This article is more than 5 years old. Information may no longer be current.
DXA assessment predicts metabolic risk better than BMI in postmenopausal women
Click Here to Manage Email Alerts
Measuring the android to gynoid fat ratio via DXA scan provides comparable data to visceral adipose tissue measurements for assessing cardiometabolic risk in postmenopausal women, according to recent study findings presented at The North American Menopause Society Annual Meeting.
In a study of postmenopausal women who underwent whole-body DXA scans and BMI assessments, researchers found that BMI measurements are a poorer measure of cardiometabolic risk vs. visceral adipose tissue (VAT) measurements.
“People who simply go by BMI, or height and weight [measurements], at least 20% of the time, may be missing something,” Steven R. Goldstein, MD, co-director of the bone densitometry and body composition unit at NYU Langone Medical Center in New York, told Endocrine Today. “DXA is less X-ray [exposure] than flying cross-country and is readily available. Even total body fat, as measured by DXA, we think is a much better assessment than standard BMI.”
Steven R. Goldstein
Goldstein and colleagues analyzed data from 113 postmenopausal women who underwent DXA scans to measure body fat in the android and gynecoid regions, as well as VAT measurements. Researchers calculated BMI for each woman. Due to the non-normality of data, researchers used Spearman nonparametric correlation coefficient test when comparing measurements; statistical analysis was performed using the statistical computing software R.
The Spearman correlation coefficient between VAT and android to gynoid fat ratio (A/G ratio) was 0.82 for both fat mass in pounds and fat volume. The correlation between percent body fat in the android region and VAT in pounds or volume was also 0.82. However, the correlation between BMI and VAT was 0.45.
“We were interested in seeing what the correlation was between standard A/G ratio, which is easily obtained, and this newer concept of VAT that is software generated,” Goldstein said. “And we found a very good correlation coefficient of 0.82. When we tried to compare BMI with VAT, the correlation coefficient was 0.45, which is really like flipping a coin, underscoring once again that BMI is a very poor indicator of metabolic risk, and that standard A/G ratio can be utilized as a surrogate for VAT.” – by Regina Schaffer
Reference:
Goldstein SR, et al. P-80. Presented at: The North American Menopause Society Annual Meeting; Sept. 30-Oct. 3, 2015; Las Vegas.
Disclosure: Goldstein reports no relevant financial disclosures.