正常中國婦女的髖部骨密度研究

Studies of Hip Bone Mineral Density in Normal Chinese Women

本研究之主要目的在於探討正常中國婦女的髖部骨密度，此研究係使用Norland 2600雙光子骨密度測量儀來測量266位正常中國婦女兩側髖部骨密度，其平均年齡為52歲（範圍18〜88歲），其中150位已停經；本研究所測定髖部骨骼部位依儀器電腦所設定之標準區域測得，包括股骨頸部，轉子間區，和Ward氏三角區。結果顯示髖部骨密度值會隨著年紀增高而降低。利用線性迴歸分析結果顯示，兩側髖部骨密度值具有高相關性，兩側股骨頸部，轉子間區，和Ward氏三角區的骨密度相關係數(r)分別為0.899，0.892和0.935，估計值標準誤分別為0.029，0.027和0.022 g/cm2。進一步分析顯示65歲以上婦女的平均相關係數為0.854，而65歲以下婦女則為0.887，顯示年紀對兩側髖部骨密度的對稱性並無影響。此外，探討慣用手側與其對側的髖部骨密度差異性時，發現慣用手側對髖部骨密度並無影響，因此，在實際作業上可用單側髖部骨密度代表對側，但是正常人兩側髖部骨密度的差異值頗大，在利用單側髖部骨密度值作臨床判定時，應小心注意。髖部骨密度降低時，髖部骨折的機會即增高，因此本研究進一步分析用單側髖部骨密度預測對側髖部骨密度時的準確性。此研究分為兩組：A組（18〜59歲，共189人），B組（60〜88歲，共77人）。若以0.6 g/cm2作分界點，所計算的陰性預測值為：A組股骨頸部(97.7%)，轉子間區(92.0%)，和Ward氏三角區(95.3%)；B組股骨頸部(82.1%)，轉子間區(86.4%)，和Ward氏三角區(44.9%)；綜合三個部位共同判定時並未改善陰性預測值。若將分界點提高為0.65 g/cm2時，陰性預測值會增高，但是準確度會明顯下降。因此，對於年紀較輕者（A組）可採用0.6g/cm2作為篩選對側髖部骨密度值的分界點，但對年紀較大者（B組）則建議採用0.65g/cm2作分界點為宜。

The purpose of this study was to assess the bone mineral density (BMD) of both hips in normal Chinese women. We used Norland 2600 dual-photon absorpiotmeter (DPA) (Norland Corp., Wisconsin U.S.A. Model 2600) to study 266 normal Chinese women. Their ages ranged from 18 to 88 years (mean, 52 years) and 150 women had experienced menopause. Three regions of interest (ROI) of the hip, i.e. the femoral neck, trochanter and Ward's triangle, were automatically selected by the Norland computer software. The results showed that the BMDs of the proximal femur decreased with the increasing age. Linear regression analysis of these the results showed an excellent correlation between bilateral hip BMD values. The correlation coefficients for femoral neck, trochanter, and Ward's triangle regions were 0.899, 0.892, and 0.935 g/cm2, and the standard errors of estimate (SEE) were 0.029, 0.027, and 0.022 g/cm2 respectively. Further analysis showed a mean correlation coefficient of 0.854 for elderly subjects (age ± 65 years, n = 51) and 0,887 for younger subjects (age<65 years, n = 215). Therefore, the symmetry of the hip BMDs was not affected by age. In addition, study of the effect of dominance on this symmetry revealed that dominance had little effect on the proximal femur BMD. Hence, assessment of the BMD of either hip alone may represent the BMD of the contralateral hip, and can reduce cost and radiation exposure during mass surveys for osteoporosis. However, the wide range of paired difference and the ratio of asymmetry of the proximal femur BMD in normal subjects should be considered in the interpretation of the proximal femoral BMD findings. Since the BMD of proximal femur is usually measured unilaterally, we studied the risk of underestimation with measurement of unilateral proximal femur BMD in the same 266 normal Chinese women: Group A (18〜59 years, n= 189) and Group B (60〜88 years, n= 77). Using 0.6 g/cm2 as a cutoff point, the negative predictive value (NPV) in Group A was 97.7% for femoral neck, 92.0% for the trochanter and 95.3% for Ward's triangle. In Group B the NPV was 82.1% for the femoral neck, 86.4% for the trochanter and 44.9% for Ward's triangle. The NPV increased at the expense of accuracy when the cutoff point was raised up to 0.65 g/cm2. Combining the BMDs of these areas does not significantly improve NPV, and decreases the accuracy. We suggest that measurement of unilateral proximal femur BMD is sufficient for screening the contralateral hip BMD in patients under 60 years of age at a cutoff point of 0.6 g/cm2. However, a cutoff point of 0.65 g/cm2 is recommended for Patients 60 years of age and older, and data of Ward's triangl should be in interpreted with caution.