INTRODUCTION: As one of the choices for limb salvage in the treatment of malignant bone tumors, a widely-resected bone mass may be irradiated in vitro during the operation and reinstituted as an autograft. Clinical trials in some institutions have proved that a single in-vitro X-irradiation of 50 to 300 Gy, together with the effect of remaining devascularization, affords sufficient local control of the tumor. A problem remains, however, as to how the reinstituted grafts will be remodeled. The present study was designed to quantitate the rate of cortical bone remodeling and to compare the rates between irradiated and nonirradiated autografts. MATERIALS AND kiETHODS: Forty adult rabbits were used. In each animal, a 15 mm-long diaphyseal segment was resected off the right tibia. The animals were devided into four groups of ten according to the doses given to their separated tibia1 segments; 0, 50, 100 and 200 Gy, respectively. The segments were placed in plastic dishes and X-irradiated with a full build-up in a sterile and moistured condition. Whether irradiated or not, all the segments were kept in the dishes for a total of 30 minutes. Then, they were reinstituted and fixed with a loose intramedullary rod so as to allow both the medullary and periosteal-side regeneration. Two animals from each group were killed at 6 and 12 weeks, and the remaining animals at 26 weeks postoperatively. After taking the soft X-rays, both decalcified and undecalcified transverse specimens of the mid-segment and of the same portion of the contralateral tibia were prepared. With the use of a microdigitizer, the area of the remaining irradiated segment and its proportion to the corresponding contralateral cortical area were obtained. By tetracycline fluorescence, the number of the growing osteons was counted. RESULTS: Soft X-rays at 6 weeks showed fusion of the segments by external bridging callus in all groups. Those at 26 weeks revealed advanced remodeling of the segments, but no apparent difference between the groups. The rate of remodeling at 26 weeks as obtained by area analysis and osteon count was only slightly, but not significantly, higher in the non-irradiated group than the irradiated groups, and there was no difference among the irradiated groups. DISCUSSION: In radiation-sterilized allografts and autoclaved autografts, excessive radiation or heat not only destroys the collagen architecture but also denatures bone morphogenetic protein and TGFs which play essential roles in bone remodeling. These disadvatages have limited their clinical use especially in weight bearing portions. On the other hand, irradiated autografts with several hundred Gy are assumed to have intact collagen architecture and undamaged bone growth factors. Our results in rabbits revealed that irradiated autografts are remodeled almost as readily as the ordinary autografts. Although some apparent disadvantages of the irradiated autografts exist, we believe that this charasteristic together with the unweakened initial strength is a valuable advantage in the skeletal reconstruction.