Femoral neck cortical bone in female and male hip fracture cases: Differential contrasts in cortical width and sub-periosteal porosity in 112 cases and controls.

OBJECTIVES: To quantitate differences between cases of hip fracture and controls in cortical width around the mid-femoral neck in men and women.

METHODS: Over 5 years, 64 (14 male) participants over age 55 (mean 79) years, who had never taken bone-active drugs and suffered intra-capsular hip fracture treated by arthroplasty, donated their routinely discarded distal intra-capsular femoral neck bone for histomorphometry. After embedding, complete femoral neck cross sections from the cut surface near the narrowest part of the neck were stained with von Kossa and cortical width was measured radially every 5 degrees of arc. Control material (n = 48, 25 male) was available through consented post mortems prior to the year 2000. Cortical widths were averaged for circumferential octants, each representing 45 degrees of arc. Divergence of individual cortical widths from their means was also examined.

RESULTS: Because sections were required to have a complete cortex, sampling was biased towards cases with sub-capital versus trans-cervical fractures. Compared to sex- and age matched controls, male cases showed larger relative differences in cortical widths than female cases. Unexpectedly, cortical widths in female but not male cases also showed marked over-representation of extremely narrow (<0.1 mm) cortical widths, located mainly posteriorly. The numbers of these very narrow cortical widths observed per subject retrospectively predicted female fracture status in logistic regression independently of mean cortical width values. Together with mean cortical width differences, the numbers of measured cortical widths <0.1 mm (out of 72 measured) raised the sensitivity of predicting fracture status in women from 48 to 80% at 80% specificity. In almost all cases, very narrow cortical widths were identified in regions enclosing a cortical pore roofed on its endosteal surface by thin structural bone defined a priori as trabecular.

CONCLUSIONS: Cortical widths <0.1 mm probably reflect zones where endosteal cortex has been trabecularised through expansion of an un-refilled sub-endosteal canal close to the periosteum. Persistent cortical defects occurring near the periosteal surface, where mechanical loading exerts its greatest stresses, are likely to result in extremes of localized concentrations of stress during a fall, unknown in young normal fallers. Such defects have the potential to help explain the excess of hip fractures among elderly women. Prevention of sub-periosteal tunnelling by osteoclasts might explain in part the additional benefits, beyond an increase in bone density, of treatments that reduce excessive bone resorption or else stimulate new bone formation on previously resorbed surfaces.