The Evaluation of Different Osteotomy Drilling Speed Protocols on Cortical Bone Temperature, Implant Stability and Bone Healing: An Experimental Study in an Animal Model

Implants have become part of mainstream dentistry, offering functional oral rehabilitation for fully and partially edentulous patients as originally reported by Brånemark et al.¹  Integration of a titanium implant within bone depends on several factors, but the primary circumferential bone healing around the implant will be of utmost importance.

Implant companies offer suggested drilling protocols for their dental implant models, but the surgeon has the opportunity to modify the surgical technique used. In doing this, the surgeon is able to influence the healing outcome.

The preparation of an implant osteotomy, due to friction of the drill with the host bone, will generate heat. This heat might have a detrimental impact on the bone, by impairing bone remodeling or even causing bone necrosis.²  This might be of specific interest in the deep cancellous bone areas, which will be encountered with greater drilling depth. Drill speed is often thought to be the most important factor when it comes to heat generation during implant site preparation. However, the load and drilling force is not to be underestimated.³  To date, there is no consensus on ideal drilling speed for implant osteotomy preparation. Some studies report high speed drilling (>2500 rpm) to decrease the risk for osseous damage,⁴  whereas others consider extreme low drilling speed (50 rpm) as suitable to preserve bone-cell viability.⁵  Definitely, this extreme low speed drilling without irrigation might offer some advantages from a surgical point of view; for example, collection of bone particles that could be used as autogenous bone graft⁶  and better intraoperative visibility⁷ ; however, the operation time will significantly increase. The drill design, drilling depth, drill wear, predrilling, bone type, thickness, and adequate saline irrigation are additional important variables that contribute to heat generation.⁸  The actual temperature of the saline might be more important as some studies have shown that saline prechilled to 10°C might be more effective in cooling host bone.⁹  Nevertheless, the optimal drilling protocol including the drilling speed, drill design, and irrigation is still yet to be established.

Micro-computerized tomography (micro-CT) is a nondestructive technique that allows 3-dimensional (3D) quantitative evaluation of implant osseointegration^10–12  offering more comprehensive and precise information compared with traditional histological methods.^10,13  Several studies have shown that bone parameters such as bone area (BA) and bone-implant contact (BIC), bone volume (BV)/tissue volume (TV), and bone area (BA)/tissue area (TA) data obtained from micro-CT correlates well with histomorphometry.^12,14 

Studies evaluating heat generation during implant osteotomy have been performed on cadaveric bovine and porcine bone blocks^15,16  or on synthetic blocks/resin models.^17,18  In these studies, heat was measured by thermocouples and infrared thermography.^16,18  To our knowledge, there is no study evaluating heat generation during implant osteotomy drilling in an in vivo model and analyzing the effects of heat changes on bone healing and osseointegration.

The aim this animal study is threefold:

• Measure the effect of drilling speed on heat generation in the cortical bone.

• Measure the effect of drilling speed on primary and secondary implant stability.

• Measure the effect of drilling speed on early and late bone healing with micro-CT.

The research was conducted according to the guidelines for the care and use of laboratory animals, and the study was approved by the ethical committee of Cukurova University (28.02.2017-2-3). The pre- and postoperative care and monitoring as well as housing and husbandry of the animals were performed in accordance with ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines.

Six healthy and mature male, Turkish Merino sheep, each with a weight of approximately 50 kg, were used in the present study. All surgical procedures were performed under general anesthesia (20 mg/kg ketamine, 2 mg/kg xylazine, and 0.5 mg atropine sulfate) and strictly sterile conditions in a veterinary surgical theater. For maintenance of general anesthesia, 15 mg/kg thiopental sodium IV injection was used. In addition, 4 mL local anesthetic (bupivacaine) was injected into the surgical site. In order to reduce the risk of postoperative infection, pre- and postoperative antibiotics (1 g cefazolin sodium IM) were administered. The animals also received analgesics (75 mg diclofenac sodium, IM) for 5 days.

The operation room temperature was kept at 25°C. All animals were immobilized in a ventral position. The pelvic area was shaved, and surgical area disinfected (1% iodine in 70% of ethanol). A transverse skin incision was made starting from the upper medial side of the iliac crest toward the anterior superior iliac spine in lateral direction. Afterward the incision was continued until the periosteum was reached. The periosteum was elevated and pushed aside, exposing the iliac crest.

Osteotomies were performed according to 5 different approaches:

• Osteotomy preparation according to an “extreme slow drilling protocol” at 50 rpm without saline cooling (group 1),

• Osteotomy preparation at 400 rpm with saline cooling (group 2),

• Osteotomy preparation at 800 rpm with saline cooling (group 3 – standard protocol as proposed by the implant company, Dentium Co, Seoul, Korea),

• Osteotomy preparation at 1200 rpm with saline cooling (group 4), and

• Osteotomy preparation at 2000 rpm with saline cooling (group 5).

Twelve implants (4 mm diameter, 10 mm length) were placed for each group. Saline cooling was performed at 30 mL/min at a temperature of 25°C. As proposed by the company manual, all implant drills were used (with a diameter of 2.2, 2.6, 2.85, and 3.3 mm in sequence) before placing the implant. Ten implants, including 2 per experimental groups, were placed in each animal. In every other animal, new burs were used to exclude bias from bur wear. Mean cortical bone temperatures (in degrees Celsius) during drilling were recorded and changes in temperature of the cortical bone measured before and during each bur use in order to calculate the delta (Δ) of heat change. The evaluations of the bone temperature were done using infrared thermography (Fluke, Everett, WA). The time of preparation with every bur (contact with bur and host bone) and implant insertion time (time until final implant insertion) were measured using a standard chronograph.

The infrared thermometer (accuracy: ±1% degrees Celsius, measurement range: −40°C to 550°C, repeatability: ±0.5% degrees Celsius and response time: 500 ms) was placed on an especially constructed tripod for standardization purposes. In order to enable the beam to be precisely aligned with the operation area and bone; the tripod was used to adjust the height of the device. The tripod was kept separate and isolated from the operation area to avoid any errors due to movements or vibrations during drilling.¹⁹  The thermometer was placed 30 cm away from the measurement area, as recommended by the manufacturer (Figure 1a).

Implant insertion (Implantium II, Dentium Co) was performed with 30 N.cm torque. Implant stability was measured by resonance frequency analysis (RFA) using an Osstell Mentor system (Integration Diagnostics AB, Göteborg, Sweden) (Figure 1b). This evaluation was performed at the installation of implants (primary stability) and at the time of animal sacrifice (secondary stability).

After implant placement, the soft tissues were closed in layers with resorbable 2.0 sutures (Dogsan, Trabzon, Turkey). The eating habits, ambulatory activities, and health status of the animals were monitored on a daily basis.

Three animals were sacrificed at each time points 4 weeks (T1) and 8 weeks after surgery (T2) using an overdose of sodium thiopental. Harvesting of the iliac wings and subsequent division into smaller pieces was performed immediately after euthanization using a carbide bur. As a result, the obtained specimen contained 1 implant with the surrounding bone.

All specimens were kept in formalin until micro-CT examination. The specimens were placed on the sample holder of the micro-CT imaging system with long axis of implant perpendicular to the X-ray beam and scanned in a SkyScan (SkyScan 1272, Bruker MicroCT) micro-tomograph (90 kV and 111 μA; filter: Al 0.5 mm + Cu 0.038; rotation step: 0.5-mm; acquisition time: 1 hour and 46 minutes). The software NRecon (Skycan, 2011, Version 1.6.6.0) and Dataviewer (Skycan, 2011, Version 1.4.4 64 Bit) were used for the quantification of the bone formation.²⁰ 

The region of interest (ROI) was determined and standardized by the area corresponding to the bone–implant interface of all implant threads that were inserted into iliac crest as previously described.²¹  The bone was identified in a threshold of 1000–2500 Hounsfield units in the ROI. The ROIs were limited to circular band of a 190-μm and stretched 60–250 μm from the implant surface and outward. It has been demonstrated that 60 μm from the implant surface is sufficient from the metal to minimize measurement artifacts. After determining the ROI, images were converted to grayscale for the 3D calculation by using an interpolated ROI tool. 3D volume of interest (VOI) was determined and bone structural parameter indicating the ratio between relative bone and tissue volume (BV/TV) in 2D and 3D were evaluated.

The statistical methodology was reviewed by an independent statistician. The continuous variables were analyzed by 1-way ANOVA with post-hoc test. The categorical variables between the groups were analyzed using the χ² test. The results are reported as mean ± standard deviation, median (minimum–maximum) and number (n) and percent (%). Statistical significance was set at P < .05 based on a 2-tailed comparison (SPSS version 22.0).

There were no complications observed during the healing period after surgical procedures. All inserted implants (half of them after 4 weeks [T1] and the other half after 8 weeks [T2]) were analyzed for the evaluation for the drilling protocol effects on bone healing.

The total osteotomy preparation time (OPT; time between initial contact of bur with bone until final implant position) in seconds, for each study group are shown in Table 1.

The longest OPT was observed in the 50-rpm group (70.0 ± 16.9 seconds). This OPT was statistically significantly different (P < .05) from the other groups. OPT times were similar for Group 2 (40.6 ± 9.9 seconds), Group 3 (35.0 ± 7.9 seconds), Group 4 (34.5 ± 9.8 seconds), and Group 5 (33.1 ± 4.7 seconds), with no statistically significant differences.

Mean cortical bone temperatures for each group during the whole drilling process (in degrees Celsius) are shown in Table 1. The highest cortical bone temperature was found in the Group 1 (31.5 ± 1.0) reaching statistical significance compared with other groups (P < .05).

The mean Δ temperatures (the amount of heat change before and during each bur use—in degrees Celsius) for the study groups are also shown Table 1. Due to the temperature of the irrigation solution (25°C), temperatures decreased by 2.8 ± 0.6°C in Group 5, followed by decreases of 2.6 ± 0.5°C in Group 3, 2.4 ± 0.7°C in Group 4, and 2.2 ± 0.9°C in Group 2. No statistically significant differences between these decreases were seen. In contrast, when no irrigation was used at 50 rpm, the temperature rose by 0.4 ± 0.5°C and this difference was statistically significant compared with other the groups (P < .05).

The ISQ values at T0, T1, and T2 are shown in Table 1. Although there were no statistically significant differences among groups for primary stability, the highest value was found in Group 1 (84.3 ± 2.8).

Group 4 showed the highest values (88.5 ± 1.6) at T2 (secondary stability), followed by Group 2 (87.3 ± 2.3), Group 1 (86.9 ± 1.0), and Group 3 (85.5 ± 2.6). The ISQ values of Group 5 at T2 (85.0 ± 2.1) were statistically lower compared with other groups (P < .05).

All groups had statistically higher ISQ values at T2, when compared with T0 (P < .05).

The results of micro-CT analysis are shown in Table 2.

2D Analysis: All groups had statistically increased BV/TV at T2 when compared with T1. The highest BV/TV volumes at T1 were noted for Group 1 (74.5 ± 1.7) followed by Group 3 (800 rpm as recommended by the implant manufacturer) (73.8 ± 0.5), Group 4 (73.8 ± 1.2), Group 2 (73.4 ± 0.6), and Group 5 (70.5 ± 1.3). At T2, Group 3 (91.9 ± 4.5) showed the highest BV/TV values.

3D Analysis (Figure 2): There were no statistically significant differences between groups for early healing period (T1) in terms of BV/TV (P > .05). At T2, Group 4 (40.8 ± 4.2) showed the highest BV/TV.

The comparison of BV/TV at T1 and T2 revealed a statistically significant increase only in the 2000-rpm group (P < .01).

Various factors can influence the temperature rise in bone during implant osteotomy preparation, including drill geometry,¹⁶  sharpness of the cutting tool,²²  drill reusage,²³  drilling depth,²⁴  drilling speed,²⁵  pressure applied to the drill,²⁶  use of graduated versus one-step drilling,²⁷  and use of internal or external irrigation.²⁸  These factors have been reported to affect the early bone healing response and subsequent osseointegration process.

In the current study, the real-time heat changes were measured simultaneously in an in-vivo sheep model. This model may provide more accurate and reliable results compared with in-vitro or ex-vivo studies (such as bone blocks or resin models) as heat absorption by the cells and the extracellular substance within the living tissue and capillary blood circulation will affect thermal conductivity and rapid heat release. In accordance with this, the heat generation values in the present study were always lower than those reported in previous in-vitro studies regardless of the drilling speed.^29,30  None of the drilling protocols used in this study produced heat generation to critical detrimental levels. Mean cortical bone temperatures were similar for all protocols including Group 1, in which no irrigation was used. The difference in our findings compared with previous studies that have reported higher temperature rises can possibly be explained by the above mentioned in-vivo design of this study.

In this study, temperatures before and during drilling procedures were measured by an infra-red laser thermometer. The commonly used thermocouple technology may enable direct measurement^29,31  while infrared thermography provides an indirect estimate.^22,32  Although thermocouple technology has commonly been used in previous studies, research on its use are not standardized due to the variations in distance and settings of the devices. In addition, thermocouples can measure the heat within the bone block; however, the heat in the coronal portion of the cortical bone (from which the peri-implant bone loss begins) cannot be measured. Recently Harder et al³³  have shown that the use of infrared technology may provide more accurate and reliable analysis compared with thermocouples.

The highest drilling speed protocol (2000 rpm with irrigation) used in our study required less operation time and produced less frictional heat; however, the primary (T0) and secondary (T2) stability values together with the 2D and 3D bone volume measurements were significantly lower in this group. Tabrizi et al³⁴  have demonstrated histologically that high drilling speeds may result in a reduction of vital bone percentage. In addition, while irrigation acts as a coolant reducing the heat, it also removes blood, fibrin clots, and osteogenic bone chips from the osteotomy,^7,35  and these bone chips are highly osteogenic contributing to new bone formation around implants.³⁶  Therefore, Aghvami et al³⁵  have suggested the use of lowest possible drilling speed without irrigation. In support of these previous findings, our results showed that the 50-rpm speed without irrigation gave better primary and secondary stabilities and BV/TV values compared with the other groups with only an approximate 1°C mean cortical bone temperature difference.³⁵  This finding is also in agreement with other previous studies that found that slow drilling speeds of different values, such as 50^5,37 and 100,³⁸  and resulted in minimal temperature variations.³⁹  Slightly higher values of secondary stability (T2) and subsequent 2D and 3D BV/TV values were found at 8 weeks in our 1200-rpm group with irrigation. This latter drilling protocol also is recommended by many implant companies and widely used. It should be mentioned that although all groups had sufficient ISQ values at 8 weeks for loading, the micro-CT analysis showed varying percentages of bone tissue around implants.

Sheep were used in the current study as they are considered to have similar bone healing rates, mineral composition, and blood supply as humans,⁴⁰  and have been previously established as useful models for human bone turnover and remodeling activity.⁴¹  Many implants can be inserted in 1 animal (which is more suitable for ethical considerations). One limitation of the study is the lack of information obtained on pressure applied on the drill during osteotomy preparation. Although all surgical procedures were performed by the same clinician for standardization purposes, previous studies have shown that even a slight difference of pressure on the drill can modify the temperatures and operation time.⁴  In addition, the laser thermometer used in the study can only reflect superficial temperatures, but heat generated in the deeper tissues of the osteotomy may cause more important effects on the bone. Therefore, further studies with novel heat techniques to assess the real-time temperature changes are needed for further clarification.

The results of this study suggest that the influence of drill speed and irrigation during implant osteotomy preparation is minimal when considering changes in temperature of the cortical bone, primary and secondary implant stability, and BV/TV.

Abbreviations

BA:

bone area

BIC:

bone-implant contact

BV:

bone volume

BV/TV:

bone volume and tissue volume

ISQ:

implant stability quotient

Micro-CT:

micro-computerized tomography

OPT:

osteotomy preparation time

RFA:

resonance frequency analysis

ROI:

region of interest

TA:

tissue area

TV:

tissue volume

VOI:

volume of interest

Professor Gulsah Seydaoglu, PhD, DMSc, Cukurova University, Faculty of Medicine, Department of Biostatistics, Turkey, was the independent statistician.

All the authors declared there is no conflict of interest.