Introduction

Dental enamel manifests high transparency in the near-IR (NIR). Our previous work demonstrated that NIR light at 1310-nm is ideally suited for the transillumination of interproximal dental caries (dental decay in between teeth) and that it can also be used to image decay in the pits and fissures of the occlusal (biting) surfaces of posterior teeth where most new dental decay occurs, and even image interproximal lesions from the occlusal surface [1-3]. Transillumination systems for dentistry have been under development for many years and such systems can be used for imaging through the thin enamel of many anterior teeth such as incisors. However, the strong light scattering of sound dental enamel at visible wavelengths, 400-700 nm inhibits imaging through more than a mm or two of the tooth in the visible range, as we have shown in transillumination measurements performed on simulated interproximal lesions comparing visible light, 830-nm and 1310-nm wavelengths for different thicknesses of enamel from 2-7-mm [4-8]. Healthy enamel is transparent in the NIR near 1310 nm and does not strongly scatter light whereas demineralized enamel, cavitated and carious lesions manifest reduced transmission and exhibit strong light scattering respectively [9].

Studies over the past 25 years have demonstrated that lasers can be used for the precise removal of carious and noncarious dental hard tissue [10-12]. The “free-running” Er:YAG, Er:YSGG, and Nd:YAG solid state lasers with pulse durations ranging from 150-300 µs have been approved by the FDA for caries removal. Another laser, such as the CO₂ laser operating at 9.3 and 9.6-µm has very strong absorption in the apatite mineral phase of hard tissue allowing efficient ablation. CO₂ lasers are available with a large range of operating parameters and can be operated efficiently at high repetition rates. The thermal relaxation time for CO₂ laser pulses absorbed at the enamel surface at 9.3 and 9.6-µm is on the order of a few microseconds, therefore CO₂ laser pulses should be in the microsecond range to minimize peripheral thermal damage and for the optimum ablation efficiency. If excessive peripheral thermal damage is produced during the ablation process, thermal damage can cause stress cracks, discoloration of the enamel and disproportion of the crystalline apatite structure that leads to the formation of white asperities around the ablation crater composed of a more acid soluble calcium phosphate phase composition [13]. A water-spray helps prevent peripheral thermal damage and the accumulation of such undesirable phases on the enamel surface. In order to effectively develop CO₂ lasers it is important to understand the laser conditions that cause peripheral thermal damage and establish the safe operating parameters such as laser pulse energy and repletion rate that can be used safely along with the required rate of water delivery needed to alleviate any thermal effects.

Enamel is 2% by volume protein and lipid, 8% by volume water and 90% by volume carbonated hydroxyapatite mineral. The mineral phase is localized in acicular enamel apatite crystals organized into prism or rod-like structures a few microns across that run the length from the tooth surface. Most of the water and organic content are localized to the interprismatic boundaries and about half of that water is mobile. Since water is the only mobile phase, any reversible optical changes in enamel must be due to varying levels of hydration. Thermal damage to the enamel prismatic structures and the interprismatic protein and lipid causes irreversible changes to the enamel birefringence and the magnitude of light scattering.

Peripheral thermal and mechanical damage after laser irradiation is typically viewed by taking cross sections through ablation craters and incisions in the hard tissue and measuring changes using polarized light microscopy, infrared spectro-microscopy and micro-Raman spectroscopy [14-18]. This approach however requires thin sectioning of the samples and can cause mechanical damage. Dehydration of sample thin sections and stresses generated during handling can also cause microcracks to form. Therefore, it is a challenge to determine with any certainty whether the observed defects are caused by the laser or by the sectioning procedure.

During a recent study involving the NIR imaging of simulated interproximal lesions that were created by drilling holes into the proximal surface of the tooth with a high-speed hand-piece [19], we noticed that peripheral stress cracks were clearly visible within the tooth around the area of the holes caused by excessive frictional heating from the drill. Those observations suggest that NIR imaging may be ideal for the visualization of peripheral thermal damage during laser ablation as the ablation craters evolve in real-time. This approach is analogous to the old practice of using transparent blocks of plexiglass to acquire 3-D profiles of the ablation craters and CO₂ laser mode patterns [20]. However, we have the capability of examining crater formation directly in the tissue of interest, dental enamel and we can directly visualize any peripheral thermal changes and crack propagation. Moreover, there is typically strong emission during ablation of hard tissues due to the very strong calcium atom and ion emission lines in the visible particularly near 600-nm. Such strong emission does not interfere at 1310-nm in the NIR. Therefore, the NIR is better suited for imaging crater evolution in enamel.

In this paper we have coupled NIR imaging with a high-speed scanning CO₂ laser ablation system to demonstrate that ablation crater formation in enamel can be imaged with high contrast as the laser selectively removes the hard tissue. These real-time images are useful: for the rapid evaluation of laser ablation rate and efficiency, can be used to indicate if stalling occurs during multiple pulse irradiation, and can show any peripheral crack formation due to acoustic and thermal transients and allows visualization of reversible and permanent changes in enamel due to thermal modification and dehydration.

1. Materials and methods

2.1 Tooth samples

Twenty one tooth sections approximately 3-mm thick were prepared from extracted human third molars that were sterilized by gamma irradiation and stored in water with only 0.1% thymol added to inhibit bacteria and fungal growth. Surfaces were serially polished to a finish of 0.1-µm using embedded diamond polishing discs. A sample section is shown in Fig. 1 along with a video taken at multiple angles that demonstrate the high transparency of enamel at 1310-nm.

2.2 Near infrared imaging (NIR)

A three component, modular 7X NIR precision zoom lens consisting of a focusable lower module, a manual upper zoom module and a 1.5 TV tube from Edmund Scientific (Barrington, NJ) was used with the InGaAs focal plane array (FPA). The field of view was adjustable from 0.81 to 5.6-mm, the magnification from 1.7X to 11.8X, and the Numerical aperture from 0.036 to 0.12. Light from a 150-W fiber-optic illuminator FOI-1, E Licht Company (Denver, CO) coupled to an aperture and a 90-nm bandpass centered at 1310-nm (filter # BP-1300-090-B) Spectrogon US Inc., (Parsippany, NJ) was used to illuminate the samples. An InGaAs FPA (318×252 pixels), the Alpha NIR™ (Indigo Systems, Goleta, CA) was used to acquire all the images. The acquired 8-bit or 12-bit digital images were analyzed using IRVista™ software (Indigo Systems, Goleta, CA).

 

Fig 1. (1.4MB) Movie of a NIR image of a 3 mm rotating tooth section showing the high-transparency of dental enamel. [Media 1]

Download Full Size | PDF

2.3 Ablation apparatus

The setup showing the respective positions of the tooth section, laser hand-piece, light-source, camera and lens and water-spray nozzle are shown in the image of Fig. 2. A transverse excited atmospheric pressure (TEA) CO₂ laser, Impact 2500 (GSI Lumonics, Rugby, United Kingdom) operated at 9.3 µm, was used to irradiate tooth samples with incident fluence up to 45 J/cm², and energies of 20 – 30 mJ per pulse with a pulse duration of 16 µs. The laser energy was calibrated and measured using a laser energy/power meter, EPM 1000, Coherent-Molectron (Santa Clara, CA) with an ED-200 Joulemeter from Gentec (Quebec, Canada).

The laser was focused with a plano convex ZnSe lens of 125-mm focal length to a beam diameter of approximately 300-µm. The laser beam diameter (1/e²) at the position of irradiation was determined by scanning with a razor blade across the beam. Two and three dimensional images of the laser spatial profile was acquired using a Spirocon Pyrocam™ I pyroelectric array (Logan, UT). Both laser beam profile and spatial profile showed that the laser was operated in a single spatial mode, i.e., Gaussian spatial beam. Holes were produced in samples at either a fixed position or by scanning the laser-beam at a rate of 1.5-mm/sec by fixing the laser-hand-piece to a computer-controlled linear stage. Repetition rate of 20 to 300-Hz were used for the experiments. Sub-ablative measurements were performed by defocusing the laser beam incident on the tooth surface. A low volume/low pressure air-actuated fluid spray delivery system consisting of a 780S spray valve, a Valvemate 7040 controller, and a fluid reservoir from EFD, Inc. (East Providence, RI) was used to provide a uniform spray of fine water mist onto the enamel surfaces at 2 mL/min.

 

Fig. 2. Imaging Setup on left and schematic diagram on right. (A) InGaAs FPA with zoom lens, (B) water spray nozzle, (C) laser hand-piece, (D) fiber-optic illuminator, (E) sample holder and tooth section.

Download Full Size | PDF

2.4 Experimental procedures

Twelve ablation holes were drilled in the first five teeth with the CO₂ laser at a fluence of 40 J/cm² and repetition rates ranging from 20 Hz to 100 Hz without a water spray. Sixteen incisions were produced on nine teeth by scanning the laser beam across the tooth section at a rate of 1.5 mm/sec at repetition rates ranging from 20 to 300 Hz with and without water spray. The CO₂ laser was defocused so that non-ablative irradiation intensities were incident on the tooth surface ablation causing only localized heating on seven tooth sections that were irradiated with a repetition rate of 100 Hz.

2. Results

For the initial samples, the position of the laser was fixed and conical holes were drilled into the tooth to a depth of approximately 1 - 2 mm without a water spray. One such hole is shown in Fig. 3. Videos were acquired at ten frames per second during ablation and the rate of progression of the holes and peripheral changes in the opacity of the enamel could be followed in real-time. Thermal emission was observed from the area of ablation and the rate of propagation of the crater slowed considerably after penetrating to a depth greater than a mm as the intensity of the incident laser beam was spread out over the large conical area of the ablation crater. The hole in Fig. 3(C) was ablated at a repetition rate of 20 Hz for approximately 25 seconds. The ablation crater extends approximately 1.5 mm into the enamel. There is high contrast between the sound enamel and the ablation crater and there is no peripheral thermal damage visible.

Figure 4 shows incisions made with the CO₂ laser at a repetition rate of 30 Hz without a water spray. The laser beam was scanned back and forth over a scanning distance of 2 mm at a rate of 1.5 mm/sec. The growth of the crater was more rapid at each end of the incision due to the longer residence time of the scanning stage at those endpoints. There were no visible changes in the surrounding sound enamel during ablation at this repetition rate. However, if the water spray was not employed at higher repetition rate, the surrounding enamel became opaque after ablation and cracks were clearly visible propagating away from the site of ablation. At repetition rates of 60 Hz or greater without a water spray, extensive cracking and thermal damage was observed in the three samples ablated at those conditions. Incisions were made at repetition rates of 50 and 100 Hz employing a water spray and no cracking or thermal damage was observed in the NIR and reflected light images.

 

Fig. 3. (2.7 MB) Movie (A) NIR image of side of tooth section before ablation. (B) NIR image of side of tooth section with CO₂ laser irradiation plume. (C) NIR image of side of tooth section with an ablation craters produced at a repetition rate of 20 Hz without a water spray. The entrance crater diameter is approx. 300 µm and the depth is approx. 1.5 mm. (9.5 MB version). [Media 2][Media 3]

Download Full Size | PDF

 

Fig. 4. (2.3 MB) Movie (A) NIR image of the side of a 3 mm thick tooth section as it is being drilled by the CO₂ laser with a repetition rate of 30 Hz, a fluence of 40 J/cm² without a water spray. (B) Final frame of a NIR image extracted from the video showing that there are no optical changes in the sound enamel peripheral to the incision. (C) Reflected light image of the approx. 1 mm incision length and approx. 2 mm depth after ablation showing no thermal damage. (5.2MB version). [Media 4][Media 5]

Download Full Size | PDF

Histological examination of ablation cross sections in prior ablation studies have never shown permanent thermal damage extending such a large distance from the zone of ablation. Therefore, we suspected that optical changes observed peripheral to the ablation craters might be due to the rapid dehydration of the highly mobile water present in the enamel. To test this hypothesis we moved the hand-piece away from the tooth samples to defocus the laser beam so that the laser pulses were non-ablative and could only heat and melt the enamel surface. The tooth sections of eight samples were subsequently irradiated at a repetition rate of 100-Hz without a water spray to melt and deposit a large amount of heat at the site of irradiation. During laser irradiation, waves of thermal emission are visible in the video propagating in a radial pattern out into the enamel from the site of energy deposition on the surface. After passage of the waves of enhanced thermal emission, the enamel area irradiated became opaque. This can be seen in the video and the frames extracted at different time points shown in Fig. 5. In the first image, Fig. 5(A), the enamel is completely transparent. A large crack was formed and the propagation of that crack was visible in the acquired video as it took approximately one second to propagate across the enamel section to terminate at the dentinal-enamel junction as shown in Fig. 5(B). In Fig. 5(C), the enamel is brighter than the surrounding enamel due to the thermal emission coming from within the sample. The thermal emission was spread out over the entire enamel area and the enamel became opaque after irradiation due to changes in the optical properties. In the last frame of the video, shown in Fig. 5(D), the entire enamel has become darker and the area underneath the irradiated surface is completely opaque. In order to determine if that opaque area represents dehydration of the mobile water from the enamel, the sample was placed in water for 24-hours and NIR images were acquired for comparison with the images collected before and after laser irradiation. The NIR images are shown in Fig. 5 before and after laser irradiation and after rehydration in water for 24-hours. After rehydration, Fig. 5(E), shows that most of the opaque zone has become transparent again with the exception of the immediate area below the irradiated enamel surface that still remains dark. Moreover, there are radial cracks circumscribing the irradiated area that are now visible close to the surface in Fig 5(F). Therefore, large areas of enamel have become opaque in the NIR at 1310-nm due to dehydration of the mobile water and that dehydration is reversible.

 

Fig. 5. (1.4 MB) Movie (A) NIR image showing the high initial transparency of the enamel through a 3 mm thick tooth section prior to CO₂ laser irradiation. (B) NIR image of crack extracted from the video during the first second of irradiation with non-ablative CO₂ laser pulses at a repetition rate of 100 Hz. (C) Thermal emission outlined by yellow lines is visible propagating through the tooth enamel. (D) NIR image after passage of that thermal emission, the enamel becomes opaque due to tissue dehydration or permanent changes to the protein, lipid or mineral. (E) NIR image after tooth section was left for 24 hours in deionized water showing rehydration of most opaque areas. (F) Reflected light images of tooth showing cracks propagating outward in a radial pattern from the irradiated area. (3.6 MB version). [Media 6][Media 7]

Download Full Size | PDF

4. Discussion

Laser ablation craters could be imaged in real-time through sound human tooth sections with considerable detail during their formation due to the high-transparency of enamel at 1310-nm. The evolution of the ablation craters and laser-incisions were clearly visible until they reached the dentin-enamel junction. The propagation of thermally driven cracks were observed and thermally induced changes in the enamel such as dehydration and permanent changes to protein, lipid and mineral were observed around the ablation craters during laser irradiation. Samples were also irradiated with non-ablative laser pulses and subsurface changes in the enamel opacity were visible induced by thermal and stress transients in the samples. Thermal changes occurred to depths exceeding 1-2 mm and some opacities were reversible after rehydration of the samples by immersion in water for 24-hours. This observation suggests that the observed reversible changes were caused by dehydration of the mobile water located along enamel prism/rod boundaries. To the best of our knowledge these are the first observations of dehydration causing changes in the optical attenuation of light in dental enamel. Since water is the only mobile phase in dental enamel, the increased opacity is likely due to the formation of pores in between the enamel prisms after loss of water and those pores may act as scattering centers due to the high refractive index mismatch between air and water. Permanent changes of enamel transparency are most likely due to changes in the tightly bound water (waters of hydration) and denaturation of the lipid and protein. The most severe thermal changes on the surface of the tooth lead to disproportionation of the mineral phase and loss of the apatite crystalline structure. Therefore, this study shows that there is great potential for NIR imaging to monitor laser-ablation events in real-time to: assess safe laser operating parameters by imaging thermal and stress-induced damage, and help elaborate the mechanisms that play a role during ablation, e.g., dehydration of the mobile water in the enamel. Additional studies are needed to determine the nature of the temperature gradients present in the enamel that give rise to the permanent and reversible optical changes occurring in the NIR.

There have been several studies regarding tissue dehydration and the effect on the Er:YAG laser ablation rate [21-23]. Water absorption and diffusion studies in enamel indicate that only approximately half of the water is actually diffusible ²⁴ and the rate for water diffusion is quite slow, on the order of several hours to days. Thermal analysis studies indicate that the tissue has to be heated to temperatures exceeding 200–300°C before all the diffusible water is removed [25]. Higher temperatures of up to 800°C are required to remove the more tightly bound water [25]. None of these studies have reported the rapid dehydration that we observe in these imaging studies.

One of the challenges in designing safe laser ablation systems is to determine the optimum-operating regimen for efficient hard tissue removal. This requires a balance of the rate of energy deposition primarily determined by the single laser pulse energy and the pulse repetition rate with the rate of water-cooling. However, the incident laser fluence, the depth of cut, crater morphology, geometry and the amount of water present on the sample surface or in the ablation crater can profoundly influence the ablation rate and efficiency and lead to stalling that in turn leads to increased heat accumulation and thermal damage. The influence of all these parameters is difficult to predict and NIR imaging provides a very rapid method for assessing the potential for thermal effects. For example, the use of 50-Hz with water-cooling produced clean ablation with little thermal damage. However, there was an obvious decrease in the ablation rate and efficiency with increasing depth due to the increasing aspect ratio of the ablation crater causing the incident laser fluence to decrease as the energy is spread over a greater area of the conical hole. At higher pulse repetition rates thermal changes can be seen spreading from the base of the crater indicating excessive heat buildup as ablation stalls and the heat accumulation accelerates. Thermocouple measurements in whole teeth carried out using pulse repetition rates from 50-300-Hz with the laser scanned along one axis confirm the observations in this study, namely that ablation stalls as the crater reaches a certain depth and aspect ratio and we found that the higher repetition rates with a similar length of cut the volume of tissue removed did not increase as anticipated, since stalling occurred and the additional laser pulses contributed to heat accumulation. These results indicate that the laser needs to be scanned in two dimensions if high repetition rates are used, not only to ensure efficient ablation but to also avoid excessive heat accumulation that may cause pulpal overheating and injury. Further work will elaborate on the rate of water delivery on the ablation process and peripheral thermal damage at high repetition rates.

As new imaging modalities become available to the dental community such as fluorescence based methods, optical coherence tomography and the new NIR method presented here, there is increasing potential to integrate these methods with laser ablation systems for controlled and selective removal of dental caries (dental decay). The promising results of this study demonstrate the potential of NIR imaging for this task. Since the position of the laser beam and carious lesions are visible in the transparent enamel, it is likely that NIR imaging can be used to monitor caries removal with the laser, and future studies will be carried out to investigate this approach to monitor the removal of demineralized enamel. One can envision that a NIR assisted vision system analogous to image-intensified viewing systems could be developed to aid the clinician in selective removal.