Osteonecrosis is a focal infarct of bone that may be caused by specific etiologic factors or may be idiopathic. It can cause pain, limitation of motion, joint collapse, and secondary osteoarthritis. Diagnosis is by radiographs and MRI. In early stages, surgical procedures may slow or prevent progression. In later stages, joint replacement may be required for relief of pain and maintenance of function.
In the United States, osteonecrosis affects approximately 20,000 new patients annually (1). The hip (femoral head) is most commonly affected, followed by the knee and shoulder (humeral head). The wrist and ankle are less often involved. It is unusual for osteonecrosis to involve the shoulder or other less commonly affected sites without the hip also being involved. Medication-related osteonecrosis of the jaw (MRONJ) has been associated rarely with antiresorptive therapy for osteoporosis. MRONJ has characteristics that differ from osteonecrosis at other sites.
General reference
1. Moya-Angeler J, Gianakos AL, Villa JC, Ni A, Lane JM. Current concepts on osteonecrosis of the femoral head. World J Orthop 2015;6(8):590-601. doi:10.5312/wjo.v6.i8.590
Etiology of Osteonecrosis
Osteonecrosis can be idiopathic, caused by trauma, or due to nontraumatic factors. The most common cause of osteonecrosis is trauma (1, 2).
Traumatic osteonecrosis
The most common cause of traumatic osteonecrosis of the femoral head is a displaced subcapital fracture of the proximal femur; osteonecrosis is uncommon after intertrochanteric fractures. The incidence of osteonecrosis after hip dislocation is related primarily to the severity of the initial injury but may be higher if the dislocation is not promptly reduced. Fracture or dislocation may cause osteonecrosis by grossly disrupting or compressing nearby blood vessels.
Nontraumatic osteonecrosis
Nontraumatic osteonecrosis affects men more often than women and occurs primarily in patients between ages 30 and 50 years (1). Nontraumatic osteonecrosis of the hip is bilateral in approximately 60% of patients (2).
The most common factors causing or contributing to nontraumatic osteonecrosis are (3, 4, 5)
Corticosteroid use (particularly with prolonged use and/or higher doses)
Excessive alcohol consumption
Other nontraumatic risk factors for osteonecrosis include
Chemotherapy
Coagulation disorders (eg, antiphospholipid syndrome, inherited thrombophilia, hypofibrinolytic disorders)
Hemoglobinopathy (eg, sickle cell disease)
Liver disease
Organ transplantation
Radiation
Smoking
Systemic lupus erythematosus (SLE) and other systemic rheumatic diseases
Tumors
Miscellaneous disorders (eg, chronic kidney disease, rare inherited metabolic disorders)
About 20% of cases of osteonecrosis are idiopathic.
> 20 mg/day for several weeks or months, or the cumulative dose is > 2000 mg (6). However, osteonecrosis has also been reported subsequent to much less corticosteroid exposure. Notably corticosteroid-associated osteonecrosis is often multifocal (affecting multiple different anatomic locations) and can affect weight-bearing joints as well as non–weight-bearing joints like the shoulders.
The risk of osteonecrosis is also increased when > 400 mL ethanol/week are consumed for more than 6 months (4).
Some genetic factors increase susceptibility to osteonecrosis. Subtle clotting abnormalities due to deficiencies in protein C, protein S, or antithrombin III, or anticardiolipin antibodies (see Overview of Thrombotic Disorders) can be detected in a high percentage of patients with osteonecrosis.
Some disorders that are common comorbidities of osteonecrosis are treated with corticosteroids (eg, systemic lupus erythematosus). Evidence suggests that the risk of osteonecrosis in many of these disorders is related primarily to the corticosteroid use rather than to the disorder.
Medication-related osteonecrosis of the jaw
The term "spontaneous osteonecrosis of the knee" (SONK) is a misnomer. This condition normally occurs in older patients and is localized to the femoral condyle or tibial plateau (7). SONK is thought to be caused by an insufficiency fracture (a type of fragility fracture caused by normal wear and tear on osteoporotic bone that occurs without direct trauma). True osteonecrosis of the knee can result from trauma or any of the nontraumatic risk factors of osteonecrosis.
Etiology references
1. Ikeuchi K, Hasegawa Y, Seki T, Takegami Y, Amano T, Ishiguro N. Epidemiology of nontraumatic osteonecrosis of the femoral head in Japan. Mod Rheumatol 2015;25(2):278-281. doi:10.3109/14397595.2014.932038
2. Hauzeur JP, Malaise M, de Maertelaer V. A prospective cohort study of the clinical presentation of non-traumatic osteonecrosis of the femoral head: spine and knee symptoms as clinical presentation of hip osteonecrosis. Int Orthop 2016;40(7):1347-1351. doi:10.1007/s00264-015-3079-x
3. Yoon BH, Jones LC, Chen CH, et al. Etiologic Classification Criteria of ARCO on Femoral Head Osteonecrosis Part 1: Glucocorticoid-Associated Osteonecrosis. J Arthroplasty 2019;34(1):163-168.e1. doi:10.1016/j.arth.2018.09.005
4. Yoon BH, Jones LC, Chen CH, et al. Etiologic Classification Criteria of ARCO on Femoral Head Osteonecrosis Part 2: Alcohol-Associated Osteonecrosis. J Arthroplasty 2019;34(1):169-174.e1. doi:10.1016/j.arth.2018.09.006
5. Fukushima W, Fujioka M, Kubo T, Tamakoshi A, Nagai M, Hirota Y. Nationwide epidemiologic survey of idiopathic osteonecrosis of the femoral head. Clin Orthop Relat Res. 2010;468(10):2715-2724. doi:10.1007/s11999-010-1292-x
6. Mont MA, Pivec R, Banerjee S, Issa K, Elmallah RK, Jones LC. High-Dose Corticosteroid Use and Risk of Hip Osteonecrosis: Meta-Analysis and Systematic Literature Review. J Arthroplasty 2015;30(9):1506-1512.e5. doi:10.1016/j.arth.2015.03.036
7. Sibilska A, Góralczyk A, Hermanowicz K, Malinowski K. Spontaneous osteonecrosis of the knee: what do we know so far? A literature review. Int Orthop 2020;44(6):1063-1069. doi:10.1007/s00264-020-04536-7
Pathophysiology of Osteonecrosis
Osteonecrosis involves the death of cellular elements of the bone marrow. Mechanisms of nontraumatic osteonecrosis may include embolization by blood clots or lipid droplets, intravascular thrombosis, and extravascular compression (1).
After the vascular insult, the repair processes attempt to remove necrotic bone and marrow and replace them with viable tissue. For example, if an infarct in the hip is small, particularly if it is not subject to major weight bearing, these processes may succeed and the femoral head remains round. However, in approximately 80% of patients, especially if the infarct is large and in the weight-bearing area, collapse of the infarcted area outstrips attempts at repair, and the infarcted area collapses (2). The femoral head is no longer round.
Because osteonecrosis usually affects the ends (epiphysis and metaphysis) of long bones, the overlying articular cartilage surface becomes flattened and irregular, with areas of collapse that may eventually lead to osteoarthritis and increased pain.
Pathophysiology references
1. Shah KN, Racine J, Jones LC, Aaron RK. Pathophysiology and risk factors for osteonecrosis. Curr Rev Musculoskelet Med 2015;8(3):201-209. doi:10.1007/s12178-015-9277-8
2. Mont MA, Zywiel MG, Marker DR, McGrath MS, Delanois RE. The natural history of untreated asymptomatic osteonecrosis of the femoral head: a systematic literature review. J Bone Joint Surg Am 2010;92(12):2165-2170. doi:10.2106/JBJS.I.00575
Symptoms and Signs of Osteonecrosis
General symptoms
Areas affected by osteonecrosis may remain asymptomatic for weeks to months after the vascular insult. Usually pain then develops gradually, although it may be acute. With progressive collapse of the joint, pain increases and is exacerbated by motion and weight bearing and is relieved by rest.
Because many of the risk factors for the development of osteonecrosis act systemically (eg, corticosteroid use, excessive alcohol intake, sickle cell disease), osteonecrosis may be multifocal. In sickle cell disease, osteonecrosis may occur throughout different long bones and cause sudden episodic pain.
Joint-specific symptoms
Osteonecrosis of the hip causes groin pain that may radiate down the thigh or into the buttock. Motion becomes limited, and a limp usually develops.
SONK usually causes sudden knee pain without preceding trauma; the sudden onset and the location of pain may help differentiate it from classical osteonecrosis. This pain is most often on the medial side of the femoral condyle or tibial plateau and manifests with tenderness, noninflammatory joint effusion, painful motion, and a limp.
Osteonecrosis of the humeral head usually causes less pain and disability than hip or knee involvement, but pain and disability may be significant in patients who use crutches to bear weight.
With advanced osteonecrosis, patients have pain and decreased motion, although passive range of motion is less affected than active range of motion. Symptomatic synovial effusions can occur, especially in the knee, and the fluid is noninflammatory.
Diagnosis of Osteonecrosis
Radiographs
MRI
Osteonecrosis should be suspected in patients with the following:
Previous fractures (eg, displaced subcapital femoral head fractures), joint dislocations (eg, hip dislocations), or other risk factors (eg, corticosteroid use, hemoglobinopathies), particularly if pain persists or worsens
Persistent spontaneous hip, knee, or shoulder pain, particularly if risk factors for osteonecrosis are present
Radiographs should be done initially (1). They may show no abnormalities for months. The earliest findings are localized areas of sclerosis and lucency. Later, a subchondral crescent sign due to a fracture in the epiphysis may appear. Then, gross collapse and flattening of the articular surface is seen, followed by advanced degenerative changes. When the cause is systemic, lesions may be multiple.
When clinical suspicion is high but radiographs are normal or nondiagnostic, an MRI, which is much more sensitive and more specific, should be done early in the course of diagnosis to avoid ongoing damage to weight-bearing joints that would limit the success of joint-saving procedures (2). Both hips should be imaged. CT is rarely needed, although it may occasionally be of value to detect joint collapse, which does not appear on radiographs and sometimes may not appear on MRI. Bone scans are less sensitive and less specific than MRI and are generally not done unless MRI and CT are contraindicated or unavailable.
Staging systems may be used to classify osteonecrosis. For example, the Association Research Circulation Osseous (ARCO) classification system for femoral head osteonecrosis is based on the size and location of the lesion, the spherical nature of the femoral head, and the presence of subchondral collapse or arthritis of the hip (3, 4).
Laboratory studies are usually normal and of little value in detecting osteonecrosis. However, they might help detect an underlying disorder (eg, coagulation defects, hemoglobinopathies, lipid abnormalities).
Diagnosis references
1. Expert Panel on Musculoskeletal Imaging, Ha AS, Chang EY, et al: ACR Appropriateness Criteria® Osteonecrosis: 2022 Update. J Am Coll Radiol 19(11S):S409-S416, 2022. doi:10.1016/j.jacr.2022.09.009
2. Boontanapibul K, Steere JT, Amanatullah DF, et al: Diagnosis of osteonecrosis of the femoral head: too little, too late, and independent of etiology. J Arthroplasty 35(9):2342-2349, 2020. doi: 10.1016/j.arth.2020.04.092
3. Yoon BH, Mont MA, Koo KH, et al: The 2019 Revised Version of Association Research Circulation Osseous Staging System of Osteonecrosis of the Femoral Head. J Arthroplasty 35(4):933-940, 2020. doi:10.1016/j.arth.2019.11.029
4. Koo KH, Mont MA, Cui Q, et al: The 2021 Association Research Circulation Osseous Classification for Early-Stage Osteonecrosis of the Femoral Head to Computed Tomography-Based Study. J Arthroplasty 37(6):1074-1082, 2022. doi:10.1016/j.arth.2022.02.009
Treatment of Osteonecrosis
Symptomatic measures (eg, rest, physical therapy, nonsteroidal anti-inflammatory drugs [NSAIDs])
Surgical decompression or other procedures to preserve the joint
Joint replacement
Nonsurgical treatments
Small, asymptomatic lesions, especially those not in weight-bearing areas, may heal spontaneously and may not need treatment. However, lesions that are symptomatic or medium-sized or greater are often treated (for example, with a simple procedure such as core decompression) to improve the chance that lesions heal without joint collapse (1). With or without other treatment, contributing factors should be mitigated (eg, limiting corticosteroid use) to promote healing.
Larger lesions, both symptomatic and asymptomatic, have a poor prognosis if untreated, especially when in the femoral head. Therefore, early treatment to slow or prevent progression and save the joint is desirable. No completely effective treatment is available. Medical therapy (eg, anticoagulation, vasodilators, oral bisphosphonates) and physical modalities (eg, electromagnetic fields,extracorporal shock waves) have shown promise in limited studies but are not routinely used (1). Limited weight bearing or non–weight bearing alone has not been shown to improve long-term outcome.
SONK is usually treated without surgery because the fracture heals with time. Analgesics and crutch use to control the pain, progressive weight bearing, and physical therapy to maintain strength and range of motion are useful adjuncts.
Surgical treatments
Surgical treatments for osteonecrosis are most effective when done before joint collapse. They have been used most often in treating osteonecrosis of the femoral head where the prognosis without treatment is worse than that for osteonecrosis in other regions.
Core decompression is the procedure most frequently done; a drill is used to make one or more small tracks or perforations are made in the osteonecrotic area in an attempt to decrease intraosseous pressure and stimulate repair. Core decompression is technically simple, and the complication rate is very low if the procedure is done correctly. Protected weight bearing (bearing weight only as tolerated and with a mobility aid, such as crutches, a walker, or a cane) is needed for about 4 to 6 weeks. Most reports indicate satisfactory or good results in 65% of patients overall (2); however, reported outcomes and, thus, the need for total hip replacement, vary widely. For example, in one study of patients with mostly early stage osteonecrosis of the femoral neck, approximately 38% of patients required total hip replacement after an average 26 months after the core decompression (3).
The injection of concentrated autologous cells harvested from the iliac crest into the core decompression track at surgery may improve the clinical outcome and imaging appearance of osteonecrosis of the femoral head (4, 5, 6, 7).
Other established procedures include various proximal femoral osteotomies and bone grafting, both vascularized and nonvascularized. These procedures are technically demanding, require protected weight bearing for up to 6 months, and have not been done often in the United States. Reports vary as to their indications and effectiveness. They should be done primarily at selected centers that have the surgical experience and facilities to achieve optimal results.
If extensive collapse of the femoral head and degenerative changes in the acetabulum cause sufficient pain and disability, an arthroplasty usually is the most reliable way to effectively relieve pain and increase range of motion. The conventional approach is total hip replacement. Good to excellent results are achieved in 95% of total hip and total knee replacements, complication rates are low, and patients resume most activities of daily living within 3 months.
Osteonecrosis of the knee and shoulder can be managed nonsurgically more often than osteonecrosis of the hip. Limited experience with core decompression and bone grafting has been promising. In advanced stages, partial or total joint replacement may be indicated.
Treatment references
1. Zhao D, Zhang F, Wang B, et al: Guidelines for clinical diagnosis and treatment of osteonecrosis of the femoral head in adults (2019 version). J Orthop Translat 21:100-110, 2020. doi:10.1016/j.jot.2019.12.004
2. Hua KC, Yang XG, Feng JT, et al: The efficacy and safety of core decompression for the treatment of femoral head necrosis: a systematic review and meta-analysis. J Orthop Surg Res 14(1):306, 2019. doi:10.1186/s13018-019-1359-7
3. Andronic O, Weiss O, Shoman H, Kriechling P, Khanduja V: What are the outcomes of core decompression without augmentation in patients with nontraumatic osteonecrosis of the femoral head? Int Orthop 45(3):605-613, 2021. doi:10.1007/s00264-020-04790-9
4. Atilla B, Bakırcıoğlu S, Shope AJ, et al: Joint-preserving procedures for osteonecrosis of the femoral head. J.EFORT Open Rev 4(12):647-658, 2020. doi: 10.1302/2058-5241.4.180073
5. Hernigou P, Dubory A, Homma Y, et al: Cell therapy versus simultaneous contralateral decompression in symptomatic corticosteroid osteonecrosis: A thirty-year follow-up prospective randomized study of one hundred and twenty-five adult patients. Int Orthop 42(7):1639−1649, 2018. doi: 10.1007/s00264-018-3941-8
6. Zhang C, Fang X, Huang Z, et al: Addition of bone marrow stem cells therapy achieves better clinical outcomes and lower rates of disease progression compared with core decompression alone for early stage osteonecrosis of the femoral head: A systematic review and meta-analysis. J Am Acad Orthop Surg, 2020. doi: 10.5435/JAAOS-D-19-00816
7. Mont MA, Salem HS, Piuzzi NS, et al: Nontraumatic osteonecrosis of the femoral head: where do we stand today? A 5-year update. J Bone Joint Surg Am 102(12):1084-1099, 2020. doi: 10.2106/JBJS.19.01271
Prevention of Osteonecrosis
Risk of osteonecrosis caused by corticosteroids can be minimized by using them only when essential and by giving them in as low a dose as needed and for as short a duration as possible. Patients should be informed of this complication when long-term and/or high-dose corticosteroid use is anticipated.
To prevent osteonecrosis caused by decompression sickness, people should follow accepted rules for decompression when diving and when working in pressurized environments.
Excessive alcohol use and smoking should be discouraged. Evidence for the efficacy of various medications (eg, anticoagulants, vasodilators, lipid-lowering medications) for prevention of osteonecrosis in patients at high risk is limited and inconclusive.
Key Points
Osteonecrosis can be idiopathic, caused by trauma, or due to nontraumatic factors; the most common nontraumatic causes are corticosteroid use and excessive alcohol use.
Osteonecrosis should be suspected in patients with unexplained nontraumatic pain in the hip, knee, or shoulder (sometimes the wrist or ankle) and after certain fractures if pain persists or worsens.
Although radiographs may be diagnostic, MRI is more sensitive and specific.
Smaller lesions may heal spontaneously, but most larger lesions, especially in the hip, progress without treatment.
Nonsurgical treatments are not widely used because efficacies are not clearly proved.
Surgical treatment is often indicated to limit progression and/or relieve symptoms, particularly for osteonecrosis of the hip.
