Maintaining femoral neck cortical thickness may help prevent hip fracture. Fracture initiation probably starts superiorly at flaws, ie where the cortex is thinnest. Whole body computed tomography (QCT) is now being used to study cortical thickness but limited resolution (> 300 micrometers) makes in vivo estimates imprecise, whereas microscopy s resolution approaches 1 micrometer. We have therefore extended our microscopic studies on femoral neck biopsies to include men (14 cases, 26 controls) and women (50 cases, 23 controls), and here provide data on true cortical thickness in subjects with and without hip fracture.

Whole femoral neck cross-sections obtained at hemiarthroplasty (or at post-mortem in controls) were embedded in methacrylate, cut, stained and imaged at medium power. Image-J was used to define cortical boundaries and to measure cortical thicknesses at 5 degree intervals of arc from the cross-sections centre of area.

We confirmed that the mid-femoral neck (or narrow neck) site, defined as where the ratio of maximum to minimum neck diameter (max:min) is 1.4, shows great asymmetry, with the thick inferior cortical octant averaging over 3mm thickness (mean age 79 years inter-quartile range 74-85). In the superior 3 octants cortical thickness averaged 26% of that seen inferiorly. To assess statistical determinants of cortical thickness, the data were modelled with linear regression in octants after adjusting for subjects age, sex, max:min, and hip fracture status. To achieve normality of residuals the cortical thickness data were log-transformed. 95% of measured cortical thicknesses fell between 45% and 220% of the mean for octant. In the thinner, superior three octants, minimum thicknesses were just under 0.3 mm in the fracture cases ie close to 35% of the subjects mean for octant. Cases had about 17% thinner cortical thicknesses in all octants than controls, while female controls had cortical thicknesses that uniformly averaged 90% of male. In conclusion, compared to gender and age-matched controls, intra-capsular hip facture cases had generalized cortical thinning in all mid-neck octants. This disease effect contrasts markedly with the effect of normal ageing, which thins preferentially the mechanically under-loaded superior cortex and spares the infero-anterior cortex.

Areal BMD (aBMD) is relatively poor at discriminating those patients at risk of hip fracture. This study tested the hypothesis that a measure of bending resistance, cross section moment of inertia (CSMI) and section modulus, derived from 3D peripheral quantitative computed tomography (pQCT) images made ex-vivo, would discriminate cases of hip fracture from controls better than areal bone mineral density.

The biopsies were from (n = 20, F) subjects that had suffered an intracapsular hip fracture. The control material (n = 23, F) was from post-mortem subjects. Serial pQCT 1mm thick cross-sectional images using the Densiscan 1000 pQCT clinical forearm densitometer were obtained, and matched for location along the neck. The image voxels were converted to units of bone mass, which were then used to derive the mass weighted CSMI (MWCSMI), section modulus and areal bone mineral density, (see Table).

The aBMD results showed that the difference between the means of the fracture cases compared to the controls was 9.9% (−0.061g/cm²; +0.0055g/cm², −0.127g/cm²; 95% confidence interval). However, the MWCSMI was 29.5% (−5966mm⁴; −8868mm⁴,−3066mm⁴; 95% confidence interval) lower in the fracture cases compared to the controls, while section modulus was 32.5% (−242mm³; −133mm³, −352mm³ 95% confidence interval) lower. When presented as Z scores the fracture cases had considerably lower section modulus Z scores (mean −1.27 SD, p=0.0001) than aBMD – Z scores (mean −0.5 SD, p=0.07). To simulate the forces experienced during a sideways fall, the model’s neutral axis was rotated by 210°. The results were similar for section modulus to those at 0°.

This study suggests that biomechanical analysis of the distribution of bone within the femoral neck may offer a marked improvement in the ability to discriminate patients with an increased risk of intracapsular fracture. Progress towards implementing this form of analysis in clinical densitometry should improve its diagnostic value.