Kenji Ohzono

Natural history of nontraumatic avascular necrosis of the femoral head

We studied the natural history of nontraumatic avascular necrosis of the femoral head (ANFH) in 115 hips in 87 patients, 69 steroid-induced, 21 related to misuse of alcohol and 25 idiopathic. The average length of follow-up was over five years. Collapse occurred most often when the focus of bone necrosis occupied the weight-bearing surface of the femoral head. Flatness of the head due to subchondral fracture was an early manifestation of collapse. Classification into six types based upon the radiographic findings provided an accurate prognosis for individual cases of ANFH which is useful in planning treatment and in assessing its outcome.

Measurements of pelvic flexion angle using three-dimensional computed tomography.

The purpose of the current study was to evaluate whether safe acetabular component position depends on differences in pelvic location between the supine, standing, and sitting positions. The subjects of the current study were 101 patients who had total hip arthroplasty. Anteroposterior radiographs of the pelvis with the patients in the supine, standing, and sitting positions were obtained preoperatively and 1 year after total hip arthroplasty. Computed tomography images of the pelvis were obtained preoperatively. Using image matching between the three-dimensional computed tomography model and anteroposterior radiograph, pelvic flexion angles with the patient in the supine, standing, and sitting positions were calculated. The mean preoperative pelvic flexion angle was 5° ± 9° (range, −37°–30°) in the supine position, 3° ± 12° (range, −46°–33°) in the standing position, and −29° ± 12° (range, −62°–10°) in the sitting position. Because there was much intersubject variability in pelvic flexion angle, it is not appropriate to determine orientation of the acetabular component from anatomic landmarks. In 90% of the cases, the difference in pelvic flexion angle between the supine and standing positions preoperatively was 10° or less. In 90% of the cases, there was 20° or greater extension of the pelvis from the supine position to the sitting position preoperatively, and the safe range of flexion of the hip from anterior prosthetic impingement in the sitting position was 20° or greater than that in the supine position. Preoperative pelvic position in each case was almost completely maintained 1 year after total hip arthroplasty. It is reasonable to regard the pelvic position in the supine position as the functional pelvic position and proper pelvic reference frame in determining optimal orientation of the acetabular component in 90% of cases before and 1 year after total hip arthroplasty, although an adjustment of orientation of the acetabular component was needed for the remaining cases.

Joint-preserving operations for idiopathic avascular necrosis of the femoral head. Results of core decompression, grafting and osteotomy

We have reviewed 54 hips in 46 patients from 2 to 14 years after a joint-preserving operation for idiopathic avascular necrosis of the femoral head. The choice between core decompression (17 hips), bone grafting (18), rotation osteotomy (15) or varus osteotomy (4) was determined by the stage and location of the area of necrosis. The overall success rate was unexpectedly low at 60%. Core decompression and bone grafting by our techniques gave poor long-term results, but those of rotation or varus osteotomies, performed with care for the correct indications, were better. The indications for each procedure are discussed: osteotomy is best when the area of necrosis is shallow and localised in the medial or anterior portion of the femoral head.

Prognostication of osteonecrosis of the femoral head in patients with systemic lupus erythematosus by magnetic resonance imaging.

To detect and prognosticate osteonecrosis of the femoral head in the preradiographic stage, 60 patients with systemic lupus erythematosus who had normal hip radiology were followed prospectively for a mean period of 5 years (range, 3–7 years) using magnetic resonance imaging (MRI). The first MRI scans showed a low intensity band in the femoral head of normal fat intensity on Tl weighted images in 16 hips of 9 patients. The MRI findings were used to classify the lesions into three categories. Type A (six hips): the lesions occupied the medial one third or less of the weight bearing portion. Type B (two hips): the lesions occupied the medial two thirds or less of the weight bearing portion. Type C (eight hips): the lesions occupied more than the medial two thirds of the weight bearing portion. At the final followup, all of the Type A and one of the Type B hips were classified as being in Stage 1, and one Type B and two Type C hips had progressed to Stage 2. The MRI appearance of six Type C hips had changed from a band to an inhomogeneous pattern with the femoral head progressing to collapse on radiographs 2–5 years after the diagnosis of systemic lupus erythematosus. The remaining hips, which had been classified as normal at the first MRI, maintained a normal appearance, except for one hip that developed a Type A lesion. The presence of a low intensity band on Tl weighted images was an early specific finding of osteonecrosis of the femoral head, and extensive lesions demarcated by band images signified a poorer prognosis in systemic lupus erythematosus patients. If no MRI abnormalities appeared after 1 year of the startup treatment for systemic lupus erythematosus, there was little risk of femoral head collapse based on the subsequent clinical course of the patients followed in the current study.

Rotational osteotomy for non-traumatic avascular necrosis of the femoral head

We reviewed 41 hips in 40 patients at three to 11 years (average 6.3 years) after Sugioka transtrochanteric rotational osteotomy for non-traumatic avascular necrosis of the femoral head. The clinical results were excellent or good in 23 hips (56%) and the radiological success rate was 56%. Failure was due to fracture of the femoral neck, nonunion of the osteotomy, secondary collapse, or osteoarthritis. Nonunion and femoral neck fracture were more common after the use of the large screws described by Sugioka than with AO blade plates. Secondary collapse was significantly more common when less than one-third of the posterior articular surface was intact (p = 0.002). Postoperative degenerative changes were seen in cases with stage III avascular necrosis. We conclude that success depends to a large extent on the amount and stage of necrosis of the femoral head, but that careful technique and the use of AO hip plates may increase the likelihood of a satisfactory result.

The fate of nontraumatic avascular necrosis of the femoral head. A radiologic classification to formulate prognosis.

One hundred fifteen hips in 87 patients with non-traumatic avascular necrosis of the femoral head (ANFH) (men, 54; women, 33) (steroid induced, 49; alcoholic, 21; idiopathic, 17) were radiologically classified into six distinct types (Types 1A, 1B, 1C, 2, 3A, and 3B) based on the following: (1) the size and location of the necrotic area in relation to the weight-bearing surface of the acetabulum as seen on anteroposterior views in the standing position and (2) initial roentgenographic abnormalities. The natural course of the disorder in each group were observed for more than two years (range, two to 18 years; mean, five years). Of the 79 hips without collapse of the femoral heads (Stage II) at the beginning of follow-up evaluation, 42 femoral heads subsequently collapsed. These collapses took place predominantly in cases involving Types 1C, 2, and 3B. Conversely, the incidence of collapse was significantly low in the other groups (Types 1A, 1B, and 3A). Collapse of the femoral head occurred in 78 femoral heads of 115. The incidences of collapse by group was 0% for Type 1A, 19% for Type 1B, 94% for Type 1C, 100% for Type 2, 12% for Type 3A, and 100% for Type 3B. These data indicate that this radiologic classification of necrosed femoral heads is useful for evaluation of the risk of collapse or for prognosis of the affected hip joints as well as for choosing an appropriate treatment modality, either conservative or surgical, during the early stages of ANFH.

Significance of lesion size and location in the prediction of collapse of osteonecrosis of the femoral head: a new three-dimensional quantification using magnetic resonance imaging

Size and location of a necrotic lesion are considered important factors predicting collapse of the femoral head in the early stages of osteonecrosis. However, few analytical studies have performed a three‐dimensional quantification of lesions to clarify how these two morphological factors are related to the occurrence of collapse. We evaluated the relevance of lesion size and location for prediction of collapse quantitatively using new three‐dimensional indexes. Magnetic resonance (MR) imaging was performed in 65 hips in a consecutive series of 47 patients with osteonecrosis without radiological evidence of collapse. Lesion volume as well as latitude and longitude of the center of gravity of the lesion within the femoral head were calculated. Thirty‐three hips developed radiological collapse, while in the remaining 32 hips collapse did not occur over 2 years. Multiple logistic regression analysis showed a significant relationship between lesion volume and radiological collapse. In 35 hips in which the lesion volume was less than 30% of the femoral head, only 9 collapsed. In comparison with non‐collapsed hips, collapsed hips had a significantly higher combined value for latitude and longitude of the lesion, corresponding to the anterosuperior portion of the femoral head. Quantitative analysis of lesion morphology demonstrated that lesion volume is strongly correlated with risk of collapse, and that lesion location is an important prognostic indicator of collapse in small necrotic lesions.

Computed-Tomography-Based Computer Preoperative Planning for Total Hip Arthroplasty

For precise preoperative planning in total hip arthroplasty (THA), we developed a technique of computed tomography (CT)-based computer preoperative planning and compared this technique with the single X-ray and template method generally used. The subjects of this study were 42 hips in 38 patients who underwent THA using a cementless total hip system. Preoperatively, a standard anteroposterior X-ray of the hip was taken, and conventional preoperative planning was done with a template of the total hip system. Transverse images were obtained using a helical CT scanner, and a CT-based computer preoperative plan was performed on true coronal slice images of the proximal femur reconstructed from CT data. Postoperatively, 29 hips (69%) showed good proximal fit of the femoral component to the medial endosteal line. Of the 20 hips with good proximal fit on preoperative X-ray planning, 12 hips had good proximal fit on postoperative X rays. Sensitivity and specificity of the proximal fit on X-ray templating were 41 and 23%, respectively. In 27 of 28 hips with good proximal fit on reconstructed CT images preoperatively, the postoperative X ray revealed good proximal fit. Sensitivity and specificity of the proximal fit on computer planning were 93 and 86%, respectively. Twelve hips with good proximal fit on preoperative templating, the reconstructed images, and the postoperative X ray had 20 degrees or less of combined femoral neck anteversion and external rotational contracture of the hip on the X-ray table. Eight hips with good proximal fit on preoperative templating and proximal poor fit on the reconstructed images had 17-65 degrees of combined version and rotational contracture. In 16 hips with poor proximal fit on preoperative templating and good proximal fit on the reconstructed images, the combined version and rotational contracture ranged from 17 to 69 degrees. When combined femoral neck anteversion and external rotational contracture of the hip is less than 15 degrees, the simple X-ray and template method might be sufficient for THA planning. Otherwise, the CT-based method of preoperative planning is recommended.

Progression and cessation of collapse in osteonecrosis of the femoral head.

Thirty-five patients (54 hips) with osteonecrosis of the femoral head without collapse or only with a crescent sign were followed up for at least 5 years to clarify the natural course of osteonecrosis. During the followup, 28 hips (52%) in 21 patients collapsed including nine hips in nine patients with small necrotic lesions occupying less than the medial ⅔ of the weightbearing area. Cessation of collapse then was observed in 15 (14 patients) of the 28 hips (54%), especially in eight of the nine hips (89%) with small necrotic lesions. Of the 15 hips in 14 patients with cessation of collapse, 11 hips (73%) in 11 patients had less than 2 mm collapse and 10 hips in nine patients became asymptomatic. The analysis indicated that collapse of the femoral head does not necessarily determine a poor prognosis, and even after collapse occurs, subsequent cessation of collapse can be expected in a certain percentage of hips. Hips with less than 2 mm collapse and necrotic lesions occupying less than the medial ⅔ of the weightbearing area have a high chance of cessation of collapse and improvement of symptoms with no surgical intervention.

Experimental steroid-induced osteonecrosis in adult rabbits with hypersensitivity vasculitis.

Osteonecrosis (ON) was experimentally induced in rabbits by employing a combined protocol of hypersensitivity vasculitis and administration of high-dose corticosteroids. Thirty-five adult rabbits were used: five were injected twice with horse serum (Group A), five were injected three times with methylprednisolone acetate (Group B), 20 were treated with a combination of horse serum and methylprednisolone acetate (Group C), and five were used as a control (Group D). Both femurs of each rabbit were obtained one to five weeks after the final treatment and were histologically examined. There was no evidence of ON in Groups A, B, and D, whereas vasculitis was prominent in the femurs of Group A rabbits. In Group C, 14 of 20 specimens (70%) showed histologic evidence of ON in the femoral metaphysis: seven showed marrow necrosis and seven marrow and trabecular necrosis. Intramedullary hemorrhage was detected in eight animals. All specimens that showed ON or marrow necrosis revealed arteriopathy (i.e., severe damage to the vascular wall structure of arterioles). These findings were similar to those observed in early ON of clinical materials. The authors conclude that arteriopathy plays an important role in the pathogenesis of ON.