Femoral neck bone density: Direct measurement and histomorphometric validation

Abstract

The purpose of this study was to develop a method for directly measuring bone density of femoral neck sections. Three types of density were measured. Real density equals wet weight divided by the actual volume of bone tissue (real volume). Apparent density equals wet weight divided by the total volume occupied by the bone plus the pore spaces (total sample volume). Ash density equals the ash weight divided by the real volume. Corticocancellous cross-sections of the femoral neck were analyzed for density at two levels: level 1, proximal neck and level 3, distal neck. Density measurements were compared with histomorphometric measurements performed on cross-sections at the midportion of the femoral neck (level 2) and with a clinical radiographic measure of bone density (cortical index 3 cm below the lesser trochanter). No correlation was found between apparent density and either real ( r = .12, P = .62) or ash density ( r = −.09, P = .72) within a given femoral neck section. There was, however, a strong correlation between real and ash density ( r = .93, P = .0001). This was expected because real and ash densities are both reflections of bone mineralization. Apparent density showed better correlation, when comparing level 1 with level 3 sections ( r = .76, P = .0001), than did ash ( r = .57, P = .01) or real density ( r = .55, P = .01). There was no correlation between either real or ash density with any histomorphometric parameter. Apparent density was moderately correlated with total bone area expressed as a percentage of cross-sectional area ( r = .66, P = .008). This finding tends to validate the direct measurement of apparent density in that both apparent density and total bone area are measurements of the concentration of bone in space. No significant correlation was found between any of the density measurements and the cortical index at 3 cm. This underscores the necessity for precisely qualifying any definition or discussion of bone quality. The success or failure of hip implants may be at least partially determined by the ability of the bone to withstand the insult of implantation of the prosthesis and to adapt successfully to the new mechanical environment. This study represents an early phase of defining parameters that may have prognostic value in long-term implant fixation