Counterclockwise Drilling with Different Tapered Drills Condenses the Implant Bed—An Optical Coherence Tomography In Vitro Study

Background and Objectives: To evaluate the condensation and the microarchitecture of implant bed walls of sites prepared with counterclockwise drilling with tapered implant drills using optical coherence tomography. Materials and Methods: Four drill designs with different wall and tip angles were used. Polyurethane laminas resembling type IV bone microarchitecture were superimposed and clamped with a vice to simulate the coronal, middle, and apical aspects of the implant site. Twenty implant beds were prepared at 1200 rpm in clockwise (control) and counterclockwise (test) directions (N = 160). Optical coherence tomography (OCT) was used to evaluate the condensation and microarchitecture characteristics of the implant bed walls. The relative condensation was calculated using the Image J software Bone application. The microarchitecture was evaluated in reconstructed 3D volumes in XY, XZ, and YZ sections. Statistical analysis was performed using one-way ANOVA. Dunnet test was applied to determine differences between groups. Significance was set as p < 0.05. Results: Counterclockwise drilling (Test) condensed and changed the microarchitecture of the apical regions for all the implant beds in all of the groups when compared to clockwise drilling (control). The apical region of test groups showed the highest relative bone condensation (p = 0.026) when compared to controls. Conclusions: The direction of rotation (counterclockwise drilling) and not the design of tapered drills (tip and wall angles) is responsible for the condensation at the apical area observed in polyurethane blocks. The OCT method can be used for the evaluation of changes in density and microstructure of polyurethane blocks.


Introduction
The evaluation of bone quantity (BQT) and bone quality (BQL) is a fundamental step during implant planning. The BQT analysis evaluates the bone availability for the placement of dental implants outlined by different anatomical structures [1]. The BQL is a relatively vague term that assesses the characteristics of the cortical bone (thickness, density, and porosity) and cancellous bone (trabecular number, thickness, and connectivity) [2].
These factors can predict dental implant stability and serve as relative prognosis factors in cases of immediate implant loading [3]. Regarding BQL, it appears that implant survival is similar in patients with low and average bone density [4]. However, late implant failures have been correlated to poor quality of cancellous bone [5], patients with osteoporosis have shown increased peri-implant bone loss compared with healthy patients after 5 years [6], initial low BQL has been linked with increased bone remodeling around dental implants [7], and implants inserted in bone of poor quality present an increased risk of failure compared to implants inserted in bone of good quality [8].

Drilling Parameters
The following parameters were standardized: drilling speed 1200 rpm, drilling depth of ±10 mm, and profuse irrigation. In total, 160 implant bed preparations were completed, with 40 implant bed preparations per drill group (20 clockwise and 20 counterclockwise).

Clockwise Drilling Method (Control)
The drill rotation was set in the clockwise direction. A pumping method was used to apply slight downward pressure to produce perforation of the bone laminas followed by drill withdrawal to allow stress relief until the drill reached a depth of ±10 mm.

Drilling Parameters
The following parameters were standardized: drilling speed 1200 rpm, drilling depth of ±10 mm, and profuse irrigation. In total, 160 implant bed preparations were completed, with 40 implant bed preparations per drill group (20 clockwise and 20 counterclockwise).

Clockwise Drilling Method (Control)
The drill rotation was set in the clockwise direction. A pumping method was used to apply slight downward pressure to produce perforation of the bone laminas followed by drill withdrawal to allow stress relief until the drill reached a depth of ±10 mm.

Counterclockwise Drilling Method (Test)
The drill rotation was set in the counterclockwise direction. A slight downward pressure was applied to allow the perforation of the bone laminas; then, the drills gradually advanced into the implant bed preparation. Afterwards the drills were pulled out (for stress relief), and then downward pressure was repeatedly applied (in/out fashion) until a drilling depth of ±10 mm was achieved.

Evaluation of the Bone Density and Bone Structure
After the completion of the implant bed preparations, the bone laminas were separated, and the relative bone density and bone structure characteristics were analyzed in each lamina for each drill group. The areas of analysis were the 3 mm coronal laminas, 3 mm middle laminas, and 3 mm apical laminas that simulated the implant bed.
The OCTG-900 spectral-domain optical coherence tomography (OCT) device (Thorlabs, Newton, NJ, USA) was used for the evaluation of the bone density and structure. Cross-sectional (2D) and volumetric (3D) data of the implant beds were obtained using an A-scan depth of 2.9 mm and a pixel size of 21.04 ± 2.32 µm to achieve a transverse resolution of 4 µm and a vertical resolution of 6 µm. The refractive index of the polyurethane foam was set as 1.5, and the scanning was completed at low-speed and high-sensitivity recording (5.5 kHz). The light intensity limits were a lower value (LV) of 22.0 and a higher value (HV) of 80.3.
For the cross-sectional scanning, a 10 mm length scanning line was traced. The line passed through the center of the implant bed preparation, dividing the implant bed into two equal semicircles. The line was extended by 3 mm per side (thus extending the scanning line by 3 mm to both sides of the implant beds). Line scanning (for the cross-sectional evaluations) was completed for each osteotomy (160 times in total).
For 3D volumetric scanning, squared areas of 10 mm × 10 mm contained in the center of each of the implant bed preparations were traced. In total, 160 volumetric analyses were

Counterclockwise Drilling Method (Test)
The drill rotation was set in the counterclockwise direction. A slight downward pressure was applied to allow the perforation of the bone laminas; then, the drills gradually advanced into the implant bed preparation. Afterwards the drills were pulled out (for stress relief), and then downward pressure was repeatedly applied (in/out fashion) until a drilling depth of ±10 mm was achieved.

Evaluation of the Bone Density and Bone Structure
After the completion of the implant bed preparations, the bone laminas were separated, and the relative bone density and bone structure characteristics were analyzed in each lamina for each drill group. The areas of analysis were the 3 mm coronal laminas, 3 mm middle laminas, and 3 mm apical laminas that simulated the implant bed.
The OCTG-900 spectral-domain optical coherence tomography (OCT) device (Thorlabs, Newton, NJ, USA) was used for the evaluation of the bone density and structure. Cross-sectional (2D) and volumetric (3D) data of the implant beds were obtained using an A-scan depth of 2.9 mm and a pixel size of 21.04 ± 2.32 µm to achieve a transverse resolution of 4 µm and a vertical resolution of 6 µm. The refractive index of the polyurethane foam was set as 1.5, and the scanning was completed at low-speed and high-sensitivity recording (5.5 kHz). The light intensity limits were a lower value (LV) of 22.0 and a higher value (HV) of 80.3.
For the cross-sectional scanning, a 10 mm length scanning line was traced. The line passed through the center of the implant bed preparation, dividing the implant bed into two equal semicircles. The line was extended by 3 mm per side (thus extending the scanning line by 3 mm to both sides of the implant beds). Line scanning (for the cross-sectional evaluations) was completed for each osteotomy (160 times in total).
For 3D volumetric scanning, squared areas of 10 mm × 10 mm contained in the center of each of the implant bed preparations were traced. In total, 160 volumetric analyses were completed. A-scan/line rates were set as 5-248 kHz with a 101-102 dB sensitivity for cross-sectional and volumetric evaluations.
The following quantitative and qualitative variables were evaluated: -Implant bed wall condensation/density (quantitative): Two cross-sections passing through the center of the implant bed preparation were obtained, thus allowing the evaluation of the four walls of the implant bed preparation. Afterward, the recorded files were transferred to the analysis software Image J (National Institutes of Health, Bethesda, MD, USA). Pixel size was determined based on the refractive index of the trabecular bone and the trabecular spacing. The implant bed wall with high density was characterized by high refraction, and a region of interest (ROI) was traced per section. The percentage of highly refractive bone was traced, and the percentage of bone in relation to the whole area was extracted ( Figure 4A-D). Qualitative changes, including trabecular space narrowing and trabecula fracture, as well as no change, were described. The mean condensation of each osteotomy at the coronal, middle, and apical region were collated per group, and mean ± SD deviations were recorded for each drill design. Results are expressed in percentages.
Trabecular structural changes (qualitative): The trabecular structure adjacent to the 3 mm of the perimeter of the implant bed preparation was evaluated. ThorImageOCT software (Thorlabs, Newton, NJ, USA) was used to mark a square containing the implant bed preparation. The dimensions of the square were 10 mm by 10 mm. The trabecular structure characteristics surrounding 360 grades of the implant bed site were detected. Vertical and horizontal scanning of the area was completed to create a 3D reconstruction ( Figure 5). Thus, 3D evaluation of the implant bed walls and proper description of the trabecular structure characteristics were completed.
The following quantitative and qualitative variables were evaluated: -Implant bed wall condensation/density (quantitative): Two cross-sections passing through the center of the implant bed preparation were obtained, thus allowing the evaluation of the four walls of the implant bed preparation. Afterward, the recorded files were transferred to the analysis software Image J (National Institutes of Health, Bethesda, MD, USA). Pixel size was determined based on the refractive index of the trabecular bone and the trabecular spacing. The implant bed wall with high density was characterized by high refraction, and a region of interest (ROI) was traced per section. The percentage of highly refractive bone was traced, and the percentage of bone in relation to the whole area was extracted ( Figure 4A-D). Qualitative changes, including trabecular space narrowing and trabecula fracture, as well as no change, were described. The mean condensation of each osteotomy at the coronal, middle, and apical region were collated per group, and mean ± SD deviations were recorded for each drill design. Results are expressed in percentages. Trabecular structural changes (qualitative): The trabecular structure adjacent to the 3 mm of the perimeter of the implant bed preparation was evaluated. ThorImageOCT software (Thorlabs, Newton, NJ, USA) was used to mark a square containing the implant bed preparation. The dimensions of the square were 10 mm by 10 mm. The trabecular structure characteristics surrounding 360 grades of the implant bed site were detected. Vertical and horizontal scanning of the area was completed to create a 3D reconstruction ( Figure  5). Thus, 3D evaluation of the implant bed walls and proper description of the trabecular structure characteristics were completed.

Statistical Analysis
The sample size of a confidence level of 95%, with an odds ratio of 10% and a power of 90%, was established as n = 20. Twenty implant bed preparations in clockwise rotation and twenty preparations in counterclockwise rotation for each implant drill geometry were prepared. Forty osteotomies were completed per group for a total sample size of N = 160 osteotomies. The normality of the samples was evaluated with the Shapiro-Wilk test, and one-way ANOVA with Dunnet post-test were used to compare differences between groups. The significance of differences was set as p < 0.05.

Coronal
The range of bone condensation at the coronal section produced by tapered drills with clockwise rotation (61.19 ± 4.58%) was similar to that produced by tapered drills with

Statistical Analysis
The sample size of a confidence level of 95%, with an odds ratio of 10% and a power of 90%, was established as n = 20. Twenty implant bed preparations in clockwise rotation and twenty preparations in counterclockwise rotation for each implant drill geometry were prepared. Forty osteotomies were completed per group for a total sample size of N = 160 osteotomies. The normality of the samples was evaluated with the Shapiro-Wilk test, and one-way ANOVA with Dunnet post-test were used to compare differences between groups. The significance of differences was set as p < 0.05.

Coronal
The range of bone condensation at the coronal section produced by tapered drills with clockwise rotation (61.19 ± 4.58%) was similar to that produced by tapered drills with counterclockwise rotation (62.57 ± 5.52%). When comparing all of the tapered drills and drilling directions at the coronal area, only the Anker drill showed significant increased bone condensation with counterclockwise drilling when compared to clockwise drilling (p = 0.04) (Figures 6 and 7; Tables 2 and 3).

Coronal
The range of bone condensation at the coronal section produced by tapered drills with clockwise rotation (61.19 ± 4.58%) was similar to that produced by tapered drills with counterclockwise rotation (62.57 ± 5.52%). When comparing all of the tapered drills and drilling directions at the coronal area, only the Anker drill showed significant increased bone condensation with counterclockwise drilling when compared to clockwise drilling (p = 0.04) (Figures 6 and 7; Tables 2 and 3).

Middle
The range of bone condensation at the middle section produced by tapered drills with clockwise rotation (58.05 ± 7.45%) was similar to that produced by tapered drills with counterclockwise rotation (59.91 ± 5.73%) ( Table 4). When comparing all of the tapered drills and drilling directions at the coronal area, there was no difference in the percentage of bone condensation between counterclockwise drilling and clockwise drilling (p > 0.05) (Figures 8 and 9; Tables 4 and 5).

Middle
The range of bone condensation at the middle section produced by tapered drills with clockwise rotation (58.05 ± 7.45%) was similar to that produced by tapered drills with counterclockwise rotation (59.91 ± 5.73%) ( Table 4). When comparing all of the tapered drills and drilling directions at the coronal area, there was no difference in the percentage of bone condensation between counterclockwise drilling and clockwise drilling (p > 0.05) (Figures 8 and 9; Tables 4 and 5).

Apical
The range of bone condensation at the apical section produced by tapered drills with counterclockwise rotation (87.52 ± 3.01%) was higher than that with tapered drills with clockwise rotation (69.61 ± 3.625%). When comparing all of the tapered drills and drilling directions at the apical area, there was a significant difference in the percentage of bone condensation between counterclockwise drilling and clockwise drilling (p < 0.05) without differences between drill designs (p > 0.05) (Figures 10 and 11; Tables 6 and 7).

Apical
The range of bone condensation at the apical section produced by tapered drills with counterclockwise rotation (87.52 ± 3.01%) was higher than that with tapered drills with clockwise rotation (69.61 ± 3.625%). When comparing all of the tapered drills and drilling directions at the apical area, there was a significant difference in the percentage of bone condensation between counterclockwise drilling and clockwise drilling (p < 0.05) without differences between drill designs (p > 0.05) (Figures 10 and 11; Tables 6 and 7).   Clockwise drilling with tapered drills with various tip and wall angles did not produce condensation of the implant bed walls, nor did it change the trabecular spacing. Meanwhile, counterclockwise drilling with tapered drills produced condensation of the implant bed at the apical area and produced a reduction in the trabecular spacing. The degree of condensation at the apical area of implant beds prepared with counterclockwise drilling was similar for all of the drill designs and was not related with variations in tip and wall angles.  Clockwise drilling with tapered drills with various tip and wall angles did not produce condensation of the implant bed walls, nor did it change the trabecular spacing. Meanwhile, counterclockwise drilling with tapered drills produced condensation of the implant bed at the apical area and produced a reduction in the trabecular spacing. The degree of condensation at the apical area of implant beds prepared with counterclockwise drilling was similar for all of the drill designs and was not related with variations in tip and wall angles.

Discussion
The goal of the present study was to determine any possible differences between tapered drills with different designs (tips and wall angles) and different directions of rotation (clockwise versus counterclockwise) and to evaluate the changes of the density of implant beds prepared in polyurethane blocks representing type IV bone.
The results of the present work showed that the direction of rotation, but not the drill design (tips and wall angles), increased the density of the implant bed. The bone density was significantly increased at the apical area in all the groups with counterclockwise drilling rotation, while no changes in bone wall density were observed at the coronal and middle regions.
The influence of the direction of rotation of implant drills on the bone density was first described by Huwais and Meyer [22]. The authors used a porcine tibia model and a tapered drill with multiple flutes that rotated clockwise or counterclockwise. The authors noted that counterclockwise drilling produced increased bone density around the implant osteotomies evaluated with microcomputed tomography (µCT), among other findings. The authors identified compacted bone with a thickness of 0.1 mm to 0.3 mm at the edges of the osteotomy and 0.5 mm to 1.0 mm thickness at the apical area of the osteotomy.
However, the authors did not compare their drill design with other drill designs. Furthermore, Antonelli et al. [21] and Delgado-Ruiz et al. [25], showed in experimental models in pig ribs that the counterclockwise drilling technique increased the bone density of the implant bed walls and that bone density was higher at the apical region (evaluated with µCT). Their results are in agreement with the results of the present work in regard to the apical area; however, no evidence of bone condensation was observed with any of the drills and rotation direction used in the present work. These differences could be produced by the different viscoelastic properties of pig bone when compared to the viscoelastic properties of polyurethane blocks [26].
During counterclockwise drilling, fracture and a simultaneous displacement of the bone occurred in the apical direction. Meanwhile, counterclockwise drilling produced a pure fracture phenomenon of the implant bed walls without lateral displacement.
This difference explains the increased bone condensation observed at the apical region but not at the bed walls, which is in agreement with the theories of Frost [27,28].
An OCT device was used in the present study to evaluate the structure and condensation of the trabecular structure of polyurethane blocks. This method was selected given its high resolution, high contrast, and the three-dimensional reconstruction capabilities that OCT systems have shown [33,34]. Moreover, this method allowed for the preparation of cross-sectional images to disclose the internal structure of the polyurethane blocks non-destructively [35]. Furthermore, bone mineral density has been evaluated by light scattering methods (which are the basis of OCT), and it was found that OCT is efficient in evaluating and detecting regional changes of bone density [36].
OCT is comparable with the micro-CT technique [37], with differences in higher penetration depth for micro-CT (the OCT device penetration depth is limited to between 2.5 mm to 3 mm) and better contrast and resolution for OCT [37]. To overcome the depth penetration limitations of OCT in the present study, bone laminas with a width of 3 mm were used. Thus, the evaluations were always within the range of a resolution depth of 3 mm of the OCT. In addition, polyurethane blocks with semi-transparent qualities and a known refractive index were used.
One drawback of the present work is that the trabecular pattern of the polyurethane foams is too homogeneous. As a result, it does not reproduce the heterogeneous structure of the human trabecular bone or other experimental animal models (pig and bovine ribs). This issue could be minimized by using layered or multi-structured (cortico-cancellous) polyurethane foams in future studies.
To date, experimental studies have been focused on the comparison of counterclockwise drilling with a single type of drill design (multifluted tapered drills). The present study results demonstrated that the tapered drill design has no influence on bone condensation. Instead, the direction of the rotation (counterclockwise drilling) of tapered drills with various tips and wall angles is responsible for the observed apical bone condensation.

Conclusions
Within the limitations of this experimental in vitro study, it can be concluded that the direction of rotation (counterclockwise drilling) and not the design of tapered drills (tip and wall angles) is responsible for the condensation at the apical area observed in polyurethane blocks. The OCT method can be used for the evaluation of changes in the density and structure of polyurethane blocks.