Kashiwazaki H
Department of Human Sciences, School of Health Sciences, University of Occupational and Environmental Health, Kitakyushu, Japan.
J Nutr Sci Vitaminol (Tokyo). 1999 Jan;45(1):79-94. doi: 10.3177/jnsv.45.79.
Energy expenditure (TEE) determined by the doubly labeled water technique (DLW) in 10 adult subjects (5 males and 5 females) was used to evaluate the energy expenditure estimated from 24 h heart rate monitoring records (HR method) for 2 d randomly sampled during 2 weeks of DLW study. Individual data on HR and oxygen consumption, obtained during a step test (resting conditions, and up and down at 3 or 4 stepping rates) and postabsorptive conditions (resting metabolic rate: RMR), were used to calculate three types of calibration regression line, i.e., straight-linear regression (EE-HR: A for TEEhr-A), log-linear regression (lnEE-HR: B for TEEhr-B), and two-linear regression (flex-HR method: C for TEEhr-C). When the 24 h HR records were applied, these calibration regressions provided three estimates of TEE (mean +/- SD kcal/24 h): TEEhr-A (3,059 +/- 1,246 in all subjects), TEEhr-B (2,472 +/- 843), and TEEhr-C (2,759 +/- 1,228). Mean TEE determined by the DLW method (TEEdlw) was 2,544 +/- 378 kcal/24 h. Although no mean values estimated by HR methods were significantly different statistically from the mean value of TEEdlw, the variances of the estimates (e.g., SD) by the HR method were much greater than that of TEEdlw (between twofold and threefold). TEEhr-B estimated by lnEE-HR regression provided the smallest differences from that of TEEdlw (mean difference of -3.1% with a range of -35.1(-)+36.6%). From these observations, the following conclusions were made: 1) The estimates of TEE by HR are useful as a group mean, but interpretation of the individual TEE estimates requires caution because of great deviations from the reference values. 2) Among the calibration methods tested, the log-linear calibration regression (lnEE on HR) gives the best estimates of TEE by the HR method and is recommended for use in future studies.
在10名成年受试者（5名男性和5名女性）中，采用双标记水技术（DLW）测定的能量消耗（TEE），用于评估在DLW研究的2周内随机抽取的2天中，根据24小时心率监测记录估算的能量消耗（HR法）。在阶梯试验（静息状态，以及以3种或4种阶梯速率上下运动）和吸收后状态（静息代谢率：RMR）期间获得的心率和耗氧量的个体数据，用于计算三种校准回归线，即直线回归（EE-HR：TEEhr-A的A）、对数线性回归（lnEE-HR：TEEhr-B的B）和双线性回归（flex-HR法：TEEhr-C的C）。当应用24小时心率记录时，这些校准回归提供了TEE的三种估计值（平均值±标准差kcal/24小时）：TEEhr-A（所有受试者中为3059±1246）、TEEhr-B（2472±843）和TEEhr-C（2759±1228）。通过DLW法测定的平均TEE（TEEdlw）为2544±378 kcal/24小时。虽然HR法估计的平均值与TEEdlw的平均值在统计学上无显著差异，但HR法估计值的方差（如标准差）比TEEdlw的方差大得多（两倍至三倍）。通过lnEE-HR回归估计的TEEhr-B与TEEdlw的差异最小（平均差异为-3.1%，范围为-35.1（-）+36.6%）。基于这些观察结果，得出以下结论：1）HR法对TEE的估计作为群体平均值是有用的，但由于与参考值有很大偏差，对个体TEE估计值的解释需要谨慎。2）在所测试校准方法中，对数线性校准回归（HR上的lnEE）通过HR法对TEE给出了最佳估计值，建议在未来研究中使用。
