The aim of this study was to evaluate the dynamics of bone healing after piezosurgical and drilling osteotomy. Our study hypothesized that when compared to conventional drilling, bone healing after piezosurgery is faster due to enhanced expression of proteins involved in bone regeneration and reduced expression of proteins involved in inflammation and apoptosis.
To test our hypothesis we chose to study the bone healing process of a subcritical bone osteotomy since subcritical bone defects spontaneously and consistently heal by complete regeneration of the missing bone as their bone regenerative potentials are always fully exploited . The only variables that can influence their healing process are the level of inflammation that occurs immediately after the creation of the defect and the speed by which the regeneration process may occur. By means of this strategy, in a rat model of tibial subcritical size bone defect we therefore were able to follow and compare the events associated with the regeneration of bone defects created by drilling (Drill group) or piezosurgery (Piezo group).
We analyzed the bone formation by means of histology and histomorphometry at several time points, going from the early inflammatory stage (3 days after surgery) to the latest time point when complete bone regeneration and remodeling has occurred (60 days after surgery). By means of immunohistochemistry we also analyzed protein expression of early bone healing markers such as VEGF and CAS-3 at the early stages of the regenerative process (starting from day 3 up to day 14) and protein expression of bone remodeling markers such as OPG, RANKL, and OC at 7 and 14 days, when maximal regenerative activity occurs. Gene expression analysis of 21 genes expressing osteoblast differentiation regulators, osteogenic markers, inflammatory cytokines, and apoptotic factors was performed at 3, 7, and 14 days after surgery to validate and substantiate the immunohistochemical analysis. Among the osteoblast differentiation regulators, we tested several genes representative of the BMP canonical pathway  and of the Wnt canonical and non-canonical pathways  because of their relevance with the cellular activity that occurs during bone regeneration. Not all genes tested were consistently expressed throughout the healing process. We chose to utilize and show data regarding those genes that consistently presented with reproducible results.
The data collected showed that in our animal model the bone healing dynamics after piezosurgery are comparable to those observed with conventional drilling, with no evident signs of bone healing acceleration in the Piezo group versus the Drill group. At all the time points analyzed, histological analysis showed no differences between the defects created by piezosurgery and drilling. Histomorphometrical analysis also showed no differences’, with the exception of higher levels of newly regenerated bone at 30 days after piezosurgery. However, this difference disappeared at 60 days, when the amount of newly regenerated bone was equal for both groups. This result could be indicative of a better ability to regenerate bone of the Piezo group at 30 days of healing. However, our subsequent analyses indicate that this temporarily higher amount of bone formation is not due to healing acceleration during the early stages of healing. In fact, immunohistochemical analysis at 3 days after surgery showed no statistically significant difference in terms of expression of both VEGF and CAS-3 and gene expression analysis of 21 different genes, including Vegf and Cas-3, showed no significant differences for all the genes analyzed. Also, seven days after surgery, immunohistochemistry showed no differences in expression of VEGF, CAS-3, OPG, RANKL, and OC, indicating no changes during early healing in terms of vascularization, apoptosis, and bone regeneration and remodeling. The gene expression analysis at 7 days also showed no difference in expression of Vegf, Cas-3, Opg, and Oc. However, a significant reduction in expression of Runx2, Wnt10b, and Sclerostin was detected in the Piezo group at this time point. Since activation of BMP and Wnt signaling have been demonstrated to be essential at the early stages of bone repair [25, 26], this data may indicate a reduction in terms of number of osteoprogenitor cells (reduction of Runx2) as well as a lower level of Wnt activity (lack of detection of Wnt10b indicating a direct down-regulation of the canonical Wnt pathway and reduction of expression of Sclerostin indicating the unnecessary expression of an inhibitor because of the already occurred down-regulation of the pathway) that may be interpreted as a deceleration rather than an acceleration of the healing process in the Piezo group. Furthermore, at 14 days after surgery, the immunoreaction also showed no significant differences between the two groups and gene expression analysis also showed no differences in expression of Vegf, Cas-3, Opg, and Oc. However, a significant reduction (lack of detection) of expression of Wnt10b and IL-6 in the Piezo group was detected at this time point. Thus, data collected at each time point may be indicative of a deceleration rather than acceleration of the healing process associated to piezosurgery.
It could be speculated, however, that the decreased levels of expression of Runx2 and Wnt10b seen at 7 and 14 days in the Piezo group are indicative of a diminished need for a full-speed regenerative process at these time points because in this group healing has already progressed to later and more advanced stages. However, considering that the histomorphometrical and immunohistochemical analyses did not show any difference in terms of amount and quality of bone regeneration at all early time points tested, we believe that this is not necessarily the case and that the gene expression changes observed in the Piezo group may simply represent normal variability of a complex process that is not regulated by few genes only.
Our data is not in contrast with the results shown by previous studies that compared piezosurgery with other traditional osteotomy methods. For instance, a study by Ma et al.  found no statistically significant differences in terms of histomorphometry but higher degree of formation of vascularized tissues, of provisional matrix, and of bone remodeling activity at 7 and 14 days after piezosurgery. These results may appear different from those shown in the present study. However, the animal model utilized by Ma and coworkers used bone defects smaller than those used in the present study and therefore the difference between the two studies may be due to the size of the bone defects utilized. Preti and co-workers  concluded that piezoelectric surgery appears to be more efficient in the first phases of bone healing than traditional osteotomy. Once again, these results may appear in contrast to those presented in this work. However, in their study Preti and co-workers analyzed the effects of piezosurgery on osseointegration of implants and not on regeneration of bigger bone defects. It is possible that the different conclusions are due to the different microenvironments studied. Similar consideration may be made between the bone defect microenvironment analyzed in our study and the periodontal defect microenvironment analyzed by Vercellotti et al. . Thus, it remains possible that piezosurgery accelerates osseointegration of titanium implants and facilitates periodontal regeneration without being advantageous in terms of regeneration of bigger bone defects.