Bilateral comparison of femoral bone density and hip axis length from single and fan beam DXA scans.

TLDR: It is concluded that measurement of a single proximal femur will usually be sufficient for clinical evaluation of BMD and/or hip axis length, and bilateral BMD measurements are indicated in subjects where unilateral degeneration or disease are suspected.

Abstract: Dominant/nondominant differences in bone mineral density (BMD) have been observed in the upper extremities. However for the proximal femur, the distinction between dominant and nondominant hips is not clear. The purpose of this study is to evaluate left/right variations in femoral BMD and hip axis length (HAL) in both single beam and fan beam dual x-ray absorptiometry (DXA) scans. A total of 36 women aged 41–76 years (average age 60±10 years) received single beam and fan beam DXA scans of both proximal femora with a Hologic QDR-2000 scanner. Femoral BMD and hip axis length were determined for each scan. Left/right and single beam/fan beam correlations were determined and differences were evaluated using a two-way analysis of variance. Femoral BMD at corresponding measurement regions in opposing femora were highly correlated (r=0.81–0.96). No significant left/right differences were detected. At the femoral neck, the mean BMD difference (± standard deviation) was 1.5%±4.7% in a single beam mode and-0.6%±6.3% in fan beam mode. Though mean values of femoral BMD were equivalent, the observed individual left/right differences were occasionally large (as high as 26% in the femoral neck). The hip axis length of the left and right hips were highly correlated and statistically equivalent. However, hip axis length using fan beam was significantly larger (7.5%) than the single beam measurement with a larger observed variation. We conclude that measurement of a single proximal femur will usually be sufficient for clinical evaluation of BMD and/or hip axis length. However, bilateral BMD measurements are indicated in subjects where unilateral degeneration or disease are suspected. If possible, hip axis length should be measured in single beam mode to avoid magnification errors.