*3.3. Finite Element Model*

Temperature distribution at the center section is shown in Figure 5. The calorific value of the normal model was 30.8 W/m3. The calorific values of the upper fixture and bottom fixture were 30.8 W/m<sup>3</sup> and 60.4 W/m3, respectively. The calorific value of fixture for the undersized drilling model showed 1.96 times greater than that for the non-undersized drilling model. The bone temperature at the interface with the implant surface was calculated for both models.

**Figure 5.** (**a**) The temperature distribution calculated on the Finite Element Model (FEM). The undersized drilling model and non-undersized model are displayed on the left and on the right, respectively. The temperature is displayed in ◦C. Note how the heat is poorly distributed around the implant surface. This means that the overheating risk is limited to the bone in the proximity to the implant. (**b**) Calculation of the bone temperature at the implant interface according to the FEM. This value was calculated by entering the value recorded with the thermocouple in the experimental part.

#### **4. Discussion**

In the present in vivo study, the intraosseous temperature was measured during dental implant installation into cortical bone, following two different drilling protocols and with two different implant surfaces. In addition, histomorphometric parameters of osseointegration were evaluated in relation to the bone temperature recorded during the implant installation and a computational model was created to examine the thermal distribution. To the knowledge of the authors, the current investigation represents the first in vivo experiment with this setup.

### *4.1. Bone Temperature*

In the present study, it was discovered that the intraosseous temperature during implant installation was influenced by the drilling protocol. During implant installation, a certain amount of energy is also dissipated into heat [25]. As previous in vitro studies have observed, the rotational torque is positively related to bone heating during implant installation [24,34]. Accordingly, in this study, the installation of implants into undersized sites developed a great friction resistance, resulting in both in a higher ITV (Figure 3) and temperature increase, compared with non-undersized osteotomies. Specifically, the undersized drilling protocol groups resulted in a median increase of temperature of approximately 8 ◦C, while in the non-undersized drilling protocol groups, it was approximately 4 ◦C. The maximum recorded temperature of 45.3 ◦C exceeded the limit of cell damage, which is 45 ◦C according to Ludewig [35], but it was lower than the critical value for bone necrosis, which is 50 ◦C according to Lundskog [36]. In the early 1980s, Eriksson, whose doctoral thesis greatly contributed to the knowledge on bone tissue regeneration, stated that the threshold level for bone survival was 47 ◦C

for 1 min [37]. However, the temperature recorded in the present study was detected at 1 mm from the implant surface. To have a better understanding of the temperature behavior in the proximity of the implant, a FEM was designed. It was estimated that the temperature at the bone-implant interface for the undersized groups and non-undersized groups reached 58.7 ◦C and 52.0 ◦C respectively (Figure 5). Thus, according to the FEM calculation, the installation of the implant caused a frictional heat over the critical temperature for bone injury at the bone-implant interface. One may expect that implants installed with an undersized drilling protocol would create a major extent of tissue damage. Still, such an overheating condition is restricted to the proximity of the implant, as shown by the model. Since the temperature change was affected by the drilling protocol (*p* < 0.001), but was not influenced by the surface topography (*p* = 0.879), the surface characteristics (parameters Sa and Sdr) were not included in the design of the FEM.

Such results are confirmed by histomorphometry. In effect, it was demonstrated that the temperature generated during implant installation has a tangible biologic impact, since the amount of bone between the threads, namely BAFO, is negatively affected by the temperature increase. This finding is in accordance with Eriksson's study, in which they observed a loss of 10% of bone tissue after 30 days when a rabbit tibia was heated to 47 ◦C for 1 min [22]. This study supports the conclusion that the portion of the bone in proximity with the implant surface might be the most sensitive to heating at the implant installation. Moreover, based on the FEM analysis and histomorphometric results, this heating might induce bone damage if the undersized drilling protocol is applied. The influence of the heat generated by the drilling procedures was likely to be excluded since one minute elapsed before the implant installation. Previous research indicated that the bone returned to baseline temperature after approximately 30 s [38].

Nevertheless, due to the low thermal conductivity properties of cortical bone, we could expect a low grade of heat distribution through the bone [39]. Thus, the risk of bone overheating is limited to the bone in the proximity to the bone-to-implant interface.

It has to be said that Eriksson observed how bone cells are susceptible to the exposure time, other than peak temperature [40]. In the present experiment, the maximum temperature endured for a few seconds, then it gradually descended. However, the actual temperature/time curve was not recorded, representing a limitation of the study.
