应用FRAX预测绝经后女性骨折风险的回顾性研究  

Fracture Risk Assessment Tool on predicting fracture risk in postmenopausal women:a retrospective study

作　　者：陈浩[1] 侍管 唐海[1] 贾璞[1] 包利[1] 冯飞[1] 陈萌萌[1] CHEN Hao;SHI Guan;TANG Hai;JIA Pu;BAO Li;FENG Fei;CHEN Meng-meng(Department of Orthopaedics,Beijing Friendship Hospital,Capital Medical University,Beijing 100050,China)

机构地区：[1]首都医科大学附属北京友谊医院骨科,北京100050

出　　处：《中华骨质疏松和骨矿盐疾病杂志》2023年第4期297-304,共8页Chinese Journal Of Osteoporosis And Bone Mineral Research

基　　金：国家重点研发计划(2021YFC2501700)。

摘　　要：目的探讨骨折风险评估工具(Fracture Risk Assessment Tool,FRAX)对绝经后女性骨折风险的预测价值。方法回顾性分析750名北京地区绝经后女性的FRAX评分,收集每位受试者股骨颈骨密度(bone mineral density,BMD)及FRAX算法中需要的危险因素,并随访收集10年内是否新发骨折,计算年龄及体质量指数(body mass index,BMI),用FRAX工具计算10年内骨质疏松引起主要部位骨折(脊柱、前臂、肩部骨折)概率(the 10-year probability of major osteoporotic fractures,PMOF)和10年内髋骨骨折概率(the 10-year probability of hip osteoporotic fractures,PHF)。分析FRAX预测值与年龄、BMI、股骨颈BMD的相关性,建立二分类Logistic回归方程研究10年内新发骨折的显著危险因素,并建立模型得到预测值,以10年内是否新发骨折为标准绘制受试者工作曲线(receiver operating characteristic curve,ROC),计算FRAX预测值、BMD及模型预测值曲线下面积(area under curve,AUC),以AUC评估FRAX工具的预测价值。结果研究对象平均年龄为(62.49±6.59)岁,平均体质量为(63.03±9.63)kg,平均身高为(156.11±5.10)cm,平均股骨颈BMD(T值)为-1.12±0.95。危险因素与FRAX预测值的相关性分析显示,除吸烟和饮酒外,余危险因素与PMOF的差异均有统计学意义(P<0.05);除饮酒和风湿性关节炎外,余危险因素与PHF的差异均有统计学意义(P<0.05)。年龄、BMI和股骨颈BMD与FRAX评分的相关性分析显示,年龄与PMOF和PHF呈正相关,差异有统计学意义(P<0.05),而BMI、股骨颈BMD与PMOF和PHF均呈负相关,差异有统计学意义(P<0.05)。以10年内是否新发骨折为因变量生成Logistic回归方程评估危险因素,得出骨折史、吸烟、股骨颈BMD、体重、身高对女性骨折风险的影响较大并得到预测模型。ROC曲线分析PMOF、PHF、股骨颈BMD和回归模型预测10年内骨折的AUC分别为0.616、0.600、0.573和0.670(P<0.05),但诊断能力较低。结论FRAX可应用于北京地区绝经�Objective To explore the predictive value of the Fracture Risk Assessment Tool(FRAX)in assessing the risk of fractures in postmenopausal women.Methods A retrospective analysis was conducted in 750 postmenopausal women in Beijing area.The FRAX scores,bone mineral density(BMD)at the femoral neck for each participant,and the risk factors required by the FRAX algorithm were collected.Follow-up data on whether new fractures occurred within 10 years were also collected.Age and body mass index(BMI)were calculated,and the FRAX tool was used to estimate the 10-year probability of major osteoporotic fractures(PMOF)(spine,forearm,shoulder)and the 10-year probability of hip osteoporotic fractures(PHF)caused by osteoporosis at any site.The correlation between FRAX predictions and age,BMI,and femoral neck BMD were analyzed.A binary logistic regression equation was established to study significant risk factors for new fractures within 10 years and to create a predictive model.The receiver operating characteristic curve(ROC)was used with new fractures occurring within 10 years as the standard to calculate the area under the curve(AUC)for FRAX predictions,BMD,and the model s predictions,evaluating the predictive value of the FRAX tool.Results The average age of the subjects was(62.49±6.59)years,average body weight was(63.03±9.63)kg,and average height was(156.11±5.10)cm.The mean femoral neck BMD(T-score)was-1.12±0.9.The correlation analysis between risk factors and FRAX predictions showed that,except for smoking and alcohol consumption,all other risk factors had statistically significant correlations with PMOF(P<0.05).Similarly,except for alcohol consumption and rheumatoid arthritis,all other risk factors had statistically significant correlations with PHF(P<0.05).In the correlation analysis between age,BMI,femoral neck BMD and FRAX scores,age showed a positive correlation with PMOF and PHF(P<0.05).However,BMI and femoral neck BMD showed a negative correlation with PMOF and PHF(P<0.05).Using new fractures occurring within 10 y

关 键 词：骨质疏松 骨折风险评估工具 绝经后女性 

分 类 号：R681[医药卫生—骨科学]