Age-specific reference values of hip geometric indices from a representative sample of the Japanese female population: Japanese Population-based Osteoporosis (JPOS) Study

Abstract

We analyzed 2,107 hip dual-energy X-ray absorptiometry (DXA) images from the Japanese Population-based Osteoporosis Study with the Hip Structure Analysis (HSA) program to obtain age-specific reference values of HSA indices for the Japanese female population. These references may help physicians accurately assess HSA results and aid researchers in making interracial comparisons of the indices.

Introduction: Hip geometry is expected to improve hip fracture risk assessment, which is usually assessed by bone mineral density (BMD) alone. We aimed to establish a reference database for Japanese women.

Methods: We studied 2,107 Japanese women (15-79 years old) with no history of bone metabolism-related diseases from the Japanese Population-based Osteoporosis Study performed in 1996. Hip geometry was conducted on DXA images with the HSA program, which yielded data for cross-sectional area (CSA), subperiosteal diameter (PD), endocortical diameter (ED), mean cortical thickness (CT), section modulus (SM), and buckling ratio at the narrow neck (NN), intertrochanter (IT), and femoral shaft (FS) regions. Mean HSA indices were determined for each 5-year age group after adjustment for height and weight based on most recent Japanese population values.

Results: Age-related changes in HSA indices were evident for the 50-54 year group in the NN and IT regions and for the 55-59 year group in the FS region; these changes increased with age thereafter. Age-related changes in CSA and CT were almost identical to that of BMD. Japanese subjects exhibited BMD and CT values similar to those reported for US non-Hispanic white women, but had 16-23% smaller SM values. CSA and CT were highly correlated with conventional BMD, whereas ED, SM, and PD showed lower correlations.

Conclusions: Age-specific reference values of HSA indices for the Japanese female population were obtained. This database will form the foundation for accurate HSA result evaluation.