Treatment of Avascular Necrosis of the Femoral Head With Vascularized Fibular Transplant

Abstract

Two hundred twenty-eight hips in 187 patients with avascular necrosis of the femoral head were treated with vascularized fibular transplant from March 1989 to March 2000. The etiologic factors associated with the disease included corticosteroids in 84 patients (44%; 101 hips, trauma in 25 patients (13%; 29 hips), alcohol abuse in 24 patients (12%; 28 hips), and 41 hips (18%) were classified as idiopathic. Systemic disorders, including systemic lupus erythematosus, sickle cell anemia, inflammatory bowel disease, pregnancy, and dysbaric disease were observed in 12, nine, four, three, and one hip(s), respectively. Of the 228 hips operated on, 184 hips (152 patients) were assessed postoperatively with followup ranging from 1 to 10 years (mean, 4.7 years). Using the Steinberg classification system, 39 hips (21%) were in Stage II; 45 hips (25%) were in Stage III; 77 hips (42%) were in Stage IV; and 23 hips (12%) were in Stage V. Of the 184 hips treated, 101 (54%) remained stable postoperatively, whereas 69 (38%) had progression, and 14 hips (8%) were converted to total hip arthroplasty. Of the 69 hips that had progression, 44 (64%) did not progress until 6 to 10 years after the procedure, whereas 25 (36%) progressed within the first 5 years postoperatively. The best results were obtained in patients with Stage II osteonecrosis in whom 95% of the hips did not progress postoperatively. In contrast, only 39% of the hips in patients with Stage V osteonecrosis remained stable. Preoperative and postoperative clinical evaluation using the Harris hip score showed an increase from 85 to 96 points in hips with Stage II disease; from 74 to 91 points in hips with Stage III disease; from 69 to 85 points in hips with Stage IV disease; and from 61 to 76 in hips with Stage V disease. The current results show that the vascularized fibular graft is an excellent procedure for the precollapse stages and a valuable alternative for patients with Stages III, IV, and V of the disease.

Avascular necrosis of the femoral head constitutes a potentially disabling disease, accounting for as many as 18% of total hip arthroplasties performed in patients in western countries. ^6,21,34 The etiology and pathogenesis of the disease remains elusive, despite the recognition of various associated factors, such as trauma, corticosteroid administration, alcoholism, dysbaric phenomena, pregnancy, and various pathologic entities, including autoimmune disorders, sickle cell anemia, and inflammatory bowel disease. ^22,23 Hypercoagulability is considered a major pathway by which the underlying abnormalities lead to impairment of the vascular supply and subsequently necrosis of the affected area. ¹⁴ Thrombophilia, an increased tendency for thrombus formation, and hypofibrinolysis, a reduced ability for thrombus degradation, have been described in a significant proportion of patients with osteonecrosis. ^10,16,18 Not all patients, however, have recognizable risk factors and are designated as having idiopathic osteonecrosis. Moreover, not all individuals with these risk factors have osteonecrosis develop. These observations show a multifactorial model of pathogenesis that has yet to be clarified.

Because of the ambiguity concerning its pathogenesis, the treatment of osteonecrosis is focused not on the mechanisms of the disease process, but rather on management of the end-stage bone changes. The aim is to halt those changes that sequentially progress to fracture of the necrotic subchondral bone, articular cartilage collapse, and finally to hip osteoarthrosis. ⁸ Total hip arthroplasty replaces the necrotic femoral head by an implanted prosthesis and offers pain alleviation and improvement of function. However, because a large proportion of patients with osteonecrosis are younger than 50 years, total hip arthroplasty is associated with an inferior long-term prognosis. ²³

Many techniques have been introduced to salvage the femoral head including core decompression, rotational osteotomies, and application of bone grafts. Results have been less than ideal and no method clearly has been shown to arrest the disease process before, or to slow the progression of hip osteoarthrosis after subchondral fracture. ³¹ Despite this, free vascularized fibular grafting has been proposed as a promising, femoral head-salvaging modality with satisfactory results reported by many authors. ^{2,4,20,25,31,35} The current study reports on the authors’ 11-year experience in the treatment of avascular necrosis of the femoral head with vascularized fibular grafts in 228 hips (187 patients).

MATERIALS AND METHODS

Study Population

Free vascularized fibular grafting was performed at the authors’ institution for treatment of femoral head osteonecrosis in 187 patients (228 hips), during the past 11 years (March 1989–March 2000). One hundred twenty-nine patients were male (69%) and 58 patients were female (31%) with a mean age at the time of surgery of 32 years (range, 16–54 years).

Of the 228 hips, 184 hips (152 patients) were available for followup at an average of 4.7 years (range, 1–10 years). Clinical evaluation was based on the presence of pain and the functional outcome assessed by the Harris hip score. ¹² Radiologic examination and comparison with radiographs obtained before surgery revealed the fate of the femoral head and the hip after vascularized fibula transplant.

Diagnosis

The diagnosis was based on the history, thorough clinical evaluation, and imaging modalities including plain anteroposterior (AP) and lateral radiographs, magnetic resonance imaging (MRI), and bone scintigraphy. Pain was present in all affected hips. Pain usually was gradual at the beginning and despite temporary improvement with rest, progressive deterioration usually occurred. In some patients, radiation of the pain to the thigh and knee was observed. Range of motion of the hip was influenced to varying degrees, according to the stage of the disease, with internal rotation of the first movement restricted.

Plain radiographs served as the first diagnostic step, serving not only to exclude other pathologic disorders, but to establish the diagnosis of osteonecrosis with the presence of subchondral collapse (crescent sign) and/or a wedge-shaped defect and flattening of the femoral head. Magnetic resonance imaging was done in most patients to more accurately assess the location and extent of the necrotic lesion. If the diagnosis remained obscure on plain radiographs, as in early stages of the disease where no alterations or only obscure lucent and sclerotic changes in the femoral head were observed, bone scintigraphy or MRI were used routinely. Presence of a cold or cold-in-hot spot attributable to reduced radionuclide uptake by the lesion was evident in bone scans, whereas MRI images disclosed signal alterations of the necrotic bone. A band of low signal intensity on T1-weighted MR images surrounding the lesion, representing new bone formation in the reactive interface, and a double line on T2-weighted MR images constituted characteristic patterns of osteonecrosis. ¹³

The disease was bilateral in 111 patients (59%), whereas it involved only one hip in 76 patients (41%). The interval between the two procedures in patients with bilateral involvement, ranged from 3 months to 5 years, with the average time being 9 months.

Classification

The disease was classified based on changes seen on plain radiographs according to Steinberg et al. ²⁷ In 53 hips (23%), the disease was characterized as Stage II, in 56 hips (25%) the disease was characterized as Stage III, in 92 hips (40%) the disease was characterized as Stage IV, and in 27 hips (12%) the disease was characterized as Stage V. The distribution according to the different age subgroups was variable. In patients who were 20 years or younger, Stage II accounted for 26% (25 patients) of the cases, Stage III accounted for 17% (four patients), Stage IV accounted for 22% (six patients), and Stage V accounted for 35% (nine patients) of the cases. In the patients who were 21 to 30 years, 16 hips (19%) were classified as Stage II, 21 were classified as Stage III (25%), 37 were classified as Stage IV (47%), and eight were classified as Stage V (9%). In the patients who were 31 to 40 years, 21 hips (26%) were classified as Stage II, 16 hips (21%) were classified as Stage III, 36 hips (46%) were classified as Stage IV, and six hips (7%) were classified as Stage V. Finally, in patients who were older than 40 years, the percentages for Stages II, III, IV, and V were 23% (10 patients), 33% (14 patients), 31% (13 patients), and 13% (five patients), respectively.

In the first 5-year period using vascularized fibular transplants for osteonecrosis of the femoral head, Stages II, III, IV, and V accounted for 11%, 18%, 50%, and 21% of cases, respectively, whereas in the last 5-year interval, the relative incidence of each stage was 31%, 28%, 35%, and 6%, respectively.

Etiology

In the current series, osteonecrosis was considered secondary to various risk factors in 187 hips (82%). Traumatic events, such as subcapital fracture of the femur and hip dislocation, were responsible for the disease in 29 hips (13%). The risk factors most commonly associated with nontraumatic osteonecrosis were corticosteroid administration and excessive alcohol consumption, present in 84 patients (101 hips; 44%) and 24 patients (12%; 28 hips), respectively. Conditions associated with the disease in the remaining hips with secondary osteonecrosis included systemic lupus erythematosus in nine patients (12 hips; 59%), sickle cell disease in eight patients (nine hips; 4%), inflammatory bowel disease in four patients (four hips; 2%), pregnancy in three patients (three hips; 1.5%), and dysbaric disease in one patient (one hip; 0.5%). In 41 hips (18%), osteonecrosis was characterized as primary or idiopathic, because of the absence of known risk factors in the otherwise apparently healthy patients.

Indications–Preoperative Evaluation

Once the diagnosis of osteonecrosis was established, the presence of a painful hip, the absence of severe osteoarthrotic changes (Stage VI) and an age younger than 55 years were prerequisites for considering the patient as a candidate for the procedure. In addition, the general health status of the patient was assessed with specific measures undertaken. For example, several patients had pathologic conditions that warranted specialist evaluation and careful perioperative treatment patients with immunosuppression who were receiving corticosteroid treatment, patients with sickle cell anemia or patients with excessive alcohol consumption who may have occult hepatic function impairment). In addition, laboratory examination of hematologic and biochemical factors and evaluation of the lipid profile and the coagulation mechanism were performed to identify associated factors that predispose disease development.

Each patient was evaluated to document the presence of detectable pulses of the posterior tibial and dorsalis pedis arteries of the ipsilateral leg to determine if acceptable as the donor site for the procedure. Detailed computed tomography (CT) and MRI studies of the affected hip were undertaken, and a three-dimensional reconstruction of the lesion was performed after computerized analysis of CT and MR images. The exact angle of insertion, length of the fibular graft, and the position of the guide pin used in the procedure were calculated. Subsequently, a custom-made targeting device was designed and manufactured. The Ioannina technique described by Beris and Soucacos ³ enables accurate placement of the graft at the ideal position within the lesion.

Operative Technique

The operative procedure used was a modification of that described by Urbaniak. ³⁰ The patient was placed in the lateral decubitus position and the hip and lower leg were prepared and draped as one sterile field. Two surgical teams (the hip and fibula teams) simultaneously were involved with hip preparation and fibular graft dissection, respectively, to diminish operative time. The approach to the hip was through a curved, anterolateral incision, splitting the interval between the tensor fascia lata and the gluteus medius and exposing the lateral aspect of the proximal part of the femur. The deep plane of dissection was between the rectus femoris and the vastus intermedialis, which permitted identification and preparation of the recipient site vessels, namely the ascending branch of the lateral circumflex femoral artery with its two accompanying veins.

The ipsilateral fibula was approached through a lateral incision, as described by Gilbert. ⁹ Preparation of the graft involved proximal and distal osteotomies, and dissection of the vascular pedicle with release of the flexor hallucis longus and posterior tibial muscles. Care was taken to identify the posterior tibial vessels, which could be absent in the case of congenital variations or easily mistaken for the peroneal vessels. After exposing and freeing the origin of the peroneal vessel pedicle, it generally was preferable to deflate the tourniquet and let the graft remain in situ until preparation of the hip had proceeded, instead of harvesting it at this point. This modification reduced the total time of graft ischemia and is thought to have had a beneficial effect on the survival of the cellular elements of the graft.

Under direct fluoroscopic control, a Steinmann pin is directed into the center of the lesion. Anteroposterior and lateral images were used to confirm that there was adequate space in the femoral neck for drilling an 18 mm-diameter tunnel centered on the pin. Progressive reaming, with the Steinmann pin serving initially as a guide, was used to form a cylindrical tunnel extending from the inferolateral aspect of the greater trochanter to the necrotic area and as much as 5 mm of the femoral head articular surface. The Ioannina technique was introduced for accurate targeting of the necrotic area and to optimize the graft position with the aid of a preoperative computer analysis. The innovation involves the use of a custom-made targeting device that enables accurate placement of the guide pin in an ideal position with significant reduction in the use of the C-arm fluoroscopic unit and associated irradiation exposure. After completion of the tunnel, the necrotic bone was removed from the femoral head by a curette. Cancellous healthy bone, harvested during the reaming procedure, was packed into the cavity and the length of the tunnel was determined.

The fibular graft then was freed from the leg just before completion of hip preparation. After examination for bleeding, the subcutaneous layer and skin were closed over a drain. Final preparation of the graft included microsurgical dissection of the pedicle and osteotomy of the graft to the optimal length. In addition, the periosteum of the end of the graft to be implanted in the lesion was reflected and sutured so as to expose the inner cambial layer of periosteum containing mesenchymal cells. This was done to enhance bone formation during the repair process.

The fibula then was inserted into the cylindrical hole with the triangular cross section of the graft allowing the pedicle to remain uncompromised. Stabilization of the graft was accomplished in the first cases done using this technique with a 3.5 mm cortical screw, whereas currently a 1.5 mm Kirschner (K) wire is preferred. Titanium implants are preferred because they do not interfere with future MRI examinations. Vein and artery end-to-end anastomoses were performed using magnifying loops with 8–0 or 9–0 interrupted nylon sutures; the vein was sutured first to reduce bleeding. The patency of the anastomoses was evaluated and endosteal bleeding from the fibula confirmed vascularization of the graft. The gluteal fascia and skin were closed under drainage. The operative time in the first 3-year period with this procedure averaged 4.5 hours (range, 3–6 hours). Currently, the operative time averages 3.5 hours (range, 2.5–5 hours).

Postoperative Treatment

Postoperatively, prophylactic antibiotics were administered intravenously for 2 days, an anticoagulation regimen of low molecular weight heparin was administered subcutaneously for 3 weeks, followed by low-dose aspirin per os for 2 months. Mobilization is initiated on the second postoperative day after drain removal. The patient is instructed to walk using two crutches with minimal weightbearing. Gradual increase of the weight leads to full weightbearing after 6 months.

RESULTS

Complications

No systemic complications, such as uncontrollable hemodynamic derangement, deep venous thrombosis, or pulmonary embolism were observed. Donor site morbidity was minimal in the series. A transient weakness of ankle eversion with mild discomfort was observed, which resolved within 3 months. Flexion contracture of the interphalangeal joints of the great toe, attributable to adhesions of the flexor hallucis longus, occurred in four patients and was corrected surgically. There were no sensory abnormalities or peroneal nerve palsy, compartment, or tibia fracture that have been reported in the literature as donor site complications. ^11,33 Recipient site complications consisted of three cases of deep infection, two of which resulted in septic arthritis and finally to destruction of the joint and subsequent total hip arthroplasy.

Disease Progression

Of the 184 hips treated with vascularized fibular grafting, 101 (54%) remained stable postoperatively, whereas 69 hips (38%) had progression and 14 (8%) were converted to total hip arthroplasty. Of the 69 hips that had progression, 44 (64%) did not progress until 6 to 10 years after the procedure, whereas 25 hips (36%) progressed within the first 5 years postoperatively. The best results were obtained in Stage II where 95% of the hips did not have progression postoperatively. In contrast, only 39% of the hips with Stage V osteonecrosis remained stable.

More specifically, according to the initial stage of the disease, progression was observed as follows: Of the 39 hips in Stage II, 37 (95%) remained stable and two (5%) progressed; of the 45 hips in Stage III, 29 (64%) were stable and 16 (36%) progressed; of the 77 hips in Stage IV, 40 (52%) were stable and 37 (48%) progressed; and of the 23 hips in Stage V, nine (39%) were stable and 14 (61%) progressed (Figs 1,2).

Fig 1.: The relative incidence of stability, progression, and conversion to total hip arthroplasty (THA) for each disease stage of avascular necrosis of the femoral head is shown.

Fig 2 A–D.: (A) Preoperative radiograph showing Stage II avascular necrosis of the femoral head secondary to alcohol abuse and unsuccessful core decompression in a 35-year-old man. Postoperative views obtained at (B) 3 years and (C) 6 years show stability of the articular cartilage and femoral head, whereas (D) a radiograph taken at 9 years shows moderate narrowing of the joint space although the patient did not have pain. Total hip replacement was not required.

Conversion to Total Hip Arthroplasty

Conversion to total hip arthroplasty was performed in 14 of the 184 hips (7.6%). The incidence of conversion varied considerably regarding the stage of the disease. None of the 39 procedures performed for Stage II lesions were converted to total hip arthroplasty (0%), whereas one (2%) of the 45 hips with Stage III disease, nine of 77 (12%) hips in Stage IV, and four of 23 (17%) hips in Stage V underwent conversion.

Clinical Outcome

The vascularized fibular grafting procedure was associated with marked pain relief and improvement of function in hips with all stages. Excluding hips that finally were treated with total hip arthroplasty, the Harris hip score 12 increased from 85 to 96 points for hips with Stage II disease, from 74 to 91 points for hips with Stage III disease, from 69 to 85 points for hips with Stage IV disease, and from 61 to 76 points for hips with Stage V disease (Figs 3–5).

Fig 3A–B.: (A) Preoperative radiograph of a 30-year-old-man with bilateral osteonecroses (Stage IV) of the hip associated with steroid administration for the treatment of malignant lymphoma. (B) Radiograph obtained 3 and 4 years after vascularized fibular transplant for the left and right hip, respectively shows that the disease stage has remained stable. The patient does not have pain and participates in normal daily activities.

Fig 4A–B.: (A) Preoperative radiograph of a 22-year-old-woman with Stage II avascular necrosis of the femoral head that was associated with steroid administration because of neurilemoma of the acoustic nerve. (B) Radiograph taken at 10-year followup shows excellent results with no additional deterioration of the joint. The patient did not have pain and had full range of joint motion.

Fig 5A–D.: (A) Preoperative radiograph of a 36-year-old-woman with Stage II osteonecrosis of the left hip secondary to pregnancy. The patient refused surgery and (B) on return 9 months later, the disease had progressed to Stage IV radiographically. Marked cystic formation, bone sclerosis, flattening of the femoral head and articular cartilage collapse can be seen. The patient underwent a vascularized fibular graft. (C) Advanced osteoarthritic changes were seen 5 years postoperatively. (D) Although additional destruction of the hip was obvious on radiographic followup at 9 years, conversion to total hip replacement was delayed.

DISCUSSION

The high prevalence of osteonecrosis in the young population and the obscure pathogenesis of the disease process raise growing concern regarding effective treatment of this clinical entity. Moreover, the natural history of the disease is characterized by gradual deterioration in the majority of cases. Merle d’Aubigne et al ²³ reported that 75% of their patients had collapse of the femoral head within 3 years of followup. The presence of symptoms is considered an influencing factor, because the painful hip almost invariably progresses to femoral head collapse without treatment, whereas the asymptomatic hip may have a variable course. ⁷

The advent of microsurgery made free tissue transfer a possibility and introduced free vascularized fibular grafting to the armamentarium of the orthopaedic surgeon for the treatment of patients with avascular necrosis of the femoral head. The outcome of this procedure, which does not violate the joint capsule, has been consistently satisfactory, as reported by many authors. ^{5,9,11,15,20,21,25,30–32} Despite the variety and versatility of the femoral head salvaging procedures used in the treatment of patients with osteonecrosis, none has proven to be ideal. However, although the results from different studies vary considerably, vascularized fibular grafting seems to be the most promising technique with a consistently high incidence of satisfactory clinical results and low rate of failure reproducible in most studies. ^{1,5,9,11,15,20,21,25,30–32}

Urbaniak and Harvey ³² observed an 83% survival of the graft of 646 hips after a followup of 1 to 17 years. A long-term study from the same center evaluated hips with a minimum followup of 5 years (mean, 7 years) and documented failure rates as low as 11% for hips without subchondral collapse. ³¹ Moreover, significant clinical improvement took place in all disease stages and patient satisfaction from fibular graft was high with 81% of patients satisfied even when the patients who subsequently underwent arthroplasty were included. ³¹

Sotereanos et al ²⁵ reported that conversion to total hip arthroplasty was required in only 23% in hips after a mean of 5 years and the clinical outcome was good or excellent in 90% of the remaining cases. Good or excellent outcome after a similar followup was reported by Yoo et al ³⁵ in 91% of hips and by Brunelli et al ⁴ in 78% of hips. In addition to these observations, the general health status of patients with osteonecrosis of the femoral head treated by fibular graft was assessed in study in which it was concluded that the procedure provides pain relief, functional improvement, and a satisfactory general health status. ¹⁹

In a review of 1303 vascularized fibula grafts performed at seven different centers with a minimum followup of 2 years, a survival rate of 83% was observed, which increased to 88% for patients who were treated before femoral head collapse. ⁵ The 10-year experience of the current authors with fibular graft was similarly positive with noteworthy clinical improvement in patients with all stages of the disease and a rate of conversion to total hip arthroplasty of only 8%, confirming the initial encouraging results. ^2,20

Studies comparing vascularized fibular grafting with core decompression established the superiority of the fibular graft. In a prospective study, Kane et al ¹⁵ observed conversion to total hip arthroplasty at a significantly lower rate of 20% in the hips treated with vascularized fibula transplants, compared with 58% in hips treated with core decompression. Similarly, a retrospective comparison showed an improved outcome with the fibular graft with 4-year survival rates of 89% compared with 65% for hips with Ficat Stage II osteonecrosis and of 81% versus 21% for hips with Ficat Stage III disease. ²⁴

In comparison with total hip arthroplasty, the fibular graft may not provide complete pain alleviation and is associated with a longer rehabilitation period. However, arrest of the disease process and salvage of the femoral head is highly possible if the procedure is done before subchondral collapse. In subsequent stages, the progression to osteoarthrosis of the hip seems to be inevitable but at considerable time delay. Despite this, conversion to total hip arthroplasty would be preferable to revision hip surgery, which frequently may be needed in young patients initially treated by total hip arthroplasty. Moreover, very little implant material is used with fibular grafts with minimal risk of infection.

The favorable outcome of fibular grafts is attributed to decompression, which leads to reduction of the high intraosseous pressure and facilitates perfusion of the femoral head. In addition, necrotic bone, which acts as a barrier to the reparative process, is removed and is substituted with a cancellous autograft that has osteoconductive and osteoinductive properties. The cortical strut reinforces and protects the weakened subchondral bone, and reconstitution of the graft’s intrinsic vascularity by anastomosis of the pedicle to the recipient site vessels enhances the blood supply and delivery of nutrients to the reparative zone.

Various parameters influence the outcome of the fibular graft procedure. The stage of the disease has been documented in several studies as a crucial prognostic factor, where initiation of treatment in early stages has been associated with improved outcome in all the femoral head-salvaging techniques. ^{5,7,20,21,24,25,30–32,35} Results in hips with Steinberg Stage II disease, before development of a subchondral fracture and appearance of a crescent sign in plain radiographs, generally are outstanding. At a mean of 7 years, Urbaniak et al ³¹ observed disease progression in only 37% of the hips with Stage II disease and conversion to total hip arthroplasty in 11%. In the current series, progression of the disease took place in only 5% of hips, without need for conversion to total hip arthroplasty. In Stages III and IV lesions, Urbaniak et al ³¹ reported a conversion rate of 23% and 43%, respectively. The current authors report similar results. Moreover, this experience is reflected in the increased proportion of hips with Stage II and Stage III osteonecrosis that were treated in the second ½ of the reported 10-year period (59% versus 29%). However, even in hips with Stages III and IV disease, vascularized fibula transplants were associated with clinical improvement and high survival rates. In the current series, 98% of hips with Stage III disease and 88% of hips with Stage IV disease survived, suggesting that fibula grafting is an effective means for postponing total hip arthroplasty in the majority of patients with Stage III or Stage IV disease. Urbaniak et al ³¹ reported somewhat lower survival rates of 77% and 57% in hips with Stage III and Stage IV disease, respectively.

The size of the lesion also has been related to outcome. ^25,26 Using a method for more accurate quantification of the extent of the necrotic lesion using MR images showed an association between lesion size and collapse in osteonecrosis. ¹⁷ In addition, the location of the necrotic area also may be of importance, because increased collapse rates were reported for lesions involving the weightbearing portions. ^28,29

The presence of a symptomatic Stage II to Stage IV lesion in a patient younger than 50 years is considered the main indication for fibula grafting. ³² Patients with Stage V lesions have the worst outcome. The current series showed disease progression in 61% of the patients, but conversion to total hip arthroplasty was done in only 17% to date. Very young patients with Stage V lesions still may benefit from the procedure. In the current series, Stage V lesions were present in 35% of the patients who were 20 years of age or younger, compared with 12% overall. In addition, the size of the lesion should be considered in decision making. Total hip arthroplasty may be preferable in a patient older than 40 years who presents with a Stage IV lesion with femoral head involvement of more than 50%. ³² Moreover, the extension of the indications for fibular graft to asymptomatic hips may be justified.

Vascularized fibular grafting constitutes a safe and efficient procedure for treatment of patients with avascular necrosis of the femoral head. It has the potential to arrest or substantially delay the disease process and is the method of choice in young patients without severe osteoarthritic changes of the hip.