Methods: For simulating BRONJ, zoledronic acid was administered to 40 rats for 5 weeks. Two weeks later, a right first molar was extracted from each rat. The rats were randomized into four groups of socket treatments: control (empty extraction socket) or with sockets filled with ACS, HA, or HA+ACS (n=4×10). After 2 weeks, 5 rats in each group were sacrificed and subjected to histopathologic and histomorphometric evaluation. Eight weeks post-surgically, the rest of rats were euthanized and histologically examined. The Kruskal-Wallis test was used to compare the four treatments at each time point (α=0.05). Results: Six rats were lost overall. In the second week, vascularization was higher in ACS group (P<0.05); osteoclast activity was not different between groups (P>0.05); empty lacunae were the most and fewest in control and HA+ACS groups, respectively (P<0.05); eosinophil infiltration was maximum in HA group (P<0.05); lymphocyte counts were maximum and minimum in the HA+ACS and ACS groups, respectively (P<0.05); the highest and lowest neutrophil counts were seen in ACS and control groups, respectively (P<0.05); and the extent of live bone did not differ between groups (P>0.05). In the eighth week, vascularization was not different in groups (P>0.05); the highest and lowest osteoclast activities were seen in the control and HA+ACS groups, respectively (P<0.05); empty lacunae were the most and fewest in control and HA+ACS, respectively (P<0.05); maximum and minimum numbers of eosinophils were in control and HA+ACS groups, respectively (P<0.05); HA and control groups exhibited the highest and lowest lymphocyte counts, respectively (P<0.05); the lowest and highest neutrophil counts were observed in HA+ACs and control groups, respectively (P<0.05); and the highest and lowest extents of the live bone were observed in HA+ACS and control groups, respectively (P<0.05). Conclusions Within the limitations of this preliminary animal study, HA and especially HA+ACS seem a proper method for preventing or treating BRONJ.