Quantitative three-dimensional (3D) imaging of angiogenesis during bone repair remains an experimental challenge. We developed a novel contrast agent containing 0.07-to 0.1-mm particles of zirconium dioxide (ZrCA) and established subtraction mCT using synchrotron radiation (sSRCT) for quantitative imaging of angiogenesis and bone repair. This method was applied to a rat model of tibial bone repair 3 days (DAY3; n ¼ 2), 5 days (DAY5; n ¼ 8), or 10 days (DAY10; n ¼ 8) after drill-hole injury. Using the same drill-hole defect model, its potential use was illustrated by comparison of bone repair between hindlimbs subjected to mechanical unloading (n ¼ 6) and normal weight bearing (n ¼ 6) for 10 days. Following vascular casting with ZrCA, the defect site was scanned with 17.9-and 18.1-keV X-rays. In the latter, image contrast between ZrCA-filled vasculature and bone was enhanced owing to the sharp absorption jump of zirconium dioxide at 18.0 keV (k-edge). The two scan data sets were reconstructed with 2.74-mm voxel resolution, registered by mutual information, and digitally subtracted to extract the contrast-enhanced vascular image. K 2 HPO 4 phantom solutions were scanned at 17.9 keV for quantitative evaluation of bone mineral. Angiogenesis had already started, but new bone formation was not found on DAY3. New bone emerged near the defect boundary on DAY5 and took the form of trabecular-like structure invaded by microvessels on DAY10. Vascular and bone volume fractions, blood vessel and bone thicknesses, and mineralization were higher on DAY10 than on DAY5. All these parameters were found to be decreased after 10 days of hindlimb unloading, indicating the possible involvement of angiogenesis in bone repair impairment caused by reduced mechanical stimuli. In conclusion, the combined technique of sSRCT and ZrCA vascular casting is suitable for quantitative 3D imaging of angiogenesis and its surrounding bone regeneration. This method will be useful for better understanding the linkage between angiogenesis and bone repair.