Elsevier

Bone

Volume 40, Issue 1, January 2007, Pages 169-174
Bone

Proximal femoral density and geometry measurements by quantitative computed tomography: Association with hip fracture

https://doi.org/10.1016/j.bone.2006.06.018Get rights and content

Abstract

Introduction

Bone mineral density and geometry measurements by volumetric quantitative computed tomography (vQCT) have been utilized in clinical research studies of aging, pharmacologic intervention and mechanical unloading, but there is relatively little information about the association of these measures with hip fracture. To address this issue, we have carried out a study comparing vQCT parameters in elderly Chinese women with hip fractures with measurements in age-matched controls.

Materials and methods

Forty-five women (mean age 74.71 ± 5.94) with hip fractures were compared to 66 age-matched control subjects (mean age 70.70 ± 4.66). vQCT was employed to characterize the volumetric bone mineral density in cortical, trabecular, and integral volumes of interest in the proximal femur. In addition to the volume of interest measurements, we computed the cross-sectional areas of the femoral neck and intertrochanteric planes, the femoral neck axis length, indices of femoral neck bending and compressive strength, and measures of femoral neck cortical geometry. To determine if cortical geometry measures were associated with hip fracture independently of trabecular vBMD, we carried out multi-variate analyses including these parameters in a logistic regression model.

Results and conclusions

All vQCT measurements discriminated between fractured subjects and age-matched controls. There was no significant difference in predictive strength between volumetric and areal representations of BMD and trabecular and integral vBMD showed comparable discriminatory power, although both of these measures were more correlated to fracture status than cortical vBMD. We found that fractured subjects had larger femoral neck cross-sectional areas, consistent with adaptation to lower BMD in these osteoporotic subjects. The larger neck cross-sectional areas resulted in bending strength indices in the fractured subjects that were comparable or larger than those of the control subjects. In multi-variate analyses, reduced femoral neck cortical thickness and buckling ratio indices were associated with fracture status independently of trabecular vBMD.

Introduction

Areal bone mineral density (aBMD) measured by dual X-ray absorptiometry is strongly associated with hip fracture both cross-sectionally [1] and prospectively [11], [23], but is limited in the information it provides on the pathophysiology of hip fracture because it is a hybrid measure combining bone density and size [10], which may be independently related to hip fracture risk. Volumetric quantitative computed tomography (vQCT), on the other hand, provides measures of trabecular and cortical volumetric BMD (vBMD) as well as measures of bone size such as tissue volume and cross-sectional area [5], [7], [17], [18]. vQCT has become an increasingly important clinical research tool in analyzing the effect of age [22], drug therapy [3], [4], mechanical unloading [16], and other phenomena on the density and geometry of bone. The density, geometry, and strength measures derived from vQCT images have been correlated to bone strength [6], [8], [14], [17], thereby providing a theoretical link to fracture risk. However, there is little direct information regarding their association with hip fracture, the key clinical endpoint in osteoporosis.

The paucity of data relating vQCT measures to hip fracture is due to the fact that hip vQCT measures have only recently been included as primary measurements in large population studies and because metallic artifacts and cohort effects complicate retrospective studies. An alternative, but logistically challenging approach is to image individuals immediately after their fracture and prior to surgery. In the only published case–control study of this type, Cody et al. compared DXA and hip vQCT measurements in elderly male and female African- and Caucasian-American subjects with femoral neck fractures to age-, gender-, and ethnicity-matched controls [7], [9]. They found that in men, cervical fractures were associated with lower bone density throughout the hip, but that in women they were associated with reduced femoral neck cancellous BMD, with reduced cortical BMD in the greater trochanter and with increased femoral head width.

In order to derive more comprehensive information regarding the association of vQCT bone density and hip geometry measures with hip fracture, we have carried out a case–control study comparing elderly Chinese women imaged within 48 h of their hip fracture to age-matched controls. We utilized a vQCT image analysis program that has been widely employed in a range of clinical research studies [3], [4], [16], [17]. In this study, we examined the associations of volumetric bone density measures, areal bone density measures, geometry measures and indices of strength, and cortical thickness with fractures of the proximal femur.

Section snippets

Study subjects

Forty-five women with atraumatic hip fractures (34 cervical, 11 trochanteric), aged 65 or older were recruited from the emergency room, Department of Traumatology and Orthopedic Surgery, Beijing Ji Shui Tan hospital. Atraumatic fracture was defined as resulting from a low-energy fall from standing or sitting height. In order to minimize changes in BMD due to the fracture, only those subjects whose fractures had occurred within the last 48 h were accepted into the study. Potential hip fracture

Descriptive statistics

Controls were not significantly different than subjects in weight, and height, but were on average 4 years younger than fracture subjects (p < 0.05) (Table 1).

BMD measurements

Results for BMD are summarized in Table 2. Subjects with fractures had 23–24% lower aBMD (p < 0.0001) than age-matched controls in the femoral neck, trochanteric, and total femur compartments as well as lower vBMD in all compartments and all regions (0.00001 < p < 0.001). The percentage difference between the fracture and controls was larger for

Discussion

Although the prospective association of areal BMD measures with hip fractures is well-known, and the cross-sectional association of volumetric trabecular and cortical BMD measures with hip fracture was reported by Cody et al. [7], our study was novel in its comprehensive analysis of differences in proximal femoral geometry in subjects with hip fractures and controls. We observed that the cross-sectional area of the femoral neck was larger in the subjects with hip fracture than in the controls.

Acknowledgments

This study was funded by a grant from Eli Lilly.

References (23)

  • J.J. Camp et al.

    A system for the analysis of whole-bone strength from helical CT images

    Proc.-SPIE

    (2004)
  • Cited by (108)

    • Finite element analysis informed variable selection for femoral fracture risk prediction

      2021, Journal of the Mechanical Behavior of Biomedical Materials
    View all citing articles on Scopus
    View full text