Journal Facts

Journal
BoneKEy Knowledge Environment
ISSN
1940-8692
Volume
12
Year
2015

Article Facts

Title
Predicting hip fractures using quantitative CT data
DOI
10.1038/bonekey.2015.6
Author
Online Date
01/21/2015

Article Links

IBMS BoneKEy | BoneKEy Watch

Predicting hip fractures using quantitative CT data


DOI:10.1038/bonekey.2015.6

Bredbenner et al. investigated the potential of quantitative computed tomography (QCT) data from the femur in predicting hip fracture risk.

Baseline QCT data were obtained for 513 participants of the Osteoporotic Fractures in Men (MrOS) study. Over the course of follow-up for a mean of 6.9 years, 45 of the men sustained a hip fracture. Analysis of 40 fracture cases and 410 controls revealed that the structural differences in the femoral bone between the two groups were extremely complex.

Statistical shape and density modeling on the QCT data created a prediction model. When adjusted for age, this SSDM-based model accurately predicted hip fracture in 55% of the participants. Standard clinical methods using areal bone mineral density data, by comparison, correctly predicted only 10% of fractures. The authors suggest that QCT data used in such a model could use structural differences at the proximal femur to more accurately assess fracture risk.

Editor’s comment: The authors used statistical shape and appearance models to capture the principal variation modes of the human proximal femur and to predict fracture risk of a cohort from the MrOS study. Beyond this elegant and powerful analysis of femoral anatomy, the improvement in classification of the cases is moderate and suggests that we should abandon this obsession of using a few random fractures to evaluate the improvement of surrogates of bone strength.

Creative Commons License This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.