Risk factors for lateral compartmental cartilage wear in varus knee osteoarthritis / 中华外科杂志

<p><b>OBJECTIVE</b>To analyze the gross features of articular cartilage wear in varus knee osteoarthritis, and discuss the risk factors for lateral compartmental cartilage erosion.</p><p><b>METHODS</b>Data prospectively collected from the dissection of 286 total knee arthroplasties (223 patients) with varus knee osteoarthritis from January 2013 to December 2013 were analyzed. At the operation, the gross assessments of articular cartilage, ligament and meniscus were recorded, and then the slices were evaluated for histologic analysis. Parameters of the patients with lateral compartmental cartilage erosion were compared with those without lateral compartmental cartilage erosion using the univariate analysis. Logistic regression analysis was used to analyze the risk factors associated with lateral compartmental cartilage erosion.</p><p><b>RESULTS</b>There were 223 patients with 286 knees were included,including 37 male patients (47 knees) and 189 female patients (239 knees), with an average age of (66±8) years (range 50-86 years), body mass index (BMI) was (27±5) kg/m2 (18.0-40.0 kg/m2). Varus degree was 8°±4° (1°-34°). Range of motion was 103°±21° (0°-143°), and Hospital for Special Surgery (HSS) score was 53±12 (29-76). Seventy-five knees (60 patients) showed lateral compartmental cartilage wear (26.2%). Environmental factors showed no differences in age, side, gender, BMI, range of motion,and HSS score (P>0.05). Factors significantly increasing the risk of lateral compartmental cartilage wear by univariate analysis included varus degree, activity level, duration of onset, meniscus, Weidow grade, Kellgren-Lawrence grade, collateral ligament and anterior cruciate ligament (P<0.05). Multiple Logistic regression analysis revealed the factors most highly associated with the increase risk for lateral compartmental wear were high activity level (OR=2.843, 95% CI: 1.010-8.002) and longer duration of onset (OR=1.216, 95% CI: 1.115-1.325). However, intact lateral meniscus (OR=0.012, 95% CI: 0.003-0.048) and anterior cruciate ligament (OR=0.406, 95% CI: 0.192-0.857) were associated with the protection of lateral compartmental.</p><p><b>CONCLUSIONS</b>In varus knee osteoarthritis, the wear incidence of lateral compartmental is low. High activity and increased duration of onset are risk factors of lateral compartmental wear, and intact meniscus and anterior cruciate ligament are protective factors.</p>

Treatment of patellofemoral osteoarthritis with patello-femoral joint arthroplasty / 中华外科杂志

<p><b>OBJECTIVE</b>To study the outcome and surgical technique of patello-femoral joint arthroplasty (PFJ) for osteoarthritis of the knee.</p><p><b>METHODS</b>From January 2010 to June 2012, 17 patients (24 knees) with patello-femoral compartmental osteoarthritis treated by PFJ were reviewed retrospectively. There were 2 male patients (4 knees) and 15 female patients (15 knees), with an average age of (64 ± 9) years (52-77 years) . The mean body mass index was (24 ± 3) kg/m² (18.0-30.1 kg/m²) . Patients were asked to return for follow-up examinations at 3, 6 months and at every year after PFJ. The range of motion (ROM), visual analogue scale(VAS), Hospital for Special Surgery score (HSS score) , Feller patella score, quadriceps muscle strength were evaluated before and after PFJ. The paired sample t-test and one-way analysis of variance (ANOVA) with replicate measures were used to determine whether there were statistically significant differences between the mean data.</p><p><b>RESULTS</b>All of the patients were followed up for 18-47 months, with a mean time of (29 ± 10) months, 75.0% patients were satisfied with the outcome of this surgical procedure 1 year postoperative, and 87.5% were satisfied 2 years postoperative. HSS score was increased from 61 ± 11 to 90 ± 6 at the final follow-up (t = 12.24, P = 0.000). VAS score was reduced from 6.7 ± 1.0 to 2.4 ± 1.0 (t = 15.84, P = 0.000). The mean post-operative ROM of the knees was 126° ± 7° (t = 3.25, P = 0.003). Feller patella scores were 18.0 ± 3.2 before operation, and 18.5 ± 4.5, 19.7 ± 3.4, 24.0 ± 3.8, 26.0 ± 3.3, 26.6 ± 2.5 at 3, 6, 12, 24 months after operation, final follow-up, respectively. Quadriceps muscle strength were 3.9 ± 0.5 before operation, and 3.7 ± 0.5, 3.9 ± 0.5, 4.2 ± 0.5, 4.3 ± 0.5, 4.3 ± 0.5 at 3, 6, 12, 24 months after operation, final follow-up, respectively. Repeated measures ANOVA found significant time effects for Feller patella scores (F = 38.97, P = 0.000) and quadriceps muscle (F = 6.89, P = 0.000). Feller patella scores and quadriceps muscle strength were low at 3, 6 months after operation, with no significant differences compared with pre-operation data (P > 0.05). The improvements of Feller patella scores and quadriceps muscle strength after 6 months were of significant difference compared with pre-operation and postoperative 6 months data (t = 5.65-10.65 and t = 2.18-2.73, P < 0.05) . Three knee reported continuing pains with quadriceps muscle strength less than 4 level.</p><p><b>CONCLUSIONS</b>PFJ is an effective method for patello-femoral compartmental osteoarthritis with less trauma. The early term outcome of PFJ is encouraging. Quadriceps muscle strength should be enhanced.</p>

Ability of lower teardrop edge to restore anatomical hip center height in total hip arthroplasty / 中华医学杂志(英文版)

<p><b>BACKGROUND</b>The acetabular teardrop is often used to guide acetabular component placement in total hip arthroplasty (THA). Placing the lower acetabular component aspect at the same level as the lower teardrop edge was assumed to restore the hip center of rotation. Here we radiographically analyzed the relationship between cup center and normal contralateral acetabulum center height on unilateral THA using this placement method.</p><p><b>METHODS</b>A total of 106 unilateral THA cases with normal contralateral acetabula were reviewed and the vertical and horizontal distances in relation to the lower acetabular teardrop edge from both hip joint centers, cup inclination, and anteversion were measured radiographically. The paired t-test was used to compare left and right hip center heights. Scatter plots and Pearson's correlation coefficients were used to evaluate differences in hip center heights, cup anteversion, inclination angles, and medialized cup center distance compared to the contralateral hip joint.</p><p><b>RESULTS</b>Cup center height was significantly greater (P < 0.01) than contralateral hip joint center height (93.4% in the 0-5 mm range, 6.6% >5 mm). There was a weak correlation between hip center height difference and inclination (r = 0.376, P < 0.01) and between difference and anteversion (r = 0.310, P < 0.01) but no correlation between difference and outer cup diameter (r = 0.184, P = 0.058) or difference and medialized cup center distance (r = -0.098, P = 0.318).</p><p><b>CONCLUSIONS</b>Although this method did not exactly replicate anatomic hip center height, the clinical significance of cup center height and anatomic hip center height differences is negligible. This acetabular component placement method has high simplicity, reliability, and stability.</p>

Fourth-generation ceramic-on-ceramic total hip arthroplasty in patients of 55 years or younger: short-term results and complications analysis / 中华医学杂志(英文版)

<p><b>BACKGROUND</b>The incidence of total hip replacement in the younger and more active patients is ever increasing. The ceramic-on-ceramic (COC) bearing was developed to reduce wear debris-induced osteolysis and loosening and to improve the longevity of hip arthroplasties. Few studies have reported the clinical results and complications of the new zirconia-toughened ceramic total hip arthroplasty (THA).</p><p><b>METHODS</b>A consecutive series of 132 young patients (177 hips) that underwent primary cementless THAs between January 2010 and December 2012 were included in this study. These arthroplasties all had fourth-generation COC bearings performed through a posterolateral approach. The average age was (41.8 ± 8.3) years (ranging from 22 to 55 years), and the mean follow-up period was (24.5 ± 9.4) months (ranging from 12 to 47 months). The results were evaluated both clinically and radiographically. Harris hip score (HHS) was determined before surgery and at the time of each follow-up. Presence of postoperative groin or thigh pain and squeaking were recorded. Other complications such as dislocations, periprosthetic fractures, and ceramic components fractures were diagnosed and treated in emergency.</p><p><b>RESULTS</b>The average HHSs improved from preoperative 60.3 ± 10.7 (ranging from 29 to 76) to 91.0 ± 5.1 (ranging from 74 to 100) at the final follow-up (t = 45.064, P < 0.05), and 97.7% of cases were scored as excellent and good results. At the last follow-up, incidental inguinal pain was found in three hips (1.7%) and thigh pain in 11 hips (6.2%). Radiographs showed a high rate of new bone formation around the acetabular and stem components. No obvious osteolysis or prosthesis loosening was detected. Complications occurred in six hips (3.4%): posterior dislocation in two hips (1.1%), periprosthetic femoral fracture in one hip (0.6%), asymptomatic squeaking in two hips (1.1%), and ceramic liner fracture in one hip (0.6%).</p><p><b>CONCLUSIONS</b>The fourth-generation COC THA showed excellent clinical results in younger active patients with no osteolysis-related prosthesis failure at a short-term follow-up study. Surgeons should still be aware of the potential risks of complications such as dislocation, periprosthetic fracture, squeaking, and ceramic components fracture.</p>

Porous bioceramic beta-tricalcium phosphate for treatment of osteonecrosis of the femoral head / 中国组织工程研究

BACKGROUND:It is a clinical difficult in the treatment of osteonecrosis with joint preservation, and to solve this problem, a variety of bone graft substitutes are at the exploration stage. OBJECTIVE:To evaluate the clinical outcome of lightbulb operation with porous bioceramic β-tricalcium phosphate in a consecutive series of patients with osteonecrosis of the femoral head. METHODS:From January to December 2008, 58 patients (88 hips) who had undergone lightbulb operation with porous bioceramic β-tricalcium phosphate were involved in this study. Al patients were evaluated both clinicaly and radiographicaly at postoperative 3, 6, 12 months and annualy. Functional improvement was assessed with the Harris hip score. RESULTS AND CONCLUSION: Among these patients, 56 patients (85 hips) were folowed up for 2-5 years. According to the ARCO staging system, there were 27 hips of stage II, 40 hips of stage IIIa, 18 hips of IIIb. According to the hospital’s classification, type C was in 4 hips, L1 in 15 hips, L2 in 28 hips, and L3 in 38 hips. According to the Harris hip score system, excelent outcome was in 55 hips, good in 12 hips, fair in 5 hips and poor in 13 hips. Nine of 11 patients who failed to preserve their own joints were subjected to hip replacement. The  mean preoperative and postoperative Harris scores were 61.2 and 85.3, respectively, with a mean improvement of 24.1 points (P < 0.001). All hips were radiologically stable, with no progress of osteonecrosis, and bone density in the bone graft area increased obviously. The replacement time of porous bioceramic β-tricalcium phosphate was 1-1.5 years. These findings suggest that the porous bioceramic β-tricalcium phosphate provides an option to treat osteonecrosis of the femoral head with satisfactory clinical outcomes, and profits the repair and reconstruction of femoral head osteonecrosis. When in the lateral column of femoral head, the porous bioceramic β-tricalcium phosphate can play a supporting role in the lateral column of the necrotic area, and further prevent collapse, which is suitable for patients with osteonecrosis of the femoral head, especially for those with joint preservation.

The classification of osteonecrosis of the femoral head based on the three pillars structure: China Japan Friendship Hospital (CJFH) classification / 中华骨科杂志

Objective To explore the regular progressive pattern of nontraumatic osteonecrosis of the femoral head (ONFH) in order to establish the reliable and convenient new classification of ONFH.Methods The coronal section of the femoral head was divided into three pillars (medial,central and lateral).The mid-coronal section of the femoral head on MRI was selected.The China-Japan Friendship Hospital (CJFH)classification of ONFH was established according to the site of necrotic focus in three pillars.A total of 153hips with ONFH were classified according to CJFH classification and Japanese Investigation Committee (JIC)classification,respectively.The collapse rate was observed and compared between both classifications of ONFH.Results The CJFH classification for ONFH consists of 3 types:type A,the medial pillar was involved; type B,the medial and central pillars were involved; type C,the lateral pillar was involved.According to site of necrosis focus in the lateral pillar,the type C was divided into 3 types:C1,there pillars were involved but there still was some normal tissue in lateral pillar;,C2,partial central pillar and all lateral pillar were involved; C3,the whole femoral head was involved.The natural history of the ONFH showed the collapse rate of type C2 and C3 in CJFH classification (95.3％) was higher than that (72.3％) of type C2 in JIC classification.Conclusion The CJFH classification of ONFH based on three pillars is more sensitive than JIC classification in predicting collapse of the femoral head.Moreover,the CJFH classification is convenient to use.

Application of CT and MRI in volumetric measurement of necrotic lesion in patient with avascular necrosis of the femoral head / 中华放射学杂志

Objective To investigate the feasibility and accuracy of volumetric measurement of necrotic lesion using CT and MRI,and to assess the value of necrotic lesion volume in predicting collapse of the femoral head in patients with avascular necrosis of the femoral head(ANFH). Methods Comparison among CT,MRI and gross section was performed in 25 femoral heads of 18 patients who underwent total hip replacement for established ANFH.The volume of necrotic lesion was measured using fluid displacement.CT and MRI data were transferred to a computer to calculate the volume of necrotic lesion using software.One way ANOVA was used to compare the volumes of necrotic lesion measured by CT,MRI and gross section.A total of 62 patients (92 hips) who were diagnosed with ANFH but without collapse by CT were followed up 24 months.Student t-test was used to compare the ratio of the volumes of the necrotic lesion and entire femoral head in the hips with and without collapse and ROC curve analysis was carried out.Results CT and MRI coincided with gross section in the necrotic area,proliferative area and extralesional area.The volumes of the necrotic lesion measured by CT,MRI and gross section were ( 20.5 ± 5.2 ),( 21.4 ± 4.8 ),( 20.9 ± 5.2 ) cm3,respectively.There was no significant difference among the necrotic volumes measured by the three methods(F =0.185,P =0.831 ).In fifty-seven out of 92 hips,collapse of the femoral head occurred during the follow-up.The ratio of the volumes of the necrotic lesion and entire femoral head was higher in hips with collapse than in hips without collapse[ (34.5 ±9.3)％ vs.(23.4 ±8.4)％ ;t =5.749,P=0.000].The area under the ROC curve was 0.808. Conclusions The volume of the necrotic lesion plays an important role in the collapse of femoral head in patients with ANFH.Both CT and MRI can identify the shape and location of the necrotic lesion intuitively and stereospecifically and can determine the volume of the necrotic lesion accurately.

Early and middle-term results after surgical treatment for slipped capital femoral epiphysis / 中华骨科杂志

Objective To evaluate the early and middle-term results of in-situ single-screw fixation and subtrochanteric osteotomy of femur with external fixator for slipped capital femoral epiphysis(SCFE).Methods From June 1998 to July 2008, eleven patients (12 hips) with SCFE were treated, including 7males (8 hips) and 4 females (4 hips) with an average age of 14.3 years (range, 9-18). The average BMI was 31.1 g/m2. According to the Southwick measurement, mild (＜30°) in 6 cases (7 hips), moderate (30°-50°) in 2 cases (2 hips), and severe (＞50°) in 3 cases (3 hips). All the mild and moderate cases were undergone insitu single screw fixation, and three severe cases were performed subtrochanteric osteotomy of femur with external fixator. All the cases were evaluated at 3 months, 6 months and every year postoperatively. Results All the cases were followed up for average 5.6 years (range, 2-12). The Harris hip score increased from mean 74.8 points preoperatively to 90.6 points postoperatively, with 10 hips excellent, 1 hip good, 1 hip fair and 0 hip poor. The excellent and good rate was 91.7% (11/12). The X-ray showed the epiphyseal closure in in-situ single screw fixation cases, and no further increase in epiphyseal-shaft angle. Partial reduction of slipped epiphysis with normal joint spaces was observed in subtrochanteric osteotomy of femur with external fixator cases. Conclusion In-situ single screw fixation provides a surgical treatment option to treat mild and moderate SCFE with satisfactory clinical outcome, and profit to the prevention further slip. Subtrochanteric osteotomy of femur with external fixator is suitable for severe SCFE with late deformity. The realignment procedure can correct deformity and postpone or avoid the occurrence of osteoarthritis.

Application of phase-contrast imaging with synchrotron radiation hard X-ray in repair of traumatic femoral defect / 中华创伤杂志

Objective To compare the effectiveness of nano-hydroxyapatite/collagen (nHAC)and autologous mesenchymal stem cells (AMSCs) for the repair of femoral defect in a rabbit model with femoral defect under the monitoring of the synchrotron radiation hard X-ray. Methods The rabbit models of traumatic bone defect were established and completely randomized into three groups. The femoral defects filled with nothing were used as control group (Group A) , the femoral defects filled with nHAC as Group B and the femoral defects filled with nHAC + AMSCs as Group C. Phase-contrast imaging with synchrotron radiation hard X-ray was applied to detect the degradation and repair process of each group at postoperative weeks 4, 8 and 12, respectively. Results Phase-contrast imaging with synchrotron radiation hard X-ray could display the reparative process. Four weeks after operation, there was collapse in some defect areas in Group A, and the degradation of nHAC and new bone formation were observed in Groups B and C. Eight weeks after operation, fibrous tissues were observed in the defect area in Group A, while osteogenesis and nHAC degradation were more obvious in Groups B and C. Twelve weeks after operation, the defect areas were still unhealed and were substituted by fibrous tissues in Group A, tissue densities of defect areas in Group C were identical with periphery areas, and trabecular bones were formed in Group C. There were statistical differences in the osteogenesis between Group A and Groups B and C,with Group C the best. Conclusion Phase-contrast imaging with synchrotron X-ray can detect the reparative process at a micro-level and plays an important role in the development of tissue engineering.

Relationship between volume, angle and extent index in non-traumatic avascular osteonecrosis on the femoral head / 中华全科医师杂志

Objective To explore the relationship between the osteonecrotic volume (lesion size), angle and the index of necrotic extent on the femoral head. Methods Fifty-one hips in 39 patients with non-traumatic avascular osteonecrosis on the femoral head were divided into 12 equal segments from the head to the neck junction (a turning point of the spherical curve of the head) with whole hip displacement, each with 30 degrees on a coronal plane of weight-bearing surface. The osteonecrotic angle of the arc of each necrotic segment from the center of the femoral head was measured at the point of one- to 12-o'clock on imaging of two dimensional reconstruction of computerized tomography (CT) scans and anteroposterior and lateral radiographs. Necrotic volume on each segment of the femoral head was calculated with fluid displacement method. The index of osteonecrotic extent on the femoral head was calculated using direct anatomical measurements. Results Osteoneerotic volume on the femoral head at the point of 12- to one-o'clock on coronal plane was (74. 5 ± 7.4)% of the sphere equivalent of the whole femoral head, which was positively correlated to its necrotic angle [ (41.9±8. 3) degrees] at the point of one-o'clock on the plane, with a coefficient of correlation of 0. 843, and that at the point of one- to two-o'clock on the plane was (73.7 ±0. 4) %, which was positively correlated to its necrotic angle [ (41.9 ± 1.8) degrees] at the point of two-o'clock, with a coefficient of correlation of 0. 543. Osteonecrotic volume on the point of 11- to 12-o'cleck was (83.6±8.6)%, and the necrotic angle at the point of 12-o'clock was (44. 9±3.9) degrees, which were positively correlated each other, with a coefficient of correlation of 0. 701 (P <0. 01 ). Osteonecrotic volume on the femoral head was positively correlated to its necrotic angle , modified index of necrotic extent, index of necrotic extent and Kerboul conjugated necrotic angle, with coefficients of correlation of 0. 798, 0. 701, 0. 377 and 0. 398 ( P < 0. 01 ), respectively, at the point of one o' clock. Conclusions Measurements of osteonecrotic volume was positively correlated to the index of necrotic extent and necrotic angle on the femoral head, respectively. Necrotic angle on the plane at 30-degree of the trochanter on the femoral head can well reflect its necrotic volume.

Post-SARS osteonecrosis and anticardiolipin antibodies / 中国矫形外科杂志

[Objective]To explore the correlation between anticardiolipin antibodies and post-SARS patients with osteonecrosis,investigate the etiology of post-SARS osteonecrosis,and select the sensitive molecular symbols for early diagnosis and distinguish the high risk population.[Method]The studied subjects were divided into two groups.Sixty-two post-SARS patients with osteonecrosis was one group,and 52 matched healthy people was as normal controlled group.Empty stomach blood samples from cubital veins were collected from both groups.IgA,IgG and IgM of anticardiolipin antibodies types were examined by ELISA.[Result]Twenty-one of 62 post-SARS osteonecrosis patients were tested at least one type of anticardiolipin antibodies positive,among them,with IgA(11.33?11.209)APL,IgG(5.127?5.927)GPL,IGM(17.821?10.606)MPL.Four of 52 control group persons were detected positive,with IGA(10.702?3.126)APL,IgG(5.184?4.780)GPL,IgM(14.684?5.516)MPL.There were significant differences between two groups,while,there were no difference in prothrombin time(PT),thrombin time(TT),prothrombin activity(PTA)and international normalized ratio(INR).[Conclusion]The study indicates an increased positive incidence of anticardiolipin antibodies in post-SARS patients with osteonecrosis.This has reflected that anticardiolipin antibodies may play a role in osteonecrosis.

Histological observation on repair of bone mesenchymal stem cell in rabbit osteonecrosis of the femoral head defect model / 中国康复理论与实践

@# ObjectiveTo evaluate the effectiveness of mesenchymal stem cell (MSC) in osteonecrosis of the femoral head defect(ONFH) repair. MethodsAniaml model of ONFH defect were established with rabbits, which were divided into 3 groups,group A did not infill anything as control, nano Hydroxyapatite/collagen(nHAC) as group B,nHAC+MSC as group C.Histology change were investigated 4,8,12 weeks after operation respectively.Results groups B and C were different with group A. The difference between groups A and C was more significant. ConclusionMesenchymal stem cell has a strong activity of osteoconduction,it has a value in repairing the bone defect of ONFH and the treatment of ONFH.

Preliminary experiences in mini-incision and minimally invasive surgery for total hip arthroplasty / 中华骨科杂志

Objective To explore the indication, to describe the key of operative technique in detail and to summarize the early results in mini-incision and minimally invasive surgery (MIS) total hip arthroplasty (THA) for hip disorders. Methods From Mar. 2003 to Mar. 2005, 43 patients (49 hips) were performed with MIS THA. The mean age of patients was 53.2 years ranging from 22 to 79 years. There were 22 males and 21 females. The average body mass index (BMI) was 23.4 ranging from 17.1 to 30.2. Using the modified postero-lateral approach, MIS THA was performed with cementless prostheses. For comparison, 35 patients (43 hips) were performed with standard THA at the same period. The age and BMI of patients in both groups were similar, but the pattern of disorders was different. The comparative items included preoperative bleeding, operative time, incisional length, postoperative functional recovery, and prosthetic position. Results 31 patients (MIS THA) and 25 patients (standard THA) were followed from 6 to 24 months (mean 13.1 months). No complications occurred in MIS THA. Dislocation appeared in one patient with standard THA at the two days after operation. The length of incision was 9.3 cm (range, 8.7 to 10.5 cm) in MIS group and 16.8 cm (range, 14 to 20 cm) in standard THA group. The difference was significant statistically (P

Prediction of collapse and selection of preserving femoral head for the osteonec rosis of the femoral head / 中华骨科杂志

Objective To explore the risk factors of collapse of osteonecro si s of the femoral head (ONFH) with a retrospective study, and put forward a treat ment protocol according to this study results. Methods From October 1993 to Apri l 2000, 40 cases (56 hips) were followed-up. The term of follow-up ranged from 1 2 to 68 months (average 29.2 months). The age of patients was from 16 to 60 year s old (average 36.8 years old). Preoperatively, the AP and frog view X-ray film, coronal and sagittal MRI were taken. The staging of ONFH proposed by ARCO was a dopted. Clinically the Harris score was used to assess the clinical outcome. The patients were divided into three groups: 1) Non-operative: 10 cases (14 hips), stage Ⅰin 7 hips(ⅠA 4,ⅠC 3),stageⅡin 7 hips (ⅡA 1,ⅡB 3,ⅡC 3); 2) Core dec ompression with vascularized iliac bone graft (VICBG): 18 cases (24 hips ), stag eⅠin 6 hips (ⅠB 4,ⅠC 2), stageⅡin 16 hips (ⅡA 4,ⅡB 8,ⅡC 4), stageⅢ in 2 hips (ⅢA); 3) Transtrochanteric rotational osteotomy(TRO): 12 cases (18 hips), stageⅡin 1 hip (ⅡC), stageⅢ in 17 hips (ⅢA 8, ⅢB 7,ⅢC 2). Harris score was 46 to 82 (mean 62.9). Results Collapse of the femoral head occurred in 9 hips a mong the 14 hips in the nonoperative group, in 10 hips among the 24 hips in the VICBG group. In TRO group, further deterioration was prevented in all but one hi p. Whether theres surgical intervention or not, no collapse occurred in both s tageⅠA and ⅡA, collapse occurred in both stage ⅠB and ⅡB was 7 hips among 15 hips (46.7%), 13 hips progressed into collapse of the femoral head with ⅠC and ⅡC (100%). The Harris score was 47 to 93(mean 77.4) postoperatively. Conclusio n With MR imaging, it is possible to predict the possibility of collapse of ONFH . The lower risk is below 30 percent of necrotic area, the higher risk is 30 to 60 percent of necrotic area, and the higherest risk is beyond 60 percent of necr otic area. The best results can be obtained if selection of treatment individual ly according to ARCO stage, patient age and necrotic extent.