基于中日友好医院分型的股骨头坏死保髋手术疗效的影响因素分析

Clinical outcomes and impact factors of hip preserving surgery for osteonecrosis of the femoral head based on China-Japan Friendship Hospital classification

目的探讨基于中日友好医院(China-Japan Friendship Hospital,CJFH)分型的保髋手术疗效的影响因素。方法2012年6月至2016年9月接受保髋手术治疗的非创伤性股骨头坏死患者325例432髋。依据CJFH分型行髓芯减压自体骨髓单个核细胞移植141髋(髓芯减压组)和头颈部开窗减压病灶清除植骨术291髋(开窗减压组)。采用Harris髋关节功能评分评估临床疗效,摄X线片观察坏死修复、股骨头塌陷及关节退变。临床失败定义为末次随访时Harris评分较术前评级降低和(或)影像学上股骨头塌陷进展(ARCO分期增加)。终点事件为疼痛明显加重、Harris评分为差(<70分)和(或)股骨头进行性塌陷。采用Cox比例回归模型分析临床失败的危险因素。结果髓芯减压组67髋临床失败(47.5%,67/141),其中CJFH分型C+M型3髋(13.0%,3/23)、L1型24髋(38.1%,24/63)、L2型14髋(82.4%,14/17)、L3型26髋(68.4%,26/38)。开窗减压组106髋临床失败(36.4%,106/291),C+M型1髋(33.3%,1/3)、L1型41髋(31.3%,41/131)、L2型22髋(84.6%,22/26)、L3型42髋(32.1%,42/131)。髓芯减压组中不同年龄(χ2=3.887,P=0.049)、不同术前CJFH分型(χ2=40.943,P=0.000)的临床失败率的差异有统计学意义;Cox回归模型分析显示年龄≥40岁[HR=2.325,95%CI(1.398,3.866),P=0.000]、术前Harris评分70~80分[HR=2.163,95%CI(1.140,4.105),P=0.018]和<70分[HR=2.597,95%CI(1.173,5.749),P=0.019]、术前CJFH分型L2型[HR=35.052,95%CI(7.721,159.133),P=0.000)和L3型[HR=13.242,95%CI(3.104,56.491),P=0.000]是临床失败的危险因素。开窗减压组中不同年龄(χ2=8.437,P=0.004)、不同术前Harris评分(χ2=19.737,P=0.000)、不同术前CJFH分型(χ2=29.265,P=0.000)的临床失败率的差异有统计学意义;Cox回归模型分析显示术前Harris评分<70分[HR=5.102,95%CI(2.339,11.129),P=0.000]、术前CJFH分型L2型[HR=32.761,95%CI(6.165,43.507),P=0.000]是临床失败的危险因素。结论股骨头坏死保髋手术治疗效果受年龄、术前Harris评分和CJFH分型的影响;术前临床症状越明显,受累股骨头负重关节面及外侧柱支撑结构未得到有效重建,保髋手术的预后越差。

Objective To evaluate the risk factors related to the mid-term outcomes of hip preserving surgery for early stages osteonecrosis of the femoral head(ONFH)basing on China-Japan Friendship Hospital(CJFH)classification system.Methods From June 2012 to September 2016,there were consecutive 325 patients(432 hips)were enrolled and divided into different preserving surgery groups,namely core decompression(CD)group 141 hips and"lightbulb"operation(LB)group 291 hips,respectively.Harris hip score(HHS)was used to evaluate the clinical outcomes.The progression of ONFH was observed by radiography.Clinical failure was defined as worsen of HHS and/or radiographic evaluation.Clinical endpoint events were marked as significant hip pain(HHS<70),and/or collapse of the femoral head requiring further interventions.Potential risk factors,including sex,age,etiology,the duration from symptom onset to treatment,preoperative CJFH type,ARCO stage and HHS,were analyzed using univariate risk analysis and Cox regression multivariate risk model.Results The rate of hip failure was 47.5%(67/141)in CD group,including type C+M 13.0%(3/23),L138.1%(24/63),L282.4%(14/17)and L368.4%(26/38),respectively.There was significant difference in age(χ2=3.887,P=0.049),type of CJFH(χ2=40.943,P=0.000)in CD group.The Cox regression analysis revealed that age≥40(HR=2.325,95%CI 1.398,3.866,P=0.000),pre-HHS 70-80(HR=2.163,95%CI 1.140,4.105,P=0.018)and<70(HR=2.597,95%CI 1.173,5.749,P=0.019),type L2(HR=35.052,95%CI 7.721,159.133,P=0.000)and L3(HR=13.242,95%CI 3.104,56.491,P=0.000)were associated with failure of core decompression.The rate of hip failure was 36.4%(106/291)in LB group,including type C+M 33.3%(1/3),L131.3%(41/131),L284.6%(22/26)and L332.1%(42/131),respectively.There were significant differences in age(χ2=8.437,P=0.004),pre-HHS(χ2=19.737,P=0.000)and type of CJFH(χ2=29.265,P=0.000)in LB group.The Cox regression analysis showed that poor pre-HHS(HR=5.102,95%CI 2.339,11.129,P=0.000),type L2(HR=32.761,95%CI 6.165,43.507,P=0.000)were associated with failure of"lightbulb"preserving surgery.Conclusion The results of hip preserving surgery for ONFH are associated with age,preoperative HHS and CJFH typing.The prognosis depends on the severity of symptoms,the residual of weight-bearing joint surface and lateral pillar of the femoral head.