NWHSU logo
Challenge of the Unknowns—Unknown 5

Diagnosis and Discussion


Spontaneous osteonecrosis of the knee (SONK).


The orthopedic surgeon suggested a total joint replacement but is very reluctant to do this surgery because of the patient's age. Arthroscopy was not recommended due to lack of mechanical "locking" symptoms. Currently the patient has not undergone surgery, sometimes uses a walker to help her walk, and takes Tylenol to manage painful exacerbation. Currently she is on a PRN basis with her orthopedic surgeon but refuses to have corticosteroid injections because she does not want to accelerate collapse of her knee joint.


You may have made the diagnosis on your first try or, as prompted, may have clicked on a couple of the other diagnostic possibilities. The difference between osteonecrosis and osteochondritis dissicans will be developed below. Chondrosarcoma is always a consideration for an aggressive bone lesion in an adult, however, the classic location and subarticular defect effectively rule-out this possibility. Neuropathic osteoarthropathy can cause significant degeneration and joint destruction which are not visualized on our case and the patient does not have a history of a disorder, such as diabetes mellitus, which would cause a neuropathic joint.

Osteonecrosis by definition is bone death. There are a number of etiologies including corticosteroid therapy, sickle cell anemia, renal transplantation, alcoholism, and a variety of vasculitites including systemic lupus. About the knee, however, there is a distinct form of osteonecrosis which may occur in a spontaneous or idiopathic fashion.

Radiographically this is very difficult to differentiate from osteochondritis dissicans, which many of you may have selected as the possible diagnosis in this case. Spontaneous osteonecrosis of the knee (SONK) is characterized by an immediate onset of pain in the knee which is almost always confined to the medial aspect of the joint. The typical patient is usually an older female with pain, tenderness, swelling, and restricted motion. Our current case demonstrated quite classic clinical findings that are often associated with SONK. It is not uncommon for the initial radiographs to be normal and a period of weeks or months may pass before the subtle flattening of the weight bearing articular surface of the medial femoral condyle is visualized. After a subtle flattening occurs, a radiolucent defect in the condyle is detected and often times within this lucent area a radiodense line, which consists of cartilage and subchondral bone, can be frequently identified. In our case, the initial trauma occurred and the radiographs were normal. Less than a year later, there is quite prominent flattening and a lucent defect in the weight bearing portion of the medial femoral condyle. In our case, the radiodense line of cartilage and subchondral bone within the defect was not visualized. On magnetic resonance imaging, the fragment was present inside this defect, however.

Typically, if this is not treated, progression will occur in the collapse and fragmentation of the weight bearing surface of the medial femoral condyle leading to quite advanced secondary degenerative joint disease in this region. Often times, the secondary degenerative joint disease can become so prominent that it obscures the underlying osteonecrosis making this diagnosis difficult to make over time. Secondary progressive varus deformity can also occur as the rapid joint space narrowing occurs at the medial joint margin. In our case, there was increased varus deformity on the follow-up weight bearing AP knee films compared to the previous study due to quite prominent moderate joint space narrowing at the medial compartment. Secondary osteoarthritis had already developed in less than a year's time.

The precise location of SONK is the weight bearing surface of the medial femoral condyle in contrast to osteochondritis dissicans which is typically present at the lateral aspect of the medial femoral condyle at a non weight bearing portion of the articular margin. Classically, osteochondritis dissicans occurs in adolescence with a rotational type of shearing trauma to the knee. Osteochondritis dissicans will lead to intra-articular cartilaginous loose bodies over time which can lead to significant symptoms such as joint locking. Since spontaneous osteonecrosis of the knee is not visualized after the initial trauma, it seems appropriate that with the classic clinical findings, advanced imaging should be considered to make the appropriate diagnosis in the early stages of true osteonecrosis in this region. In this regard, scintigraphic examination using bone-seeking radio pharmaceutical agents has shown promise in discovering spontaneous osteonecrosis about the knee when radiographic findings are not visualized. The problem with bone scanning is this is quite sensitive but not specific for a specific cause of increased bone turnover in this region. This does also not look for underlying ligamentous instability and \ or associated meniscal tears which seem to be extremely common at the medial aspect of the joint in patient's with SONK.

MR imaging is probably the most appropriate follow-up for spontaneous osteonecrosis of the knee. Classically, low signal intensity in the affected region on T1 weighted images and surrounding high signal intensity on T2 weighted images, likely representing bone marrow edema, is characteristic. Other findings on MRI include the presence of cystic lesions, bone collapse, and ,quite interestingly, tears or degeneration of the adjacent meniscus. MRI can also give an accurate measurement of the size of the lesion which may have prognostic significance in that larger legions are typically associated with the poor clinical outcome. MRI can also identify the separate cartilaginous and/or osseous fragments lying in situ within the defect, slightly displaced, or even free floating within the articular cavity.

The etiology of spontaneous osteonecrosis of the knee is still not extremely clear. Typically there is at least minimal traumatic insult which probably produced micro factors in the subchondral bone plate which can lead to overlying osseous and cartilaginous collapse. Another interesting correlation is the association with meniscal injury on the same side as the spontaneous osteonecrosis of the knee. Meniscal tears have been questioned in the pathogenesis of SONK and the impact of the articular surface against a fragmented meniscus during activity could result in the initial ischemic event at the medial femoral condyle. It is uncertain whether the sudden onset of pain which is characteristic of SONK could also be related to an acute tear of the meniscus in some cases.

In this current case, the most pertinent current concern is progressive collapse of the medial femoral condyle with continued weight bearing which may lead to continued varus angulation and could eventually force the hand of the orthopedic surgeon to consider a total knee replacement. Still, the age of the patient is a concern and the patient does not seem to be extremely interested in the surgical approach. It is uncertain whether limited weight bearing and using a walker periodically will be sufficient to reduce the progressive collapse and varus deformity at this articulation. It is uncertain whether nutritional supplementation or a conservative approach may be effective in slowing down this collapse. Certainly if you have any thoughts that may be helpful in this regard, please do not hesitate to call us here at the college as it would be refreshing to gain some added insight on this most interesting case.

Case courtesy of Jackie Rosencraz, DC.


Click here to continue to self assessment



Questions? Comments? Send e-mail to Northwestern Health Sciences University Web Coordinator.
2001 Northwestern Health Sciences University/ www.nwhealth.edu - All Rights Reserved
Copyright Information