Role of Photobiomodulation Therapy in Modulating Oxidative Stress in Temporomandibular Disorders. A Systematic Review and Meta-Analysis of Human Randomised Controlled Trials

This systematic review and meta-analysis (PROSPERO registration; ref CRD 42020198921) aimed to govern photobiomodulation therapy (PBMT) efficacy in temporomandibular disorder (TMD). PRISMA guidelines and Cochrane Collaboration recommendations were followed. Differences in pain reduction assessment by qualitative measurement with visual analogue scale (VAS), pain pressure threshold (PPT) and maximum mouth opening (MMO) were calculated with 95% confidence intervals and pooled in a random effects model with a subgroup analysis, evaluating the role of follow-up duration. Heterogeneity was analysed using Q and I2 tests. Publication bias was assessed by visual examination of funnel plot symmetry. Qualitative analysis revealed 46% of the 44 included studies showed a high risk of bias. Meta-analysis on 32 out of 44 studies revealed statistically significant intergroup differences (SSID) for VAS (SMD = −0.55; 95% CI = −0.82 to −0.27; Z = 3.90 (p < 0.001)), PPT (SMD = −0.45; 95% CI = −0.89 to 0.00; Z = 1.97 (p = 0.05)) and MMO (SMD = −0.45; 95% CI = −0.89 to 0.00; Z = 1.97 (p = 0.05)), favouring PBMT compared to control treatment strategies. Sensitivity analysis revealed SSID (SMD = −0.53; 95% CI = −0.73 to −0.32; Z = 5.02 (p < 0.0001)) with low heterogeneity (Τ2 = 0.02; χ2 = 16.03 (p = 0.31); I2 = 13%). Hence, this review, for first time, proposed suggested recommendations for PBMT protocols and methodology for future extensive TMD research.


Introduction
TMD is considered as one of the main triggers in inducing orofacial pain of non-dental origin, which can have a negative impact on a patient's functionality and psychological status [1,2], which ultimately can affect quality of life (QoL) [3]. Hence, the International Association for the Study of Pain (IASP) has defined TMD as a set of clinical conditions with signs and symptoms in the masticatory muscles, temporomandibular joint (TMJ) and associated structures (fatigue or stiffness of the jaws) and pain on palpation of the masticatory muscles [4].
The pathogenesis and aetiology of TMD are complex and not clearly understood. The production of free radicals, proinflammatory cytokines, nociceptive neuropeptides [5] and growth factors can lead to inflammation, pain and progressive tissue changes [6]. Due to the complexity of TMD pathophysiological, several treatment modalities have been Photobiomodulation (PBM) (laser or light-emitted diodes (LEDs)) therapy has gained interest as a non-invasive tool with immunomodulatory, anti-inflammatory and bioregenerative effects in stimulating healing, relieving pain and reducing inflammation [18,19]. The light of the optical window between 600 and 1200 nm is absorbed by cells' chromophores. The cytochrome c oxidase (CCO) of mitochondrial electron chain transport complex IV absorbs photonic energy of wavelengths from 600 to 900 nm, whereas longer wavelengths are understood to be absorbed by water and light-sensitive ion channels. PBM can also reduce the oxidative stress by dissociating the inhibitory nitric oxide (NO) from CCO [20]. PBM modulates OS and reactive oxygen species (ROS)-mediated signalling in TMD management. PBM laser therapy is proven to enhance perfusion to bone and joint structures, stimulate osteoblasts [21] and chondrocytes, reduce the inflammatory cytokines and chemokines, increase anti-inflammatory cytokines and reduce nerve stimulation [21,22], promote endogenous opioids release, enhance tissue healing, increase angiogenesis, augment muscle tensile strength [23], increase the pain threshold by affecting the cellular membrane potential and decrease inflammation, possibly due to the reduction in prostaglandin E2 and suppression of cyclooxygenase 2 (COX-2) levels [24,25] (Figure 1). Moreover, PBM enhances the lymphatic system response [26].
Utilisation of PBM therapy (PBMT) for TMD management can be an effective treatment in modulating pain intensity by reducing inflammation while exhibiting regenerative and analgesic effects [27,28]. However, its clinical efficacy has been controversial due to the diversity of the reported results. Many systematic reviews and meta-analyses have analysed PBMT effectiveness in musculoskeletal disorders [29,30], but only three relatively focused on TMDs in which effective dosage and power density were analysed to evaluate the reduction in TMD-induced pain [31][32][33]. Additionally, the use of LED PBM has captured the interest of many investigators, due to its easy administration over a large surface area [34]. Hence, it is significantly important to explore its use for TMD, as different light sources and intensities might produce different or better results [35,36]. As little light has been shed on this matter, it was included in the present review. Clinicians, scholars and investigators have collectively concluded within the current available RCTs that further robust double-blind RCTs with vigorous standardised methodology and laser dosimetry are needed.
The rationale of conducting the present systematic review and meta-analysis was in the line of the above-mentioned key factors.
The present systematic review and meta-analysis aimed to evaluate the existing contextual scientific evidence, justifying the gaps in the literature and building a conceptual framework to govern the efficacy of PBMT in TMD management.
The objectives of this research review are listed below: 1.
To explore the basis of and extrapolate the reasons for the inconsistencies among the data.

2.
To evaluate the sensitivity of the results' methods of assessment and obtain vigorous standardised methodology.

3.
To attempt to propose a preliminary empirical consensus of PBM laser and LED dosimetry and treatment protocols.
The highlights of this research review are listed below: 1.
The pathogenesis and aetiology of temporomandibular disorder (TMD) are complex and not clearly understood.

2.
Oxidative stress and reactive oxygen species play a vital part in TMD pathogenesis and its progression. 3.
Photobiomodulation (PBM) therapy is an effective treatment modality, as monotherapy of various light sources of single wavelength or in combination of two wavelengths, in improving chronic pain, functionality, anxiety/depression and, subsequently, quality of life in patients with TMD.

4.
This review, for the first time, addressed the standardisation of methodology and PBM protocols by proposing suggested recommendations, which can only be used to pave the roadmap for future extensive research in the management of TMD chronic symptoms.

Protocol and PROSPERO Registration
This systematic review was conducted in accordance with the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement and Cochrane Handbook for Systematic Reviews (Supplementary File S1) [37,38]. Review protocol is published in Prospective Register Of Systematic Reviews (PROSPERO); ref CRD 42020198921.

Focused Questions of Review Search
The capsulised focused research review questions were as follows: 1. Does PBM with laser or LEDs or combined therapies have superior effects compared to placebo or TMD standard care, or combined therapies (PBMT and standard care), in reducing pain intensity or improving patients' functionality and psychological status, as well as QoL? 2.
Does combined laser PBM therapy of red and IR wavelengths provide synergistic effects compared to placebo? 3.
Is it possible to propose clinical guidance and recommendations of PBMT (LEDs and laser) for TMD management?
"Temporomandibular Joint Disorders" (MeSH Major Topic) AND "Myofascial pain" [MeSH Major Topic]) AND "Humans" (MeSH) AND "LLLT" (MeSH Major Topic) OR "Photobiomodulation" (MeSH Major Topic) OR "PBM therapy" (MeSH Major Topic). Studies of in vivo human randomised controlled trials (RCTs) (split-mouth, parallel or prospective), comparing the efficacy of PBMT to any other standard care treatment modalities or combined therapies (PBM and one standard care treatment).

3.
Studies investigating the effects of PBMT on TMD symptoms; chronic pain for ≥3 months in TMJ or masticatory muscles, loss of movement or masticatory malfunction for at least 3 months were included. 4.
Light sources: laser or light-emitting diodes (LEDs) with no wavelength restrictions.

5.
Studies reporting at least one of the following parameters as an outcome variable: pain score, functionality score, qualify of life or immunological profile. 6.
Studies reporting any of the following outcomes: immediately after treatment, middle of treatment and end of treatment. 7.
If multiple terms of outcomes were reported within one period, a period the closest to two weeks, one month, three months and six months for each follow-up timepoint respectively used. 9.
All timepoints assessments of additional outcomes ( Studies utilised home or stellate ganglion or acupuncture PBM (laser or LEDs) approach.

2.
Other neuropathic orofacial pain conditions not related to TMD.

3.
Studies utilised pharmacotherapy as a primary outcome.

5.
Physiological or systematic conditions contributing to the pain.
Subjects with mental illnesses which could affect the clinical picture of patients, cervical disc herniation, history of trauma, TMJ surgery, musculo-articular pathologies, history of facial trauma, TMJ disc or condyle erosion, fibromyalgia, removable denture, missing more than one tooth in each quadrant and major malocclusion (anterior open bite, maxillary unilateral lingual cross-bite and overjet greater than 6 mm). 8.
Subjects with active head and neck malignant tumours. 9.
Pregnant and lactating women. 10. Subjects who underwent treatment for headache or bruxism in last 6 months prior to their enrollment in RCT study 11. Studies utilised homeopathic therapy as a comparative therapy. 12. Narrative and systematic reviews, case reports, in vitro studies, in vivo animal studies, commentaries, interviews, updates or case series. 13. A necessity of initiating the use of any type of medications during any of phase of the study. 14. Studies investigating acute TMD or acute versus (vs.) chronic TMD.

1.
Functional improvement (muscles movements: mouth opening and closing and chewing) from baseline up to the end of follow-up.

2.
Reduction in anxiety/depression and improved QoL from baseline up to the end of follow-up.

Data Extraction
A careful selection of the eligible studies from the search engines was carried out by two reviewers independently (R.H. and S.D.). They performed the review, assessment and data extraction for each eligible study. Each study received an identification with the name of the first author, year of publication and origin. A tabular representation of additional relevant information, such as the impact factor of the journal, study design, sample size, participants' demographical data, baseline characteristics, intervention and comparator groups, type of light sources (laser or LEDs), number and location of trigger points (TP), utilised laser parameters, treatment frequency, follow-up duration, methods of outcome evaluation, statistical tests performed, results and conclusions, were harvested from each eligible study.

Qualitive Analysis
Two review authors (R.H. and S.D.) independently assessed the risk of bias for each RCT study included in this review, using the criteria described in the revised Cochrane Risk-of-Bias (RoB) tool for randomised trials, version 2.0 (RoB 2) [42][43][44]. The performed assessment of risk of bias was based on the following domains:

1.
Bias arising from the randomisation process.
Bias due to missing outcome data.

4.
Bias in measurement of the outcome.

5.
Bias in selection of the reported result.
Contingent on fulfilment of the above criteria, the chosen studies were governed as low, moderate or high RoB. Any incongruities between the two review authors' assessments were settled via discussion with a third author (S.B.), as well as the "discrepancy check" feature in RoB 2, which was used to obtain consensual answers for the quality assessment of included studies.

Statistical Analysis of Data
A meta-analysis of the data of interest extracted from the eligible studies was performed, using RevMan (Version 5.4.1) [45]. A random effects meta-analysis for continuous outcomes was conducted to assess the heterogeneity amongst the included studies. Relevant data on the primary outcome VAS and secondary outcomes, maximum mouth opening (MMO) and PPT, was extracted from the included studies. Data were collected from the baseline evaluation up to the final follow-up evaluation of the eligible studies. Treatment effects were calculated through pooled standardised mean differences (SMDs) with associated 95% confidence intervals (95% CIs) and pooled overall effect was considered statistically significant when p < 0.05 [46]. Forest plots were visually inspected to identify statistical heterogeneity through the presence of outlier studies [46]. Additionally, I 2 statistics for homogeneity that ranged from 0-100% with the following interpretation: 0% = no evidence of heterogeneity; 30-60% = moderate heterogeneity; and 75-100% = high heterogeneity, were calculated [47]. In order to evaluate the results after negation of heterogenous studies, a sensitivity analysis was performed [48]. The presence of publication bias was analysed by visual assessment of funnel plot symmetry [49].

Racial Background
For the purpose of this systematic review and meta-analysis, relevant reported data on patients' racial background in the included studies was sought as follows: Black, Black/Caucasian, non-Caucasian. All the included studies failed to provide information regarding the phenotypic characteristics of the recruited subjects, apart from one study [88], which reported their subjects' racial backgrounds under the category of Black/Caucasian (n = 44, 42 "White" patients, 2 "Black" patients).
With regards to the duration of presented pain symptoms prior to treatment, seven studies reported chronic pain over six months [57,62,65,69,71,74,90], whereas two studies reported pain for three to six months [93,96].
In terms of allocation of symptoms/affected area at baseline, four out of 44 studies specified unilateral TMJ symptom presentation [65,67,72,92], whereas three studies specified bilateral TMJ involvement [58,76,77]. Interestingly, three studies reported inclusion of patients with either unilateral or bilateral TMJ involvement [57,88,100], which were as follows; in one study, five patients in each group (two groups with 15 patients in each, n = 30) presented with unilateral TMJ involvement, and the remaining 20 patients showed bilateral TMJ involvement [57]; in one study, seven patients (n = 44) presented with unilateral TMJ involvement and the rest of the patients showed bilateral TMJ involvement [88]; and in one study, 16 patients (n = 44) presented with unilateral TMJ involvement and the rest of the patients showed bilateral TMJ involvement [100].

Study Design
All the included studies were conducted using a parallel study design (Table S1).

Intervention Group
Several inconsistencies were observed amongst the included studies , in terms of the intervention groups, which are outlined in Table S1.
In terms of LED PBMT, one study utilised 660 nm [94]. Two studies [75,95] utilised LEDs (red and infrared) combined with infrared laser, as follows: 630 ± 10 and 850 ± 10 nm red LEDs and 780 nm infrared laser [75], whereas the other study utilised a 905 nm laser diode combined with LEDs of eight clustered heads, of which four were 640 nm and the other four were of 875 nm infrared [95].  [63,77,78] and super-pulsed mode only in one study [70].
Pulse Width (s, µs) and Frequency (Hz) The frequency for laser-PBMT in pulsed mode was mentioned in three studies [63,67,77] in the range between 80 and 1500 Hz. The frequency for laser PBMT in super-pulsed mode was mentioned in one study [70] and ranged between 1 and 50 kHz. The frequency values for a combined super-pulsed laser and pulsed LED PBM that were found in one study [95] were 1000 and 2-16 Hz, respectively. Only two studies [70,77] reported a pulse width of <200 nanoseconds (ns) and 1 microsecond (µs), respectively.

Duration of the Treatment
The total duration of the PBMT was reported in all included studies with a range of four days to eight weeks.

Spot Size/Spot Area/Beam Diameter/Fibre-Tip Diameter Parameters
These parameters were addressed in the included studies, as follows; spot size in the range of 0.02 to 2.8 cm 2 in nine studies [32,33,35,36,67,73,77,[82][83][84]100] and 5 mm in one study [84], spot area of 0.028 cm 2 in one study [80], beam diameter in six studies [69,71,74,81,97,100] in the range of 3.5 to 8 cm, beam area in three studies [61,67,97] in the range of 0.09 to 1 cm 2 and tip diameter 300 µ in one study [85]. Additionally, two studies [59,95] mentioned the probe aperture size, which ranged from 4 to 0.2 cm 2 . An illuminated area of 0.5 cm 2 was mentioned in one study [75], while the irradiated area was mentioned in two studies [97,100] in the range of 0.5 to 0.7 cm 2 /point.
Around the TMJ region, the number of the TP were between three and five points/side in three studies [75,89,90], of which in one study [89] the five-TP distribution was as follows: superior, anterior, lateral, posterior and postero-inferior to the condyle, whereas the distributions of five points/side of the other study [90] were related to the lateral pole of the mandible head as follows: lateral, superior, anterior, posterior and inferior regions. One study did not provide the exact allocation of three TP on TMJ [75].
In terms of the TMJ itself, the distribution of TP was as follows: the joint capsule (five points (lateral, posterior, superior, anterior, inferior)/side) in one study [73], the condylar region (five points (anterior, superior, posterior and posteroinferior points of the condylar position and in the external auditory meatus)/side) in one study [74] and pre-auricular region and external auditory meatus in four studies, of which two studies mentioned the application of one point inside the external auditory duct towards the retrodiscal region [60,61], one study applied four points/side (TP and allocation unspecified) [88] and one study failed to provide any of the above relevant information [59]. Furthermore, one study applied one point/side on superior SCM [71] while one study applied irradiation on SCM without specifying the exact allocation or number of TP [79].
In the line of the extrapolated data of laser documentations, it is noteworthy that five studies [57,60,64,70,97] recruited subjects with mixed TMD aetiology (Section 3.2.6), and did not apply different laser protocols for each category. Additionally, they did not identify the number of subjects according to aetiology category.

Qualitative Assessment
All included studies were assessed for their quality, using the RoB 2 tool designed for in vivo human RCTs, as shown in Figures 3 and 4. This tool is the latest version, which was used to execute a qualitative assessment [34][35][36][37][38][39][40][41]. Figure 3 is an illustration of the RoB assessment summary of all the eligible studies, whereas Figure 4 is a domain-wise graphical representation of percentage RoB score evaluated using this tool. Both figures represent the consensual answers verified using the "discrepancy check" feature of the RoB 2 tool, across two independent reviewers (R.H. and S.D.) (inter-reviewer agreement, κ = 0.92). Of the included trials, 40% were at low risk of inadequate randomization, whereas 32 and 28% of the included trials had some concerns or were at high risk, respectively. Of the included studies, 59% were at low risk of deviations from intended interventions, whereas 39 and 2% were at high risk or had some concerns, respectively.
3.6. Quantitative Assessment 3.6.1. Outcome Variables The treatment outcomes were broadly based on three categories; pain reduction, functionality improvement and anxiety reduction/QoL improvement (Table S3). These outcomes were assessed with qualitative and quantitative measures.
A meta-analysis on other secondary outcomes could not be conducted owing to the disparity in scoring methodology and incomplete or incomparable numerical data.
A test of subgroup differences revealed very low heterogeneity amongst the shortand long-term evaluation protocols (χ 2 = 1.06 (p = 0.30); I 2 = 5.3%), which indicated that the length of follow-up evaluation was not a contributing factor towards the highly heterogenous results obtained in the abovementioned meta-analysis (Figure 8).

Publication Bias
Visual inspection of funnel plots for pain reduction assessment by VAS, PPT and functionality improvement assessment by MMO revealed only mild asymmetries, respectively, indicating a not significant risk of publication bias in this meta-analysis (Figures 10-12).

Discussion
This systematic review and meta-analysis scrutinised RCTs that recruited subjects diagnosed with TMD, according to RDC/TMD [32,33], and treated them with PBMT of various wavelengths, as a single or dual treatment modality compared to the placebo or several conventional standards of care, or different laser PBM or LED PBM. Despite the inconsistencies and diversities in the reported PBM parameters, variable assessment tools, missing data and variable study designs, the majority of the included studies reported positive results in favour of the effectiveness of PBMT. Hence, this systematic review and meta-analysis, for the first time, has addressed the methodology and PBM protocols standardisation by proposing suggested recommendations, which can only be used to pave the roadmap for future extensive research in TMD management. In this context, our in-depth review has revealed the following important lacunae and the evidence-based science and practice to bridge them.
In terms of racial background, only one study by Borges et al. (2018) reported the characteristics of the recruited subjects' racial backgrounds under the category of Black/Caucasian (n = 44: 42 "White" and 2 "Black" patients) [88]. However, they failed to utilise different laser parameters and the number of subject allocations was uneven. The remaining 43 studies failed to report the racial background of their recruited subjects. It is noteworthy that skin colour plays an important role in the scattering and absorption of the photonic energy [110,111], which should be considered when a PBM (laser or LEDs) parameter protocol is formulated. Regrettably, this was not employed by Borges et al. (2018); hence, their results cannot be reproducible [88].
The above notes highlight the possibility of discrepancy and diversity in the included studies' findings. Hence, standardisation of an equal number of recruited subjects of the same racial background, gender, suitable age range and muscle volume are the most fundamental key factors for recruited TMD population characteristics. Based on this, the PBM laser parameters can be formulated.

Evaluation of Study Design
The investigators need to pre-determine the eligibility criteria for the population included in the trial. Interestingly, three RCT studies [57,88,100] included in their inclusive criteria TMD subjects with unilateral and bilateral TMJ symptoms; however, the numbers of recruited subjects were uneven. The distribution of the recruited subjects in relation to these parameters were as follows: 10 [100]. This can have a great impact on the reported outcomes; however, the findings were positive. The majority of the high risk of bias in the quality of available evidence was derived from randomisation processes (28%, arising from 12 out of 44 studies) [58,59,62,[76][77][78]80,[84][85][86][87]98]. Hence, a robust randomisation process is required to ensure validation of the findings.

Diagnostic Criteria
Within our review eligibility criteria, we included all the studies that utilized RDC/TMD tools to assess TMD symptoms. In this review, two studies [61,84] mentioned symptoms related to the neck; however, these were not addressed for PBM applications. Interestingly, two studies [71,79] did not report the neck symptoms of their participants, but the TP of the cervical muscles were irradiated. The remaining 40 studies neither reported neck symptoms nor if the TP of the cervical muscles were irradiated.
Based on the above notes, it is important to highlight the factors that contributed to the inconsistency of findings, as follows:

1.
Despite RDC/TMD being the most common TMD diagnostic tool utilised by various researchers, there are three fundamental downsides to it, which are as follows [112]: (a) Relatively limited use due to the diversity of clinically presented TMD symptoms. Therefore, this tool can be utilised but many of the presented symptoms will not fit in one or any category. (b) They do not account for cervical spine involvement, which is crucial to thorough evaluation and management. (c) Various identified patients' experiencing pain symptoms such as hyperalgesia and/or allodynia should be managed accordingly, despite the fact that these variables are not addressed by RDC/TMD criteria.

2.
Regardless of if TMD participants have symptoms in the cervical regions, a full clinical assessment is required and the cervical muscles' TP need to be addressed in the irradiation protocol. This is due to an altered neuro-biomechanical function of the cervical spine, which can apply stress on the TMJ, causing TMD. There is evidence to support that the craniomandibular region and upper cervical spine are related from anatomical, biochemical and neurophysiological standpoints [113,114], due to a neuroanatomical link between the orofacial and cervical regions, as well changes in the isometric strength of cervical flexors, according to the bite position of TMD patients [115,116]. Hence, the manual muscle test (MMT) is a reliable and useful clinical diagnostic tool for cervical muscles assessment to be considered [117].
Based on the above notes, a combination of RDC/TMD and MMT tools in TMD diagnosis and measurement of the presence and severity of symptoms could offer a standardised consensus among researchers and clinicians.

Assessment of the Outcome Measures
In many previous studies of PBMT in TMD pain patients, the focus has been on jaw movements or self-reported pain intensity; however, a few studies have determined changes in patients' somatosensory function using more objective testing methods [118,119]. Simple rating scales, such as NRS or VAS have been considered the most reliable tools to evaluate self-reported pain intensity in clinical practice; however, this approach may be an oversimplification of complex biopsychosocial pain problems and may possibly result in an underestimation or overestimation of reported pain [19]. All the included studies in this review utilised either of the above pain assessment tools. The present review's authors recommend utilisation of the core outcome measures for clinical trials of chronic pain treatment efficacy and effectiveness by considering the Initiative on Methods, Measurement and Pain Assessment in Clinical Trials (IMMPACT) recommendations when studying patients with chronic pain (pain, physical functioning, emotional functioning, participant ratings of improvement and satisfaction with treatment, symptoms and adverse effects and participant disposition) [120,121].
None of the included studies in this review considered utilising immunological and quantitative synovial fluid analysis, beta-glucuronidase, IgA or IgG, which demonstrates elevated levels of inflammatory mediators in diseased joints compared with asymptomatic non-diseased joints [124,125].
Comprehensive instruments of outcome measure assessment are vital to augment the potential of optimal clinical outcomes. Hence, this review's authors categorised the variables' evaluation tool instruments into quantitative and qualitative, summarised in Table S2. Additionally, the IMMPACT II tool was used to develop consensus reviews and recommendations for improving the design, execution and interpretation of clinical trials of treatments for pain [126]. This can provide insight for investigators to employ in future studies.

Role of Tissue Optical Properties in Determining the Therapeutic Dosimetry
Light transportation in the tissues is a complex process, owing to multiple scattering and absorption, as well to the optically anisotropic biological tissue [127]. This largely limits the ability to focus a beam into deeper tissue, yet sometimes this scattering is desired [128]. In therapeutic applications, absorption is the primary mechanism to generate physical effects, and light scattering arises from the presence of heterogeneities within a bulk medium [127]. The scattering signals can be utilised to determine optimal light dosimetry [128]. The principal parameters that affect scattering are as follows: wavelength of red to near infrared (NIR) in the range 650 to 1200 nm, relative refractive index and particle radius [129,130]. An in vivo study by Alvarenga et al. (2018) utilised semiconductor 660 nm laser irradiation trans-gingivally, which resulted in an attenuation of light intensity of 50% at 5 mm in depth [131]. Hence, a great consideration of light attenuation is transmucosal, whilst setting optimal parameters is required. In this context, all the included studies in the present review failed to consider the optical properties of the affected trigger point areas with different structural organisation, when PBM irradiation was applied with an EO or cervical (transcutaneous) or IO (transmucosal) approach or any combination. This would ultimately have a great impact on the optimal irradiation doses and a significant effect on the clinical outcome optimisation. Surprisingly, the results are positive in favour of PBMT.
In lieu of energy loss as described, we can anticipate that the results of the included studies were either overinterpreted or underrepresented. In this context, in order to achieve standardisation and reproducibility of PBM parameters, an increase in the surface applied dose of around 10 times for sub-surface targets is recommended [132]. It is projected that at a depth of 1 cm, there will remain around 5-10% of the surface NIR photons arriving at the target tissue at that level [133,134].

Evaluation of EO (Transcutaneous) and IO (Transmucosal) PBM Therapy Approaches
EO applications of PBM have some clinical advantages regarding convenience of use due to intraoral discomfort and treatment of a large surface area. However, the dose delivery of EO approaches as well as the treatment dosimetry to the oral tissues remain challenging and require further investigation. PBMT increases the lymphatic flow, which reduces oedema and decreases prostaglandin E2 and cyclooxygenase-2 levels [25]. Under the surface of the skin at 1 cm depth, the intensity of a laser is reduced to 10% of its value. Therefore, a laser with a power density of 100 mW/cm 2 at the skin surface would be 10 mW/cm 2 at 1 cm below, and at 2 cm below would be 1 mW/cm 2 [135]. This would provide an insight in formulating the therapeutic power output and effective fluence, taking into consideration the utilised light source.

Evaluation of Reported Laser Treatment Parameters
Some of the key factors determining the light depth of penetration are as follows: target optical properties (consistency, structure, thickness, skin colour, absorption/scatter coefficient), light source wavelength, shape of laser beam, duration of irradiation exposure and tissue pressure. Table 2, reported by Jenkins et al. (2011), highlights the essential and desirable laser treatments that should be reported to standardise the clinical examination of TMD patients, improve methodology reproducibility among clinicians and facilitate the comparison of results among researchers [136]. Table 2. The essential and desirable PBM parameters that should be reported to standardise the laser protocol for TMD patients, improve methodology reproducibility among clinicians and facilitate the comparison of results among researchers [136]. All the abbreviations in this A further impact of the acceptance of reproducibility is the incomplete reporting of parameters in the literature. Bearing this is in mind, the authors of this review aimed to provide clinical PBM protocols guidance based on evidence derived from the literature and evidence-based and expert opinions, and is intended only for further research standpoints. We synthesised an eclectic assortment of experimental laser parameter protocols from 11 studies of this review, which utilised power meters to measure the therapeutic power output [61,65,67,70,73,75,77,83,88,89,95] as outlined below.

Utilisation of a Single Wavelength in Test Group
Out of the 11 studies, eight [61,65,70,73,77,83,88,89]  The subjects' symptoms in this study were associated with TMJ, temporalis and masseter muscles, as well as neck regions associated with mandibular dysfunction (restriction of MM); however, of the five TP, four were mapped in the shape of "a cross" in the pre-auricular region and the other one in the external auditory meatus, bilaterally. The authors only addressed the TMJ region but failed to address the TP of the masticatory region (IO and EO), as well the cervical muscles, knowing the patients experienced restriction in MM. Moreover, the beam area of 0.039 cm 2 is a very small spot size to utilise for a large area to irradiate. Nevertheless, a significant reduction (p < 0.05) in pain level in PBM group was observed, compared to the placebo.
It is noteworthy that placebo/sham PBM, as a comparable arm in TMD study design, is essential to validate the optimal outcome [137].
A study by Marini et al. (2010) [70] utilised the following protocol: 910 nm, superpulsed, 400 mW, range of 1 to 50 KHz (three steps on TMJ region: 1.2 KHz for 10 min, 1.2 KHz for 5 min and 3.16 KHz for 5 min) and pulse width of <200 ns, five days per week for two weeks (10 sessions). There were many missing data, such as dose, spot size and irradiance. Moreover, there were bilateral PBM applications on the TMJ region without specifying the number and distribution of the TP. The PBM group's findings showed a statistically significant improvement in pain intensity on VAS and functional movements (MMO), compared to drug (ibuprofen) and sham LLLT groups at all timepoints and at the one-month follow-up. Despite this, the study showed potential in their findings but the missing data compromised the potential of its reproducibility. Only one of the included studies (power meter usage was not specified) conducted by Lassemi et al. (2008) [63], utilising 980 nm, 80 Hz and 2 J/point, reported pain reduction and jaw clicking, which were maintained up to one year, but regrettably, essential parameters such as power output, dose/point, spot size, pulse width and irradiance were unreported.
Interestingly, a study by Emshoff et al. (2008) [65] utilised the following laser parameters: 632.8 nm, CW, 30 mW, contact, 1.5 J/cm 2 /point and 120 sec, two to three times a week for eight weeks, and their findings showed no difference between LLLT and placebo at eight weeks in reducing pain, for which LLLT was no better than the placebo at reducing pain. These findings were supported by Venacio et al. (2005) [57]. Nevertheless, taking into account the tissue optical properties, and the 632.8 nm shallow penetration depth in tissue, not surprisingly, insignificant pain reduction in the long term was noted. It is noteworthy that the following data were unreported in Emshoff et al. (2008): energy, beam area, spot size, irradiance, tissue-spot distance and neither the number of TP nor the allocations specified. Hence, it is difficult to utilise this protocol for future studies.
A study by De Carli et al. (2012) [73] used combined therapy of PBM and piroxicam and both therapies alone. The PBMT protocol was as follows: 808 nm, 2.8 J/point (total 56 J), 100 mW/point, 0.028 cm 2 , 100 J/cm 2 /point and 28 sec, twice a week (four sessions), with 30 days follow-up. Bilateral EO PBM applications of 10 TP/side were as follows: TMJ and muscle points on each side; joint capsule (lateral, posterior, superior, anterior, inferior), masseter (origin, insertion) and temporal (anterior, middle, posterior). The findings revealed a significant reduction in pain on VAS (p < 0.05) but combined therapies were not more effective than single therapy in TMD. This study was at a low risk of bias. Interestingly, Brochado et al. (2018) [89] utilised 808 nm as well comparing combined PBMT and manual therapy (MT) to a single therapy, with the following laser protocol: CW, 4, 48 or 576 J/point, 100 mW, contact, 0.03 cm 2 , 13.3 J/cm 2 (total: 133 J/cm 2 ), 3.33 W/cm 2 , 40 s/point, 12 TP/side and bilateral application employed (five in the TMJ region (superior, anterior, lateral, posterior and postero-inferior to the condyle) and seven for masticatory muscles (temporalis (anterior, middle and posterior), masseter (upper, middle and lower portion) and insertion of the medial pterygoid), three times a week, for four weeks (12 sessions) with various timepoints of treatment assessment (7th, 14th, 21st and 28th day) and follow-up (two and three months post-treatment). The study's findings showed a reduction in pain in all treatment groups compared to baseline, but this was maintained only in the PBMT and PBMT and MT groups over the follow-up period. Depression and anxiety were reduced in the both the PBMT alone and PBMT and MT groups. Nevertheless, the latter did not show a superior effect, compared to PBMT alone. This study was at a low risk of bias.
Studies Interestingly, in both studies there were no follow-up timepoints after the completed treatment. They only assessed the variable during the treatment duration, which was one day [83] and three weeks [88]. The findings, on the other hand, reported no statistical difference between PBM and the placebo groups in terms of passive or active mouth opening (p ≥ 0.05), pain improvement over the masseter muscle and total pain [83], whereas a study by Borges et al. (2018) [88] reported pain reduction in all the groups; however, a statistically significant increase in MOM and an improvement in mandibular protrusion was only observed with fluence values of 1-8 J/cm 2 . The latter study is of significance, as it tested three PBM doses to identify the optimal results' values. This study's drawback was a lack of follow-up timepoints to determine the long-term effects of PBMT. In the context of PBM dose, a study by da Silva et al. (2012) [74] (power meter usage was not specified) tested two fluence values (780 nm, 70 mW, CW, contact, beam diameter-5 mm, GI: 52.5 J/cm 2 , 30 s, GII: 105 J/cm 2 , 60 s, twice a week for five weeks) to identify optimal clinical outcome compared to the placebo. The findings show a statistically significant reduction in pain on VAS and an increase in the mandibular range of movements, favouring a high fluence of 105 J/cm 2 ; the drawback of this study was a lack of follow-up after the completed treatment.

Utilisation Two Laser Wavelengths in Test Group
One study out of 11 conducted by Shirani et al. (2009) [67] used two combined laser wavelengths; 660 and 890 nm, in CW and pulsed mode, respectively, in the test group, compared to the placebo, based on the following laser protocols: spot size 0.6 cm 2 , 1 cm 2 irradiated area, extraoral masseter muscle and intraoral MPM and LPM TP (number and distribution of TP unspecified), in contact, mild tissue pressure applied, twice a week for three weeks (six sessions in total), 660 nm (CW, 6-10 J/point, 17.3 mW, six mins exposure time) and 890 nm (pulsed, 1500 Hz, 9.8 W, 1 J/cm 2 , 10 min exposure time). In this study, the recruited subjects were only females with a mean age of 23.8 years old. Positive results were reported in using combined GaAs and InGaAlP lasers, usually applied for deep-lying disorders and superficial disorders, respectively, for pain reduction, compared to placebo. This was supported by a study conducted by Wang et al. (2011) [138] applying GaAlAs at 650 and 830 nm, implying that the combination of two laser wavelengths may be beneficial to patients with TMD.
Based on the above notes, red wavelengths are more easily absorbed by tissue surface components than infrared wavelengths. The absorbed energy can be dissipated in the form of heat around the skin surface. Conversely, the infrared wavelengths can penetrate deeply into the body tissues, which have lower scattering and absorption properties [139]. In this context, different wavelength utilisation, red and infrared radiation, can act on different sites of the tissue. Red light acts on the mitochondria, whereas IR acts on both mitochondrial and cellular membranes [140]; the combined effects of both wavelengths can be advantageous for tissue biomodulation. Future studies should be performed combining red and infrared radiation for TMD treatment.

Mixed light Sources in the Test Groups
Out of the 11, two studies [75,95] utilised the following various light sources in the test groups. A study by Panhoca et al. (2013) [75] employed the following protocols for three recruited groups: red (630 ± 10 nm) (9 J/point,18 J/cm 2 ,150 mW, 300 mW/cm 2 /point, 0.5 cm 2 ), IR (850 ± 10 nm) LEDs, (9 J/point, 18 J/cm 2, 150 mW, 300 mW/cm 2 /point, 0.5 cm 2 ) and IR laser (780 nm) positive control (4.2 J/point, 70 mW, 0.04 cm 2 , 1.7 W, 7 cm 2 /point, 105 J/cm 2 ), 60 s/point exposure time, five EO TP and contact, twice a week for eight weeks (total of eight sessions). A significant reduction in pain and an increase in MOM for all groups (p < 0.05) was reported. There was no significant difference in the pain scores and maximum oral aperture between groups at baseline or any periods after treatment (p ≥ 0.05). All the fluences of 18 and 105 J/cm 2 were shown to be effective. Interestingly, Panchoca et al. (2015) [75] chose the applied dose based on typical doses used in regular practice of around 100 J/cm 2 , which was shown to increase the muscle activity when irradiated with LED PBM therapy compared with the placebo. In this context, a study by Pöntinen et al. (2000) showed that a fluence of 4 J/cm 2 at skin level has maintained irradiance at depths in the range of 0.5 to 2.5 cm [141]. Following irradiation of joints or muscles, a fluence of 100-300 J/cm 2 was attenuated to 2 J/cm 2 , and irradiance can be maintained at certain depths [141,142].
A recent study by Herpich et al. (2020) [95] utilised a nine-clusters head of the following various light sources: one super-pulsed diode laser (905 nm), four red LEDs (640 nm) in pulsed mode and four IR LEDs (875 nm) in pulsed mode (Table S1). This study aimed to assess the influence of an intraoral approach of PBMT in a female cohort (30 in total with mean age of 31.7 ± 5.2 years old). Significant differences between groups were found regarding pain (p ≤ 0.01) and functioning (p ≤ 0.04), compared to sham. However, no statistically significant difference was found regarding range of mandibular motion (p > 0.05). This implies the advantages of utilising combined laser and LED PBM of red and IR, with different emission modes of super-pulsed and pulsed, in patients with TMD. Important to note is that the aperture size measured 4 cm 2 , but an adapter with an aperture of 0.394 cm 2 was used for IO application.
As all the included studies did not specify the beam profile, we assume it is Gaussian, which is of a very small spot size, and the distribution of the energy is uneven at the peripheral area, receiving <50% of that at the centre. To overcome this challenge, PBM handpieces were developed to optically correct the beam profile to a flat-top shape with colimitation distribution, an evenly uniform delivered energy over 1 cm 2 [132].
The above included studies [61,65,67,70,73,75,77,83,88,89,95] in this review made efforts in utilising the power meter to determine therapeutic power output to achieve consistent and optimal outcomes; however, they have proven elusive, due to the diversity of their findings. However, the dual wavelengths and the use of laser and LED light sources have provided a perspective for suggested clinical guidance for future extensive studies, taking into consideration the necessity for the development of LED and laser prototype devices with the same geometry and parameters, in order to achieve standardisation and results validation.
Based on the above-gathered notes, the authors have answered the focused review questions and have proposed suggested recommendations for clinical PBMT protocols for future extensive TMD RCT studies, which are grounded in the current available evidencebased clinical practice and experts in the field (Table 3). Additionally, a summary of the influencing factors that are important for TMD study design and recommendations suggested for future reproducible methodology are illustrated in Table 4. The latter would ultimately assist in paving the roadmap for scientific consensus in TMD management.

Randomisation and blinding processes
Two independent blind investigators to assess the variables at all timepoints (double-blind) and record the data. Robust randomised process. Parallel arm study design.
Refer to Section 4.2.1

Comparable arms of the study
Placebo/sham PBM, as a comparable arm in TMD study design, is essential to validate the optimal outcome. It assists in providing standarised and reproducible data [137]

Standardised laser protocol
Based on gathered evidence-based practice and science, 11 out of 44 studies utilised power meter and placebo/sham PBM; recommendations of PBMT protocols suggested for further research Refer to Table 4 and Section 4.5

Light device standardisation
Standardised prototype development for each involved light source in the study

Route of delivery of PBM irradiation
Identifying the consistency, structure, thickness, skin colour and absorption/scattering coefficient of the target tissues (optical properties) of each of the following PBM delivery rout(s), prior to setting up the PBM parameter protocols: Transmucosal approach (IO) Transcutaneous approach (EO) Combination of the above approaches Refer to Section 4. 4 and Table 4  Table 4. Cont.

Key Factors Suggested Recommendation
Description and Citation of Scientifc Evidence (Manuscript)

Unilateral/bilateral PBM irradiation
Regardless of unilateral or bilateral TMD symptoms, PBM irradiation needs to be applied bilaterally, due to the concept of compensation effects [147] Trigger Points (TP) Affected sites Number of the TP/site All the TP need to be addressed as follows: EO and IO muscles' contribution to TMD, including the masticatory muscles, as well as the cervical (palpable or unpalpable). The number of the TP depends on the origin and insertion of each muscle, its location and volume Refer to Section 4.5.4 and Figure 13 The investigated variables and follow-up timepoints

Evaluation of the Trigger and Palpable Points and Applications of Irradiation
It is very clear from our results that there are inconsistencies and discrepancies in the number, allocation and distribution of TP among the majority of the included studies. Additionally, there is a lack of consideration of the cervical muscles, such as SCM, trapezius and belly of the digastric muscles noted, despite the fact there is a correlation between the level of tenderness of masticatory and cervical muscles and jaw dysfunction and neck disability [143,144]. Disability has a great impact on patients' QoL [145]. Hence, irradiation of the cervical muscles needs to be part of the TMD PBM irradiation protocol.
Based on the above notes, the authors of this review have extrapolated from this review's findings the required TP and their allocations for PBM irradiation to be considered, as a guide for future extensive studies ( Figure 13).
The results of this meta-analysis show that in terms of reducing pain intensity and improvement in mouth opening, PBMT shows statistically significant superior results in comparison to control intervention groups. Presence of a potential conflict of interest was observed in two out of 44 studies [59,95].

Role of Meta-Analysis Outcome
The present systematic review and meta-analysis was based on the hypothesis that PBMT with LEDs or lasers is as beneficial as or superior to placebo (sham PBM), the pharmacological approach, the cognitive approach, physiotherapy, conservative treatment modalities (occlusal splint), ultrasound, TENS, alpha-lipoic acid or needle therapy, in terms of pain intensity reduction (chronic pain), functional improvement (masticatory malfunction) or anxiety/depression improvement, or QoL in patients with TMD. Consequently, a critical appraisal of the available scientific evidence was conducted. Meticulous scrutinisation of the available scientific literature resulted in the inclusion of 44 studies in this systematic review and meta-analysis .
Additionally, a meta-analysis on four out of 32 studies of assessment of PPT levels [69,71,72,81] and 11 out of 32 studies of MMO assessment [58,70,73,74,77,[83][84][85]95,96,100] was performed, which showed statistically significant improvement in the PBMT group, compared to the control group with low and no signs of heterogeneity, respectively. No significant risk of publication bias for pain reduction assessment by VAS, PPT and functionality improvement assessment by MMO was noted in this meta-analysis.
QoL is an important tool for assessment of how an individual's emotional, social and physical well-being are affected by disease/disorder/disability in their daily life [147]. However, only four out of 44 studies [79,89,97,98] assessed this parameter with heterogeneity in the presented data, thereby resulting in a lack of a possible meta-analysis to assess QoL.
Petrucci et al. (2011) conducted a systematic review and meta-analysis on six RCTs to assess the efficacy of LLLT in the treatment of TMD [31]. The meta-analysis findings revealed no statistically significant difference in pain reduction assessment by VAS for the LLLT group as compared to the placebo [147]. A systematic review and meta-analysis by Chang et al. (2014) assessed the use of LLLT on the masticatory muscle or joint capsule for TMJ pain in seven randomised, single-or double-blind studies and showed that the former had a moderate analgesic effect in comparison to placebo/sham LLLT [148]. However, the authors were unable to deduce an optimal LLLT protocol to treat TMJ pain [148]. Chen et al. (2015) conducted a meta-analysis of 14 randomised controlled trials and evaluated the effectiveness of LLLT for patients suffering from TMDs, and concluded that in comparison to the placebo, LLLT was unable to reduce chronic TMD pain, but provided a significantly better functional improvement in comparison to the placebo [118]. Xu et al.
(2018) conducted a systematic review and meta-analysis to evaluate the effect of LLLT vs. placebo in 31 RCTs performed on patients with TMD [149].
In the present systematic review and meta-analysis is an up-to-date appraisal of the available scientific literature with a robust search strategy ranging from 2005 to 2021, resulting in the inclusion of a large cohort of studies. The results of this review show that PBMT effectively relieves pain assessment by VAS and improves the functional outcomes in patients with TMD at the short-term follow-up [118]. The authors also performed dosage analysis, which showed inconsistent results regarding the effects of high or low doses for patients with TMD [118]. The results of our systematic review and meta-analysis contradict the findings of Petrucci et al. (2011) [31] and Chen et al. (2015) [118] (pain reduction assessment by VAS); however, they are in accordance with Chang et al. (2014) [148], Chen et al. (2015) [118] (functional improvement) and Xu et al. (2018) [149].

Conclusions and Future Direction
To date, this is the first extensive systematic review and meta-analysis (44 eligible RCTs, of which 32 studies were eligible for meta-analysis) synthesising an eclectic assortment of experimental protocols. The majority of the available evidence suggests that PBM of laser or LEDs, or combined treatment modalities, have fundamental and substantial effects on improving TMD chronic pain, functionality and QoL. Notably, both red light and NIR, as well red/NIR combination light sources, were utilised to target superficial and deep target tissues, assisting to improve pain and functionality. Despite the heterogeneity of the eligible studies' clinical outcomes, the majority of them were of low-risk bias. Hence, researchers must hone in on specific clinical PBM protocols, therapeutic outcome measures, underlying mechanisms and replicability. Hence, for the first time, suggested recommendations for clinical PBMT protocols for future extensive TMD RCT studies were proposed by this review's authors, grounded in the current available evidence-based clinical practice and experts in the field, as well as a summary of TMD guidance study design for future reproducible methodology, which is ultimately the first steppingstone for evidence-based consensus. Additionally, there is still a considerable gap between the potential findings of PBMT in TMD management and understanding their underlying biological and molecular mechanisms, which require the scientific community's attention.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/antiox10071028/s1, Table S1: Tabular description of all the selected eligible in vivo RCTs human studies of TMD, in terms of demography, study design, symptoms, affected areas, functional problems, intervention groups, methods of assessment, evaluation period and outcomes. All the abbreviations in this table are listed in Supplementary File S2; Table S2: Tabular representation of lasers/LEDs parameters utilised in the chosen eligible in vivo RCTs human studies related to temporomandibular dysfunction syndrome (TMD). All the abbreviations in this table are listed in Supplementary File S2; Table S3