Hip Osteonecrosis

Updated: Mar 16, 2021
Author: Michael Levine, MD; Chief Editor: William L Jaffe, MD 


Practice Essentials

Osteonecrosis of the femoral head involves the hip joint, with osteocytes of the femoral head dying along with the bone marrow; resorption of the dead tissue by new but weaker osseous tissue can then lead to subchondral fracture and collapse. There are 2 forms of osteonecrosis: traumatic (the most common form) and atraumatic. Other terms to describe this disorder are avascular necrosis and ischemic necrosis, to denote vascular etiology. The term aseptic necrosis also has been used to indicate that infection does not play a causative role.

Osteonecrosis is now a commonly recognized disorder with significant morbidity. The end stage of the process is severe destruction of the femoral head with resultant degeneration of the hip joint. In many patients, even early identification and intervention do not alter the result. Unfortunately, patients who are affected with osteonecrosis are young, usually in the third to sixth decades of life.

Traumatic and atraumatic osteonecrosis are essentially 2 distinct problems. The traumatic form has a definitive causal event and is isolated to the particular injured bone. The atraumatic form has multiple etiologies (eg, excessive alcohol consumption, chronic corticosteroid use, autoimmune and chronic inflammatory disorders[1] ) and can involve multiple bones. The main focus of this article is atraumatic osteonecrosis.

The natural history of atraumatic osteonecrosis is still not well understood. Different etiologies of the disease often have different clinical courses. It has been reported that approximately 56% of asymptomatic patients eventually become symptomatic.[2] Steroid-induced disease has the worst prognosis, with most cases progressing to collapse of the femoral head. 

Plain radiographic findings frequently are normal. Therefore, history and physical examinations are paramount for diagnosis; see Presentation.

Treatment is indicated after diagnosis is confirmed with radiographic studies. Along with biophysical modalities, agents that have been used for nonoperative management include  bisphosphonates, anticoagulants, vasodilators, and statins.[3] Most studies indicate that the risk for disease progression is greater with nonsurgical treatment than with surgical intervention. Prostheses with novel bearing surfaces (ie, metal-on-metal, ceramic-on-ceramic) are being investigated, to increase the success rate for total hip replacements in patients with osteonecrosis.

For patient education information, see Avascular Necrosis (Aseptic Necrosis or Osteonecrosis).


Traumatic osteonecrosis is a direct result of disruption of the blood supply to the femoral head. Death of bone marrow occurs within 6-12 hours after vascular insult. Death of the bone becomes apparent several days later.

The pathophysiology in atraumatic osteonecrosis remains controversial.[4] Fat cell hypertrophy with resultant pressure increase within the femoral head, leading to vascular collapse and then necrosis, has been proposed as a mechanism for steroid-induced osteonecrosis. A fat embolism phenomenon with resultant vascular occlusion is another proposed mechanism. A hyperlipidemic state seems to be related to causation, but the exact mechanism is unknown. Similarly, the lipid hypothesis has also been applied to cases associated with alcohol abuse.

Studies have shown evidence of acquired hypercoagulabiltiy. This effect appears to be augmented by tobacco abuse.[5] Studies have also shown elevated cryofibrinogen levels in atraumatic osteonecrosis.[6]

In caisson disease, circulating nitrogen bubbles occlude blood vessels in response to reduction in ambient pressure during decompression. Sickle cell anemia results in bone death secondary to the sickling process and subsequent vascular occlusion.

Increased intraosseous pressure contributes directly to the propagation of necrosis, regardless of etiology. As bone death occurs, a repair process takes place as dead bone is removed and replaced by new bone. During this phase, the bone underlying the joint surface is weakened. In most patients, subchondral fracture alters the articular surface, resulting in abnormal mechanics and arthritic alterations to the joint.

The disease affects both sides of the joint, as confirmed by PET scan imaging showing earlier involvement in the acetabulum than is discernible by other radiographic modalities.


As the name implies, traumatic osteonecrosis is secondary to direct injury to the femoral head with resultant damage of the blood supply. Fracture of the femoral head or neck and hip dislocation are the primary mechanisms of injury.

Atraumatic osteonecrosis has many risk factors. The 2 most commonly associated problems are corticosteroid use and alcohol abuse.[7, 8] The idiopathic cases make up the third most common category. Other factors include sickle cell anemia, Gaucher disease, systemic lupus erythematosus, coagulopathies, hyperlipidemia, organ transplantation, caisson disease, and thyroid disorders. Genetic factors may also play a role.

Hip osteonecrosis resulting from corticosteroid use or alcohol abuse is associated with the worst prognosis. Frequently, steroid-induced osteonecrosis involves multiple bones and, in the case of the hip, results in nearly 100% bilateral involvement. The exact dose required to induce osteonecrosis remains an enigma, but most studies indicate that higher doses, even over a short period of time, present the highest risk. Often, patients on steroids have other associated risk factors.

Osteonecrosis associated with alcohol abuse usually occurs in those who drink more than 400 mL of alcohol per week. It is more common in those with a long-term history of heavy consumption.


Approximately 10,000-20,000 new cases are identified each year in the United States. At least 50% of patients with atraumatic hip osteonecrosis are thought to have bilateral involvement. Other bones often are involved in the atraumatic form, including the shoulder, knee, and talus. 

The traumatic form of hip osteonecrosis occurs in 10% of undisplaced femoral neck fractures, 15-30% of displaced femoral neck fractures, and 10% of hip dislocations. Corticosteroid use contributes to the atraumatic form of osteonecrosis in 25-50% of patients and alcohol-associated osteonecrosis has been reported in 20-45% of cases.[9, 10]

 The male-to-female ratio is about 4:1. The typical age ranges from 35 to 50 years old, with the average age of presentation being 36 years. 

A national epidemiologic survey in Japan estimated an annual incidence of 2200 new patients diagnosed with osteonecrosis of the femoral head. Osteonecrosis was associated with steroid use (51%), heavy alcohol consumption (31%), both (3%), and neither (15%). Steroid-induced osteonecrosis was diagnosed in 34% of male patients and in 76% of females patients. The underlying diseases requiring steroid administration included systemic lupus erythematosus (SLE), nephritic syndrome, polymyositis/dermatomyositis, asthma, and thrombocytopenic purpura.[11]   


The success rate in patients not treated by arthroplasty in stage 0 or I approaches 90% in some series. Once femoral head collapse occurs, these treatments offer limited benefit. Procedures such as the trapdoor procedure potentially may improve results in stage II and III, but presently, total hip replacement remains the treatment of choice once collapse has occurred. If not treated, 80% of femoral heads collapse within 4 years of diagnosis. Location and extent of the necrotic lesion appear to be good indicators of collapse of the femoral head.

The risk of femoral head collapse can be stratified into three groups based on the modified Kerboul combined necrotic angle calculated by the summation of the arc of femoral head necrosis on mid-sagittal and midcoronal MRI, as follows[12] :

  • Low risk: combined necrotic angle less than 190 degrees
  • Moderate risk: combined necrotic angle between 190 and 240 degrees
  • High risk: combined necrotic angle greater than 240 degrees

Core decompression success rates are better than those with conservative treatment, with approximately 70% success in stages before radiographic collapse and limited morbidity.[13] In a systematic review of core decompression without augmentation in patients with nontraumatic osteonecrosis of the femoral head, which included pooled results from 1134 hips, nearly 80% of which were in an early stage of osteonecrosis, approximately 38% of patients underwent a total hip replacement at an average of 26 months after the procedure.[14]

Complications of core decompression are minimal in the hands of experienced surgeons. The most severe complication is fracture, which can occur if core is drilled below the trochanteric ridge. Core decompression has been shown to be a highly cost-effective alternative when a total hip replacement is delayed by 5 years or more.

Complications of bone grafting procedures include donor-site morbidity; peroneal sensory neuropathy, contractures of the flexor hallucis longus, and deep venous thrombosis. Retrieval studies have shown little bone ingrowth, insufficient mechanical support of subchondral bone, and a significant rate of femoral head collapse with tantalum implants.[15]  

For trapdoor, 83% good or excellent results were demonstrated in 1 study.[16]  Limited complications are reported, aside from deep venous thrombosis.

Cup arthroplasty, unipolar arthroplasty, and bipolar arthroplasty have poor success rates; as disease appears to affect both sides of the hip joint.


Total resurfacing arthroplasty has a greater than 90% survivorship at greater than 3 years, however, femoral neck fracture is the most common and critical complication.

Results are poor for arthrodesis in terms of achieving fusion and patient satisfaction. Total hip replacement early results were poor with early cement techniques, with failures up to 25% or higher. Studies with current techniques have shown success rates at over 90%, making it the treatment of choice following collapse or failure of less-invasive procedures. Complications include infection, peroneal nerve palsy, deep venous thrombosis, intraoperative fracture, and postoperative dislocation; risk-benefit ratio strongly reflects success of procedure.




Patients with osteonecrosis usually are men in the sixth decade of life who experience pain primarily in the groin but occasionally the buttocks. Pain usually is deep and throbbing and is worse with ambulation, but it also is significant at night. Onset often can be described as acute. Patients frequently describe a catching or popping sensation with motion. A history of trauma, steroid use, alcohol abuse, and other risk factors should be sought.


Physical Examination

Physical examination reveals pain with range of motion and ambulation. Limitation of internal rotation in both flexion and extension are prevalent, with passive internal rotation in extension being particularly painful. A Trendelenburg gait often is present.



Approach Considerations

In 2019, the Association Research Circulation Osseous (ARCO) published consensus-based diagnostic criteria for glucocorticoid- associated osteonecrosis of the femoral head (GA-ONFH)and alcohol-associated ONFA. The diagnosis is made by reference to symptoms, signs, imaging, and/or histological examination[9, 10]

Criteria for a diagnosis of GA-ONFH included the following[9] :

  • The patient should have a history of glucocorticoid use > 2 g of prednisolone or its equivalent within a 3-month period
  • Osteonecrosis should be diagnosed within 2 years after glucocorticoid usage
  • The patient should not have other risk factor(s) besides glucocorticoids.

Criteria for a diagnosis of alcohol-associated ONFH included the following[10] :

  • The patient should have a history of alcohol intake > 400 mL/wk (320 g/wk, any type of alcoholic beverage) of pure ethanol for more than 6 months
  • ONFH should be diagnosed within 1 year after intake of this amount of alcohol
  • The patient should not have other risk factor(s)

Laboratory Studies

Lab tests have limited utility in the diagnosis of osteonecrosis, with exceptions as follows:

  • Sickle cell testing in African Americans
  • Lipid profile
  • Screening for hypercoagulability

Imaging Studies

Anteroposterior (AP) radiographs (see image below) and frog lateral radiographs of both hips are the primary diagnostic modalities.

Osteonecrosis, hip. Anteroposterior radiograph sho Osteonecrosis, hip. Anteroposterior radiograph showing Ficat stage III disease.

AP and frog lateral tomograms are indicated if patients have evidence of disease on radiographs but have no collapse. These images are often helpful in staging.

Sensitivity and specificity of magnetic resonance imaging (MRI) is greater than 98%, which is higher than all other modalities. This study is ideal if x-ray findings are normal and clinical suspicion is high. MRI should be performed in all patients with osteonecrosis to assess the extent of the disease. Three-dimensional MRI scanning with image registration may be used to assess changes in lesion size.

Osteonecrosis is classically delineated on MRI by serpiginous hypointense signal with or without associated subchondral collapse or secondary osteoarthritis.[17] MRI is recommended to identify bilateral disease when 1 hip has radiographic signs of disease and the other is normal (see image below).

MRI showing osteonecrosis of right hip, normal lef MRI showing osteonecrosis of right hip, normal left hip.

Bone scanning is a low-cost alternative when index of suspicion is low. They can be helpful when x-ray findings are normal if MRI cannot be obtained (see image below).

Bone scan showing osteonecrosis of right hip. Bone scan showing osteonecrosis of right hip.

Diagnostic Procedures

Core biopsy and interosseous pressure measurement

An open biopsy of a 10-mm core of bone from the femoral head can be used for diagnosis (see image below). Measurement of interosseous pressure can be obtained before and after biopsy to confirm decompression of the intraosseous space.

Osteonecrosis, hip. Anteroposterior radiograph cor Osteonecrosis, hip. Anteroposterior radiograph core biopsy.


Injection of contrast under image intensification has been used as part of the functional evaluation of bone when measuring intraosseous pressure. This can be used to confirm presence of the needle within the head and venous congestion.

Histologic Findings

The first histologic findings are marrow and adipocyte necrosis. Next, liquefaction necrosis and interstitial edema occur. Pyknotic nuclei with empty lacunae are identified as osteocyte necrosis occurs. Eventually, the zone of necrosis is surrounded by repair tissue as revascularization proceeds. During this phase, the subchondral plate is weakened as resorption occurs faster than reformation, leading to subchondral collapse and eventual cartilage damage.


Nearly 20 classification systems have been developed to stage osteonecrosis and provide information on prognosis, guide treatment selection, and facilitate outcome comparison. The following are the three most widely used classification systems:

  • Ficat classification [18]
  • University of Pennsylvania classification of osteonecrosis (Steinberg system) [19, 20]
  • Association Research Circulation Osseous (ARCO) system [21]

Ficat classification

The Ficat classification of osteonecrosis was developed before the MRI and is based on radiographic findings.[18] There is no assessment of the size or extent of the lesion. However, it still is widely utilized because it is simple and easy to use. The stages are as follows: 

  • Stage 0 - No pain, normal radiographic findings, abnormal bone scan or MRI findings
  • Stage I - Pain, normal x-ray findings, abnormal bone scan or MRI findings
  • Stage IIa - Pain, cysts and/or sclerosis visible on x-ray, abnormal bone scan or MRI findings, without subchondral fracture
  • Stage III - Pain, femoral head collapse visible on x-ray, abnormal bone scan or MRI findings, crescent sign (subchondral collapse) and/or step-off in contour of subchondral bone
  • Stage IV - Pain, acetabular disease with joint space narrowing and arthritis (osteoarthrosis) visible on x-ray, abnormal MRI or bone scan findings

The University of Pennsylvania classification of osteonecrosis (Steinberg system)

The University of Pennsylvania classification was developed by Steinberg and the important features of this system is that it includes measurement of lesion size and extent of joint involvement using MRI.[19, 20]  The stages are defined in the table below.

Table 1. University of Pennsylvania classification of osteonecrosis (Open Table in a new window)




Normal radiograph, bone scan, and magnetic resonance images


Normal radiograph. Abnormal bone scan and/or magnetic resonance images

 A: Mild (< 15 % of femoral head affected)

 B: Moderate (15–30 % of femoral head affected)

 C: Severe (>30 % of femoral head affected)


Cystic and sclerotic changes in femoral head

 A: Mild (< 15 % of femoral head affected)

 B: Moderate (15–30 % of femoral head affected)

 C: Severe (> 30 % of femoral head affected)


Subchondral collapse without flattening (crescent sign)

 A: Mild (< 15 % of femoral head affected)

 B: Moderate (15–30 % of femoral head affected)

 C: Severe (>30 % of femoral head affected)


Flattening of femoral head

 A: Mild (< 15 % of surface and < 2 mm of depression)

 B: Moderate (15–30 % of surface and 2–4 mm of depression)

 C: Severe (>30 % of surface and > 4 mm of depression)


Joint narrowing or acetabular changes

 A: Mild

 B: Moderate

 C: Severe


Advanced degenerative changes

Association Research Circulation Osseous (ARCO) system

The Association Research Circulation Osseous (ARCO) system was an attempt to establish an internationally accepted classification with uniform definition and terminology.[21]  The stages are defined in the table below.

Table 2. Association Research Circulation Osseous (ARCO) system (Open Table in a new window)




Bone biopsy results consistent with osteonecrosis; other test results normal


Positive findings on bone scan, MRI, or both

A: < 15% involvement of the femoral head (MRI)

B: 15-30% involvement

C:  > 30% involvement


Mottled appearance of femoral head, osteosclerosis, cyst formation, and osteopenia on radiographs; no signs of collapse of femoral head on radiographic or CT study; positive findings on bone scan and MRI; no changes in acetabulum

A: < 15% involvement of the femoral head (MRI)

B: 15-30% involvement

C: > 30% involvement


Presence of crescent sign lesions classified on basis of appearance on AP and lateral radiographs

A: < 15% crescent sign or < 2-mm depression of femoral head

B: 15-30% crescent sign or 2- to 4-mm depression

C: > 30% crescent sign or > 4 mm depression


Articular surface flattened; joint space shows narrowing; changes in acetabulum with evidence of osteosclerosis, cyst formation, and marginal osteophytes





Approach Considerations

Surgery is the mainstay of treatment for osteonecrosis. Obvious disorders aside (eg, severe systemic disease, systemic sepsis), these patients often are young and have few surgical contraindications. Numerous procedures are available, indicating that no single procedure is distinctly advantageous. 

Nonsurgical treatment of osteonecrosis is limited. Observation and protected weight bearing are options. Certain cases of early-stage disease (eg, Ficat stage 1) can be treated successfully with this option. However, most studies indicate that the risk of disease progression is greater with nonsurgical treatment than with surgical intervention.

Cell-based therapies for the treatment of osteonecrosis of the femoral head have been reported to be safe and suggest improved clinical outcomes with lower disease progression rate, particularly when employed at early disease stages.[22, 23, 24, 25, 26, 27, 28]  However, there has been substantial heterogeneity in the cell-based therapies used and studies vary widely with respect to cell sourcing, cell characterization, adjuvant therapies, and assessment of outcomes. Specific clinical indications and cell-therapy standardization have not yet been determined.[29]

Medical Care

Nonsteroidal anti-inflammatory drugs (NSAIDs) can be used to reduce pain and inflammation in patients who cannot have surgery for medical or other reasons or for patients who are undergoing surgical treatment. Physical therapy can be helpful to restore motion and improve gait. Electrical stimulation has been used in several centers. In some studies, it has been helpful in treatment prior to femoral head collapse. 

Pharmacotherapy that addresses the pathophysiology of the disease has had mixed results. Examples include gemfibrozil (Lopid) for hyperlipidemias and nifedipine for vascular disorders. Alendronate has been suggested as an option to avoid or delay progression of the disease clinically and radiographically. However, one randomized study showed no significant difference in radiographic and MRI data between the alendronate and control groups.[30]  Short-term follow-up (about 24 months) of patients in alendronate studies have demonstrated delayed femoral head collapse.[31]

Extracorporeal shockwave treatment has shown some promise in treating early disease by promoting angiogenesis and bone remodeling.[32, 33]

Surgical Care

The choice of procedure is based on preoperative staging. Core decompression and cancellous and cortical bone grafting procedures usually are indicated in Ficat stage IIa or earlier stages. The trapdoor procedure and allograft procedures are indicated for stage IIb or stage III lesions. Osteotomies are used for stage II and stage III disease. Arthrodesis and arthroplasty are utilized primarily for stages III and IV but occasionally are used for stages I and II. Limited femoral resurfacing for young patients with intact acetabular cartilage and a collapsed femoral head is a valuable alternative to total hip replacement.

The growth factors Op-1 (osteogenic protein-1) and rhBMP-2 (recombinant human bone morphogenetic protein-2) may be useful bone grafting adjuncts.[34]  Platelet-rich plasma therapy is also used as an adjunctive treatment.[35, 36]   

Core decompression and bone grafting

The objective in core decompression is to stimulate revascularization and decrease pressure within the femoral head. The patient is placed supine on a fracture table. Using image intensification through a lateral incision above the trochanteric ridge, a 10-mm core of bone is removed from the femoral necrotic lesion. Other techniques include multiple drilling into the lesion

For cancellous bone grafting, the defect is filled with cancellous bone graft material, usually iliac crest or allograft. Cancellous iliac crest graft is placed in a channel in the infracted region and covered by a graft with the quadratus femoris muscle attached. Following surgery, non-weightbearing ambulation for 6-12 weeks, then gradual resumption of normal activities as tolerated.

With cortical bone grafting, a strut graft is placed in the defect under the weightbearing surface of the femoral head. Iliac crest or fibula has been used. Strut grafting with a tantalum implant, a highly porous metallic cylinder placed in a channel to support subchondral bone, is also an option.[37, 38]  Following surgery, no weight bearing for 6 weeks, with progressive weight bearing to 6 months.  Recent retrieval studies have shown little bone ingrowth, insufficient mechanical support of subchondral bone, and a significant rate of femoral head collapse.[15]  

Use of a vascularized free fibular graft harvested from the ipsilateral leg with a vascular pedicle inserted into the proximal femoral defect and anastomosed with the lateral circumflex artery has become popular. The procedure is technically difficult with increased morbidity and has questionable benefit compared with core decompression.

Trapdoor procedure and allograft 

Trapdoor is indicated more in stage III disease, in which above procedures are unsuccessful. The trapdoor procedure involves open excision of the necrotic bone by elevation of the cartilage and cancellous grafting. In an osteochondral allograft procedure the necrotic area is replaced with a nonvascularized free allograft. During recovery, the patient is 20% weight bearing for 6 weeks, 50% weight bearing to 10 weeks, then progresses to full weight bearing.


Osteotomy is technically very difficult and complications include nonunion and malunion. Total hip replacement is technically more difficult following osteotomy thus it is used in cases in which total hip replacement is not advisable. 

The concept in osteotomy is to rotate the diseased area of the femoral head away from the weightbearing surface. Several different techniques are available:

  • Angular osteotomy: Varus or valgus flexion usually is performed intertrochanterically and fixed with a plate.
  • Rotational osteotomy: The head is rotated transtrochanterically, moving the weightbearing surface away from the necrotic lesion.

Recovery requires protected weight bearing for 6 weeks with gradual progression to full weight bearing.

Arthrodesis and arthroplasty

Arthrodesis is fusion of the hip joint. The joint is denuded of articular cartilage, and the femoral head and acetabulum are fixed to create a solid interface. No weight bearing is allowed following the procedure, with full weight bearing initiated at 3 months.

In arthroplasty, conventional techniques are used with either cemented or cementless implants. Resection arthroplasty involves excision of the femoral head. Mold or cup arthroplasty involves resurfacing of the articular surface of the femoral head with a prosthetic device. Resurfacing arthroplasty involves a cup-type arthroplasty on the femoral side with a metal-on-metal acetabular component.

Unipolar prosthetic arthroplasty involves replacement of the femoral head with a nonmobile bearing head and bipolar arthroplasty involves replacing the femoral head with a mobile bearing component.

Total hip replacement is primarily with cementless devices, with metal-on-polyethylene, ceramic-on-polyethylene, or ceramic-on-ceramic bearings. Weight bearing is allowed as tolerated immediately following surgery, depending on surgeon preference.



Guidelines Summary

Japanese Orthopaedic Association guidelines for osteonecrosis of the femoral head (ONFH) include the following recommendations and statements[39] :

  • Unloading for ONFH with orthotics (canes and crutches) is useful for alleviating alleviate pain and improving walking function, but cannot be expected to prevent the progression of collapse of the femoral head or reduce the need for surgical treatment in the long term.

  • Extracorporeal shock waves, electromagnetic field stimulation, and hyperbaric oxygen therapy for ONFH may be effective in alleviating pain. It is unclear whether these therapies can prevent the progression of femoral head collapse or reduce the need for surgical treatment.

  • The long-term effect of bisphosphonate (alendronate, zoledronate) administration for ONFH on the prevention of femoral head collapse is unclear.

  • Short-term results of core decompression for Ficat stage I ONFH are good; however, this surgery should not be selected for ONFH with Ficat stage II or higher. As there have been several cases with poor improvement of pain and progression to collapse of the femoral head even in Ficat stage I, it is necessary to consider the surgical indications based on detailed evaluation of the size and position of the necrotic area. Core decompression combined with bone marrow-derived cells and/or growth factors is expected to improve clinical outcomes compared to core decompression alone; however, clinical outcomes are still poor for cases of Ficat stage III or more.

  • The results of vascularized bone grafting vary among reports, but good clinical outcomes can be expected in 60%–94% of cases when arthritic changes have not appeared.

  • Femoral varus osteotomy is useful to relieve symptoms and prevent the progression of ONFH with sufficient intact area at the lateral femoral head. Patients with a postoperative intact ratio of more than 34% generally have a good clinical outcome.

  • Transtrochanteric rotational osteotomy is useful to relieve symptoms and prevent the progression of the stage of ONFH with a wide necrotic area.

  • The long-term results of contemporary cementless total hip arthroplasty have been generally good, and it is an effective treatment option for patients with low levels of osteolysis around the implant, dislocation, and deep infection.

  • The long-term results show that cemented total hip arthroplasty have using modern cementing techniques is a generally good and useful treatment; however, there are fewer long-term reports on cemented acetabular components than on femoral components. The longevity of cemented acetabular components is slightly inferior to that of the femoral components.

  • Bipolar hemiarthroplasty is indicated for ONFH at stage 3 or earlier without osteoarthritic changes. The mid-to long-term results of bipolar hemiarthroplasty for ONFH in stage 3 and earlier are generally good, making it a useful treatment; however, postoperative buttock and groin pain and migration of the outer head may occur.

  • The short-to mid-term results of hip resurfacing arthroplasty for ONFH are generally good; however, there are few reports on its long-term results. An increase in serum metal ion concentration and the occurrence of femoral neck fracture have been reported. Indications must be strictly considered.

  • Hip replacement for young people (age 50 years or less) is one of the most useful treatments for ONFH, with good mid-term results; survival rates in highly active young patients are 100% at 7–10 years postoperatively, when ceramic-on-ceramic or highly cross-linked polyethylene is used for bearing. However, long-term results need further verification. In addition, if a blood disorder such as sickle cell disease is involved in the occurrence of ONFH, the incidence of complications may increase and the implant survival rate may decrease.