Introduction
The purpose of this 2-arm-parallel split-mouth trial was to investigate the effect of low-level laser therapy (LLLT) on the repair of orthodontically induced inflammatory root resorption (OIIRR).
Methods
Twenty patients were included in this study, with 1 side randomly assigned to receive LLLT, and the other side served as a sham. Eligibility criteria included need for bilateral maxillary first premolar extractions as part of fixed appliance treatment. OIIRR was generated by applying 150 g of buccal tipping force on the maxillary first premolars for 4 weeks. After the active force was removed, the teeth were retained for 6 weeks. LLLT commenced with weekly laser applications using a continuous beam 660-nm, 75-mW aluminum-gallium-indium-phosphorus laser with 1/e 2 spot size of 0.260 cm 2 , power density of 0.245 W/cm 2 , and fluence of 3.6 J/cm 2 . Contact application was used at 8 points buccally and palatally above the mucosa over each tooth root for 15 seconds with a total treatment time of 2 minutes. After 6 weeks, the maxillary first premolars were extracted and scanned with microcomputed tomography for primary outcome OIIRR calculations. Subgroup analysis included assessment per root surface, per vertical third, and sites of heaviest compressive forces (buccal-cervical and palato-apical). Randomization was generated using www.randomization.com , and allocation was concealed in sequentially numbered, opaque, sealed envelopes. Blinding was used for treatment and outcome assessments. Two-tailed paired t tests were used to determine whether there were any statistically significant differences in total crater volumes of the laser vs the sham treated teeth.
Results
Total crater volumes were 0.746 mm 3 for the laser treated teeth and 0.779 mm 3 for the sham. There was a mean difference of 0.033 ± 0.39 mm 3 (95% CI, −0.21 to 0.148 mm 3 ) greater resorption crater volume in the sham group compared with the laser group; this was not statistically significant ( P = 0.705). No harm was observed. Conclusions: No significant difference was found between LLLT and sham control groups in OIIRR repair.
Highlights
- •
The effect of low-level laser therapy (LLLT) on healing of root resorption was assessed on extracted teeth.
- •
Root resorption crater volumes were similar in the LLLT and sham teeth.
- •
No differences were found per vertical thirds and tooth surfaces.
- •
LLLT did not seem to influence root repair after orthodontic force cessation.
Orthodontically induced inflammatory root resorption (OIIRR) has been described as an unavoidable pathologic consequence of orthodontic tooth movement. It is a result of collateral damage to the root surface by the action of clastic cells in the removal of necrotic tissue formed from periodontal ligament compression. Manifestations may be cemental or dentinal mineral loss, whereas circumferential resorption may precede apical root shortening. It has been reported that up to 91% of teeth can experience some form of root length loss during orthodontic treatment. A classification of severe resorption (>3 mm) has been reported in 10% to 20% of patients. When diagnosed, a 2- to 3-month treatment pause may be recommended, since significantly reduced resorption has been found in patients who had their treatment interrupted compared with those who had not. The continuation of OIIRR is thought to be related to the persistence of hyalinized tissue. The cessation of orthodontic force allows for the removal of necrotic tissue, while minimizing further formation, and also enables cemental repair.
Repair has been observed histologically to begin with the migration of fibroblast-like cells and cementoblasts into the resorption lacunae and the deposition of an unmineralized cementoid matrix with subsequent mineralization. This has been reported to occur at locations from the periphery, central part of the lesion, and all directions. A rapid healing potential was noted up to 7 weeks; then a plateau in the rate was seen.
A modality that is regaining popularity in research in the promotion of wound healing is low-level laser therapy (LLLT). It has had many synonyms such as “cold laser,” “soft laser,” and “LLLT.” Its initial discovery was in the late 1960s with the observation of accelerated hair regrowth and wound healing in mice. The proposed mechanism of action involves the stimulation of cellular metabolism near infrared or infrared light at a low energy density. Common delivery sources include laser or light-emitting diodes, and many parameters influence the quality, quantity, and density of the light energy delivered to the target tissue. These include wavelength, light source-power, and spot size, number and frequency of applications, application time, application energy, and fluence.
In response to LLLT, the effects of the in-vitro level have demonstrated improved cellular metabolism, and differentiation and proliferation of progenitor cells, osteoblasts, and cementoblasts. Clinically, LLLT has been shown to be favorable in the healing of aphthous ulcerations and effective in the prevention of oral mucositis. Hard tissue studies have indicated an overall positive effect for bone regeneration and repair and secondary dentinogenesis.
The effects of LLLT on cemental repair or remodeling have been investigated in a few studies with mixed results. Alsulaimani et al showed favorable histologic results with negligible radiologic changes, and Altan et al observed improved cemental repair in a rodent model after LLLT application during a retention phase after cessation of orthodontic force. However, no study has investigated the effects of LLLT on the repair of OIIRR in a human model after orthodontic force cessation.
Specific objectives or hypothesis
The aim of this study was to investigate the effect of LLLT application on cemental repair during a simulated treatment pause in a human model after a 4-week period of 150 g of buccal tipping force. The volumes of tooth root resorption craters were measured on extracted premolars by microcomputed tomography (micro-CT).
Material and methods
Trial design and any changes after trial commencement
This was a 2-arm parallel, split-mouth trial, with randomization of 1:1 right and left sides as LLLT treated or sham control. There were no changes after trial commencement.
Participants, eligibility criteria, and settings
Ethics approval was granted by Sydney Local Area Health District, Royal Prince Alfred Hospital Zone (ethics approval numbers X16-0276 and Human Research Ethics Committee/16/RPAH/347). Patients were screened and selected from February to November 2015 in the Department of Orthodontics, Sydney Dental Hospital in Australia, based on previously described criteria by all authors except one (P.P.). These included need for bilateral maxillary first premolar extractions (necessitating moderate anchorage) and fixed appliance treatment; permanent dentition; completion of apexification; similar minimal crowding on each side of the maxillary arch; no previous orthodontic or orthopedic treatment; no unilateral or bilateral posterior crossbites; no craniofacial anomaly; no history of trauma, bruxism, or parafunction; no past or present signs and symptoms of periodontal disease; no significant medical history that would affect the development or structure of the teeth and jaws and any subsequent tooth movement; and no history of asthma. Written informed consent was obtained, pretreatment records were collected, and standardized oral hygiene instruction was provided.
Interventions
Self-ligating 0.022-in SPEED brackets and tubes (Strite Industries, Cambridge, Ontario, Canada) were bonded on the first premolars and first molars, respectively, and 150 g of buccal tipping forces was applied to the first premolars via a 0.017 × 0.025-in beta-titanium cantilever (3M Unitek, Monrovia, Calif) from the first molar to the first premolar, bypassing the second premolar ( Fig 1 ). The force produced was verified with a strain gauge (Dentaurum, Ispringen, Germany). Additionally, occlusal stops (Transbond Plus Light Cure Band Adhesive; 3M Unitek) were placed in the occlusal grooves of the first and second molars to allow uninhibited tipping of the maxillary first premolars and to prevent any interference.
After 4 weeks, the brackets were removed, and the patients were switched to a retention phase with an 0.018-in stainless steel fixed retainer (3M Unitek) with loops bent at the terminal ends cemented to the buccal surfaces of the maxillary first molars, first premolars, and second premolars with composite resin (Filtek Supreme XTE Universal Restorative; 3M ESPE, North Ryde, Australia) ( Fig 2 ). This retention phase lasted 6 weeks, with the experimental side receiving a laser application at the start of each week. The laser used was a 660-nm, 75-mW aluminum-gallium-indium-phosphorus laser (Thor Photomedicine, Buckinghamshire, United Kingdom). In total, 8 points of contact application were used, with 4 on the buccal side, and 4 on the palatal side on the mucosa directly above the root surface of each tooth. The configuration of the points of application consisted of 2 points at the cervical portion (mesial and distal), 1 at midroot, and 1 at the apex of the tooth per buccal and lingual sides. Each application was 15 seconds with a total treatment time of 2 minutes. A continuous beam was used. The 1/e 2 spot size area was 0.260 cm 2 , and 1/e 2 power density was 0.245 W/cm 2 . Total time was 120 seconds, total power was 0.0637 W, total energy was 7.6 J, and fluence was 3.6 J/cm 2 . The LLLT regimen was performed at weekly sessions over a period of 6 weeks. The positive controls had a sham laser applied with identical settings on the laser unit except that output was blocked via a sham function on the device.
After 6 weeks in retention, the first premolars were extracted by the main operator (C.M.A.K.) under local anesthesia (lignospan lidocaine hydrochloride 2.2 ml, 2%, 1:100,000 adrenaline; Septodont, Saint-Maur-des-Fossés, France) only with forceps to minimize root damage caused by luxators. The teeth were stored in 10% neutral buffered formalin (Sigma-Aldrich, Castle Hill, Australia) and scanned with a desktop micro-CT machine (SkyScan 1172; Bruker, Aartselaar, Belgium). This unit produces high-resolution 3-dimensional images from multiple 2-dimensional x-ray shadow transmission images of the specimen at differing angulations. The specimens were scanned from at least the cementoenamel junction to the root apex, with rotation of 180° around the vertical axis and an image resolution of 17.6 μm. A copper-aluminum filter with x-ray source voltage of 100 kV and source current of 100 μA were used. Reconstruction of the SkyScan image data was undertaken with its proprietary software, NRecon (version 1.6.9.18; Bruker). This software produced a slice-by-slice axial reconstruction based on a modified Feldkamp cone-beam algorithm. The output produced was a 16-bit tagged image file format based on standardized settings in NRecon.
Outcomes (primary and secondary)
Root resorption crater detection and quantification were performed as previously described. This involved use of the imaging software program Fiji (version 2.0.0-rc-15/1.49k; available at https://imagej.net/Fiji ) and a custom macro (Enigma; Australian Centre for Microscopy and Microanalysis, University of Sydney). Detection of craters was done by manual examination of all axial slices of the image stack per specimen. Once detected, a crater was highlighted. The horizontal extent of the crater was recorded via an x-y coordinate system that was standardized for all axial slices. The number of slices of the image stack that the crater spanned indicated the corono-apical extent of the crater.
After this, axial slices within the image stack were segmented into binary format (black and white), differentiating tooth structure from air ( Fig 3 ). With the crater coordinate data in 3 planes, the macro Enigma calculated the volumes of each crater from the segmented axial images using an algorithm based on the convex hull method.
Furthermore, craters were categorized by tooth surface—buccal, palatal, mesial, or distal—and also by their vertical position—cervical, middle, or apical third of the root. This allowed calculation of total crater volumes for each tooth, each surface, each vertical third, and specific surfaces per vertical third such as the buccal-cervical and palato-apical, which receive the highest compressive forces during buccal tipping.
Sample size calculation
Sample size calculation was based on our previously published study, where more frequent LLLT applications were delivered to the treated subjects. Accordingly, a power calculation was carried out to determine an appropriate sample size for 2-sided paired t tests, significance level of 0.05, and power of 0.85 to ascertain a difference in mean root resorption of 0.15 mm 3 with a standard deviation of 0.21 mm 3 , representing a 30% difference between teeth using a within-subject split-mouth design. The sample consisted of 40 maxillary first premolars extracted from 20 patients (8 male, 12 female; mean age, 15 years 9 months; range, 13 years 8 months to 18 years 7 months). The extractions of the maxillary first premolars were required as part of their orthodontic treatment plan.
Interim analysis and stopping guidelines
Not applicable.
Randomization (random number generation, allocation concealment, implementation)
A split-mouth design was used. Maxillary first premolars were randomly allocated to 1 of 2 groups, with the left and right sides allocated into group A or B. The randomization scheme was generated by using www.randomization.com . Block sizes of 4, 6, 6, and 4 were used to maintain equal numbers of laser and sham treated sites between the left and right sides of the patients. All patients received LLLT on 1 side; by default, the sham was the other side. The letter (A or B) denoted in the randomization sequence which switch on the laser unit was to be set when applying the LLLT on the right side of the patient’s mouth. By default, the patient then received the alternate setting on the left side of the mouth.
The laser device (Thor Photomedicine) had a sham function that allowed the laser to beep when activated with no energy output. The mechanism consisted of 2 dials: an external dial switched between A and B, corresponding to the side of the mouth to be irradiated ( Fig 4 ), and a secondary internal dial allowed for the selection of whether A or B would be the sham setting. The internal dial was set once by an assistant other than the main operator and was only revealed at the end, after data analysis, allowing for allocation concealment, and subject and operator blinding ( Fig 5 ).
Orthodontic procedures and laser application were performed by 1 operator (C.M.A.K.) according to the randomization table generated by switching between A and B on the external dial according to whether the left or right side was being irradiated.
Blinding
Blinding of patient and operator was performed for LLLT application. Extraction of teeth and assessment were performed by the same LLLT operator. However, this person was blinded for treatment allocation, and the teeth were placed in coded vials. Codes were kept by another person who was not involved in these processes.
Statistical analysis (primary and secondary outcomes, subgroup analysis)
Two-tailed, paired t tests were used to determine whether there were any statistically significant differences in total crater volumes of LLLT vs sham treated teeth. Subgroup comparisons included total crater volumes per root surface (buccal, palatal, mesial, and distal), total volumes for each vertical third (cervical, middle, and apical), and buccal-cervical and palato-apical sites. The level of statistical significance set at P = 0.05 for all comparisons. Error of measurement was calculated by repeated measurements of 7 randomly selected teeth 30 days after the initial measurements to determine the overall standard error of measurement and the coefficient of variation. SPSS software, a statistics program (version 21; IBM, Armonk, NY), was used for statistical analysis.
Material and methods
Trial design and any changes after trial commencement
This was a 2-arm parallel, split-mouth trial, with randomization of 1:1 right and left sides as LLLT treated or sham control. There were no changes after trial commencement.
Participants, eligibility criteria, and settings
Ethics approval was granted by Sydney Local Area Health District, Royal Prince Alfred Hospital Zone (ethics approval numbers X16-0276 and Human Research Ethics Committee/16/RPAH/347). Patients were screened and selected from February to November 2015 in the Department of Orthodontics, Sydney Dental Hospital in Australia, based on previously described criteria by all authors except one (P.P.). These included need for bilateral maxillary first premolar extractions (necessitating moderate anchorage) and fixed appliance treatment; permanent dentition; completion of apexification; similar minimal crowding on each side of the maxillary arch; no previous orthodontic or orthopedic treatment; no unilateral or bilateral posterior crossbites; no craniofacial anomaly; no history of trauma, bruxism, or parafunction; no past or present signs and symptoms of periodontal disease; no significant medical history that would affect the development or structure of the teeth and jaws and any subsequent tooth movement; and no history of asthma. Written informed consent was obtained, pretreatment records were collected, and standardized oral hygiene instruction was provided.
Interventions
Self-ligating 0.022-in SPEED brackets and tubes (Strite Industries, Cambridge, Ontario, Canada) were bonded on the first premolars and first molars, respectively, and 150 g of buccal tipping forces was applied to the first premolars via a 0.017 × 0.025-in beta-titanium cantilever (3M Unitek, Monrovia, Calif) from the first molar to the first premolar, bypassing the second premolar ( Fig 1 ). The force produced was verified with a strain gauge (Dentaurum, Ispringen, Germany). Additionally, occlusal stops (Transbond Plus Light Cure Band Adhesive; 3M Unitek) were placed in the occlusal grooves of the first and second molars to allow uninhibited tipping of the maxillary first premolars and to prevent any interference.
After 4 weeks, the brackets were removed, and the patients were switched to a retention phase with an 0.018-in stainless steel fixed retainer (3M Unitek) with loops bent at the terminal ends cemented to the buccal surfaces of the maxillary first molars, first premolars, and second premolars with composite resin (Filtek Supreme XTE Universal Restorative; 3M ESPE, North Ryde, Australia) ( Fig 2 ). This retention phase lasted 6 weeks, with the experimental side receiving a laser application at the start of each week. The laser used was a 660-nm, 75-mW aluminum-gallium-indium-phosphorus laser (Thor Photomedicine, Buckinghamshire, United Kingdom). In total, 8 points of contact application were used, with 4 on the buccal side, and 4 on the palatal side on the mucosa directly above the root surface of each tooth. The configuration of the points of application consisted of 2 points at the cervical portion (mesial and distal), 1 at midroot, and 1 at the apex of the tooth per buccal and lingual sides. Each application was 15 seconds with a total treatment time of 2 minutes. A continuous beam was used. The 1/e 2 spot size area was 0.260 cm 2 , and 1/e 2 power density was 0.245 W/cm 2 . Total time was 120 seconds, total power was 0.0637 W, total energy was 7.6 J, and fluence was 3.6 J/cm 2 . The LLLT regimen was performed at weekly sessions over a period of 6 weeks. The positive controls had a sham laser applied with identical settings on the laser unit except that output was blocked via a sham function on the device.
After 6 weeks in retention, the first premolars were extracted by the main operator (C.M.A.K.) under local anesthesia (lignospan lidocaine hydrochloride 2.2 ml, 2%, 1:100,000 adrenaline; Septodont, Saint-Maur-des-Fossés, France) only with forceps to minimize root damage caused by luxators. The teeth were stored in 10% neutral buffered formalin (Sigma-Aldrich, Castle Hill, Australia) and scanned with a desktop micro-CT machine (SkyScan 1172; Bruker, Aartselaar, Belgium). This unit produces high-resolution 3-dimensional images from multiple 2-dimensional x-ray shadow transmission images of the specimen at differing angulations. The specimens were scanned from at least the cementoenamel junction to the root apex, with rotation of 180° around the vertical axis and an image resolution of 17.6 μm. A copper-aluminum filter with x-ray source voltage of 100 kV and source current of 100 μA were used. Reconstruction of the SkyScan image data was undertaken with its proprietary software, NRecon (version 1.6.9.18; Bruker). This software produced a slice-by-slice axial reconstruction based on a modified Feldkamp cone-beam algorithm. The output produced was a 16-bit tagged image file format based on standardized settings in NRecon.
Outcomes (primary and secondary)
Root resorption crater detection and quantification were performed as previously described. This involved use of the imaging software program Fiji (version 2.0.0-rc-15/1.49k; available at https://imagej.net/Fiji ) and a custom macro (Enigma; Australian Centre for Microscopy and Microanalysis, University of Sydney). Detection of craters was done by manual examination of all axial slices of the image stack per specimen. Once detected, a crater was highlighted. The horizontal extent of the crater was recorded via an x-y coordinate system that was standardized for all axial slices. The number of slices of the image stack that the crater spanned indicated the corono-apical extent of the crater.
After this, axial slices within the image stack were segmented into binary format (black and white), differentiating tooth structure from air ( Fig 3 ). With the crater coordinate data in 3 planes, the macro Enigma calculated the volumes of each crater from the segmented axial images using an algorithm based on the convex hull method.
Furthermore, craters were categorized by tooth surface—buccal, palatal, mesial, or distal—and also by their vertical position—cervical, middle, or apical third of the root. This allowed calculation of total crater volumes for each tooth, each surface, each vertical third, and specific surfaces per vertical third such as the buccal-cervical and palato-apical, which receive the highest compressive forces during buccal tipping.
Sample size calculation
Sample size calculation was based on our previously published study, where more frequent LLLT applications were delivered to the treated subjects. Accordingly, a power calculation was carried out to determine an appropriate sample size for 2-sided paired t tests, significance level of 0.05, and power of 0.85 to ascertain a difference in mean root resorption of 0.15 mm 3 with a standard deviation of 0.21 mm 3 , representing a 30% difference between teeth using a within-subject split-mouth design. The sample consisted of 40 maxillary first premolars extracted from 20 patients (8 male, 12 female; mean age, 15 years 9 months; range, 13 years 8 months to 18 years 7 months). The extractions of the maxillary first premolars were required as part of their orthodontic treatment plan.
Interim analysis and stopping guidelines
Not applicable.
Randomization (random number generation, allocation concealment, implementation)
A split-mouth design was used. Maxillary first premolars were randomly allocated to 1 of 2 groups, with the left and right sides allocated into group A or B. The randomization scheme was generated by using www.randomization.com . Block sizes of 4, 6, 6, and 4 were used to maintain equal numbers of laser and sham treated sites between the left and right sides of the patients. All patients received LLLT on 1 side; by default, the sham was the other side. The letter (A or B) denoted in the randomization sequence which switch on the laser unit was to be set when applying the LLLT on the right side of the patient’s mouth. By default, the patient then received the alternate setting on the left side of the mouth.
The laser device (Thor Photomedicine) had a sham function that allowed the laser to beep when activated with no energy output. The mechanism consisted of 2 dials: an external dial switched between A and B, corresponding to the side of the mouth to be irradiated ( Fig 4 ), and a secondary internal dial allowed for the selection of whether A or B would be the sham setting. The internal dial was set once by an assistant other than the main operator and was only revealed at the end, after data analysis, allowing for allocation concealment, and subject and operator blinding ( Fig 5 ).
