up MRI revealed partial lesion healing (Figure 7B) and a 6-month MRI and CT scan showed restoration of subchondral bone without any signs of cartilage pathology (Figure 7C and E).

Figure 7

(A—E). A 13-year-old patient with a stable OCD lesion was revealed on MRI (A) and CT scan (D). Bone marrow high-intensity-signal on PDFS MRI without any signs of instability, and the low-intensity line between progeny and parent bone can be found representing ossification impairment (A). The subchondral bone defect is shown on CT scan (D). Retroarticular drilling and PRP intralesional injection procedure were performed. The 2-month follow-up frontal PDFS MRI revealed relatively low bone marrow signal intensity (B). Moreover, the low-intensity line became shorter representing bone restoration process (B). Frontal PDFS MRI and CT scan at 6 months after the operation showed almost full subchondral bone restoration (E) and intact cartilage with minimal residual changes in subchondral bone (D).

This method was assessed in a comparative study that is yet unpublished. Fifty-five patients with femoral condylar OCD were included in the study and divided into three groups based on treatment method—retroarticular drilling and intralesional PRP injection, transarticular drilling without any additives, or transarticular drilling with intraarticular PRP injection. In total, 12 out of 15 OCD lesions in the retroarticular drilling+PRP group healed completely and three patients with shallow but wide and long OCD lesions did not have signs of healing at a median 10-month follow-up. Despite no statistically significant differences in OCD lesion size between groups were found, PRP implementation led to decreased time-to-heal: 6 months for complete healing in PRP groups versus median 10 months in the no-PRP group. We suggest retroarticular drilling with PRP intralesional injections for deep and short stable OCD lesions.

11. Defining healing of OCD lesion

To days, treatment guidelines for OCD of the knee are lacking evidence. One of the reasons is the lack of commonly adopted healing criteria and the absence of correlation between symptoms and OCD stage except stage 4 when mechanical symptoms usually guide the diagnosis [93]. Parikh et al. reported low interrater reliability for radiographic assessment of healing at 6 months after operation [94]. Wall et al. developed a reliable radiographic method of OCD lesion healing assessment based on a subjective assessment of five parameters on a continuous slider scale from −100 to +100 points [93]. Radiographic features assessed included articular surface shape, boundary, sclerosis, ossification, lesion size, and overall healing. Authors reported substantial to excellent interrater reliability at 2–24 months with ICC values from 0.77 to 0.88 [93]. This reliable method is perfectly fit for the assessment of OCD lesion healing independent from place of residence. Limitations include radiation exposure and inability to visualize cartilage.

Based on the study by Wall et al. [93] group of specialists in pediatric sports medicine and musculoskeletal radiologist of N.F. Filatov Children's City Hospital of Moscow Healthcare Ministry and the department of pediatric surgery of Russian National Research Medical University made an adaption of the abovementioned scale for MRI [95]. An expert group was conformed consisting of a 6-year medical university student, three 2nd-year residents, one pediatric orthopedic surgeon, and one musculoskeletal radiologist. After two rounds of learning by PowerPoint presentation for scale assessment rules, 34 knee PDFS MRI was assessed at different time points before and after operation. Five MRI-based features were determined (Figure 8): the degree of bone edema (Figure 8A), fragment consolidation (Figure 8A), subchondral bone structure (Figure 8B), articular cartilage damage (Figure 8C), and general lesion healing. Unlike the original Wall et al.’s study, general healing was calculated as a mean of four previous features assessed by experts. Excellent reliability was found for all parameters with ICC values of 0.97–0,99 except the degree of bone marrow edema on the latest follow-up MRI—0.54.

Figure 8.

Four main parameters are assessed to calculate general healing. The bone substance edema (circled with a continuous line—A), the degree of consolidation (the line between the osteochondral fragment of the lesion and the maternal bone is indicated by an arrow in Figure A), bone structure (estimated by density and structure of the bone tissue—B), and articular cartilage structure (indicated by arrows in Figure C).

A comparative study on the treatment of stable OCD lesions is currently on go using this novel MRI healing score for OCD of femoral condyles.

12. Authors’ preferred strategy for adolescent OCD of the knee management

While most of the adolescent patients with knee OCD are active and 53% of them are involved in competitive sports activities according to statistics from the department of orthopedics and traumatology of N.F. Filatov Children's City Hospital of Moscow Healthcare Ministry, not only 100% healing rate but also as fast as possible return to the abovementioned activities is awaited by patients. Appropriate treatment usually starts with appropriate history taking and imaging. We use several obligate questions for all patients about which mechanical symptoms they have, the longevity of their complaints, and prior treatment attempts. The presence of mechanical symptoms or knee effusion episodes always makes us suppose an unstable OCD lesion. MRI usually assists in lesion-type clarification. In the case of stable OCD lesions, conservative treatment usually starts, including affected limb unbearing, physical therapy without weight-bearing exercises, pulsed electromagnetic fields application, massage, and laser therapy. MRI is reassessed at 3 months. If no healing signs are presented—elective surgery is indicated. Stable OCD lesions are managed either by transchondral drilling with one intraarticular PRP injection after joint drying or with retroarticular drilling with intralesional PRP injection. Note that long and shallow OCD lesions are not fit well for retroarticular drilling techniques because of the risk of not reaching all zones of OCD lesions, which can later lead to lesion persistence.

Unstable OCD lesions are managed by refixation using metal screws either with compression cancellous screw or with headless screw. We prefer cancellous metal compression screw for large osteochondral fragments to achieve good compression forces at the progeny-perant bone line. For small fragments headless screws are appropriate for use because of the risk of progeny bone fragmentation.

Full-thickness defects management depends on the defect’s size. Small defects less than 2 cm² usually undergo bone marrow stimulation (BMS) procedures. We prefer 1.5-mm pin drilling to a depth of no less than 1-cm and 5-mm distance between drill holes. Intraarticular PRP injection is routinely used in our department after each BMS procedure. For 2–4 cm²-sized lesions osteochondral autograft transplantation (OAT), minced cartilage (MC) with fibrin glue and BMAC implementation, autologous matrix-induced chondrogenesis (AMIC), osteochondral allograft implantation (OCA), and matrix-induced autologous chondrocyte implantation (MACI) preferable depending on its availability for patient. For those defects whose size is more than 4 cm², OCA and MACI are the only methods that have evidence in the context of safety and possibility for use.

Brief authors’ preferred algorithm for OCD treatment in adolescents is presented in Figure 9.

Figure 9.

Authors’ preferred algorithm of OCD treatment in adolescents. BMST—bone marrow stimulation techniques, MC—minced cartilage technique, OAT—osteochondral autograft transplantation, AMIC—autologous matrix-induced chondrogenesis, OCA—osteochondral allograft implantation, and MACI—matrix-induced autologous chondrocyte implantation.

13. Conclusion

Osteochondritis dissecans of the knee in adolescents is potentially harmful disease, which mostly affects active children after 12 years old, usually involved in sports activities [5]. OCD generally has four stages with the last one representing osteochondral separation [6], potentially leading to early osteoarthritis and knee replacement if not managed appropriately [43, 96]. Proper OCD imaging allows to define stability of the lesion, which is a guide for treatment [97].

Stable lesions can be treated conservatively using several methods, evidence for which is not clear yet [11]. Conservative treatment can last from 3 to 6 months, and healing signs lacking at 6 months is an indication for surgery. Arthroscopy-assisted drilling for subchondral bone reparation improvement is the gold standard for stable lesions treatment regardless of the type [45, 98, 99, 100, 101]. Biological stimulation procedures are generally safe and proven to decrease healing time and increase the healing rate for nonunion [102, 103, 104, 105]. PRP or BMAC potentially can be applied in OCD cases [88, 89].

The goal of fourth stage OCD treatment is to deal with osteochondral defects. Depending on its size different options are available nowadays [34, 52, 53, 70, 72, 106, 107, 108, 109, 110, 111, 112, 113]. Bone marrow stimulation techniques (BMST), minced cartilage implantation (MC), or autologous induced chondrogenesis (AMIC) with membranes for lesions less than 2 cm² are followed by osteochondral autograft transplantation (OAT), osteochondral allograft transplantation (OCA), matrix-induced autologous chondrocyte transplantation (MACI), or minced cartilage implantation (MC) in lesions 2–4 cm². Defects more than 4 cm² require OCA or MACI for getting better outcome [52].

All methods have different cost-efficiency, availability, and complications. The individual decision must be taken for each patient considering all influencing factors.

Acknowledgments

We would like to acknowledge coworkers of N.F. Filatov Children's City Hospital of Moscow Healthcare Ministry, department of pediatric orthopedics and traumatology: Nataliya Trusova, Ekaterina Kardash.

Conflict of interest

The authors declare no conflict of interest.

Special thanks

We would like to address special thanks to contributors of our surgical operations, coworkers of the N.F. Filatov Children’s City Hospital of Moscow Healthcare Ministry: transfusiology department head—Marina Khlebnikova; Anesthesiologists—Alexander Leshkevitch, Anna Shaginyan; operation nurses—Darya Panfilova, Nataliya Kiyayeva, Lyudmila Klyueva, Tatiana Kotova; and residents of pediatric surgery department of Russian National Research Medical University.