gms | German Medical Science

Objective: Bone continually responds and adapts to the mechanical loading it experiences. In vitro studies have shown that bone cells at the endosteum and periosteum respond differently to biological and mechanical stimulation. It is unknown how these two regions respond to mechanical stimuli in vivo. We investigated the hypotheses that under tibial compression, formation occurs in regions under higher strain than resorption, similarly in both regions.

Methods: Five female C57Bl/6J mice (10 week old) underwent two weeks of in vivo loading of the left tibia while the right tibia served as an internal control. In vivo µCT (10.5 µm resolution) was performed at day 0 and 15. The region of interest was centered at the tibia midshaft extending 5% of the bone length. To monitor bone remodeling, changes in bone morphology between two consecutive µCT images were evaluated. Using an image registration algorithm, each voxel was classified into formed, resorbed or constant. To determine the local mechanical strains, µCT-based finite element model of the entire tibia was developed based. Bone material properties accounted for regional differences in the bone mineral density. A novel method based on the nearest neighborhood algorithm was developed to relate the remodeling event at each voxel for the endosteal and the periosteal regions to the calculated strains.

Results and conclusion: Formation sites occurred in regions with a mean strain of 3570±323 µstrain, which was significantly higher than constant (2900±88 µstrain) and resorption sites (1870±460 µstrain)(p<0.05, Student's t-test). Mean strains were higher in the periosteal region (3540±2250 µstrain) compared to the endosteal (2880±1760 µstrain). Periosteal formation occurred in regions under 3900±393 µstrain, which was 28±4% higher than strains at endosteal formation sites (p<0.05). Constant sites at the periosteum exhibited 68±40% higher strains as compared to the endosteal constant sites. The strain values at resorbed sites in the two regions were not significantly different.

Earlier studies have reported that age-related bone loss occurs by an acceleration of endosteal resorption and deceleration of periosteal apposition. In this study we investigated the response of these regions to controlled mechanical stimulation. Under compression, we observed specific (re)modeling responses according to the strain level, with formation occurring at strain higher levels than resorption. Bone formation occurred in the endosteal regions at lower strain levels than in the periosteal region suggesting that bone from this region may be more mechanoresponsive. Future studies will investigate the influence of age on the remodeling response seen in these two regions.

Figure 1 [Fig. 1]