Bone densitometry in pediatric populations: discrepancies in the diagnosis of osteoporosis by DXA and CT

J Pediatr. 2005 Jun;146(6):776-9. doi: 10.1016/j.jpeds.2005.01.028.


Objectives: To test the hypothesis that because of errors associated with growth and development, osteoporosis is frequently overdiagnosed in children when using dual-energy x-ray absorptiometry (DXA). This study compared bone density values obtained by DXA with those from computed tomography (CT), which is not influenced by body or skeletal size.

Study design: Vertebral bone density was measured by using both DXA and CT in 400 children (100 each, healthy and sick boys and girls). Regression analysis was used to compare DXA and CT Z scores, and the agreement between DXA and CT classifications of Z scores below -2.0 was examined.

Results: DXA and CT Z scores were moderately related (r2 = 0.55 after accounting for age and anthropometric measures). DXA Z scores predicted CT Z scores below -2.0 with reasonable sensitivity (72%), specificity (85%), and negative predictive value (98%), but positive predictive value was low (24%). Many more subjects were classified as having bone density lower by DXA (76/400) than by CT (25/400), particularly subjects below the 5 th percentile of height and/or weight for age.

Conclusions: The inability of DXA to account for the large variability in skeletal size and body composition in growing children greatly diminishes the accuracy of this projection technique for assessing bone acquisition and diagnosing osteoporosis in pediatric populations.

Publication types

  • Comparative Study
  • Research Support, N.I.H., Extramural
  • Research Support, U.S. Gov't, Non-P.H.S.
  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Absorptiometry, Photon*
  • Adolescent
  • Anthropometry
  • Bone Density / physiology*
  • Child
  • Female
  • Humans
  • Lumbar Vertebrae
  • Male
  • Osteoporosis / diagnosis*
  • Predictive Value of Tests
  • Regression Analysis
  • Retrospective Studies
  • Sensitivity and Specificity
  • Tomography, X-Ray Computed*