Femoral Head Osteonecrosis—Nonoperative Treatment

Femoral Head Osteonecrosis—Nonoperative Treatment

Joseph A. Karam

Timothy M. DiIorio

Javad Parvizi


Osteonecrosis of the femoral head (ONFH) is a disease of the young patient with up to 25% of affected patients being less than 25 years in some institutions (1). Most patients will ultimately need total hip arthroplasty (THA) and it is estimated that ONFH constitutes about 10% of indications of THA procedures performed in the United States (1,2,3). Consequently, it represents a very large economic burden on healthcare systems, but more importantly, it is a large burden on the young patient population it affects, since THA in this population has been shown to yield relatively poor outcomes with increased risk of mechanical complications and the large majority of patients will need at least one revision procedure during their lifetime (1,4,5). Alternatives to THA have been sought and core decompression is one of the most commonly performed procedures nowadays with successful results in the early stages of the disease, as well as low morbidity and low rate of complications (<1% in most reported series) (1). However, many efforts have been spent during the past years to find still less invasive means to address the disease, especially at the very early stages or even as prophylaxis in high-risk patients. The largest difficulty in identifying the most suitable agents for such a task stems from all the unknowns in the pathogenesis of the disease and the controversies about its natural history. Many modalities have succeeded in slowing the progress of the disease process or even halt it and prevent the collapse of the femoral head and their use could be promising at the early stages of the disease. These nonoperative means of treatment currently include pharmacologic agents such as statins, anticoagulants, vasodilators, and bisphosphonates, as well as biophysical therapies like electromagnetic fields, extracorporeal shock waves (ESWs), and hyperbaric oxygen (HBO) therapy.

Conservative Treatment

Observation or reduced weight bearing (no or partial weight bearing with the use of walking aids) has been a subject of controversy for a long time in the treatment of asymptomatic or very early stage ONFH. The main reason is that reports about the natural history of the disease are controversial and present wide variations between different authors.

Mont et al. (6) in 1996 after reviewing available studies at that time identified satisfactory clinical outcomes in around 23% of cases treated conservatively, with almost 80% of patients ultimately requiring THA or other surgical interventions. This meta-analysis included 819 hips treated with nonoperative management (observation with full weight bearing, partial weight bearing, or no weight bearing) with a mean follow-up of 34 months. The observed results led the authors to discard any possible role of such conservative measures in the treatment of ONFH.

In an attempt to shed more light into the natural history of ONFH, Nishii et al. (7) followed patients with precollapse stage ONFH for at least 5 years. Almost half of the hips ended up collapsing, including the ones that had small lesions (involving less than the medial 2/3 of the weight-bearing area). However, in 54% of these lesions, collapse stopped progressing and 64% of patients with cessation of collapse (9/14) eventually became asymptomatic. Cessation of collapse was found to be significantly correlated with the extent of the necrotic lesion (89% cessation in hips that initially had necrosis of less than 2/3 of the weight-bearing area) and with the amount of collapse at final examination (0% progression in 11 hips with less than 2 mm collapse at final examination). The authors concluded that collapse of the femoral head does not necessarily indicate a relentless progression of the underlying disease. Cessation of collapse can happen, especially in small lesions, and clinical improvement can follow without the need for any invasive intervention. Limited progression of the disease in cases with small lesions has also been observed by many other authors (8,9,10,11).

Hernigou et al. conducted a prospective study with a minimum of 10-year follow-up in 40 patients with small lesions (stage I according to Steinberg et al. (12)) identified on MRI in contralateral hips of patients presenting with symptomatic ONFH (13). At latest follow-up (average follow-up was 11 years), 88% of hips had become symptomatic and 73% had evidence of collapse, all of which eventually required surgical intervention. However, the fate of bilateral disease
could be different from unilateral disease and should not be extrapolated to all cases. Indeed, Cheng et al. (14) have identified spontaneous resolution in 3/13 hips with asymptomatic unilateral ONFH detected through prospective screening in solid organ transplant patients. In this study 9/17 (53%) hips with asymptomatic ONFH in patients who had presented with contralateral symptomatic ONFH ultimately became symptomatic and were treated with core decompression, versus only 2/13 hips (15%) in posttransplant cohort.

Min et al. (15) looked at MRI-identified asymptomatic ONFH in 81 patients who presented with symptomatic ONFH on the contralateral side. Included patients had a minimum follow-up of 5 years (average follow-up was 8.3 years). Thirty-eight percent of the hips were symptomatic at the latest follow-up visit and hip survival rates (as defined by absence of collapse) were 91.4% at 5 years and 70% at 10 years. Although patient age, gender, weight, and presumed etiology of ONFH were poor predictors of collapse, extent and location of necrosis was a strong predicting factor. Type C2 lesions according to Sugano et al. (16) had the worst outcome (lesions occupying more than the medial two-thirds and extending laterally to the acetabular edge). The occurrence of pain was deemed a strong sign of impending collapse and all collapse cases were preceded by the occurrence of pain of an average of 8-month duration.

In conclusion, the natural history of ONFH is still poorly understood to help in identifying a role for conservative management. Even though small asymptomatic lesions have shown a low rate of progression in some studies, many authors argue that with extended follow-up progression eventually occurs and observation or reduced weight bearing have no place in the management of ONFH.

Pharmacologic Treatment

Diverse pharmacologic agents have been shown to have a role in management of patients with early stage ONFH. These agents target hypothesized pathophysiologic mechanisms for osteonecrosis such as increased bone marrow pressure, adipogenesis, and thrombogenesis.

Lipid Lowering Agents

Disturbance of lipid metabolism is one proposed mechanism of osteonecrosis, particularly among patients with systemic lupus erythematosus, as well as those being treated with corticosteroids (17,18). Fat emboli and fat deposition in the bone marrow may lead to vessel compression and compromise of blood flow to areas of tenuous supply such as the femoral head. This led some authors to consider statins as a potential treatment in ONFH. Cui et al. (19) in an experimental study showed that lovastatin inhibited steroid-induced expression of genes implicated in adipogenesis and counteracted steroid-induced osteoblastic gene repression. An in vivo part of their study showed that lovastatin administration prevented high-dose methylprednisolone-induced osteonecrosis in chickens. Wang et al. (20) on the other hand showed that clofibrate reduced bone marrow fat and intrafemoral head pressure in rabbits. More recently, Nishida et al. (21) showed significant risk reduction of steroid-induced osteonecrosis in rabbits with the use of the new statin, pitavastatin. Histopathologic studies demonstrated a significant reduction in size of bone marrow fat cells in animals that received the drug. Other mechanisms of action of statins are also being explored in preventing ONFH. Sakamoto et al. (22) showed for example that simvastatin suppresses the secretion of the antifibrinolytic adipokine plasminogen activator inhibitor-1 (PAI-1) and this may explain its protective effects against steroid-induced osteonecrosis.

In the clinical setting, Pritchett (23) identified 284 patients that were started on high-dose corticosteroids while on statin therapy. The risk of developing ONFH after a mean follow-up of 7.5 years was only 1%, which was viewed as very low compared to the reported 3% to 20% incidence of ONFH in patients receiving high-dose corticosteroids. On the other hand, Ajmal et al. (24) reviewed 2,881 renal transplant recipients who were on high-dose corticosteroids and looked at the occurrence of osteonecrosis in statin users versus non-statin users. They were not able to identify a statistically significant difference in the occurrence of ONFH and concluded that statin treatment did not significantly affect the occurrence of osteonecrosis in renal transplant patients under high-dose steroids. This study had many limitations, namely the specific subpopulation of renal transplant patients and the restriction of outcome measurement to self-reported cases.

Motomura et al. (25) showed the benefit of combined treatment with warfarin and a lipid-lowering agent (probucol) in preventing the development of osteonecrosis in rabbits after the injection of high-dose corticosteroids. They showed that using a combination of warfarin and probucol decreased the rate of occurrence of steroid-induced ONFH to a significantly larger extent than either agents used alone.

Anticoagulants and Antiplatelet Drugs

Thrombosis of intraosseous arteries, veins and/or bone marrow sinusoids holds a major role in the pathogenesis of ONFH. This led many teams to identify a possible role for anticoagulants and antiplatelet agents in the prevention or the management of early stage ONFH.

Norman et al. (26) induced femoral head osteonecrosis in rats by incising the periosteum at the base of the femoral neck and cutting the ligamentum teres, thus depriving the femoral head from its main blood supply. A histologic study of the femoral head of a subset of rats treated with enoxaparin showed reduced amounts of residual necrotic bone, decreased bone remodeling, and less expansive cartilage degeneration compared to the control group. These results, coupled with the pro-angiogenic and anti-inflammatory properties of unfractionated heparin and low–molecular-weight heparin led the authors to suspect a potential clinical benefit of enoxaparin in the remodeling phase of Perthes disease, which would thus prevent or delay the development of hip osteoarthritis in patients affected by this condition.

In 1995, Glueck et al. (27) studied a specific subgroup of patients presenting with ONFH who suffered from
hypofibrinolytic states: Four patients with hypofibrinolysis due to high levels of PAI and one patient with high lipoprotein A (Lp(a)) levels. The authors attempted treatment of ONFH in those patients with stanozolol, an anabolic agent capable of reducing PAI levels and thus restoring fibrinolytic activity. They were able to achieve clinical improvement of ONFH in one out of the four patients with high levels of PAI, as well as in the patient with elevated Lp(a). Ten years later, the same group used enoxaparin (60 mg/day for 12 weeks) to prevent the progression of early stage ONFH in patients with thrombophilic or hypofibrinolytic disorders (28). Their findings indicated that enoxaparin was effective in stabilizing early stage primary osteonecrosis in those patients but was not efficient in patients with ONFH secondary to high-dose corticosteroid treatment.

Wada et al. (29) showed that warfarin considerably reduced the rate of ONFH in spontaneously hypertensive rats (SHR), which were used as an experimental model for ONFH. Two years later, Nagasawa et al. (30) used warfarin to prevent the occurrence of ON in patients with systemic lupus erythematosus treated with high-dose corticosteroids. Although they were not able to demonstrate a statistically significant difference in the incidence of ON between patients who received warfarin and controls, there was a clear tendency toward decreased ON in patients who were under warfarin treatment in this lupus population.

Yamaguchi et al. (31) were recently able to significantly reduce the incidence of steroid-induced osteonecrosis in rabbits with the antiplatelet drug clopidogrel. This drug was used prophylactically before inducing ONFH through the administration of high-dose steroids in the rabbits and was able to significantly reduce the occurrence of femoral and humeral osteonecrosis, thus shedding more light into the pathogenesis of steroid-induced osteonecrosis and emphasizing the role of platelet aggregation in that process.


Several vasoactive substances have been demonstrated to reduce bone marrow pressure and have thus been explored for potential use in the treatment of ONFH, such as naftidrofuryl, a vasodilator used for painful peripheral arterial disease and other vascular disorders and iloprost, a prostacyclin (PGI2) analog used in patients with pulmonary hypertension and peripheral arterial disease (32,33). Indeed, iloprost has been shown to substantially improve cases of bone marrow edema both radiologically and clinically (34,35,36,37). Jäger et al. (38) in a recent prospective study that included 95 patients with painful bone marrow edema or osteonecrosis also demonstrated its benefit in reducing pain levels and improving functional scores. Even though there were no serious adverse effects attributable to iloprost administration, treatment had to be discontinued in five patients because of severe headaches.

Laroche et al. (39) showed that the dihydropyridine calcium channel blocker nifedipine successfully reduced pain in patients diagnosed with ONFH. Compared to iloprost, this drug has the advantage of oral intake. Moreover, a recent study on New Zealand rabbits by Drescher et al. (40) demonstrated the ability of a nitrate patch to restore femoral head histology after glucocorticoid-induced osteonecrosis with a significant reduction in the amount of adipocytes and empty lacunae.


Bisphosphonates are a class of drugs that target osteoclasts and decrease their function, thus reducing bone resorption and bone remodeling. They are most famous for their use in osteoporosis, Paget disease of bone, and malignant hypercalcemia, but they have also found indications in several other diseases of the musculoskeletal system. Even though one of the most discussed side effects of bisphosphonates is osteonecrosis of the jaw, many efforts have been spent to try and identify a possible role for them in the treatment of ONFH.

Agarwala et al. (41) postulated that inhibiting osteoclast action and necrotic bone resorption would prevent collapse of the femoral head. They followed 60 patients with ONFH treated with alendronate and found a significant improvement in pain scores, functional status, and bone edema as seen on MRI. This helped delay or even avoid surgery in many patients, as only 10% of their patients ultimately required surgery with a follow-up ranging from 3 months to 5 years. This was also supported by an experimental study conducted by Peled et al. (42

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

May 22, 2016 | Posted by in PHYSICAL MEDICINE & REHABILITATION | Comments Off on Femoral Head Osteonecrosis—Nonoperative Treatment

Full access? Get Clinical Tree

Get Clinical Tree app for offline access