Validity of predictive equations for resting metabolic rate in healthy older adults.

Background & Aims: Accurate estimation of energy expenditure in older people is important for nutritional support. The current literature contains controversial or inconsistent data regarding the resting metabolic rate (RMR, or basal metabolic rate) in older adults, including the relationship between the RMR and ethnicity. Little information about the RMR in healthy Asian older adults is available. This study was performed to examine the RMR in healthy Japanese older adults and compare it with previously established 16 equations.

Methods: Thirty-two community-dwelling, healthy, and active elderly Japanese adults were enrolled (age, 64-87 years; 14 men, 18 women; mean height, 154.9 ± 8.9 cm; mean weight, 53.5 ± 9.1 kg; mean body mass index, 22.2 ± 2.5 kg/m). The RMR was measured by indirect calorimetry. The measured RMR was compared among 16 equations. Correlation analysis, a paired t test, and a Bland-Altman plot were used to assess the agreement among the equations.

Results: The average RMR was 1132 ± 178 kcal/day with 2233 ± 437 kcal/day average total energy expenditure (TEE) measured by doubly labeled water (DLW). The smallest bias was established by De Lorenzo et al.'s equation as bias ±1.96SD = 4 ± 121 kcal/day. De Lorenzo et al. and Ikeda et al.'s equations had no significant average bias both in men and women (P > 0.05). The 1.96SD of bias in six equations was within 160 kcal/day. In contrast, residuals between the measured and predicted RMR were largely correlated with the RMR in four equations. A sex-related difference in the mean bias was observed in many equations.

Conclusion: Although the average Japanese healthy older adult has a shorter stature and lower weight than older adults in the Western population, the current data suggest that a similar predictive equation for the RMR can be applied to both Japanese and Western older adults. This study demonstrate that the De Lorenzo et al.'s or Ikeda's equation may be useful for estimating RMR in the community-dwelling, healthy, and active elderly Japanese adults without any systematic bias.