【摘要】 目的 采用欧洲脊柱体模(European spine phantom,ESP)和志愿者用常用不同品牌的双能X线吸收测量仪(dual energy X-ray absorptiometry,DXA)测量的脊柱和髋部骨密度(bone mineral density,BMD)值进行体模校准和体内交叉校准,为建立统一的中国人群骨密度参考数据库,实现全国不同品牌DXA T值的标准化提供基础数据。方法采用FUJIFILM、 GE-Lunar、 Hologic、 Medilink和Osteosys 5个品牌的DXA。体模校准实验使用同一个ESP (编号SN 145)在每个扫描仪上扫描10次,每次重新定位。体内交叉校准研究在同一天使用GE-Lunar iDXA系统和其他4家品牌之一的DXA对100名20～80岁健康志愿者进行腰椎(L1-4)和髋部扫描。采用SPSS 22. 0完成线性回归分析,将FUJIFILM、Hologic、Medilink和Osteosys的脊柱、股骨颈和全髋BMD与GE-Lunar相对应部位的BMD进行交叉校准。并根据一个采用GE-Lunar DXA的75 321名20岁及以上中国成年人全国性骨密度调查数据库计算正常参考值,采用本次体内交叉校准换算方程推导出其他4种品牌DXA的等效青年人参考数据库。结果 GELunar、FUJIFILM、 Hologic、 Medilink和Osteosys DXA的ESP L2-4 BMD (单位:g/cm2)均值分别为1. 170±0. 005、0. 930±0. 005、0. 980±0. 005、1. 000±0. 009和1. 100±0. 003,各品牌间BMD差异有统计学意义(P<0. 01)。10次重复扫描的变异系数(CV%)分别为0. 31%、0. 53%、0. 37%、0. 91%和0. 25%。FUJIFILM、Hologic、Medilink和Osteosys DXA脊柱和髋部的BMD测量值与GE-Lunar测量值高度相关,Pearson相关系数r为0. 911～0. 982,并计算出每种DXA的BMD值与GE相应部位BMD值的换算公式。列出5家DXA等效中国青年人骨密度参考范围,实现这些DXA设备扫描同一患者时计算的T值一致。结论 本研究提供了国内5种品牌DXA的ESP和体内交叉校准数据以及换算公式,获得了统一的中国人DXA骨密度正常参考值,为实现全国DXA脊柱和髋部骨密度T值的标准化提供基础数据。更多还原

【Abstract】 Objective We performed the European spine phantom( ESP) in vitro and healthy volunteersin-vivo cross-calibration studies with dual-energy X-ray absorptiometry( DXA) to compare spine and hip bone mineral density( BMD) measurements on 5 different DXA scanners used in China. The aim was to generate consistent young adult reference data to standardize the clinical reporting of T-scores produced on different DXA brands across China. Methods The study examined FUJIFILM,GE-Lunar,Hologic,Medilink,and Osteosys DXA systems. For in vitro study the same ESP( SN 145) was scanned ten times on each scanner with repositioning. For in vivo cross-calibration,groups of 100 healthy volunteers aged 20-80 years had lumbar spine( L1-4) and hip scans on the same day on a GE-Lunar DXA system and one of the other four manufacturers’ systems. SPSS 22. 0 was used for statistical analysis. Linear regression analysis was used to cross-calibrate spine,femoral neck,and total hip BMDs for FUJIFILM,Hologic,Medilink,and Osteosys relative to GE-Lunar BMDs and calibration equation was established for spine and hip for each brand DXA. Previously published young adult reference data for GE-Lunar derived from a nationwide study of 75 321 Chinese adults aged 20 years and older were used together with the in vivo cross-calibration equations to derive equivalent young adult reference data for the other four DXA brands. Results The mean ESP L2-4 BMD values for GE-Lunar,FUJIFILM,Hologic,Medilink,and Osteosys DXA were 1. 170±0. 005,0. 930±0. 005,0. 980±0. 005,1. 000±0. 009 and 1. 100±0. 003 g/cm2,respectively.The coefficient of variation( CV%) of 10 repeated scans was 0. 31,0. 53,0. 37,0. 91,and 0. 25%,respectively. In vivo BMD measurements on the FUJIFILM,Hologic,Medilink,and Osteosys systems were highly correlated with the equivalent GE-Lunar measurements with Pearson correlation coefficients between 0. 911 and 0. 982. We tabulated the equivalent Chinese young adult reference ranges for all five DXA manufacturers. The use of these ranges would ensure consistent T-score of a patient cross 5 brands of DXA. Conclusions We present ESP in vitro and in vivo cross-calibration data for five different brands of DXA used in China. From these data we derive consistent young adult reference ranges whose use will ensure the standardization of DXA spine and hip T-scores in Chinese population.更多还原