Frame size, ethnicity, lifestyle, and biologic contributors to areal and volumetric lumbar spine bone mineral density in Indian/Pakistani and American Caucasian premenopausal women.

Published data on the spinal bone mineral density (BMD) of premenopausal women originating from the Indian subcontinent (Indian/Pakistani) are few. We compared anteroposterior (AP) and lateral areal BMD (aBMD) using dual X-ray absorptiometry and calculated volumetric BMD (vBMD) in Indian/Pakistani (n = 47) vs American (n = 47) women with dissimilar statures and skeletal sizes. To account for differences, we "adjusted" lumbar aBMD separately for vertebral size (aBMD/the square root of the projected area), height (aBMD/height), and hip skeletal width (aBMD/hip width). We "corrected" bone mineral content (BMC), aBMD, and vBMD for frame size, collectively using height, hip width, and vertebral size. Unadjusted mean aBMD values for AP lumbar (L1-L4, p = 0.0086; L3-L4, p = 0.044) spine were higher in Americans than Indians/Pakistanis,whereas lateral vBMD (p = 0.56) or aBMD (p = 0.060) values were not different. After adjusting for height, hip width, or vertebral size, or correcting for frame size, differences in aBMD disappeared. Regression analyses indicated that the best measures to correct for frame size were: vertebral area for BMC, hip width for aBMD, and vertebral width for lateral vBMD. Height was not significant in any model. In correcting for frame size, we accounted for 73-85% of the variability in BMC, 22-28% in aBMD, and 27% in lateral vBMD. After frame size was corrected, we accounted for 34% of the variability in AP BMC and aBMD, in contrast with 6-9% in the lateral models. Five significant biologic and lifestyle factors remained in AP models; only body weight remained for lateral spine. Upon accounting for frame size using regression, much variability in BMD, aBMD, and vBMD was explained by lifestyle and biologic factors, not by ethnicity.