• Develop better predictive equations for BMR in healthy Chinese adults
• Evaluate factors that may influence BMR.

Definitions

• Steady state: VO₂and CO₂measurements obtained every 30-secs until equilibrium achieved for at least 5 to 6 consecutive minutes
• Harris-Benedict equation: Cited 1918 study
• Ideal body weight: Cited Huang 1992 for Chinese persons
• Body fat: Siri’s equation, 1961
• Body Surface Area (BSA) equation: used formula derived from Dubois and Dubois
• Body Cell Mass (BCM) equation: Used Moore equation
• Fat-free mass (FFM): Body weight (1-%body fat).

• Understand and give written consent
• Healthy
• Normal weight for Chinese persons (based on equations derived by Huang et al for Chinese persons).

• Refusal to consent
• Hx of current illnesses
• Recent weight loss
• Endocrine disorders
• Pharmacologic therapy
• Hormonal treatment
• Extremely underweight (<80% IBW) or obese (120% IBW).

Recruitment

Procedures not described

Design

Cross-sectional study

Subjects stratified on basis of gender and age (20-29, 30-39, 40-49, 50-59, 60+ years)

Blinding used

Not applicable

Intervention

Not applicable

Statistical Analysis

• Student's T-test used to assess differences between men and women
• Pearson correlation coefficients and linear regression analyses used to evaluate relationships between measured BMR and age, weight, height, percentage ideal body weight, percent body fat, fat-free mass, body surface area and body cell mass results expressed as mean±standard deviation (SD)
• Cross-validation study: Sample was randomly divided into two groups for development and cross-validation of predictive equations; there were no differences in body composition parameters between each sample.

Timing of measurements

One measurement time

Dependent variables

• Measured BMR: Indirect calorimetry [(VO₂, liters per minute), VCO₂ (liters per minute; ml/kg per minute)].
  • C type: Metabolic cart with a canopy system
  • Rest before measure: “Relaxed supine position”
  • Fasting length: 12 hours; Measured between 8-10 a.m.
  • Exercise conditioning 24 prior to test: Refrain 12 hours before test
  • Smoking: Refrain one hour before test
  • Room temp: Not reported
  • No. of measures were they repeated? O₂ and CO₂ taken at 30 second intervals until a steady-state was achieved and maintained for at least five to six consecutive minutes
  • Steady state: VO₂ and CO₂ measurements obtained every 30-seconds until equilibrium achieved for at least five to six consecutive minutes
  • Coefficient of variation? None reported
  • Equipment of Calibration: None reported
  • Training of measurer? Dietitians trained to use a standardized protocol
  • Subject training of measuring process? None reported
  • Sleep at the facility: Not reported
  • Monitored heart rate? None reported
  • Body temperature? None reported.

Independent variables

• Predicted REE using HB, Mifflin, Owen (men), Owen(women), Kleiber, Cunningham
  • Harris-Benedict equation: Cited 1918 study
• Height: Method not reported
• Weight: Method not reported
• Body mass index (BMI: kg/m²
• Ideal body weight: based on Huang, 1992 for Chinese persons
• Body composition (body fat and fat free mass): sum of skinfolds (biceps, triceps, subscapular, suprailiac) and bioelectrical impednace analysis, resistance and reactance
  • Body fat: Siri’s equation, 1961
  • Fat free mass (FFM): body weight (1% body fat)
• Body Surface Area (BSA) equation: Formula derived from Dubois and Dubois
• Body Cell Mass (BCM) equation: Moore equations for males and females.

Control variables

Age and gender: Sample stratified

• Initial N: Not given
• Attrition (final N): N=223 Chinese adults (N=102 males; N=121 females)
• Age: Mean age: 43.8±14.3 years; range: 20-78 years
• Ethnicity: Chinese
• Anthropometrics:

• BFS: Body fat measured by BIA; BFB- Body fat measured by bioelectrical impedance analysis; FFMS-fat-free mass measured by skinfold thicknesses; FFMB- fat-free mass measured by BIA, BCM-body cell mass; BSA- body surface area
• * Significantly different from men at P=0.0001
• # Significantly different from BFS at P=0.02.
• Location: Taipei, Taiwan.

Pearson Correlation Coefficients between:

• BMR and weight: 0.77
• BMR and height: 0.73
• BMR and age: -0.25
• BMR and  percent IBW: 0.33
• BMR and BFS: -027
• BMR and BFB: -0.34
• BMR and FFMS: 0.84
• BMR and FFMB: 0.86
• BMR and BSA: 0.80
• BMR and BCM: 0.89.

BMR best correlated with body cell mass (BCM) and BCM correlated best with fat-free mass by BIA (R=0.94). There was a negative correlation between BCM and age (-0.28).

BMR adjusted for body composition and body surface area

 *  P=0.0001

Clinical: Resting energy expenditure

• P=0.0001

Differences between measured and predicted RMR in males and females comparison measured BMR and estimate BMR (i.e., total sample) by different predictive equations

[Taken from Liu HY, Lu YF, Chen WJ. Validity of predictive equations for the calculation of basal metabolic rate in healthy Chinese adults. Chinese Nutr Soc. 1994; 19 (2): 141-150.]

Differences between measured and predicted RMR in males and females

Developing predictive equations

The men and women in the two samples (validation and cross-validation) were comparable in age, anthropometric measurements (height and weight), body composition (including percent body fat, fat-free mass and body cell mass) and body surface area.

The mean difference, r and P-value for paired T-test results between predicted BMR (X-var) and measured BMR (Y-var) using cross-validation sample (N=104):

• All of the available predictive equations overestimated BMR in healthy Chinese adults (P=0.0001)

The mean difference, r and P-value for paired T-test results between predicted BMR (X-var) and measured BMR (Y-var) in cross-validation sample (N=104)

• No significant differences were observed between measured BMR and predicted values estimated by the nine equations developed in this study

[NOTE: The four equations above are discussed due to availability of obtaining the factors needed.]

• The BMR was estimated in 10 randomly selected subjects from the cross-validation study using Mifflin et al equation and equation 8
• Significant differences were observed between BMR estimated by Mifflin and measured BMR (P=0.0001). No significant differences were observed between measured BMR and BMR estimated by equation 8.

As stated by the author in body of report:

• In our study, women had a lower metabolic rate than men because of a relatively smaller amount of body cell mass; the result of smaller skeletal muscle mass in tandem with greater fat mass.
• The major difference between men and women was the difference in amount of skeletal muscle mass
• Previous findings suggest resting energy expenditure correlated most highly with fat-free mass, which correlated highly with weight and height. Our results agree that accurate determination of height and weight, not calorimetry or assessment of body composition are needed to provide a base for estimating daily energy expenditure of individual mean and women.
• Fat-free mass was the best single predictor of BMR and could explain 75% of the variance observed among people... the same r² value was obtained when body weight and body height were used to substitute fat-free mass for predicting BMR. The use of four variables-body weight, body height, age and sex resulted in an r² value of 0.81. 81% of the variance in BMR among persons was accounted for by the covariates of weight, height, age and sex.
• A significant degree of overestimation was still observed in predicting BMR in Chinese adults using the Mifflin et al equations (P=0.0001)
• We do not know if predictive equations overestimated BMR for Chinese adults is the result of ethnic differences or other factors such as climate and lifestyle
• We recommend these [our] equations for clinical use in healthy Chinese adults who are within normal limits for body weight.

Strengths 

Selected an important ethnic group within a healthy population.

Generalizability/Weaknesses

• Large study population represented and stratified by age
• Cross-validated equations on separate stratified sample
• Study weaknesses include:
  • No subject training prior to measurement; did not sleep over at facility where measurement being taken-hence definition of “BMR” is most likely “Resting metabolic rate”
  • Room temperature not reported
  • Sample represents Chinese adults living in Taipei, Taiwan, Republic of China; Hence, considering lifestyle differences-limited generalizability to US-residing Chinese population.