Feasibility of Dual Energy X-Ray Absorptiometry Based Images for Measurement of Height, Sitting Height, and Leg...

Interpretation of pediatric bone mineral density by dual energy absorptiometry (DXA) requires adjustment for height (Ht). This is often not easily obtainable in nonambulant subjects.

To investigate the feasibility of using DXA images to evaluate measurements of Ht, sitting height (SH), and leg length (LL).

A total of 2 observers performed measurements of Ht, SH, and LL on 3 separate occasion using DXA digital images in 125 children. Intraclass correlation and relative technical error of measurement (rTEM) were performed to assess reliability of repeated measurements. In 25 children, Ht and SH were measured in clinic on the same day and Bland-Altman analysis was performed to compare DXA measured Ht, SH, LL with clinic measurements for these 25 children.

Intraclass correlation for DXA based Ht, SH, and LL measurements ranged from 0.996 to 0.998 (p < 0.0001). rTEM of Ht, SH, and LL for observer 1 was 0.0016%, 0.002%, and 0.0034%, respectively. rTEM of Ht, SH, and LL between observer 1 and 2 was 0.0047%, 0.0049%, and 0.0087%, respectively. Mean difference between clinic and DXA measurements from Bland-Altman plots were +0.57 cm (95% confidence interval [CI] -0.54 to +1.68) for Ht, +1.33cm (-1.60 to +4.24) for SH, and -0.76cm (-3.88 to +2.37) for LL.

Our study demonstrated for the first time that Ht, SH, and LL in children can be measured very precisely using DXA images. Ht can be measured accurately. We believe this may be a convenient method to obtain Ht measurements to allow size adjustment of DXA bone mineral density in immobile children with chronic conditions.