Structural analysis of the human tibia by tomographic (pQCT) serial scans

Abstract

This study analyses the evaluation of tomographic indicators of tibia structure, assuming that the usual loading pattern shifts from uniaxial compression close to the heel to a combined compression, torsion and bending scheme towards the knee. To this end, pQCT scans were obtained at 5% intervals of the tibia length (S5-S95 sites from heel to knee) in healthy men and women (10/10) aged 20-40 years. Indicators of bone mass [cortical area, cortical/total bone mineral content (BMC)], diaphyseal design (peri/endosteal perimeters, cortical thickness, circularity, bending/torsion moments of inertia - CSMIs), and material quality [(cortical vBMD (bone mineral density)] were determined. The longitudinal patterns of variation of these measures were similar between genders, but male values were always higher except for cortical vBMD. Expression of BMC data as percentages of the minimal values obtained along the bone eliminated those differences. The correlative variations in cortical area, BMC and thickness, periosteal perimeter and CSMIs along the bone showed that cortical bone mass was predominantly associated with cortical thickness toward the mid-diaphysis, and with bone diameter and CSMIs moving more proximally. Positive relationships between CSMIs (y) and total BMC (x) showed men's values shifting to the upper-right region of the graph and women's values shifting to the lower-left region. Total BMC decayed about 33% from S5 to S15 (where minimum total BMC and CSMI values and variances and maximum circularity were observed) and increased until S45, reaching the original S5 value at S40. The observed gender-related differences reflected the natural allometric relationships. However, the data also suggested that men distribute their available cortical mass more efficiently than women. The minimum amount and variance of mass indicators and CSMIs, and the largest circularity observed at S15 reflected the assumed adaptation to compression pattern at that level. The increase in CSMIs (successively for torsion, A-P bending, and lateral bending), the decrease in circularity values and the changes in cortical thickness and periosteal perimeter toward the knee described the progressive adaptation to increasing torsion and bending stresses. In agreement with the biomechanical background, the described relationships: (i) identify the sites at which some changes in tibial stresses and diaphyseal structure take place, possibly associated with fracture incidence; (ii) allow prediction of mass indicators at any site from single determinations; (iii) establish the proportionality between the total bone mass at regions with highly predominant trabecular and cortical bone of the same individual, suitable for a specific evaluation of changes in trabecular mass; and (iv) evaluate the ability of bone tissue to self-distribute the available cortical bone according to specific stress patterns, avoiding many anthropometric and gender-derived influences.

Fig. 1

Serial pQCT scans obtained from one of the studied (male) legs, numbered S5 to S95 from the distal to the proximal end, as described in the text.

Fig. 2

(A) Means ± SD of crude values of total mineral content (total BMC) of the whole set of tibial scans (numbered 5–95 as described in the text) of the male and female individuals studied. Dashed lines indicate the ToC reduction of about 33% at approximately S15 with respect to S5 and its recovery at approximately S40 as commented in the text. (B) Means ± SD of the total BMC values determined as in (A), expressed as percentages of the minimum value obtained [i.e. that corresponding to the S15 site in (A)]. No statistical difference was found between the determinations in men and women. Accordingly, a single, polynomial equation (shown in the graph) was calculated for the description of the theoretical evolution of the data for both genders. The adjusting curve is not shown. The maximum reduction in total BMC observed at S15 with respect to S5 and its recovery at approximately S40 are indicated as in (A). (C) Means ± SD’s of circularity values for the cross-sections calculated at each studied site. As male and female values did not differ statistically within sites throughout the bone, a single adjusting curve is shown for both genders.

Fig. 3

Comparative changes of means ± SD of cortical mineral content (A), cortical thickness (B) and periosteal perimeter (C) of the whole set of tibial scans (numbered 5–95 as described in the text) of the male and female individuals studied.

Fig. 4

Comparative changes of means ± SD of the moments of inertia for lateral bending CSMI (A), A–P bending CSMI (B) and torsion CSMI (C) of the tibia scans obtained between the S10 and 90 sites of the male and female individuals studied. Sketches on the right indicate the position of the reference axes considered for calculation of the corresponding CSMIs. Arrows indicate the sites at which the CSMI values begin to grow significantly in the proximal direction with respect to the virtually constant proximal values in each graph.

Fig. 5

(A) Correlation between the mean values of torsion CSMI and total BMC measured at the S65 site in males and females, closely adjusted by a single exponential equation. (B) The same data as in (A) re-analyzed assuming the existence of independent exponential relationships for each gender. An ancova test of the linearized relationships showed that the corresponding curves, both significantly positive, did not differ in slope (F = 0.14, P = 0.717) but had significantly different intercepts (F = 4.56, P < 0.05).