Osteonecrosis of the distal femur.
The incidence of ON in the distal femur is second only to that of the femoral head, however, it still represents only 10% of all cases of avascular necrosis. (2) As a result most studies contain few patients, making definitive conclusions regarding natural history and appropriate treatment difficult. Early stages are generally treated nonoperatively while more advanced stages often require surgical management.
Primary spontaneous ON has been hypothesized to have a traumatic etiology. (3-5) It is thought that microfractures in the subchondral bone result from an accumulation of minor trauma. This allows synovial fluid to enter the bone, increasing edema, marrow pressure and eventually leading to ischemia. This theory is conjectural, as only a minority of patients report any preceding trauma. At present, two main theories regarding the etiology of secondary ON exist, neither of which have been proven. Interruption of the microvascular circulation in the subchondral bone may be caused by fat emboli or microthrombi, which then causes edema and increases the bone marrow pressure. This increased pressure further impedes circulation, resulting in eventual ischemia and necrosis. Steroid treatment has been associated with secondary ON. (6) Corticosteroids may increase the size of fat cells in the marrow. This factor may increase marrow pressure leading to eventual ischemia. Many other conditions have been associated with ON including: systemic lupus erythematosus (SLE), renal transplantation, alcoholism, Gaucher's disease, hemoglobinopathies, and Caisson's decompression disease. (7-11)
Arthroscopic surgery has rarely been associated with ON of the knee. (12) Whether arthroscopy can be considered a factor in the pathogenesis of ON remains controversial. Several studies have reported on "post-arthroscopic" ON as diagnosed by MRI. (13-16) Unfortunately most cases were diagnosed retrospectively and pre-existing ON was impossible to rule out.
Reddy and Frederick (17) identified a distinct difference between the lateral and medial femoral condyles with respect to both intraosseous and extraosseous blood supply in a cadaveric dye injection study. They hypothesize that this difference may explain the increased frequency with which the medial femoral condyle becomes osteonecrotic. The superior lateral and inferior lateral genicular arteries combine to supply the lateral femoral condyle, while the superior medial genicular artery and other lesser branches of the popliteal artery supply the medial femoral condyle. The intraosseous supply to the lateral condyle consists of an arcade that supplies multiple branches to the subchondral bone, as opposed to the medial femoral condyle, which has a single nutrient vessel to the subchondral bone and a potential watershed area in the anterior femoral condyle. They note that the femoral tunnel used in PCL reconstruction is in close proximity to the major extraosseous blood supply to the medial femoral condyle and may be at risk during the operation.
Patel and colleagues recommended classifying ON of the distal femur into two categories: 1. spontaneous or idiopathic ON and 2. secondary ON. (18) Spontaneous ON typically affects patients over the age of 60. (5,19) Women are affected three times more frequently than men. Pain usually begins suddenly, is well-localized to the affected condyle, and may be worsened with weightbearing. Acutely, a mild synovitis and knee effusion with limitation of full knee motion may be noted.
Spontaneous ON has recently been reported to be a form of insufficiency fracture. Yamamoto and Bullough (20) reported on an elderly man who sustained an insufficiency fracture of his femoral head and who was treated with total hip arthroplasty. He subsequently sustained a similar insufficiency fracture with subchondral collapse in the medial femoral condyle for which they performed arthroscopic debridement. Pathologic analysis was consistent with healing callus with granulation tissue and a small amount of necrotic bone between the fracture line and the articular surface. Because there was no evidence of a well-demarcated wedge-shaped area of ON, they concluded the fracture preceded the necrosis.
Yamamoto and Bullough (20) presented the following characteristics of an insufficiency fracture: elderly patient, overweight, and osteopenic. Shortly after the onset of pain, radiographic analysis is negative. They noted that pain appears out of proportion to radiographs. Magnetic resonance (MR) imaging may show a pattern of bone marrow edema that appears as a region of low intensity on T1-weighted images. This low intensity region corresponds to the fracture line and repair tissue and is parallel to the articular surface. This differs from the shape of the low intensity band in osteonecrosis, which is typically concave to the articular surface. Typical ON is not normally associated with extensive callus formation and granulation tissue near the fracture line. This callus and granulation tissue indicates a healing process at work, which may indicate a potential for resolution of the fracture.
Secondary ON occurs as a result of another process such as steroid therapy, SLE, alcoholism, Gaucher's disease, hemoglobinopathies, renal transplantation, or decompression sickness. Most secondary ON is due to steroid therapy. Secondary ON is three to four times more common in women than in men. The typical patient is younger than 55, most commonly in the fourth decade of life. It affects the lateral femoral condyle in 60%, is bilateral in 50%, and may be multifocal. (18) Pain may be medial or lateral and is often insidious in onset. Patients may develop symptoms of internal derangement. (21)
Lotke and associates (22) classified ON of the medial femoral condyle into three groups based on the area of necrosis. They proposed the ratio between the transverse width of the lesion and the width of the affected femoral condyle. Group I patients have the typical clinical appearance with a positive bone scan but no radiographic evidence of the disease. These patients are generally asymptomatic by 15 months. Group II patients have small lesions measuring less than 50% of the diameter of the condyle. Group III patients have large lesions greater than 50% of the diameter of the condyle that lead to rapid collapse of the affected joint. As such, the recommendation is to intervene surgically prior to development of a fixed deformity in group III patients, treating others nonoperatively until late degenerative changes become symptomatic. Aglietti and coworkers (23) used a similar ratio to estimate the size of the lesion, noting that the ratio is not subject to magnification, and suggested that unfavorable outcomes are more likely with a ratio greater than or equal to 0.40. They concluded that an area above 5 [cm.sup.2] is more likely to have a poor result. Muheim and Bohne also correlated the size of the necrotic lesion with prognosis. (24) They also multiplied the width of the lesion by its length using AP and lateral radiographic views. A product greater than 5 cm2 was associated with a poor prognosis. Attempts at determining the volume of a lesion proved technically difficult.
Bone Marrow Edema Syndrome
Transient osteoporosis has been reported in the knee, (25) although less frequently than in the hip. It has been noted that the term "bone marrow edema syndrome" (BMES) may be more appropriate, as osteoporosis is not seen on biopsy of this lesion. (26) Papadopoulos and colleagues (25) reported a good result in a 45-year-old male who underwent extra-articular core decompression after failing 8 weeks of nonoperative treatment for BMES. Histologic analysis confirmed absence of osteoporosis or osteonecrosis. In this clinical entity, radiographic examination is negative, a bone scan is positive, as is MR imaging. Findings on MR images include decreased uptake on T1, and increased uptake on T2-weighted images.
The typical clinical course of BMES is indolent. Complications including pathologic fracture or progression to AVN have been reported. (27,28) Terms used to describe similar clinical entities include: transient osteoporosis, migratory osteolysis, idiopathic or regional osteoporosis, and algo-dystrophy. Although clinical and imaging similarities exist between BMES and reflex sympathetic dystrophy, atrophic changes in superficial tissues are absent in BMES. (25)
Some investigators believe BMES represents an early, reversible stage of osteonecrosis. (28-31) The BMES pattern may be more diffuse, however progression to osteonecrosis may be inhibited by abundant new bone formation. (26) Koo and associates (32) suggest that BMES and osteonecrosis may be at opposite ends of a continuum of tissue sensitivity to an ischemic event. If ischemia reaches a certain threshold, progression to AVN is more likely. Only a few case reports of BMES about the knee exist in the literature, (33-42) as such most information on BMES is related to the hip and has been extrapolated to the knee.
Bone marrow edema syndrome affects women at one-third the rate of men. Affected men are generally middle-aged. When women are affected, it is almost exclusively during the third trimester of pregnancy. (29) Hofmann (29) found patients with BMES have similar risk factors to patients with osteonecrosis, again suggesting a common etiology may exist. The lateral femoral condyle is more frequently involved. Treatment is controversial. Some believe that the risk of BMES progressing to AVN outweighs the minimal risk of core decompression and recommend performing this procedure more aggressively, (29) while others believe the good prognosis of BMES does not justify surgical intervention, reserving core decompression for true AVN. (43,44)
Bone marrow edema syndrome involves three stages. Stage one begins abruptly. Up to 30% of patients may report unusual activity preceding symptoms. Pain is improved with rest and exacerbated by weight bearing. X-rays are negative in stage one which lasts for about one month. Stage two may last up to two months during which time osteopenia may be seen on plain radiographs. In stage three, symptoms resolve and bone density is restored. This stage may be four months in duration. The average course of disease is 6 months, but may take up to 2 years. The key characteristic of BMES is the delay between symptom onset and appearance of osteopenia.
Schweitzer and White (45) were able to induce marrow edema in normal volunteers by altering weightbearing. They caused overpronation with a shoe insert and performed sequential MR imaging studies. Foot bones were most commonly affected followed by the tibia and femur. In most cases, MR imaging revealed diffuse bone marrow edema, while in two volunteers the changes were more similar to stress fractures. These investigators note, however, that because patients' symptoms were relieved immediately after removing the shoe insert and remained asymptomatic through one month follow-up, the MR image changes were not actually stress fractures. Due to the study design, histologic analysis was obviously impossible and therefore no correlation with tissue condition could be made.
Staging of Spontaneous Osteonecrosis
Koshino (46) developed the staging system for spontaneous osteonecrosis shown in Table 1. Aglietti (23) modified this staging system to include five stages (Table 2). Mont and colleagues (7) adapted the Ficat and Arlet (47) staging system used in ON of the hip for use in the knee; it is primarily a radiograph-based system consisting of four stages (Table 3).
Nonoperative treatments are most appropriate for asymptomatic patients and those in early stages of the disease. Protected weightbearing, analgesics, non-steroidal anti-inflammatory drugs (NSAIDs), and physical therapy for quadriceps and hamstring strengthening have all been used. Spontaneous ON may respond well to nonoperative treatment. Lotke and colleagues reported 35 of 36 knees functioning well when treated nonoperatively for stage I spontaneous ON. (22) Lesions occasionally may spontaneously resolve. This is more likely in early stages and with smaller lesions. Ahlback and associates (1) found lesions less than 0.24 [cm.sup.2] could resolve, even after several months.
Uchio and coworkers, in a randomized controlled trial, concluded that use of a lateral wedge insole is an effective treatment for early spontaneous osteonecrosis affecting the medial femoral condyle. (48) They found a significant improvement in the size of the necrotic area, as well as pain in response to lateral wedge insole use. They caution that outcome was not affected in knees with advanced arthritic changes or significant varus alignment. Secondary ON may not do as well with nonoperative treatment. Mont and colleagues found only 59% of knees treated nonoperatively for secondary ON had good or excellent results at only 2 years follow-up. By 6 years, only 18% of knees survived nonoperative treatment and the stage of disease was found to contribute little to the ultimate result. (7)
A variety of surgical techniques have been performed for ON including arthroscopic debridement, osteotomy, core decompression, bone grafting, and prosthetic replacement. Indications for surgery are based upon the stage and extent of disease as well as patient age and demands.
Wiedel (21) reported successful treatment of mechanical symptoms and pain in knees involved with steroid induced ON at an average follow-up interval of 40 months via arthroscopic debridement of articular cartilage flaps, loose bodies, and degenerative meniscal tears. In a small series, Miller concluded debridement does not alter the natural course of the disease, but may relieve mechanical symptoms. (49) If fluid extravasation through subchondral fissures occurs, intraosseous pressure may increase, possibly worsening ischemia.
High Tibial Osteotomy
High tibial osteotomy (HTO) may be used to shift the mechanical axis of the extremity such that the affected femoral condyle experiences decreased load. Use of HTO necessitates only one affected condyle, therefore it is more commonly used in cases of spontaneous ON. Koshino (46) compared results of drilling or bone grafting of osteonecrotic lesions of the femoral condyle with or without high tibial osteotomy to correct angular deformity. In this study, both radiographic and clinical improvement were greater in those knees in which tibial osteotomy was performed concomitantly. Because results with valgus osteotomy for medial condyle involvement were best in those knees with a postoperative femorotibial angle of 164[degrees] to 173[degrees] and 7[degrees] to 16[degrees] of anatomic valgus, Koshino recommended a goal of 10[degrees] of valgus. The postoperative knee score was inversely correlated with the preoperative Koshino stage and therefore earlier operative intervention was recommended. (46) At a mean follow-up of five years, only one patient required total knee replacement, while 95% experienced pain relief and improved function. Aglietti and coworkers reported 87% good and excellent results at 6.2 years when using HTO with or without additional bone grafting. (23) In their study, necrosis improved in 58% while articular deterioration was seen in 29%. Soucacos and colleagues, (19) in a retrospective review of 32 patients, recommended HTO in patients less than 60 years of age with primary osteonecrosis stage III involving less than 50% of the condyle. Otherwise prosthetic replacement produced better results. Use of HTO in secondary ON is usually not recommended as patients may have bicondylar or more extensive involvement.
Core decompression is thought to decrease intraosseous pressure, allowing improved vascularity and possibly slowing the progression of the disease. In a procedure similar to that used in the hip, a guide wire may be placed into the osteonecrotic bone in an antegrade direction. A 6 mm or 8 mm cannulated drill is then used to decompress the bone without violating the articular surface. (52,53) Multiple smaller drill holes, either antegrade or retrograde, may be used to decompress and stimulate osteonecrotic lesions to heal. (47) A 0.62 Kirschner wire may be used for this technique.
Lotke and associates (22) did not recommend core decompression; it did not change the natural history in their series of 79 patients with osteonecrosis of the medial femoral condyle stage I.
An extra-articular drilling procedure was used to achieve core decompression in 16 patients with primary ON in a study by Forst and coworkers. (50) Fifteen of these patients had stage I disease. This resulted in immediate pain relief in all cases and total resolution of marrow signal in all but one case by 6 months postoperatively. Nonoperatively managed patients continued to have pain for several months. These investigators concluded that extra-articular drilling for early stage ON of the femoral condyle will relieve pain, but that once flattening of the condyle appears, progression of disease cannot be prevented.
Jacobs and colleagues (51) reported their results of 28 core decompressions performed for mainly secondary AVN (26 of 28 cases, 2 idiopathic). Their results were stage dependent. All patients with Ficat stage I and II had good results at a mean of 54 months follow-up. Those with stage III osteonecrosis had more unpredictable results; however despite a 52% success rate they suggest this procedure may have a role in the young patient, as the alternative may be total knee replacement. They made the important point that core decompression should not be considered an alternative to femoral or tibial osteotomy to correct malalignment.
In a retrospective analysis of 102 knees with secondary ON from steroid use, Mont and associates (7) initially treated all patients nonoperatively; 47 knees that remained symptomatic after a minimum of three months of protected weightbearing underwent core decompression. Nonoperative treatment yielded dismal results with only 18% of knees classified as good or excellent according to the Knee Society score at six years, whereas the core decompression group had 73% good to excellent results at 11-years follow-up. This difference was even more dramatic as the Ficat stage increased (86% in the core group versus 15% in the nonoperative group). They also reported reduced radiographic progression in the core group (35% progression to stage III or IV in the core group versus 75% progression in the nonoperative group).
Goitz (52) recommended the addition of an injectable calcium sulfate-based bone putty and bone marrow aspirate into the defect created by core decompression in an attempt to expedite filling of the defect and prevent potential bony collapse.
Core decompression is a relatively minor procedure. Patients often feel immediate pain relief and it may slow the progression of disease. It should be done prior to articular collapse. Salvage operations are not compromised by this procedure.
Autologous Bone Grafting
Perez and coworkers (53) reported an excellent result in a single patient at 2-year follow-up after using an autologous iliac crest bone plug followed by CPM and 8 weeks of non-weightbearing in a patient with AVN grade 3 of the lateral condyle.
Fukui and colleagues (54) used autologous osteoperiosteal iliac crest bone graft in the treatment of femoral condyle ON secondary to steroid use. In their series, 90% of knees achieved good pain relief and maintained good range of motion at mean 79-month follow-up. (54) All seven patients in this study were able to walk an unlimited distance after rehabilitating from the procedure. Osteoperiosteal graft was harvested from the ipsilateral iliac crest and press-fit into the osteochondral defect at the site of osteonecrosis. The shape and contour of bone was chosen to match as well as possible the area of the femoral condyle affected. The periosteum was left intact, and continuous passive motion was used to stimulate chondrogenic potential postoperatively. Whether this actually resulted in chondrogenesis was not determined, however the investigators note that no deterioration of function occurred at the latest follow-up examination. The one patient in this study with a poor result had extensive areas of necrosis, which did not provide a stable base for the graft, which subsequently migrated. These investigators recommended realignment correction if mechanical axis deviation exists and urged careful planning for defects wider than 3 cm, as graft width is limited. They concluded that this technique is useful in treatment of osteonecrosis as it may allow satisfactory function, and maintain bone stock, even if a definitive surgery is needed in the future.
Kotani and associates (55) reported on 16 patients all with satisfactory results at 5 years following use of autologous osteoarticular grafting with the osteoarticular transfer system (OATS) for Lotke 2-3 lesions. They emphasized the point that these patients did not have varus alignment.
The use of osteochondral allograft to replace osteonecrotic bone has had mixed results in the literature. Bayne and coworkers reported 66% good or excellent results in posttraumatic ON, while spontaneous or secondary osteonecrotic knees did poorly with one exception. (56) They felt that elderly patients with spontaneous ON were unable to comply with postoperative restricted weightbearing leading to subsidence and fragmentation of the grafts while poor vascularization in secondary ON prevented incorporation of the grafts.
Meyers and colleagues reported on the use of fresh osteochondral allograft for a variety of diagnoses. (57) All three patients treated with fresh osteochondral allograft for osteonecrosis had excellent results at 3- to 4-year follow-up. They identified eight conditions necessary for successful results using osteochondral allograft:
1. Osteochondral allograft must be fresh, and include no more than 0.5 to 1 cm of subchondral bone;
2. The donor and recipient sizes must be precisely matched to obtain congruity of articular surfaces;
3. Graft must be "shimmed" with autologous bone to ensure congruous contact;
4. A firm press-fit must be obtained;
5. The joint must be stable;
6. The synovial membrane must be closed to contain synovial fluid, which provides nourishment;
7. Non-weightbearing must be strictly enforced until complete incorporation of graft; and
8. Joint motion must be started early.
Flynn and associates reported on the use of fresh frozen allograft in secondary ON in patients younger than 50 years of age. (58) At a mean of 4.2 years, 70% of knees had good or excellent results. The osteoarticular allografts in their study contained substantial amounts of bone, as they were used to substitute for large osteonecrotic lesions. Frozen allograft may be less immunogenic than fresh allograft bone, (59) however post-mortem histologic inspection of fresh osteochondral allografts has revealed that human chondrocytes may survive as long as 8 years. (56, 60) This technique may best be used as an adjunct with another procedure like HTO, which may protect the graft by shifting the mechanical axis of the limb, especially in a young patient facing prosthetic replacement.
Unicompartmental Knee Arthroplasty
When considering prosthetic replacement, unicompartmental arthroplasty is an attractive option as there may be less blood loss, less bony resection, quicker rehabilitation, and more normal knee kinematics. It is important to recognize that unicondylar arthroplasty is unable to make large corrections to the mechanical axis. If there is significant malalignment, a total knee replacement would be indicated.
Atsui and coworkers61 found unicompartmental knee arthroplasty to be a useful procedure in ON with the following indications: unicompartmental ON, body weight less that 155 pounds, age greater than 60 with low expected activity level, flexion contracture less than 30[degrees], less than 10[degrees] of angular knee deformity, and intact cruciate and collateral ligaments. These investigators aimed for a femoral tibial angle of 175[degrees], 5[degrees] of valgus, and reported good to excellent results in 100% of ten patients at an average of 42 months follow-up.
In a review of 34 unicompartmental arthroplasties performed for ON of the medial femoral condyle at an average of 5.5 years follow up, Marmor (62) reported 89% good to excellent results. Two of the four failures in his series were due to development of lateral condyle ON requiring revision to TKA. Cartier and colleagues (63) found that 95% of patients maintained good and excellent results at 54 months following unicondylar replacement for osteonecrosis.
Lotke and associates (22) recommend treatment of their stage II-III lesions with arthroplasty. Unicondylar arthroplasty may be adequate if the lesion is confined to a single condyle. Larger lesions may develop fixed deformities if surgical intervention is delayed and may require total knee replacement.
Unicondylar knee replacement is probably best used in spontaneous ON, affecting only one compartment in a patient who is older and lighter. It is appropriate to order a preoperative MRI when considering unicondylar replacement to rule out the presence of osteonecrotic lesions in the opposite condyle. If the other compartment is affected, or in cases of secondary osteonecrosis, total knee replacement may provide a more predictable result.
Total Knee Arthroplasty
Several studies have focused on total knee replacement for ON and results have been mixed. Ritter and associates (64) compared survivorship of total knee replacement (TKR) for primary ON to that for osteoarthrosis and reported no statistical difference with regard to pain, revision, or radiolucency between these groups. Limitations of this study included a small study population consisting only of females and a high drop out rate. Their revision rate was 17% at seven years. Aglietti and coworkers23 recommended TKR for stage III or greater primary ON. This procedure yielded 95% satisfactory results at 4.4 years. They caution against use of unicondylar replacement as they found a 28% revision rate following initial good results. Radke and colleagues (65) also concluded that although unicondylar replacement may provide better short-term results, total knee replacement would do better over the long-term.
Bergman and Rand (66) reported 87% good and excellent results with TKR for secondary ON at 4 years, with predicted 85% survivorship at 5 years. In this study, no failures were due to aseptic loosening and the authors noted that the osteonecrotic bone did not provide any problems in management. (66)
Seldes and coworkers (67) reported 84% survivorship at 5 years in steroid-induced ON of the knee. Three failures were due to loosening, while two were due to infection. They also found a 19% minor complication rate. They hypothesized this complication rate may be attributable to the fact that many patients who develop secondary ON have chronic diseases and are therefore physiologically weaker.
Mont and colleagues (68) reported a 55% success rate and a 37% loosening rate in a population of patients with secondary ON associated with lupus and steroid use. In a more recent study, they reported a discouraging 71% success rate using a variety of total knee prostheses for secondary ON over a 24-year period. (69) A follow-up study using modern prostheses with cement fixation and stems when needed yielded a 97% success rate at a minimum of 4-year follow-up. (70)
Although results of TKA for ON are not quite as good as those performed for osteoarthritis, it appears that addressing bony deficiency with cement augmentation and stems may narrow the difference. Total knee arthroplasty remains the most predictable surgical treatment modality, particularly in the elderly patient with ON.
As osteonecrosis of the knee is a relatively rare disease, the literature regarding treatment is sparse. There are no prospective, randomized trials comparing treatment options. Most investigators recommend non-operative treatment for small, asymptomatic lesions. Core decompression may provide pain relief if done prior to collapse, however it remains unclear if it changes the natural history of the disease. Prosthetic replacement remains the most predictable modality for treating advanced disease, however as compared to osteoarthritis, the complication rate may be higher and ultimate success rate slightly lower.
(1.) Ahlback S, Bauer GC, Bohne WH: Spontaneous osteonecrosis of the knee. Arthritis Rheum. 1968;11:705.
(2.) Mankin HJ: Nontraumatic necrosis of bone (osteonecrosis). N Engl J Med. 1992;326:1473.
(3.) Ahuja SC, Bullough PG: Osteonecrosis of the knee: a clinicopathological study in twenty-eight patients. J Bone Joint Surg Am. 1978;60:191.
(4.) Ecker ML, Lotk, PA: Spontaneous osteonecrosis of the knee. J Am Acad Orthop Surg. 1994;2:173.
(5.) Lotke PA, Ecker ML: Osteonecrosis of the knee. J Bone Joint Surg Am. 1988;70:470.
(6.) Cruess RL: Steroid-induced osteonecrosis: a review. Can J Surg. 1981;24:567.
(7.) Mont MA, Tomek IM, Hungerford DS: Core decompression for avascular necrosis of the distal femur: long-term followup. Clin Orthop. 1997;334:124.
(8.) McCarthy EF: The pathology of transient regional osteoporosis. Iowa Orthop J. 1998;18:35.
(9.) Isono SS, Wollson St, Shurman DJ: Total joint arthroplasty for steroid-induced osteonecrosis in cardiac transplant patients. Clin Orthop Relat Res. 1987 Apr;(217):201-8.
(10.) Hungerford DS, Zizic TM: Alcoholism associated ischemic necrosis of the femoral head. Early diagnosis and treatment. Clin Orthop Relat Res. 1978 Jan-Feb;(130):144-53.
(11.) Jones JPJ: Etiology and pathogenesis of osteonecrosis. Semin Arthroplasty. 1991;2:160.
(12.) DeFalco RA, Ricci AR, Balduini FC: Osteonecrosis of the knee after arthroscopic meniscectomy and chondroplasty: a case report and literature review. Am J Sports Med. 2003;31:1013.
(13.) Brahme SK, Fox JM, Ferkel RD, et al: Osteonecrosis of the knee after arthroscopic surgery: diagnosis with MR imaging. Radiology. 1991;178:851.
(14.) Muscolo DL, Costa-Paz M, Makino A, et al: Osteonecrosis of the knee following arthroscopic meniscectomy in patients over 50-years old. Arthroscopy 1996;12:273.
(15.) Prues-Latour V, Bonvin JC, Fritschy D: Nine cases of osteonecrosis in elderly patients following arthroscopic meniscectomy. Knee Surg Sports Traumatol Arthrosc. 1998;6:142.
(16.) Santori N, Condello V, Adriani E, et al: Osteonecrosis after arthroscopic medial meniscectomy. Arthroscopy. 1995;11:220.
(17.) Reddy AS, Frederick RW: Evaluation of the intraosseous and extraosseous blood supply to the distal femoral condyles. Am J Sports Med. 1998;26:415.
(18.) Patel DV, Breazeale NM, Behr CT, et al: Osteonecrosis of the knee: current clinical concepts. Knee Surg Sports Traumatol Arthrosc. 1998;6:2.
(19.) Soucacos PN, Xenakis TH, Beris AE, et al: Idiopathic osteonecrosis of the medial femoral condyle: classification and treatment. Clin Orthop. 1997;341:82.
(20.) Yamamoto T, Bullough PG: Spontaneous osteonecrosis of the knee: the result of subchondral insufficiency fracture. J Bone Joint Surg Am. 2000;82:858.
(21.) Wiedel JD: Arthroscopy in steroid-induced osteonecrosis of the knee. Arthroscopy. 1985;1:68.
(22.) Lotke, PA, Abend, JA, Ecker, ML: The treatment of osteonecrosis of the medial femoral condyle. Clin Orthop 171:109, 1982.
(23.) Aglietti P, Insall JN, Buzzi R, et al: Idiopathic osteonecrosis of the knee: aetiology, prognosis and treatment. J Bone Joint Surg Br. 1983;65:588.
(24.) Muheim G, Bohne WH: Prognosis in spontaneous osteonecrosis of the knee: investigation by radionuclide scintimetry and radiography. J Bone Joint Surg Br. 1970;52:605.
(25.) Papadopoulos EC, Papagelopoulos PJ, Kaseta M, et al: Bone marrow edema syndrome of the knee: a case report and review of the literature. Knee. 2003;10:295.
(26.) Plenk HJ, Hofmann S, Eschberger J, et al: Histomorphology and bone morphometry of the bone marrow edema syndrome of the hip. Clin Orthop. 1997;334:73.
(27.) Brodell JD, Burns JEJ, Heiple KG: Transient osteoporosis of the hip of pregnancy: two cases complicated by pathological fracture. J Bone Joint Surg Am. 1989;71:1252.
(28.) Turner DA, Templeton AC, Selzer PM, et al: Femoral capital osteonecrosis: MR finding of diffuse marrow abnormalities without focal lesions. Radiology. 1989;171:135.
(29.) Hofmann S, Engel A, Neuhold A, et al: Bone-marrow oedema syndrome and transient osteoporosis of the hip: an MRI-controlled study of treatment by core decompression. J Bone Joint Surg Br. 1993;75:210.
(30.) Dihlmann W, Delling G: [Is transient hip osteoporosis a transient osteonecrosis?]. Z Rheumatol. 1985;44:82.
(31.) Jones JP: Osteonecrosis and bone marrow edema syndrome: similar etiology but different pathogenesis. In: Urbaniak JR, Jones JP (eds): Osteonecrosis: Etiology, Diagnosis, and Treatment. Rosemont, IL: American Academy of Orthopaedic Surgeons, 1997, p. 181.
(32.) Koo KH, Jeong ST, Jones JPJ: Borderline necrosis of the femoral head. Clin Orthop. 1999;358:158.
(33.) Boos S, Sigmund G, Huhle P, et al: [Magnetic resonance tomography of so-called transient osteoporosis. Primary diagnosis and follow-up after treatment]. Rofo. 1993 Mar;158(3):201-6. German.
(34.) Corbett M, Colston JR, Tucker AK: Pain in the knee associated with osteoporosis of the patella. Ann Rheum Dis. 1977;36:188.
(35.) Doury P, Delahaye RP, Granier R, et al: Highly localized transient osteoporosis of the knee. Arthritis Rheum. 1978;21:992.
(36.) Fertakos RJ, Swayne LC, Colston WC: Three-phase bone imaging in bone marrow edema of the knee. Clin Nucl Med. 1995;20:587.
(37.) Festini V: [A case from practice (340). Transient regional osteoporosis of the right knee]. Schweiz Rundsch Med Prax. 1996;85:68.
(38.) Lagier R: Partial algodystrophy of the knee: an anatomico-radiological study of one case. J Rheumatol. 1983;10:255.
(39.) Lechevalier D, Eulry F, Crozes P, et al: [In situ migratory algodystrophies of the knee. Value of modern imaging]. Rev Rhum Mal Osteoartic. 1992;59:29.
(40.) McAlindon TE, Ward SA, MacFarlane D, et al: Transient regional osteoporosis presenting as a septic arthritis. Postgrad Med J. 1993;69:871.
(41.) Parker RK, Ross GJ, Urso JA: Transient osteoporosis of the knee. Skeletal Radiol. 1997;26:306.
(42.) Wambeek N, Munk PL, Lee MJ, et al: Intra-articular regional migratory osteoporosis of the knee. Skeletal Radiol. 2000;29:97.
(43.) Guerra JJ, Steinberg ME: Distinguishing transient osteoporosis from avascular necrosis of the hip. J Bone Joint Surg Am. 1995;77:616.
(44.) Komatsu T, Kadoya Y, Minoda Y, et al: Transient osteoporosis of the femoral condyle: a case report. Knee. 2002;9:241.
(45.) Schweitzer ME, White LM: Does altered biomechanics cause marrow edema. Radiology. 1996;198:851.
(46.) Koshino T: The treatment of spontaneous osteonecrosis of the knee by high tibial osteotomy with and without bone-grafting or drilling of the lesion. J Bone Joint Surg Am. 1982;64:47.
(47.) Ficat P: [Vascular pathology of femoral head necrosis (author's transl)]. Orthopade. 1980;9:238.
(48.) Uchio Y, Ochi M, Adachi N, et al: Effectiveness of an insole with a lateral wedge for idiopathic osteonecrosis of the knee. J Bone Joint Surg Br. 2000;82:724.
(49.) Miller GK, Maylahn DJ, Drennan DB: The treatment of idiopathic osteonecrosis of the medial femoral condyle with arthroscopic debridement. Arthroscopy. 1986;2:21.
(50.) Forst J, Forst R, Heller KD, et al: Spontaneous osteonecrosis of the femoral condyle: causal treatment by early core decompression. Arch Orthop Trauma Surg. 1998;117:18.
(51.) Jacobs MA, Loeb PE, Hungerford DS: Core decompression of the distal femur for avascular necrosis of the knee. J Bone Joint Surg Br. 1989;71:583.
(52.) Goitz HT: Minimally invasive treatment of non-steroid induced knee osteonecrosis of the lateral femoral condyle. Orthopedics. 2004;27:s129.
(53.) Perez CL, Gomez CFJ, Gomez dAG, et al: Core decompression and arthroscopic bone grafting for avascular necrosis of the knee. Arthroscopy. 1996;12:323.
(54.) Fukui N, Kurosawa H, Kawakami A, et al: Iliac bone graft for steroid-associated osteonecrosis of the femoral condyle. Clin Orthop. 2002;401:185.
(55.) Kotani A, Ishii Y, Sasaki S: Autogenous osteochondral grafts for osteonecrosis of the femoral condyle. J Orthop Surg (Hong Kong). 2003;11:117.
(56.) Bayne O, Langer F, Pritzker KP, et al: Osteochondral allografts in the treatment of osteonecrosis of the knee. Orthop Clin North Am. 1985;16:727.
(57.) Meyers MH, Akeson W, Convery FR: Resurfacing of the knee with fresh osteochondral allograft. J Bone Joint Surg Am. 1989;71:704.
(58.) Flynn JM, Springfield DS, Mankin HJ: Osteoarticular allografts to treat distal femoral osteonecrosis. Clin Orthop. 1994;303:38.
(59.) Gertzbein SD, Tait JH, Devlin SR, et al: The antigenicity of chondrocytes. Immunology. 1977;33:141.
(60.) Gross AE, Silverstein EA, Falk J, et al: The allotransplantation of partial joints in the treatment of osteoarthritis of the knee. Clin Orthop. 1975;108:7.
(61.) Atsui K, Tateishi H, Futani H, et al: Ceramic unicompartmental knee arthroplasty for spontaneous osteonecrosis of the knee joint. Bull Hosp Jt Dis. 1997;56:233.
(62.) Marmor L: Unicompartmental arthroplasty for osteonecrosis of the knee joint. Clin Orthop. 1993;294:247.
(63.) Cartier P, Gaggiotti G, Jully JL: [Primary osteonecrosis of the medial femoral condyle. Unicompartmental or total replacement?]. Int Orthop. 1988;12(3):229-35. French.
(64.) Ritter MA, Eizember LE, Keating EM, et al: The survival of total knee arthroplasty in patients with osteonecrosis of the medial condyle. Clin Orthop. 1991;267:108.
(65.) Radke S, Wollmerstedt N, Bischoff A, et al: Knee arthroplasty for spontaneous osteonecrosis of the knee: unicompartmental vs bicompartmental knee arthroplasty. Knee Surg Sports Traumatol Arthrosc. 2005 Apr;13(3):158-62.
(66.) Bergman NR, Rand JA: Total knee arthroplasty in osteonecrosis. Clin Orthop. 1991;273:77.
(67.) Seldes RM, Tan V, Duffy G, et al: Total knee arthroplasty for steroid-induced osteonecrosis. J Arthroplasty. 1999;14:533.
(68.) Mont MA, Myers TH, Krackow KA, et al: Total knee arthroplasty for corticosteroid associated avascular necrosis of the knee. Clin Orthop. 1997;338:124.
(69.) Mont MA, Baumgarten KM, Rifai A, et al: Atraumatic osteonecrosis of the knee. J Bone Joint Surg Am. 2000;82:1279.
(70.) Mont MA, Rifai A, Baumgarten KM, et al: Total knee arthroplasty for osteonecrosis. J Bone Joint Surg Am. 2002;84A:599.
Christopher Gorczynski, M.D., is a Sports Medicine Fellow, and Robert Meislin, M.D., is an Assistant Professor of Orthopaedic Surgery, in the NYU-Hospital for Joint Diseases Department of Orthopaedic Surgery. New York, New York.
Correspondence: Robert Meislin, M.D., Assistant Professor of Orthopaedic Surgery, NYU-Hospital for Joint Diseases Department of Orthopaedic Surgery, 301 East 17th Street, New York, New York 10003.
Table 1 Staging System of Spontaneous Osteonecrosis Developed by Koshino (46) Stage I: Incipient Pain. Normal radiographs. Positive bone scan. Stage II: Avascular Subchondral radiolucency in weightbearing area. Distal sclerosis. Minimal articular change at arthroscopy. Stage III: Collapsed Calcified plate with radiolucency surrounded by sclerotic halo. Collapse of subchondral bone. Stage IV: Degenerative Osteoarthritic changes including joint space narrowing, osteophyte formation, and sclerosis. Table 2 Aglietti's (23) Modified Staging System for Spontaneous Osteonecrosis Stage I: Radiographs normal. If no change occurs within 6 months, the patient will remain in stage I. Stage II: Area of slight flattening of femoral condyle. Stage III: Radiolucency with distal sclerosis and halo of bony reaction. Crescent sign. Stage IV: Calcified plate, sequestrum or flap fragment with radiolucency and sclerotic halo. Stage V: Osteoarthritis Table 3 Staging of Spontaneous Osteonecrosis Modified from Ficat and Arlet (47) by Mont and Colleagues (7) Stage 1: Normal joint space, normal joint contour, areas of mottles osteoporosis. Stage II: Normal joint space, normal joint contour, wedge sclerosis. Stage III: Slightly decreased joint space, subchondral collapse, sequestrum appearance. Stage IV: Decreased joint space, collapse, destruction of trabecular pattern.