摘要目的:比较带血管髂骨瓣移植术（pedicled vascularised iliac bone graft transfer，PVIBGT）与外科脱位打压植骨术（surgical hip dislocation combined with impacting bone grafts，SHD-IBG）治疗股骨头坏死国际骨循环研究协会（Association Research Circulation Osseous，ARCO）ⅢA期的临床效果及骨密度特点。方法:回顾性分析2015年1月至2022年7月于南京中医药大学附属医院骨科接受PVIBGT或SHD-IBG治疗ARCOⅢ期股骨头坏死55例患者资料。于术后3、6、12、18、24个月髋关节CT片测量植骨区和股骨距区域的HU值，计算植骨区骨密度与股骨距标准化骨密度比值（standardization bone mineral density ratio，sBMD Ratio）及骨密度变异系数（coefficient of variation，CV），通过双因素重复测量方差分析比较PVIBGT组和SHD-IBG组骨密度变化及均匀性，结合随访评估治疗效果及股骨头塌陷率。结果:测量的HU值具有良好的观察者内（ ICC=0.870）及观察者间一致性（ ICC=0.812）。术后2年时PVIBGT组股骨头塌陷率（25%，9/36）和转全髋关节置换术（total hip arthroplasty，THA）率（17%，6/36）与SHD-IBG组[32%（6/19）、21%（4/19）]的差异无统计学意义（ P>0.05）。术后2年时两组Harris评分均增加（时间效应 F=162.660， P<0.001），组别与时间交互作用显著（ F=3.574， P=0.031）；PVIBGT组由术前66.55±7.18增至术后2年83.08±6.42，SHD-IBG组由68.58±8.03增至80.68±5.91，PVIBGT组改善幅度更大。两组植骨区骨密度差异有统计学意义（组间效应 F=26.428， P<0.001；交互效应 F=9.550， P<0.001），sBMD Ratio变化趋势一致（组间效应 F=17.131， P<0.001；交互效应 F=3.808， P=0.024）。PVIBGT组植骨区CV低于SHD-IBG组（组间效应 F=5.065， P=0.030；术后6个月组间比较 P=0.042），即骨密度分布更均匀；SHD-IBG组植骨区最大长径（3.30±0.30）cm和最大宽径（1.82±0.22）cm均大于PVIBGT组（ t=4.328， P<0.001； t=3.956， P<0.001）。术后2年时PVIBGT组未塌陷者HU值为528.59±123.02，高于塌陷者的420.67±168.70（ t=2.108， P=0.046）；两组未塌陷者CV均低于塌陷者（PVIBGT组为35.48%±7.76%和50.66%±16.07%， t=3.672， P<0.001；SHD-IBG组为35.41%±8.68%和61.75%±12.93%， t=4.215， P<0.001）。 结论:PVIBGT与SHD-IBG治疗股骨头坏死ⅢA期均可获得良好临床效果。PVIBGT植骨区域呈现出更为均匀和稳定的高密度特性，但植骨面积相对较小；而SHD-IBG具有更大的植骨覆盖面积和较高的骨密度。

abstractsObjective:To compare the clinical outcomes and bone density characteristics of pedicled vascularized iliac bone graft transfer (PVIBGT) and surgical hip dislocation combined with impaction bone grafting (SHD-IBG) in the treatment of stage III primary osteonecrosis of the femoral head.Methods:A retrospective analysis was conducted on 55 patients with Association Research Circulation Osseous (ARCO) stage III osteonecrosis of the femoral head who underwent either pedicled vascularized iliac bone graft transplantation (PVIBGT) or surgical hip dislocation with impaction bone grafting (SHD-IBG) at the Department of Orthopaedics, Affiliated Hospital of Nanjing University of Chinese Medicine from January 2015 to July 2022. Postoperative hip CT scans were performed at 3, 6, 12, 18, and 24 months. Hounsfield unit (HU) values were measured in both the graft area and the calcar region. The standardized bone mineral density ratio (sBMD Ratio=HU graft/HU calcar) and the coefficient of variation (CV) of bone density were calculated. Two-way repeated measures ANOVA was used to compare changes in bone density and homogeneity between the two groups. Clinical outcomes and femoral head collapse rates were evaluated during follow-up.Results:HU measurements showed excellent intraobserver ( ICC=0.870) and interobserver ( ICC=0.812) reliability. At 2 years postoperatively, the PVIBGT group showed lower rates of femoral head collapse (25.0%, 9/36) and conversion to total hip arthroplasty (THA) (16.7%, 6/36) compared to the SHD-IBG group (31.6%, 6/19; 21.1%, 4/19) without significant difference ( P>0.05). Harris Hip Scores improved significantly in both groups (time effect: F=162.660, P<0.001), with a significant group×time interaction ( F=3.574, P=0.031), indicating a greater improvement in the PVIBGT group (preoperative 66.55±7.18 to 2-year 83.08±6.42) than in the SHD-IBG group (68.58±8.03 to 80.68±5.91). There were significant between-group differences in the changes of bone density in the graft area (group effect: F=26.428, P<0.001; interaction: F=9.550, P<0.001), and similar trends were observed for sBMD Ratio (group effect: F=17.131, P<0.001; interaction: F=3.808, P=0.024). The CV values in the graft area were significantly lower in the PVIBGT group than in the SHD-IBG group (group effect: F=5.065, P=0.030; between-group comparison at 6 months: P=0.042), indicating more homogeneous bone density distribution. The length (3.30±0.30 cm) and width (1.82±0.22 cm) of the graft area in the SHD-IBG group were significantly greater than those in the PVIBGT group ( t=4.328, P<0.001; t=3.956, P<0.001). At the last follow-up, in the PVIBGT group, patients without collapse had significantly higher CT HU values than those with collapse (528.59±123.02 vs. 420.67±168.70, t=2.108, P=0.046). In both groups, patients without collapse had significantly lower CV values than those with collapse (PVIBGT: 35.48%±7.76% vs. 50.66%±16.07%, t=3.672, P<0.001; SHD-IBG: 35.41%±8.68% vs. 61.75%±12.93%, t=4.215, P<0.001). Conclusions:PVIBGT produces a more homogeneous and stable high-density graft area due to superior revascularization, despite a relatively smaller graft size. SHD-IBG offers a larger graft coverage and higher initial bone density. A hybrid surgical approach that integrates the advantages of both techniques may provide a promising strategy for hip-preserving treatment of osteonecrosis of the femoral head.