Dental implant surface temperatures following double wavelength (2780/940 nm) laser irradiation in vitro

Abstract Objective To estimate the implant surface temperature at titanium dental implants during calibrated irradiation using double wavelength laser. Material and methods A double wavelength laser, 2780 nm Er,Cr:YSGG and 940 nm diode, was calibrated and used to irradiate pristine titanium dental implants, OsseoSpeed, TiUnite and Roxolid SLActive, representing different surface modifications. Initial calibration (21 implants; 7 implants/group) intended to identify optimal wavelength/specific output power/energy that not critically increased the temperature or altered the micro‐texture of the implant surface. Subsequent experimental study (30 implants; 10 implants/group) evaluated implant surface temperature changes over 190 s. Irradiation using a computerized robotic setup. Results Based on the initial calibration, the following output powers/energies were employed: Er,Cr:YSGG laser 18.4 mJ/pulse (7.3 J/cm2)–36.2 mJ/pulse (14.4 J/cm2) depending on implant surface; diode laser 3.3 W (1321.0 W/cm2). During double wavelength irradiation, implant surface temperatures dropped over the first 20 s from baseline 37°C to mean temperatures ranging between 25.7 and 26.3°C. Differences in mean temperatures between OsseoSpeed and TiUnite implants were statistically significant (p < 0.001). After the initial 20 s, mean temperatures continued to decrease for all implant surfaces. The decrease was significantly greater for TiUnite and Roxolid SLActive compared with OsseoSpeed implants (p < 0.001). Conclusion Calibrated double wavelength laser irradiation did not critically influence the implant surface temperature. During laser irradiation the temperature decreased rapidly to steady‐state levels, close to the water/air‐spray temperature.

However, irradiation of titanium dental implants using laser energy may lead to unintended heating of the implant potentially compromising osseointegration. Animal studies suggest a 47 C exposure over 1 min a critical threshold to inflict damage to bone (Eriksson & Albrektsson, 1983, 1984. Only a few studies examining Er,Cr:YSGG lasers and none the 940 nm diode laser or double wavelength laser have been presented relative to their thermal effects on titanium dental implants (Gomez-Santos et al., 2010;Romanos et al., 2017) or titanium discs (Ercan et al., 2014;Strever et al., 2017). By contrast, in several studies, thermal effects on different titanium surfaces have been evaluated during or after irradiation of pulsed 2940 nm Er:YAG laser (Monzani et al. 2018;Hakki et al., 2017) and diode lasers of 810, 980 and 1064 nm (Geminiani et al., 2012;Leja et al., 2013;Matys et al., 2016;Valente et al., 2017). These studies suggest that by using water-spray cooling, the implant temperature may not exceed 47 C. As each wavelength has a unique curve for the absorption coefficient in different tissues and substances, a comparison between the properties of two neighboring lasers in the electromagnetic spectrum is only approximate (Valente et al., 2017). Moreover, a study on titanium dental implants placed in porcine bone concluded that thermal conductivity as well depends on the chemical composition and diameter of the implant (Matys et al., 2016).
Literature reviews fail to consent on preferred protocols for safe laser debridement and disinfection of contaminated dental implants.
Heterogeneity in protocol, variation in included parameters, and lack of information concerning calibration of the laser equipment and measuring instruments hamper comparisons between studies (Kamel et al., 2014;Smeo et al., 2018). Most laser systems report a discrepancy between device power/energy setting and actual output power/energy. Variables including laser light transmission system, output power/energy, pulse rate, laser beam area and divergence (spread), and distance to irradiated surface need to be identified and optimized prior to clinical application (Takagi et al., 2018;Tunér & Jenkins, 2016).
The effect of double wavelength laser irradiation has to our knowledge not been reported for titanium dental implants. Acknowledging the lack of laser specific information, the overall aim of this study was to investigate whether double wavelength laser irradiation critically increases the implant surface temperature above the critical threshold of 47 C for different implant systems using a validated in vitro protocol. We hypothesized that surface temperature would rapidly increase, but not above the critical threshold of 47 C. The null hypothesis was that the final temperature would be similar for the different implant systems. This study consists of two parts: First, to identify the output handpiece fiber tip power and energy for each implant system that do not produce thermal heating or surface micro-texture alterations, and second, to evaluate implant surface temperature for principal implant systems using calibrated irradiation.

| Titanium dental implants
Titanium dental implants, representing three principal implant systems featuring different surface characteristics, were evaluated in vitro.
The sample size for the experimental part of the study, was based on findings obtained from the calibration part. To account for unexpected implant variation, we aimed at n = 10.

| Laser system
This study used a double laser that combines two wavelengths of laser light; one free running pulsed 2780 nm Er,Cr:YSGG laser and one 940 nm diode laser operating in continuous wave mode (Table 1; Biolase Inc., Irvine, CA).
The following settings were used during calibration and the experimental study: Er,Cr:YSGG-pulse duration 60 μs, repetition rate F I G U R E 1 Flow chart of study. Part 1: Calibrations of the double wavelength laser irradiation; Part 2: Experimental study using double wavelength laser irradiation 50 Hz (Al-Karadaghi et al., 2015;Gutknecht et al., 2016); 940 nm diode laser-continuous wave mode (CW). When water/air spray cooling was used, the device setting was 80% water-20% air.

| Output power
A calibrated thermal sensor (FL250A-BB-50, Ophir Photonics, Darmstadt, Germany) and universal power meter (Vega Standard, P/N 7Z01560, Ophir Photonics) were used to measure the output power from the laser handpiece fiber tip at a 1-mm distance without water/ air spray. With one laser wavelength (single) activated at each time, output power was measured for the pulsed Er,Cr:YSGG laser and for the 940 nm diode laser (CW) using 10 different power settings (1.0-4.25 W). Double wavelength irradiation was measured at 10 power settings for Er,Cr:YSGG laser from 1.0 to 4.25 W in combination with three diode laser settings (1.0 W; 2.25 W; 4.25 W) and compared to the same settings as single wavelength irradiation. In total, 50 different power settings repeated thrice were tested.

| Initial implant body temperature
One implant from each system was used for body temperature measurements following irradiation with Er,Cr:YSGG and 940 nm diode laser as single and double wavelength laser. An oscillating movement in a half turn pattern at a speed of approximately 8.2 mm/s over an area of 24 mm 2 was used to approximate clinical settings. The implants, attached to a hollow pin, were driven by an endodontic handpiece connected to an electric motor. One thermo-coupler was fixed in the center of the implant and the implant temperature adjusted to 37 C using a fan. Final temperature following a 30-s irradiation was displayed on a thermometer logger.
Three power settings for each laser wavelength (1.0 W; 2.25 W; 4.25 W) were tested at the fixed distance of 1.0 (±0.2) mm. The two wavelengths were tested as single and double. Water/air-spray was activated at all Er,Cr:YSGG irradiations. Diode laser irradiations were tested with and without water/air-spray. In total, for each implant, 18 settings repeated thrice were performed for the implant body temperature measurements.

| Implant surface micro-texture
A scanning electron microscope (SEM, Jeol JSM-7400F, Tokyo, Japan) was used to evaluate surface micro-texture alterations following irradiation with four (0.75 W-3.0 W) and two (1.00 W; 4.25 W) different power settings for the Er,Cr:YSGG and the 940 nm diode laser, respectively. Areas of non-irradiated implant were used as control.
The two lasers were tested for single and double irradiation.
Each implant was divided into four areas (Takagi et al., 2018). The 1.0-mm distance between the fiber tip end and implant surface was controlled using a calibrated USB-microscope (Dino-Lite/Europe, Naarden, Netherlands) and output power controlled using the calibrated thermal sensor/power meter. A Computer Numerical Control (CNC) (Lase-o-Matic, Viking; ILSD Sweden AB, Stockholm, Sweden) prototype device was used for repeatable oscillation of the implants.
The fixed laser handpiece in the CNC device had an angle of 90 relative to the implant surface, and a repeatable vertical movement pattern simulating clinical debridement. The speed of movement was calibrated to 3.3 mm/s, and each vertical step of every half turn was 0.5 mm. Irradiated area was 21 mm 2 (3 × 7 mm, height × width). The setup was designed to ensure that each mm 2 of implant surface was exposed to a minimum of 32 pulses from the Er,Cr:YSGG laser and 5 s of the continuous wave diode laser.
For the SEM evaluation, the implants were placed on sample studs and fully inspected at 5-10 kV at a magnification of ×50, ×190, and ×1500. Representative surface areas were recorded. Implant areas were then allocated into two groups based on the following surface descriptions: areas with no alteration (No surface alteration group) compared with control areas; areas with infractions, cracks, melting or ablations (Surface alteration group).

| Experiment
The experimental study was conducted under validated settings based on the calibrations. All implants were irradiated at an angle of 90 , four turns in horizontal-vertical direction on one side of the implants covering 42 mm 2 (7 × 6 mm, height × width). The speed of movement was horizontally 3.3 mm/s, and each vertical step of every half turn was 0.5 mm.
The start position of the fiber tip and the distance to the implant surfaces were adjusted, images recorded using the calibrated USB-T A B L E 1 Properties of the lasers (manufacturer's information)

| Statistical analysis
The methodology, results, statistical analysis and conclusions were reviewed, and all statistical analysis was performed blindly, by a professional statistician (SAL) who is also one of the study authors.
For calibration, mean output power for each power setting of double irradiation was compared with the sum of mean output power

| Implant surface micro-texture
The SEM evaluation of the implants following diode laser irradiation did not reveal any alterations of the surface texture at any power setting; two surface areas for each implant system in No surface alteration group (output power 0.9 and 3.3 W).
For Er,Cr:YSGG laser irradiation, surface micro-texture alterations appeared at different output powers pending implant system. In the No surface alteration group, there were one OsseoSpeed (1.1 W), one Roxolid SLActive (1.8 W) and two TiUnite areas (0.7 W and 0.9 W) ( Table 2).
In the Surface alteration group, there were three OsseoSpeed and

| Temperature experiment
For the experimental part, calibrated laser settings (Table 2)

| Calibration
The laser devices use optical fiber for light energy transmission with losses of light energy. In the calibration part, mean measured output power compared with device setting was approximately 20% lower for both Er,Cr:YSGG and 940 nm diode lasers. This is in accordance with manufacturer information and recent studies (Al-Karadaghi et al., 2015;Gutknecht et al., 2016). Nevertheless, we found differences related to power settings. At low power the differences were only 10% and at high power up to 24%, indicating that every output power/energy needs to be controlled and calibrated for each irradiation.
For the calibration of the Er,Cr:YSGG laser, only small differences were observed in initial temperature at different power settings following a 30-s irradiation. An inside implant temperature variation between 22 and 25 C was recorded. The findings are congruent to those in an in vitro study where a temperature of 20 C was measured following Er,Cr:YSGG laser irradiation on SLA titanium discs at a power setting of 0.5, 1.0 and 1.5 W and a distance of 0.5 mm when water/air-spray was activated (Strever et al., 2017).
With water/air-spray, 940 nm diode laser irradiation with different power settings only slightly influenced the surface temperature. In  contrast, without water cooling, a rapid increase in temperature was observed following increased output energy for all implant systems.
OsseoSpeed and TiUnite reached a temperature 45 C higher than Roxolid SLActive at output power of 3.3 W (power setting 4.25 W).
The laser energy absorption coefficient of the surface and the thermal conductivity of the core titanium material differ between the implant systems. Core material of OsseoSpeed and TiU are pure Grade IV titanium, whereas Roxolid SLActive is a titanium-zirconium alloy with a F I G U R E 4 SEM images of surface micro-texture alterations following single and double wavelength irradiation at different output powers of the Er,Cr:YSGG laser. Device setting in brackets. The 940 nm diode laser did not cause surface alterations at any output power. Minor observed signs of alteration (white circles); extensive alterations (white arrows). Magnification ×1500 T A B L E 3 Statistical analysis of mean max, mean temperatures, confidence interval and ICC values for five irradiations on 10 implants for each system, measured in the 20-190 s interval of double wavelength irradiation showing that the temperature is affected by the laser wavelength (Valente et al., 2017;Leja et al., 2013;Geminiani et al., 2012) and titanium surface composition (Giannelli et al., 2015;Matys et al., 2016).
These studies also report a positive association between increased power/energy density and increased temperature without watercooling, underlining the importance of water-cooling during irradiation of Er:YAG, Er,Cr:YSGG and /or diode lasers on titanium surfaces.
The highest Er,Cr:YSGG laser output power not causing microtexture alterations was 1.8 W for the Roxolid SLActive, 0.9 W for TiUnite, and 1.1 W for the OsseoSpeed surface. These findings indicate an implant system specific interaction between output laser energy and implant surface composition. In still another study, evaluating the effect of different laser wavelength on titanium discs, similar interaction between wavelength, surface alteration, surface chemistry and output energy was observed (Park et al., 2012).
To study thermal and decontamination effects of Er,Cr:YSGG (Gholami et al., 2018;Takagi et al., 2018) or Er:YAG lasers (Al-Hashedi et al., 2017;Larsen et al., 2017;Matys et al., 2016) on titanium surfaces, a distance of 0.5-1.5 mm between fiber tip end and implant surface has been used. Diode laser has been tested at a distance of 2-5 mm (Geminiani et al., 2012;Leja et al., 2013;Valente et al., 2017). These differences may greatly impact outcomes. In one study, robotic guidance used a custom computer-controlled program to regulate the movement in a bidirectional raster scan pattern for handpiece positioning and movement at a 0.5 mm distance (Strever et al., 2017). The present study used a computer numerical control

| Experiment
In the experimental part, following 190 s of irradiation, mean tempera-  (Romanos et al., 2017;Strever et al., 2017). Even though power/energy output was not reported, principal findings are congruent with the present study. Addressing the efficacy of debridement and disinfection of titanium surfaces described in recent studies on Er,Cr:YSGG (Ercan et al., 2014;Takagi et al., 2018) and diode lasers (Bach et al., 2000;Deppe & Horch, 2007), transmission through the fiber tip may decrease due to damages caused by disrupted particles from the titanium surface, and consequently, reduction of output energy may occur (Taniguchi et al., 2013). However, in the present study, evaluation of the fiber tip revealed only few cases with minor damage not affecting the output energy.
The authors acknowledge several limitations of the study. First, the calibration revealed that specific implant system breakdown thresholds and applied laser energies are not immediately applicable for other lasers or implant systems. Second, only three titanium dental implant surfaces out of a plethora of commercially available surface modifications were examined limiting the general applicability of the observations. Included implants were selected based on frequent use worldwide and representing different surface modifications. Third, a small sample size was applied during the calibration and the evaluation not qualitatively validated, the empirical value of the calibration is limited. In contrary, the sample size for the experimental study was oversized compared with the post-hoc sample size calculation. Moreover, surface micro-texture evaluation by SEM depicts only 2-d surface characteristics (Wennerberg & Albrektsson, 2009 Further in vitro and preclinical studies need to be undertaken before double wavelength laser irradiation is deemed reliable to be tested in clinical settings. A laboratory 3-d implant surface evaluation of the debridement and decontamination efficacy, as well as the biocompatibility of contaminated implants following the double wavelength laser irradiation, are warranted. In addition, improved fiber tip design, ensuring an optimized distance between the tip end and implant surface under sufficient water cooling, is also motivated.