Comparison of calcaneus ultrasound and dual X-ray absorptiometry in children at risk of osteopenia.

The optimal method to assess pediatric bone mass remains controversial. Dual X-ray absorptiometry (DXA) is used most commonly for clinical assessments in children, but calcaneus ultrasound (CUS) is less costly, is free of ionizing radiation, and predicts fracture as well as DXA in adults. This study was designed to compare CUS and DXA in 42 young patients (ages 9-21) with chronic disease and/or fragility fractures. Zscores for broadband ultrasound attenuation (BUA) and speed of sound (SOS) determined using the Lunar Achilles Plus ultrasonometer were compared with Z-scores for areal bone mineral density (BMD) and volumetric BMD using DXA (Hologic). Logistic regression was employed to predict low bone density measured by DXA (defined as spinal BMD Z-score < -2) from CUS measurements. Sensitivity/specificity analysis was performed to compare CUS and spinal DXA Z-scores as predictors of previous low-impact fracture. Correlations between CUS and DXA Z-scores were in the range of r = 0.3-0.6. Areas under the receiver operator characteristic (ROC) curves for BUA and SOS predicting low bone density by DXA were similar: 0.81 and 0.82, respectively. ROC curve areas for spinal DXA, BUA, and SOS predicting previous fracture were also similar: 0.85, 0.84, and 0.84, respectively. While CUS correlates only modestly with DXA, ROC curve areas indicate that CUS detects low bone mineral in children with fragility fractures as well as DXA and may be a viable initial screen for osteopenia.

Ethnic differences in bone mass of young women vary with method of assessment.

To examine ethnic differences in bone mass measured by calcaneus ultrasound (CUS) and dual energy X-ray absorptiometry (DXA), and to compare the two methodologies, CUS was performed in 904 healthy Asian, African American, Latina, and Caucasian women 20-26 yr old using the Lunar Achilles Plus ultrasonometer. CUS measurements (broadband ultrasound attenuation [BUA] and speed of sound [SOS]) were made following standard methodology (standard CUS) and repeated adjusting for foot size using shims (with-shim CUS). Areal bone mineral density (BMD) and estimated volumetric bone density (BMAD) at the spine, femoral neck, and whole body were determined using the Lunar DPX-IQ. African Americans had greater height- and weight-adjusted BUA than Caucasians, while Asians and African Americans had greater SOS than Caucasians and Latinas. Additionally, African Americans had greater height- and weight-adjusted BMD and BMAD than all other groups. CUS and DXA measurements correlated moderately (r = 0.2-0.5). With-shim CUS values were 0.9-7.8% lower than standard CUS values. In conclusion, African American women had greater DXA measurements than all others and greater CUS measurements than Caucasians. In contrast to DXA, CUS measurements in Asians and Latinas were not significantly lower than those in African Americans. Most notably, Asians had greater values for SOS than Caucasians and Latinas. Discrepancies in ethnic comparisons and modest correlations suggest that CUS and DXA methods may capture different bone qualities.