Validity of predictive equations for resting energy expenditure among Iranian women

To determine the resting metabolic rate (RMR) in a sample of Iranian women, and to evaluate the validity of predictive equations for estimating RMR in normal and obese subjects.

This cross-sectional study was conducted on a total of 187 healthy women aged 18-45 years. Anthropometric data were collected and subjects' RMR was measured by indirect calorimetry for 15 minutes following an overnight fast. RMR was also predicted using various types of formulas based on subjects' height, age, weight or fat-free mass. Body composition was estimated by bioelectric impedance analysis (BIA).

Measured RMR (mean ± SD) was found to be 1473 ± 296 kcal/day. The abbreviation formula overestimated RMR, while other formulas underestimated it. Harris-Benedict formula was valid among all BMI categories (normal weight, overweight, obese and morbidly obese). Two Schofield formulas were valid in normal weight, overweight and morbidly obese subjects; and Cunningham formula was valid only among overweight and obese women. Overweight and obese Iranian women had higher RMR in comparison with normal weight subjects (p<0.01); although after age and weight adjustment, the differences were not significant in any of the BMI categories.

The Harris-Benedict for mula provides a valid estimation of RMR at the group level in a range of normal-weight to morbidly obese Iranians. However, at the individual level, errors might be so high that using a measured value has to be preferred over an estimated value.