EffectsofAgeandBodyMassIndexontheResults of Transtrochanteric Rotational Osteotomy for Femoral Head Osteonecrosis

Transcription

1 314 COPYRIGHT Ó 2010 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED EffectsofAgeanBoyMassInexontheResults of Transtrochanteric Rotational Osteotomy for Femoral Hea Osteonecrosis By Yong-Chan Ha, MD, Hee Joong Kim, MD, Shin-Yoon Kim, MD, Ki-Choul Kim, MD, Young-Kyun Lee, MD, an Kyung-Hoi Koo, MD Investigation performe at the Department of Orthopaeic Surgery, Chung-Ang University College of Meicine, Seoul; the Department of Orthopaeic Surgery, Seoul National University College of Meicine, Seoul; an the Department of Orthopaeic Surgery, Kyungpook National University College of Meicine, Daegu, South Korea Backgroun: Avance-stage osteonecrosis an a large area of necrotic bone are known risk factors for failure of transtrochanteric rotational osteotomy of the hip in patients with osteonecrosis. The purpose of this stuy was to etermine whether there are other risk factors for failure of this osteotomy. Methos: One hunre an five patients (113 hips) unerwent an anterior transtrochanteric rotational osteotomy for the treatment of femoral hea osteonecrosis an were followe for a mean of 51.3 months postoperatively. Raiographic failure was efine as seconary collapse or osteoarthritic change. Multivariate analysis was performe to assess factors associate with seconary collapse an osteophyte formation. The Kaplan-Meier prouct-limit metho was use to estimate survival. Results: Seconary collapse occurre in twenty-seven hips (24%), an fourteen hips (12%) were converte to a total hip arthroplasty. At the time of the most recent follow-up, the hip scores accoring to the system of Merle Aubigné et al. range from 6 to 18 points (mean, 15.8 points). Multivariate analysis showe that the stage of the necrosis (III or greater) (hazar ratio = 3.28; 95% confience interval = 1.49 to 7.24), age of the patient (forty years or oler) (hazar ratio = 1.08; 95% confience interval = 1.02 to 1.14), boy mass inex ( 24 kg/m 2 ) (hazar ratio = 1.19; 95% confience interval = 1.03 to 1.38), an extent of the necrosis (a combine necrotic angle of 230 ) (hazar ratio = 1.08; 95% confience interval = 1.04 to 1.11) were associate with seconary collapse. Seven of the eighty-six hips without collapse showe progression to osteoarthritis. The survival rate at 110 months was 63.4% (95% confience interval = 51.1% to 75.7%) with total hip arthroplasty or raiographic failure as the en point an 56.0% (95% confience interval = 44.6% to 67.4%) with total hip arthroplasty, raiographic failure, or loss to follow-up as the en point. Conclusions: Our stuy showe that age, boy mass inex, an the stage an extent of the osteonecrosis were etermining factors for seconary collapse, unsatisfactory clinical results, an conversion to total hip arthroplasty. These factors shoul be consiere when selecting patients for a transtrochanteric rotational osteotomy. Level of Evience: Prognostic Level II. See Instructions to Authors for a complete escription of levels of evience. Osteonecrosis of the femoral hea occurs in young aults, an high rates of failure of total hip arthroplasty ue to excessive polyethylene wear, periprosthetic osteolysis, an aseptic loosening have been reporte in these young patients 1-4. Although total hip arthroplasty with the use of contemporary articulations is known to have a lower rate of failure 5-7, treating young patients with femoral hea osteonecrosis remains challenging. Anterior transtrochanteric rotational osteotomy is one of the surgical proceures use to preserve the hip joint. It moves the necrotic portion of the femoral hea from the weight-bearing region to a non-weight-bearing region 8. Previous stuies have emonstrate variable rates of failure after transtrochanteric rotational osteotomy Although Sugioka et al. an other surgeons reporte satisfactory outcomes 8-12, results from Western countries have not been as Disclosure: The authors i not receive any outsie funing or grants in support of their research for or preparation of this work. Neither they nor a member of their immeiate families receive payments or other benefits or a commitment or agreement to provie such benefits from a commercial entity. J Bone Joint Surg Am. 2010;92: oi: /jbjs.h.01020

2 315 satisfactory The preoperative stage of the isease an the size of the necrotic portion of the femoral hea are factors known to affect the result of the osteotomy Inappropriate patient selection, inaequate surgical technique, an failure of fixation have been associate with poor outcomes after the osteotomy 13,15. From 1993 to 2005, we performe 296 anterior transtrochanteric rotational osteotomies in 277 patients who were younger than fifty-five years of age. Although we use magnetic resonance imaging to ai in patient selection, seconary collapse of the femoral hea an seconary osteoarthritis after the osteotomy were observe in some patients. Therefore, the purpose of this stuy was to etermine risk factors associate with seconary collapse an osteoarthritic change in an effort to better efine the criteria for selecting patients for the osteotomy. Materials an Methos Patient Selection From January 1999 to December 2004, 119 patients (127 hips) who were iagnose as having femoral hea osteonecrosis an met the selection criteria for the anterior transtrochanteric rotational osteotomy were operate on. The iagnosis of osteonecrosis was base on finings on raiographs an magnetic resonance imaging 16. The selection criteria for the osteotomy were osteonecrosis classifie, with the moifie Ficat system, as stage IIB (a crescentic subchonral line an segmental flattening of the femoral hea) or stage III (a sequestrum, a break in the articular cartilage from one en of the affecte area to the other, an a normal or increase joint space) 17,18 ; an age younger than fifty-five years; a painful hip; an an aequate area of viable bone (an arc of >120 between the central vertical line of the femoral hea an the posterior margin of the necrotic portion as seen on a misagittal magnetic resonance imaging scan) (Fig. 1) 12. The esign an protocol of this retrospective stuy were approve by the institutional review boar at our hospital, an all patients provie informe consent preoperatively. Fig. 1 Misagittal T1-weighte magnetic resonance imaging scan showing measurement of the angle of the intact area (I), which is efine as the arc between the central vertical line of the femoral hea an the posterior margin of the necrotic portion. Demographic Data Of the 119 patients (127 hips), fourteen patients (fourteen hips) were lost before thirty-six months (the minimum uration of follow-up) ha elapse. The last evaluations of these fourteen patients were one at nine to thirty months (mean, twenty-five months) postoperatively. None of the patients lost to follow-up ha ha a total hip arthroplasty or seconary collapse at the time of the last evaluation. The remaining 105 patients (113 hips) were followe until the time of conversion to total hip arthroplasty or for thirty-six to 108 months (mean, 51.3 months). There were ninety men (ninety-eight hips) an fifteen women (fifteen hips). The mean patient age at the time of the osteotomy was 34.3 years (range, eighteen to fifty-five years). The mean boy mass inex was 24.0 kg/m 2 (range, 18.5 to 31.1 kg/m 2 ). The cause of the osteonecrosis of the femoral hea was iiopathic in forty-nine hips (forty-five patients), alcoholassociate in forty hips (thirty-seven patients), steroi-inuce in nineteen hips (eighteen patients), an posttraumatic in five hips (five patients). Prior to the osteotomy, nineteen patients (twenty-one hips) were seentary workers, seventy patients (seventy-six hips) performe moerately strenuous labor, nine patients (nine hips) performe labor of intermeiate strenuousness, an seven patients (seven hips) performe intensive labor 19. Preoperatively, accoring to the moifie raiographic classification system of Ficat 17,18, sixty-nine hips were in stage IIB an forty-four hips were in stage III. The Kerboul combine necrotic angle 20,21 range from 195 to 260 (mean, ) (TableI).Thepreoperativehipscoresaccoringtothe system of Merle Aubigné etal. 22 range from 9 to 16 points (mean, 14.1 points). Preoperative Measurement of the Arc of the Necrotic Portion on Magnetic Resonance Imaging Scans We measure the extent of the osteonecrosis with a moification of the combine necrotic angle of Kerboul et al. 20 as we escribe previously 21. The necrotic angle was measure on magnetic resonance imaging scans instea of raiographs. The micoronal an misagittal sections, which showe the largest iameter of the femoral hea, were use for the measurement.

3 316 TABLE I Demographic Data No. of patients (hips) 105 (113) Male:female (no. [%] of patients) 90 (86%):15 (14%) Age* (yr) 34.3 ± 8.6 (18-55) Boy mass inex* (kg/m 2 ) 24.0 ± 2.7 ( ) Cause of osteonecrosis (no. of patients [no.; % of hips]) Alcohol-associate 37 (40; 35%) Steroi-inuce 18 (19; 17%) Posttraumatic 5 (5; 5%) Iiopathic 45 (49; 43%) Stage (no. [%] of hips) IIB 69 (61%) III 44 (39%) Strenuousness of occupation (no. of patients [no.; % of hips]) Seentary 19 (21; 18%) Moerate 70 (76; 67%) Intermeiate 9 (9; 9%) Intensive 7 (7; 7%) Combine necrotic angle* (eg) ± 12.9 ( ) *The values are given as the mean an stanareviation with the range in parentheses. The angle of the necrotic portion on the micoronal image (A) an the angle of the necrotic portion on the misagittal image (B) were measure, an the combine necrotic angle was calculate as A 1 B (Fig. 2). Operative Technique an Postoperative Care The anterior transtrochanteric rotational osteotomies were performe by one surgeon (K.-H.K.) using a technique that has been escribe by Sugioka et al. 8,9. The osteotomy site was fixe with use of a 120 compression hip screw (Solco, Seoul, South Korea). Postoperative care consiste of the use of crutches to walk with protecte weight-bearing for three to six months, until there was raiographic evience of osseous union of the osteotomy site. Follow-up Evaluations Follow-up evaluations were performe at three, six, nine, an twelve months postoperatively an every six months thereafter. Patients who ha not returne for regularly scheule visits were contacte by telephone. Clinical assessment was one accoring to the system of Merle Aubigné et al. 22, which allots up to 6 points each for pain, mobility, an gait. A score of 17 or 18 points was recore as an excellent result; 15 or 16 points, as a goo result; 13 or 14 points, as a fair result; an 12 points, as a poor result. All of the patients unerwent a bone scan at two to four weeks postoperatively to assess the viability of the femoral hea. Raiographic assessment inclueetermination of whether there was postoperative collapse of the newly establishe weight-bearing area of the femoral hea, osteophyte formation aroun the femoral hea, or osteoarthritic change. We consiere the six-week raiographs as the baseline stuies to be use for comparison. Seconary collapse was efine as subsience of the articular surface exceeing 2 mm as compare with the contour on the baseline raiograph 12. Osteophyte formation was efine as spurs aroun the femoral hea with a thickness exceeing 3 mm as compare with the contour on the baseline raiograph 23. We measure the minimum with of the Fig. 2 Calculation of the combine necrotic angle on a magnetic resonance imaging scan. A = the angle of the necrotic area on the micoronal T1-weighte image, an B = the angle of the necrotic area on the misagittal T1-weighte image. The combine necrotic angle = A 1 B.

4 317 hip joint space to etermine osteoarthritic change, with a with of 2.0 mm consiere to be joint space narrowing 24. We calculate the time from the inex osteotomy to conversion to a total hip arthroplasty for patients who unerwent such a conversion uring the stuy perio. Statistical Analysis The sex, age, boy mass inex, an occupation of the patient; the cause an stage of the osteonecrosis; an the combine necrotic angle were assesse to etermine the relationships between these variables an seconary collapse, osteophyte formation, hip score, an necessity for conversion to total hip arthroplasty. We use a chi-square test or Fisher exact test for categorical variables an a t test or analysis of variance for numerical variables. All reporte p values were two-sie, an a p value of <0.05 was use to etermine significance. Multivariate analysis was performe. Variables that ha a p value of <0.10 (age, boy mass inex, stage, an combine necrotic angle) were inclue in the multivariate moel. The Cox proportional-hazars frailty moel, in which the association between failure times is explicitly moele with a ranomeffect term (commonly calle the frailty ), was carrie out to ientify inepenent factors with regar to seconary collapse an osteophyte formation 25,26. We also performe a sensitivity analysis with a Cox proportional hazars moel using just one ranomly chosen hip for the eight patients who ha unergone bilateral osteotomy, to eliminate the issue of clustering. In aition, we trie to assess age, boy mass inex, stage, an combine necrotic angle retrospectively in an attempt to fin cutoff values. The Kaplan-Meier prouct-limit metho was performe to estimate survival 27. We performe three analyses: a best-case scenario with total hip arthroplasty as the en point; a milecase scenario with total hip arthroplasty or raiographic failure (seconary collapse of the femoral hea or osteoarthritic change) as the en point; an a worst-case scenario with total hip arthroplasty, raiographic failure, or loss to follow-up as the en point. Source of Funing There was no external funing for this stuy. Results Postoperative Scintigraphy All hips showe aequate perfusion of the proximal segment, inicating preservation of the meial femoral circumflex artery. Uptake of raionuclie was increase at the site of the osteotomy. Seconary Collapse an Conversion to Total Hip Arthroplasty Seconary collapse of the newly establishe weight-bearing area of the femoral heaevelope in twenty-seven hips (24%). The mean amount of time from the osteotomy to the seconary collapse was 20.6 months (range, three to thirty-six months). Twenty of the twenty-seven hips with seconary collapse ha osteoarthritic changes, an fourteen of the twenty osteoarthritic hips were converte to a total hip arthroplasty. The mean perio of time from the onset of the seconary collapse to the total hip arthroplasty was 9.1 months (range, three to seventeen months). There were no significant ifferences between the group with collapse an the group without collapse with regar to the sex (p = 0.189) or occupation (p = 0.827) of the patient or the cause of the osteonecrosis (p = 0.586). However, there were significant ifferences between the two groups with regar to the stage of the osteonecrosis (p = 0.023), age (p < 0.001), boy mass inex (p < 0.001), an combine necrotic angle (p < 0.001). There were no significant ifferences between the group that ha conversion to a total hip arthroplasty an the group that i not with regar to the sex (p = 0.470) or occupation (p = 0.094) of the patient or the cause of the osteonecrosis (p = 0.584). However, there were significant ifferences between the two groups with regar to age (p < 0.001), the stage of the osteonecrosis (p = 0.038), the boy mass inex (p < 0.001), an the combine necrotic angle (p < 0.001). Multivariate analysis with use of the Cox proportionalhazars frailty moel showe that the stage of the osteonecrosis (hazar ratio = 3.28; 95% confience interval = 1.49 to 7.24), age (hazar ratio = 1.08; 95% confience interval = 1.02 to 1.14), boy mass inex (hazar ratio = 1.19; 95% confience interval = 1.03 to 1.38), an extent of the necrotic area (hazar ratio = 1.08; 95% confience interval = 1.04 to 1.11) were factors etermining seconary collapse. In the sensitivity analysis of one hip of each patient, the hips of patients with an age of forty years or oler (hazar ratio = 2.77; 95% confience interval = 1.08 to 7.16), a boy mass inex of 24 kg/m 2 (hazar ratio = 3.20; 95% confience interval = 1.03 to 8.87), an osteonecrosis stage of III or greater (hazar ratio = 2.66; 95% confience interval = 1.22 to 5.79), an a combine necrotic angle of 230 (hazar ratio = 7.78; 95% confience interval = 2.86 to 17.78) were associate with an increase risk of seconary collapse. Osteophyte Formation an Osteoarthritic Change in the Group without Collapse Seconary collapse of the femoral heai not occur in eightysix hips (76%), an none of them ha unergone a total hip arthroplasty as of the time of the latest follow-up. At the final evaluation, osteophyte formation (range, 3 to 12 mm; mean, 5.3 mm) was observe in thirty-seven (43%) of the eighty-six hips without seconary collapse. The mean perio from the osteotomy to the osteophyte formation was 20.6 months (range, three to thirty-six months). Of these thirtyseven hips, seven showe raiographic evience of joint space narrowing, which became efinite thirty-six to sixty-three months (mean, 46.1 months) postoperatively. There were no significant ifferences between the group with osteophyte formation an the group without osteophyte formation with regar to the sex (p = 0.726), age (p = 0.795), occupation (p = 0.664), or boy mass inex (p = 0.894) of the

5 318 Fig. 3-A Fig. 3-B Figs. 3-A, 3-B, an 3-C A forty-five-year-ol man with osteonecrosis of the right femoral hea an a boy mass inex of 28.8 kg/m 2. Fig. 3-A Preoperative anteroposterior raiograph of the hip. Fig. 3-B Anteroposterior raiograph of the hip mae six weeks postoperatively. The newly establishe viable bone in the weight-bearing area below the ome of the acetabulum is thin an beak-shape (arrows). patient; the combine necrotic angle (p = 0.663); or the cause of the osteonecrosis (p = 0.631). However, the prevalence of osteophyte formation was higher in stage-iii hips (p < 0.001). Cox regression analysis showe that an osteonecrosis stage of III or more was an inepenent preictor of osteophyte formation (hazar ratio = 4.00; 95% confience interval = 2.05 to 7.82) (p < 0.001). Clinical Results A elayeeep infection evelope in one hip at thirteen months after the osteotomy. It was successfully treate with removal of the compression hip screw, ébriement, an aministration of antibiotics. Delaye union of the osteotomy site with excessive varus occurre in one hip. Nonunion of the greater trochanter osteotomy site occurre in four hips. All of these hips were asymptomatic an ha no aitional intervention. Seventeen hips ha limitations of abuction an external rotation, making it ifficult for the patients to put on their shoes an socks. The operatively treate limb was shortene by >1 cm (range, 1 to 2 cm; mean, 1.4 cm) in twenty-seven cases. Thirty-one patients (thirty-one hips) ha a limp, which was mil in twenty-six an moerate in five. However, none of these patients use a crutch or cane. Patients who ha osteophyte formation ha significantly less flexion (p < 0.001), abuction (p = 0.040), internal rotation (p < 0.001), an external rotation (p = 0.002) than those who i not. The mean hip score accoring to the system of Merle Aubigné et al. was 15.8 points (range, 6 to 18 points) at the most recent follow-up visit or the last follow-up evaluation before conversion to a total hip arthroplasty. Fifty-seven hips (50%) ha an excellent score; thirty-eight (34%), a goo score; four (4%), a fair score; an fourteen (12%), a poor score. The mean hip score was 16.8 points (range, 8 to 18 points) in the group without collapse an 12.6 points (range, 6 to 18 points) in the group that ha collapse (p < 0.001). The mean hip score was 16.5 points (range, 15 to 18 points) in the group with osteophyte formation an 17.2 points (range, 15 to 18 points) in the group without osteophyte formation (p = 0.132). The hip scores were not significantly affecte by the sex (p = 0.830) or occupation (p = 0.135) of the patient or by the cause (p = 0.307) or stage of the osteonecrosis (p = 0.053); however, they were significantly influence by age (p = 0.012), boy mass inex (p < 0.001), an the combine necrotic angle (p < 0.001). A higher likelihoo of an unsatisfactory result was associate with an age of forty years or oler (os ratio = 8.26; 95% confience interval = 1.99 to 34.33), a boy mass inex of

6 319 Fig. 3-C Follow-up raiograph, mae at twenty-four months after the operation, showing seconary collapse of the newly establishe weight-bearing area of the femoral hea. 24 kg/m 2 (os ratio = 5.79; 95% confience interval = 1.07 to 31.36), an a combine necrotic angle of 230 (os ratio = 9.97; 95% confience interval = 2.55 to 38.91). Survival Analysis The survival rate at 110 months was 87.2% (95% confience interval = 79.3% to 92.2%) with total hip arthroplasty as the en point (best-case scenario), 63.4% (95% confience interval = 51.1% to 75.7%) with total hip arthroplasty or raiographic failure as the en point (mile-case scenario), an 56.0% (95% confience interval = 44.6% to 67.4%) with total hip arthroplasty, raiographic failure, or loss to follow-up as the en point (worst-case scenario) (Fig. 4). Discussion Transtrochanteric rotational osteotomy is theoretically an ieal metho for preserving a hip joint affecte by osteonecrosis 8,9. The necrotic portion can be remove from the weight-bearing zone an replace with intact cartilage an bone with use of this osteotomy. However, the reporte results of the osteotomy have been inconsistent Stuies from Japan an Korea have emonstrate satisfactory results 8-12, but such favorable results have not been reprouce by surgeons in Western countries The striking ifference in the results has been thought to be relate to ifferences in the preoperative stage of the necrosis, the extent of the necrosis, the surgical technique, the metho of fixation, an/or the postoperative management. Seconary collapse after the osteotomy has been the main cause of failure. To ate, there is little information regaring the association of patient age or boy mass inex with the result of anterior transtrochanteric rotational osteotomy. Our stuy showe that boy mass inex an age as well as the extent an stage of the osteonecrosis are important factors etermining seconary collapse after this osteotomy. In our stuy, thirteen patients with seconary collapse an thirty-seven patients with osteophyte formation i not require a hip replacement uring the stuy perio although they experience intermittent pain an limping. Dean an Cabanela 13 reporte that, of eighteen hips treate with the osteotomy, fifteen showe further collapse of the femoral hea an only three ha a satisfactory result. They suggeste there might be race-epenent ifferences in boy weight, boy mass inex, an/or the anatomy of the hip capsule between Korean or Japanese patients an white patients that might have influence the outcome of the osteotomy an the nee for hip replacement. Although collapse an osteophyte formation frequently lea to a poor clinical result an the subsequent nee for a total hip arthroplasty, the inications for total hip arthroplasty vary consierably among institutions, among geographical areas, an among cultures. In some cultures, patients with a notable egree of clinical an raiographic progression of osteoarthritis are much less likely to be treate with total hip replacement because of ifferences in tolerance to pain anisability as well as socioeconomic factors. Thus, we analyze three scenarios, with the mile-case scenario (a survival rate of 63.4% with a total hip arthroplasty, seconary collapse of the femoral hea, or osteoarthritic change as the en point) perhaps being the best inication of the overall results of this proceure. We coul not explain why seconary collapse was more frequent in patients who were forty years of age or oler with a boy mass inex of 24 kg/m 2. Seconary collapse may represent a stress fracture of the newly establishe intact portion of the femoral hea in the weight-bearing region (Figs. 3-A, 3-B, an 3-C) 10. This intact portion was usually thin an beak-shape. In both men an women, age-relate osteopenia begins near the age of forty years an progresses linearly at a rate of 0.5% to 1% per year, accounting for nearly 40% of the total bone loss accrue by the age of seventy years 28. Such bone loss results in increase porosity, ecrease mineralization, an ultimately increase fracture risk 28. In overweight patients, an excessive loa can be inuce on the femoral hea, which leas to a stress fracture an seconary collapse of the newly forme weightbearing portion. In our stuy, thirty-seven (43%) of eighty-six hips without collapse ha osteophyte formation aroun the femoral heauring the stuy perio. This le to seconary osteoarthritis with a narrow joint space in seven (19%). Osteophyte

7 320 Fig. 4 Kaplan-Meier survival curves with 95% confience intervals showing the best-case scenario (conversion to total hip arthroplasty), the mile-case scenario (conversion to total hip arthroplasty or raiographic failure), an the worst-case scenario (conversion to total hip arthroplasty, raiographic failure, or loss to follow-up) after anterior transtrochanteric rotational osteotomy. formation has been known to cause impingement of the hip joint an subsequent osteoarthritis 10,29. Sugano et al. reporte postoperative osteophyte formation aroun the femoral hea in 39% of the stage-iii hips in their stuy at a mean of 6.3 years postoperatively 10. Hisatome et al. observe the same phenomenonin64%oftheircasesuringameanfollow-up perio of 6.4 years 29. Limitations of our stuy inclue a relatively short followup perio, the risk of selection bias, an possible effects of unmeasure confouners. Previous stuies have shown that most of the collapse of the new weight-bearing area evelops within three years postoperatively Although the three-year follow-up perio was sufficient to assess the seconary collapse, it was not sufficient to evaluate the effect of osteophyte formation on the outcome of the osteotomy. It shoul be note that we performe the osteotomy in patients who ha marke segmental collapse when they were younger than thirty years of age. We i not evaluate alcohol intake or steroi use after the osteotomy, although these factors might have influence the results of the osteotomy. The percentage of necrotic volume has been ientifie as an important factor in the collapse of an osteonecrotic femoral hea However, we i not measure the percentage of necrotic volume in our patients. Previous stuies have shown that some patients remain asymptomatic without any treatment an also that other jointpreserving proceures can be effective in the treatment of femoral hea osteonecrosis 21,33. Anterior transtrochanteric rotational osteotomy is a technically emaning operation that requires a prolonge perio of rehabilitation. Delaye union an malunion of the osteotomy site, nonunion of the greater trochanter, an shortening of the operatively treate limb may be complications following this osteotomy. To verify the effectiveness of the anterior transtrochanteric rotational osteotomy, it shoul be compare with other proceures in a ranomize clinical trial. The success rate of anterior transtrochanteric rotational osteotomy may be improve by more efficient selection of patients. The proceure shoul be performe in theearlystagesoftheisease,beforemarkecollapseof the hea occurs. The viable portion of the femoral hea shoul be of such a size that restoration of an aequate weight-bearing articular surface is possible 8,10. The use of magnetic resonance imaging instea of plain raiographs for the selection of patients improves the success rate of the osteotomy 12. This stuy showe that age, boy mass inex, an the stage an extent of the osteonecrosis are factors etermining seconary collapse after anterior transtrochanteric rotational osteotomy an shoul be consiere in the selection of patients for this proceure. n

8 321 Yong-Chan Ha, MD Department of Orthopaeic Surgery, Chung-Ang University College of Meicine, Heukseok-ong, Dongjak-gu, Seoul , South Korea Hee Joong Kim, MD Ki-Choul Kim, MD Young-Kyun Lee, MD Kyung-Hoi Koo, MD Department of Orthopaeic Surgery, Seoul National University College of Meicine, 28 Yeongeon-ong, Jongno-gu, Seoul , South Korea. aress for K.-H. Koo: Shin-Yoon Kim, MD Department of Orthopaeic Surgery, Kyungpook National University College of Meicine, 200 Donguk-ro, Jung-gu, Daegu , South Korea References 1. Cornell CN, Salvati EA, Pellicci PM. Long-term follow-up of total hip replacement in patients with osteonecrosis. Orthop Clin North Am. 1985;16: Saito S, Saito M, Nishina T, Ohzono K, Ono K. Long-term results of total hip arthroplasty for osteonecrosis of the femoral hea. A comparison with osteoarthritis. Clin Orthop Relat Res. 1989;244: Katz RL, Bourne RB, Rorabeck CH, McGee H. Total hip arthroplasty in patients with osteonecrosis of the hip. Follow-up observations on cementless an cemente operations. Clin Orthop Relat Res. 1992;281: Kim YH, Oh JH, Oh SH. Cementless total hip arthroplasty in patients with osteonecrosis of the femoral hea. Clin Orthop Relat Res. 1995;320: Daniel J, Pynsent PB, McMinn DJ. Metal-on-metal resurfacing of the hip in patients uner the age of 55 years with osteoarthritis. J Bone Joint Surg Br. 2004; 86: Kearns SR, Jamal B, Rorabeck CH, Bourne RB. Factors affecting survival of uncemente total hip arthroplasty in patients 50 years or younger. Clin Orthop Relat Res. 2006;453: Yoo JJ, Kim YM, Yoon KS, Koo KH, Kim JW, Nam KW, Kim HJ. Contemporary alumina-on-alumina total hip arthroplasty performe in patients younger than forty years: a 5-year minimum follow-up stuy. J Biome Mater Res B Appl Biomater. 2006;78: Sugioka Y. Transtrochanteric anterior rotational osteotomy of the femoral hea in the treatment of osteonecrosis affecting the hip: a new osteotomy operation. Clin Orthop Relat Res. 1978;130: Sugioka Y, Hotokebuchi T, Tsutsui H. Transtrochanteric anterior rotational osteotomy for iiopathic an steroi-inuce necrosis of the femoral hea. Inications an long-term results. Clin Orthop Relat Res. 1992;277: Sugano N, Takaoka K, Ohzono K, Matsui M, Saito M, Saito S. Rotational osteotomy for non-traumatic osteonecrosis of the femoral hea. J Bone Joint Surg Br. 1992;74: Inao S, Ano M, Gotoh E, Matsuno T. Minimum 10-year results of Sugioka s osteotomy for femoral hea osteonecrosis. Clin Orthop Relat Res. 1999;368: Koo KH, Song HR, Yang JW, Yang P, Kim JR, Kim YM. Trochanteric rotational osteotomy for osteonecrosis of the femoral hea. J Bone Joint Surg Br. 2001;83: Dean MT, Cabanela ME. Transtrochanteric anterior rotational osteotomy for osteonecrosis of the femoral hea. Long-term results. J Bone Joint Surg Br. 1993;75: Langlais F, Fourastier J. Rotation osteotomies for osteonecrosis of the femoral hea. Clin Orthop Relat Res. 1997;343: Tooke SM, Amstutz HC, Heley AK. Results of transtrochanteric rotational osteotomy for femoral hea osteonecrosis. Clin Orthop Relat Res. 1987;224: Totty WG, Murphy WA, Ganz WI, Kumar B, Daum WJ, Siegel BA. Magnetic resonance imaging of the normal an ischemic femoral hea. AJR Am J Roentgenol. 1984;143: Ficat RP. Iiopathic bone necrosis of the femoral hea. Early iagnosis an treatment. J Bone Joint Surg Br. 1985;67: Smith SW, Meyer RA, Connor PM, Smith SE, Hanley EN Jr. Interobserver reliability an intraobserver reproucibility of the moifie Ficat classification system of osteonecrosis of the femoral hea. J Bone Joint Surg Am. 1996;78: Flugsru GB, Norsletten L, Espehaug B, Havelin LI, Meyer HE. Risk factors for total hip replacement ue to primary osteoarthritis: a cohort stuy in 50,034 persons. Arthritis Rheum. 2002;46: Kerboul M, Thomine J, Postel M, Merle Aubigné R. The conservative surgical treatment of iiopathic aseptic necrosis of the femoral hea. J Bone Joint Surg Br. 1974;56: Ha YC, Jung WH, Kim JR, Seong NH, Kim SY, Koo KH. Preiction of collapse in femoral hea osteonecrosis: a moifie Kerboul metho with use of magnetic resonance images. J Bone Joint Surg Am. 2006;88 Suppl. 3: Merle D Aubigné R, Postel M, Mazabrau A, Massias P, Gueguen J, France P. Iiopathic necrosis of the femoral hea in aults. J Bone Joint Surg Br. 1965;47: Swezey RL, Bjarnason DM. Inhibition of seconary osteophyte formation in rheumatoi arthritis. Rheumatol Rehabil. 1976;15: Jacobsen S, Sonne-Holm S, Søballe K, Gebuhr P, Lun B. The relationship of hip joint space to self reporte hip pain. A survey of subjects of the Copenhagen City Heart Stuy: the Osteoarthritis Substuy. Osteoarthritis Cartilage. 2004;12: Hougaar P. Frailty moels for survival ata. Lifetime Data Anal. 1995;1: Hougaar P. Analysis of multivariate survival ata. Springer; New York: Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53: Parfitt AM. Age-relate structural changes in trabecular an cortical bone: cellular mechanisms an biomechanical consequences. Calcif Tissue Int. 1984;36 Suppl. 1:S Hisatome T, Yasunaga Y, Takahashi K, Ochi M. Progressive collapse of transpose necrotic area after transtrochanteric rotational osteotomy for osteonecrosis of the femoral hea inuces osteoarthritic change. Mi-term results of transtrochanteric rotational osteotomy for osteonecrosis of the femoral hea. Arch Orthop Trauma Surg. 2004;124: Holman AJ, Garner GC, Richarson ML, Simkin PA. Quantitative magnetic resonance imaging preicts clinical outcome of core ecompression for osteonecrosis of the femoral hea. J Rheumatol. 1995;22: Nishii T, Sugano N, Ohzono K, Sakai T, Sato Y, Yoshikawa H. Significance of lesion size an location in the preiction of collapse of osteonecrosis of the femoral hea: a new three-imensional quantification using magnetic resonance imaging. J Orthop Res. 2002;20: Steinberg ME, Bans RE, Parry S, Hoffman E, Chan T, Hartman KM. Does lesion size affect the outcome in osteonecrosis? Clin Orthop Relat Res. 1999;367: Nam KW, Kim YL, Yoo JJ, Koo KH, Yoon KS, Kim HJ. Fate of untreate asymptomatic osteonecrosis of the femoral hea. J Bone Joint Surg Am. 2008; 90: