Texture analysis of X-ray radiographs is a more reliable descriptor of bone loss than mineral content in a rat model of localized disuse induced by the Clostridium botulinum toxin.

Botulism is a generalized paralyzing disease caused by the toxin of Clostridium botulinum (BTX). The toxin acts 3-4 days after injection by blocking the release of acetylcholine to the muscle. Six Wistar rats received a 2-U injection of BTX in the right quadriceps. Six rats were similarly injected with saline and were used as control. Paralysis of the quadriceps was obtained 4-5 days after the injection. Animals were killed 4 weeks after the BTX injection. The bone mineral content (BMC) was measured by dual-energy X-ray absorptiometry (DXA) on the femur and tibia. No side-to-side difference was observed for BMC on the whole tibia and femur in the BTX group. When subregions were selected in the bones, a significant decrease in BMC was obtained on the proximal tibia (-17.4 +/- 2.5%, p < 0.02). No significant difference could be observed on the proximal or distal femur, nor on the diaphyseal shafts. Numeric X-rays were done and a region of interest was transferred to an image analyzer. The texture of the trabecular bone was analyzed by the run length and fractal methods (skyscrapers and blanket). Significant differences were obtained on the proximal tibia for all methods except with fractal skyscrapers. On the distal femur, significant differences were obtained with the run length method, and the skyscrapers and the blanket method in the vertical direction. No differences were obtained with any method on the tibia and femur from control animals. Bone is a highly anisotropic material and its architecture at the microscopic level is conditioned by strains. The trabecular pattern differs in the proximal tibia than in the distal femural. Depending on the trabecular anisotropy, the algorithms can be more or less pertinent. BTX induced a significant bone loss on the bony subparts that are directly influenced by disuse. Texture analysis of X-ray images can reveal differences that were not evidenced by naked eyes. However, a combination of several methods appears necessary to appreciate the bone loss.