Adaptation of bone to altered loading environment: a biomechanical approach using X-ray absorptiometric data from the patella of a young woman.

Loading induced changes in the bone mineral apparent density (BMAD) and average strain magnitude (strain index) of the patella were estimated using a novel analytic method of dual energy X-ray absorptiometric (DXA) and isometric strength data obtained from repeated measurements of a 26-year-old woman. The strain index was used as a major determinant of bone adaptation to physical loading. The subject's lower limb skeleton was measured 12 times during a 3-year period including a 1-year unilateral strength training intervention, an accidental knee ligament rupture, and a 2-year rehabilitation period. Instead of standard DXA analysis, the patella, a bone practically affected by quadriceps activity only, was analyzed in terms of estimated loading induced stresses and strains. The mechanical stress is directly proportional to the effective force and inversely proportional to the area over which the given force is applied. The strain, in turn, is proportional to the given stress divided by the bone stiffness, which was assumed to be proportional to BMAD. The effective force, the contact area, and the BMAD were estimated from the muscle strength and DXA data, respectively. In addition, post-traumatic bone loss and subsequent recovery with time were assessed using an exponential model. The loading during the 1-year training period increased the average patellar strain index by 47% and did not affect the patellar BMAD. Immediately after the injury, the apparent inability to do muscle work drastically reduced the strain index and likely initiated the patellar bone loss, a loss which continued for about 4 months. During rehabilitation, the imbalance between the patellar stiffness and increased functional stress (muscular performance) became sufficiently large to level off the bone loss and stimulate bone gain. At that time, the strain index indicated that the patellar deformation more than doubled (135%) as compared to baseline level. Accordingly, the BMAD increased until an apparent balance between BMAD and the muscular strength was achieved by the 3-year end point. The exponential models explained well both the post-traumatic bone loss (R2 = 0.89) and bone gain during recovery (R2 = 0.98). In conclusion, the present observations support the concept of the nonlinear nature of the skeletal response to mechanical loading and suggest a potential utility of muscle strength measurements and DXA information for improved understanding of the effects of training, immobilization and remobilization on bone tissue.