Although bone mass is a contributory risk factor for intracapsular hip frac
ture, its distribution and porosity within the femoral neck is also, import
ant for bone strength. III femoral neck biopsies from 13 women with intraca
psular hip fracture (mean +/-SEM: 75.1 +/- 2.1 years, OP) and 19 cadaveric
samples (9 men and 10 women [control] aged 79.4 +/- 1.7 years), a segmental
analysis was used to quantify circumferential variations in the characteri
stics of cortical bone haversian systems. In female control femoral necks,
there was an increasing porosity gradient between the inferior (I) (7.7 +/-
0.6%) and superior regions (S) (16.05 +/- 1.8%, p < 0.005). In walking, th
ese regions undergo compression and tension, respectively. In mean, a simil
ar trend was observed, but the differences were not significant (I: 11.1 +/
- 1.2%; S: 14.1 +/- 1.7%; p = 0.133). This porosity gradient was not mainta
ined in the fracture group (I: 10 +/- 1%; S: 12.65 +/- 1.2%). In contrast,
porosity in the fracture group was greatest in the anterior cortex, being 4
1% higher in that quadrant than in controls (p = 0.06). The areal density o
f haversian canals ranged from 16.7 to 21.3 canals/mm(2) with no significan
t differences between fractures and controls. In the control women, mean ca
nal diameter was highest in the superior region (60 +/- 2.8 mu m), and thes
e canals were significantly larger than those in the inferior region (49.4
+/- 1.4 mu m, p < 0.05). This difference was less marked in the fracture ca
ses (I: 53.21 +/- 2.5 mu m; S: 59.1 +/- 2.8 mu m; p = 0.0878). Although the
mean canal diameter in the anterior quadrant of the fracture cases was hig
her. than in the control women this did not reach significance (OP/F: 59.5
+/- 3 mu m; control/F: 52.7 +/- 2.6 mu m; p = 0.106). However, the proporti
on of "giant" canals with diameters >385 mu m (defined as the top 0.5% in t
he controls) was doubled in the anterior region of the fracture eases (OP/F
: 1.28%; control/F: 0.69%; p < 0.005). Porosity is related to the square of
the canal radius; therefore, such canals make a substantial contribution t
o cortical porosity. Previous work has shown that the elastic modulus of bo
ne decreases approximately as the square root of porosity. Therefore, the i
ncreased porosity and the higher prevalence of "giant" canals have a marked
ly negative influence on the ability of the cortical shell to withstand str
esses associated with a fall, The mechanisms responsible for the localized
generation of "giant" haversian canals, and ultimately the "trabecularizati
on" of the cortex require further investigation. (C) 1999 by Elsevier Scien
ce Inc. All rights reserved.