Hofsteenge Geesje H, Chinapaw Mai J M, Weijs Peter J M
Department of Nutrition & Dietetics, Internal Medicine, VU University Medical Center, De Boelelaan 1117, 1081, , HV, Amsterdam, The Netherlands.
EMGO Institute for Health and Care Research, VU University Medical Center, Amsterdam, The Netherlands.
BMC Pediatr. 2015 Oct 15;15:158. doi: 10.1186/s12887-015-0476-7.
In clinical practice, patient friendly methods to assess body composition in obese adolescents are needed. Therefore, the bioelectrical impedance analysis (BIA) related fat-free mass (FFM) prediction equations (FFM-BIA) were evaluated in obese adolescents (age 11-18 years) compared to FFM measured by dual-energy x-ray absorptiometry (FFM-DXA) and a new population specific FFM-BIA equation is developed.
After an overnight fast, the subjects attended the outpatient clinic. After measuring height and weight, a full body scan by dual-energy x-ray absorptiometry (DXA) and a BIA measurement was performed. Thirteen predictive FFM-BIA equations based on weight, height, age, resistance, reactance and/or impedance were systematically selected and compared to FFM-DXA. Accuracy of FFM-BIA equations was evaluated by the percentage adolescents predicted within 5% of FFM-DXA measured, the mean percentage difference between predicted and measured values (bias) and the Root Mean Squared prediction Error (RMSE). Multiple linear regression was conducted to develop a new BIA equation.
Validation was based on 103 adolescents (60% girls), age 14.5 (sd1.7) years, weight 94.1 (sd15.6) kg and FFM-DXA of 56.1 (sd9.8) kg. The percentage accurate estimations varied between equations from 0 to 68%; bias ranged from -29.3 to +36.3% and RMSE ranged from 2.8 to 12.4 kg. An alternative prediction equation was developed: FFM = 0.527 * H(cm)(2)/Imp + 0.306 * weight - 1.862 (R(2) = 0.92, SEE = 2.85 kg). Percentage accurate prediction was 76%.
Compared to DXA, the Gray equation underestimated the FFM with 0.4 kg (55.7 ± 8.3), had an RMSE of 3.2 kg, 63% accurate prediction and the smallest bias of (-0.1%). When split by sex, the Gray equation had the narrowest range in accurate predictions, bias, and RMSE. For the assessment of FFM with BIA, the Gray-FFM equation appears to be the most accurate, but 63% is still not at an acceptable accuracy level for obese adolescents. The new equation appears to be appropriate but await further validation. DXA measurement remains the method of choice for FFM in obese adolescents.
Netherlands Trial Register ( ISRCTN27626398).
在临床实践中,需要采用对患者友好的方法来评估肥胖青少年的身体成分。因此,对生物电阻抗分析(BIA)相关的去脂体重(FFM)预测方程(FFM - BIA)在肥胖青少年(11 - 18岁)中进行了评估,并与通过双能X线吸收法测量的FFM(FFM - DXA)进行比较,同时开发了一个针对特定人群的新FFM - BIA方程。
经过一夜禁食后,受试者前往门诊。测量身高和体重后,进行双能X线吸收法(DXA)全身扫描和BIA测量。系统地选择了13个基于体重、身高、年龄、电阻、电抗和/或阻抗的预测FFM - BIA方程,并与FFM - DXA进行比较。通过预测值在FFM - DXA测量值的5%范围内的青少年百分比、预测值与测量值之间的平均百分比差异(偏差)以及均方根预测误差(RMSE)来评估FFM - BIA方程的准确性。进行多元线性回归以开发新的BIA方程。
验证基于103名青少年(60%为女孩),年龄14.5(标准差1.7)岁,体重94.1(标准差15.6)kg,FFM - DXA为56.1(标准差9.8)kg。各方程的准确估计百分比在0%至68%之间;偏差范围为 - 29.3%至 + 36.3%,RMSE范围为2.8至12.4 kg。开发了一个替代预测方程:FFM = 0.527 * 身高(cm)² / 阻抗 + 0.306 * 体重 - 1.862(R² = 0.92,标准误 = 2.85 kg)。准确预测百分比为76%。
与DXA相比,Gray方程低估FFM 0.4 kg(55.7 ± 8.3),RMSE为3.2 kg,准确预测率为63%,偏差最小为( - 0.1%)。按性别划分时,Gray方程在准确预测、偏差和RMSE方面的范围最窄。对于用BIA评估FFM,Gray - FFM方程似乎是最准确的,但63%的准确率对于肥胖青少年来说仍未达到可接受的水平。新方程似乎合适,但有待进一步验证。DXA测量仍然是肥胖青少年FFM测量的首选方法。
荷兰试验注册中心(ISRCTN27626398)