This study used a multislice peripheral quantitative computed tomography (pQCT) to measure volumetric BMD (vBMD) and cortical thickness for investigating regional adaptation in lower tibial shaft in 72 healthy postmenopausal women aged 47-60. Tomographic slices were analysed on four distinct cortical regions: the anterior, posterior, medial and lateral cortical wall. One-way analysis of variance (ANOVA) test was used to compare the vBMD in the four regions. The results showed that the posterior cortex had the highest vBMD (1923 +/- 135.3 mg/cm(3)), significantly (P < 0.001) higher than the anterior cortex (1805 +/- 110.6 mg/cm(3)), medial cortex (1863 +/- 103.6 mg/cm(3)) and lateral cortex (1815 +/- 111.6 mg/cm(3)); whereas there was no significant difference (P > 0.05) between the medial and lateral cortices located near the neutral plane of bending. The anterior cortex had the greatest thickness (2.56 +/- 0.47 mm), significantly (P < 0.001) greater than that of the posterior cortex (2.11 +/- 0.27 mm), medial cortex (2.20 +/- 0.39 mm) and lateral cortex (2.03 +/- 0.29 mm). The vBMD of the posterior cortex was a significant 6.5% higher than that of the anterior cortex (P < 0.001); whereas the anterior cortical thickness was a significant 21.3% greater than that of the posterior cortex (P < 0.001). There was no linear relationship found between cortical vBMD and cortical thickness measured at the four cortical regions (r = 0.086, P > 0.05). In conclusion, the regional differences, with higher vBMD found in posterior cortex, might be a result of mechanical adaptation, which caused the posterior cortex to sustain higher compressive loading than the anterior tensile cortex during the landing phase in the gait cycles of individuals. Nevertheless, regional geometric adaptation in anterior cortical thickness might be adapted to accommodate for the reduced vBMD and to reduce the bending stress in this region.