Ageing Effects on 3-Dimensional Femoral Neck Cross-Sectional Asymmetry: Implications for Age-Related Bone Fragility in Falling

This paper explores the effects of aging on femoral neck (FN) anatomy in a study of women aged 20-90years in relation to implications for FN fracture propensity in buckling. Five hundred and four participants were scanned by Quantitative Computed Tomography and analyzed using Quantitative Computed Tomography Pro BIT (Mindways). FN cross-section was split through geometric center into superior and inferior sectors. Bone mass, structural measurements, and bone mineral density were analyzed. Buckling ratio was calculated as ratio of buckling radius to cortical thickness. Between 2nd decade and 8th decade, age-related integral bone mass reduction in superior sector was substantially larger than in inferior sector (33% compared to 21%), especially in cortical bone superiorly compared to inferiorly (53% vs 21%; p < 0.001), principally due to reduction in cortical thickness, averaged cortical thickness (56%) with little difference in density. Superior and inferior sector trabecular bone mineral density reduction was similar at 41% and 43% respectively. Differential cortical bone loss in superior sector resulted in a 59% inferior displacement (δ) of center-of-mass from geometric center. Differences in δ and averaged cortical thickness with age accounted for a 151% increase in mean superior buckling ratio from 9 to 23. Analysis confirms significant progressive age-related superior cortical bone loss as the major age effect on FN structure with relative preservation of inferior cortex probably related to maintenance of inferior sector by regular loading as a result of standing and walking. Computation of buckling ratio may allow prediction of fracture propensity in a sideways fall.

Differences in femoral neck structure between elderly Caucasian and Chinese populations: a cross-sectional study of Perth-Beijing cohorts

Structural skeletal differences of the femoral neck of older Beijing-Chinese and Perth-Caucasian women were compared; adjusting for frame size-related differences, Beijing-Chinese have lower periosteal width; however, indices of internal bone distribution suggest that Beijing-Chinese may exhibit increased resistance to fracture that may relate to the reduced hip fracture incidence.

INTRODUCTION

Ethnic differences in skeletal structure may relate to differences in hip fracture risk in Chinese and Caucasian populations. 2D mass, size, and structural biomechanics were compared in the two populations.

METHODS

Quantitative computed tomography-derived geometric variables were compared in age-matched community-derived female populations, 196 Beijing-Chinese 76.5 ± 4.8 (mean ± SD) years and 237 Perth-Caucasians 77.1 ± 5.0 years. These included scanned area (A), periosteal width (W), bone mineral content (BMC), aBMD, bone cross-sectional area (bCSA), section modulus (Z) and buckling ratio (BR). Assumption-free measures included sigma (σ), related to the distribution of bone in the scanned image previously identified as a predictor of hip fracture, and delta (δ), the center-of-mass displacement from the geometric center.

RESULTS

Compared to Beijing-Chinese, Perth-Caucasians were heavier (Beijing-Chinese 58.7 ± 11.8; Perth-Caucasians 66.1 ± 11.0 kg), taller (154.9 ± 16.7 vs 158.9 ± 6.0 cm), and had higher BMC, A, and W. After adjustment for frame size, BMC was not significantly different but W remained higher in Perth-Caucasians. Differences in variables aBMD, Z, BR, and σ favored higher resistance to failure with Beijing-Chinese before and after adjustment for frame size. δ was similar in both populations; bCSA was higher in Beijing-Chinese before adjustment for frame size but not after.

CONCLUSIONS

Bone mass differences in two populations were related to frame size differences. However, femoral neck width remained smaller in Beijing-Chinese suggesting effects of local genetic and environmental factors. In Beijing-Chinese participants compared to Perth-Caucasians, internal bone distribution suggests increased resistance to deformation if exposed to same force that may, in-part, relate to reduced incidence of hip fracture in Beijing-Chinese.

Reconsideration of the Effects of Age on Proximal Femur Structure: Implications for Joint Replacement and Hip Fracture

OBJECTIVES

In recent years quantitative computed tomography (QCT) has allowed precise non-invasive, three dimensional, in vivo measurement of hip structure in large numbers of individuals. The effects of ageing on proximal femur structure are reported and implications for the prevention of hip prosthesis loosening and hip fracture considered.

DESIGN, SETTING AND PARTICIPANTS

An observational cross-sectional study of proximal femur QCT in 719 unselected female European descent aged 20 to 89 years recruited from US and Australian populations.

MAIN OUTCOMES MEASURES

QCT scans were obtained using software that separates cortical and cancellous bone by a thresholding technique. Voxel based mineral volume and mass was computed for the integral (external), cancellous and cortical compartments of 1 mm wide sections through the femoral neck (FN), trochanter (TR) and intertrochanter (IT) regions.

RESULTS

Over the adult life span total integral volumes at the FN, TR and IT sites expand linearly by between 18 and 37% at the same time as bone mass decreased by 22 to 25% resulting in massive reductions in true volumetric BMD (vBMD) of 40 to 50%. Cancellous volume expansion was larger at 65 to 79% at the three sites. Between the ages of 65 and 75 the average increase in cancellous volume at the IT site was 3.74 cm3 (12.1%). Voxel determined FN cortical volume decreased linearly by 43%, as did cortical bone mass so that vBMD did not change substantially. TR and IT cortical volumes decreased 54 and 28% respectively, small reductions in TR and IT cortical vBMD also occurred.

CONCLUSIONS

Large endosteal expansion in the area in which hip replacement stem placement occurs may contribute to loosening. Regarding the propensity to hip fracture, periosteal expansion contributes to increased resistance to bending but cortical thinning contributes to loss of bone to resistance to bending forces. Understanding individual hip structure may contribute to individualisation of risk and subsequent targeting of management using pharmaceutical agents.

Evaluation of a simplified hip structure analysis method for the prediction of incident hip fracture events

Many attempts have been made to improve the predictive ability of areal bone mineral density (aBMD) which integrates bone mass and area. The addition of an extra variable derived from the hip dual-energy X-ray (DXA) image TR_σ, which describes distribution of mass within the scanned area of the trochanter, improved prediction of 15-year hip fracture probability in elderly women.

INTRODUCTION

Two-dimensional DXA imaging of the proximal femur to produce an aBMD is a clinically useful predictor of future fracture risk. Further analysis of the DXA image to produce an eight-variable hip structure analysis (Beck HSA) has been developed to improve understanding of structural factors determining hip bone strength at each of three proximal femur sites, the narrow femoral neck (NN), intertrochanter (TR) and shaft (S). Recently, data on four measurements derived from the currently used eight Beck HSA variables were used to capture population variation in bone structure at each site. These include two previously used variables, the localised aBMD and the sub-periosteal width (W) applying to 5-mm sections (at each sites), and two new variables, standard deviation of normalised mineral-mass projection profile distribution (σ), and displacement between centre-of-mineral mass and geometric centre-of-mineral mass of projection profile (δ).

METHODS

Using a cohort of 1159 women, mean baseline age 75, who sustained 139 hip fractures over 15 years, we determined whether these measures significantly improved 15-year hip fracture prediction compared to current approach utilising age and total hip aBMD. To describe the most parsimonious model for hip fracture risk prediction, the 12 base measures (4 from each site), total hip aBMD and age were evaluated in stepwise logistic regression models.

RESULTS

The final model included TR_σ, total hip aBMD and age and provided improved utility for hip fracture prediction compared to total hip aBMD and age alone (C-statistic 0.73 vs. 0.69, P = 0.009 and net reclassification improvement 0.164, P < 0.001, respectively).

CONCLUSIONS

Addition of TR_σ to total hip aBMD and age substantially improved prediction of 15-year hip fracture risk in this cohort of elderly women.

Differences in structural geometrical outcomes at the neck of the proximal femur using two-dimensional DXA-derived projection (APEX) and three-dimensional QCT-derived (BIT QCT) techniques

Structural geometric parameters at neck of the proximal femur were obtained using DXA-derived hip structural analysis (APEX 3) and quantitative computed tomography-derived (BIT QCT) techniques in 237 elderly females. Linear correlations for parameters ranged from 0.45 to 0.90. The average value of the subperiosteal width, as determined by the two techniques, was the same; variables dependent on mass measurements were different.

INTRODUCTION

There has been increasing interest in using bone structural geometry to assess bone fragility to complement bone mineral mass. The objective of this study is to compare structural geometrical differences between "2D" DXA-derived and "3D" QCT-derived techniques in unselected clinical cases.

METHODS

All 237 females had both DXA and QCT assessments of femoral neck structural geometry. Variables compared were areal bone mineral density, cross-sectional area (CSA), cross-sectional moment of inertia (CSMI), section modulus (Z), averaged cortical thickness (Ct), endosteal width (ESW), subperiosteal width (W), and buckling ratio (BR).

RESULTS

Correlation of femoral neck variables ranged from 0.45 for ESW to 0.90 for CSA. APEX 3 and BIT QCT-derived femoral neck W values were numerically similar. However CSA, CSMI, Z and Ct values measured by APEX 3 were higher and ESW and BR values were lower than corresponding BIT QCT.

CONCLUSIONS

2D DXA structural analysis of neck of femur is related to but different from same parameters calculated from true 3D images obtained by CT. Femoral neck size values are similar for DXA and QCT, but structural geometrical variables dependent on mass calibration standards, location of neck ROI and mathematical derivation techniques are different.

Proximal femur structural geometry changes during and following lactation

Human lactation is associated with transient decreases in bone mineral density (BMD). Bone strength is related to both mass and structural geometry. This study investigated longitudinal changes of hip bone strength during lactation using hip structural analysis (HSA), which determines hip structural geometry (including areal BMD, BMDa) from dual-energy X-ray absorptiometry scans (DXA). Forty-eight lactating women were studied longitudinally at the proximal femur using DXA at approximately 2 weeks postpartum, peak-lactation and post-lactation. Nonpregnant, nonlactating women (NPNL, n=23) were studied concurrently at baseline and after 1 year. Hip scans were analysed using HSA at the narrow neck, intertrochanter and proximal shaft. No significant change (>0.05) was observed in NPNL women for any measurement. In contrast, for lactating women BMDa decreased significantly from 2 weeks postpartum to peak-lactation at narrow neck (-2.8%), intertrochanter (-3.2%) and shaft (-1.4%). Cross-sectional area (CSA) decreased at narrow neck (-3.4%) and intertrochanter (-2.7%). There were no significant changes in bone width. Section modulus decreased at intertrochanter (-2.1%). At shaft, cortical thickness decreased (-1.7%) and buckling ratio increased (2.3%). By post-lactation, measurements were not significantly different from 2 weeks postpartum except for decrements in BMDa (-1.1%) and CSA (-1.2%) at the shaft. During the study, lactating women lost 5% of their body weight. Adjusting for weight changes decreased the magnitude and significance of HSA changes at peak-lactation and by post-lactation there were no significant differences from 2 weeks postpartum. Calcium intake was not a significant predictor of changes in HSA variables. In conclusion, lactation is associated with significant but transient changes in hip BMD and structural geometry. Changes in body weight but not calcium intake were associated with these changes. These small changes at the hip during lactation occurred mainly at internal surfaces and had minimal impact on bending or compressive strength.

Comparison of QCT-derived and DXA-derived areal bone mineral density and T scores

Two-dimensional areal bone mineral density (aBMD) of the proximal femur measured by three-dimensional quantitative computed tomography (QCT) in 91 elderly women was compared to dual-energy X-ray absorptiometry (DXA) aBMD results measured in the same patients. The measurements were highly correlated, though QCT aBMD values were marginally lower in absolute units. Transformation of the QCT aBMD values to T score values using National Health and Nutrition Examination Survey (NHANES) DXA-derived reference data improved agreement and clinical utility.

INTRODUCTION

World Health Organization guidelines promulgate aBMD (g cm(-2)) measurement of the proximal femur for the diagnosis of bone fragility. In recent years, there has been increasing interest in QCT to facilitate understanding of three-dimensional bone structure and strength.

OBJECTIVE

To assist in comparison of QCT-derived data with DXA aBMD results, a technique for deriving aBMD from QCT measurements has been developed.

METHODS

To test the validity of the QCT method, 91 elderly females were scanned on both DXA and CT scanners. QCT-derived DXA equivalent aBMD (QCT(DXA) aBMD) was calculated using CTXA Hip software (Mindways Software Inc., Austin, TX, USA) and compared to DXA-derived aBMD results.

RESULTS

Test retest analysis indicated lower root mean square (RMS) errors for CTXA; F test between CTXA and DXA was significantly different at femoral neck (FN) and trochanter (TR) (p < 0.05). QCT underestimates DXA values by 0.02 +/- 0.05 g cm(-2) (total hip, TH), 0.01 +/- 0.04 g cm(-2) (FN), 0.03 +/- 0.07 g cm(-2) (inter-trochanter, IT), and 0.02 +/- 0.05 g cm(-2) (TR). The RMS errors (standard error of estimate) between QCT and DXA T scores for TH, FN, IT, and TR were 0.36, 0.40, 0.39, and 0.49, respectively.

CONCLUSIONS

This study shows that results from QCT aBMD appropriately adjusted can be evaluated against NHANES reference data to diagnose osteoporosis.