*4.2. Drilling Protocol*

The present findings showed that an undersized drilling protocol per se might not be detrimental to osseointegration for the Brånemark implant. In a previous study by the authors, implants were inserted in sheep mandible according to two different drilling techniques [41]. The results showed that implants inserted into an undersized osteotomy caused tissue damage to the peri-implant bone. In particular, large remodeling cavities with resorption activity were noted, and a lower amount of bone was identified up to 1.5 mm distance from the implant, both from histomorphometric μ-CT analysis. In addition, the drilling protocol seemed not to influence the amount of total BIC. Such findings are partially confirmed by the present results, since the amount of bone up to 1.5 mm from implant surface (BA1.5), was negatively influenced by the undersized drilling protocol. An explanation could be that the bone compression during the implant installation in a tight osteotomy would trigger a remodeling process at a distance from the implant interface. On the other hand, the temperature change does not influence this parameter, since the thermal conductivity of the bone would prevent the heat from being transferred at such a distance from the implant.

Compared with the previous sheep study [41], a number of differences were noted. In the former study, an implant with a micro-threaded neck was used, while the Brånemark type implant was utilized in the present investigation. A previous FEA study on the press-fit phenomena at the implant insertion, [42] affirmed that the micro-thread portion, induced more relevant strains compared to the situation without microthreads. Thus, one could expect greater bone damage after undersized drilling in such a scenario. On the contrary, the undersized drilling protocol positively influenced the amount of bone in close proximity to the implant, i.e., BIC and BAFO, in the present study. It could be assumed that a large portion of the bone in contact with the implant might be the original bone that was forced to the implant surface during the implant installation and still not removed. This finding was observed in several previous studies, which used a similar implant design [5,43,44]. In normal conditions, such tissue will be gradually resorbed with time and eventually substituted with vital bone, according to the remodeling process [45].

#### *4.3. Surface*

An interesting finding of the present study was that the type of implant surface did not influence the bone temperature. It could be expected that a moderately rough surface could increase the friction and thereby the heat [46,47]. However, no differences were noticed in the temperature increase between moderately rough and turned surfaced implants. According to the results, a moderately rough surface had a positive effect on the amount of bone in contact with the implant. This finding confirms that such surface topography is able to promote the osteoconductive properties of the implant, compared to turned surfaced implants, as reported by animal and human histologic reports by Ivanoff et al. and Zechner et al. [28,48].

#### *4.4. Clinical Applications and Limitations*

The study represents one of the first attempts to study the implant insertion temperature in an in vivo setting. Results demonstrated that an undersized drilling protocol causes an increase of intraosseous temperature during implant seating. A temperature increase negatively affected the amount of peri-implant bone. Thus, it may be suggested to reduce the friction overheating during implant installation. This would include decreasing the rotational speed [49], the use of the self-tapping implant design [24], the use of irrigation during implant installation, and the selection of the proper drilling protocol based on the implant design and the bone quality.

Considering the limitations in the present study, the record of the temperature was limited to the peak value during each insertion. Further studies are needed to evaluate the exact duration of the heat. In addition, from the present study, we cannot confirm whether the increase of the temperature caused tissue necrosis, since no specific stain for cell metabolism and tissue turnover was used. Moreover, the relationship between the compression and the heat, following undersized drilling was not explored. Future studies should be designed in order to indicate whether there is a predominant factor in the generation of tissue damage in the proximity of the implant. Finally, it must be stated that the sheep model, which has been broadly used in dental implant research, presents similarities and differences compared to human bone [50]. The thickness and density of cortical bone may approximate clinical scenarios, such as encountered in the mandible. In addition, sheep bone turnover resembles bone processes in humans, even though it is slightly more rapid. The healing period of five weeks was selected according to our previous research [41], since the influence of the surgical protocol is more evident at this stage in the peri-implant bone. However, the metatarsal bone, while it represents an accessible and convenient substrate for orthopedic and dental implant research, presents quite large anatomical and physiological differences compared with the human jaw, and it may display slightly divergent thermal properties. Therefore the present findings must be taken with reasonable caution.

#### **5. Conclusions**

Within the limitations of the present study, it was shown that different drilling protocol for the Brånemark implant system affects both the intraosseous temperature during implant installation and the peri-implant bone healing. The undersized drilling protocol provokes a greater increase of bone temperature in the proximity of the implant compared with non-undersized drilling. The temperature at the bone-implant interface may exceed the critical value for thermal necrosis, and it may have negative effects on peri-implant bone healing. The present results indicate that undersized drilling increases the amount of bone in the proximity of the implant, but it has a negative impact on the bone area far from the implant surface. The moderately rough surface does not influence the bone temperature, while it increased the bone attached to the implant. Further studies are needed to confirm the present results and to deeply investigate the thermal behavior and the biologic effect of peri-implant bone overheating during implant installation and to provide guidelines on the clinical decisions for the proper drilling protocol, based on the bone quality and implant design.

**Author Contributions:** Conceptualization—M.S. and J.B.; Data curation—M.S., Y.J. and S.Y.; Formal analysis—M.S. and S.Y.; Funding acquisition—J.B.; Investigation—M.S., Y.J., M.T., M.A. and E.P.; Methodology—M.S., Y.J., M.A., E.P. and S.Y.; Project administration—E.P. and J.B.; Resources—Y.J., M.T. and M.A.; Software—S.Y.; Supervision—J.B.; Validation—Y.J., M.T., E.P. and J.B.; Visualization—M.T. and E.P.; Writing—Original draft—M.S., Y.J. and S.Y.; Writing—Review & Editing—M.T., M.A., E.P., S.Y. and J.B.

**Funding:** This research was supported by own budget in Department of Oral & Maxillofacial Surgery and Oral Medicine, Faculty of Odontology, Malmö University, Sweden.

**Acknowledgments:** The authors are grateful to Benoit Lecuelle and Thomas Lilin from Ecole Nationale Vétérinaire d'Alfort, Paris, France and Simone Selvaggio, Italy for their valuable help during the surgical procedures.

**Conflicts of Interest:** The authors declare no conflict of interest.
