The effect of photobiomodulation on hearing loss: A systematic review

To assess outcomes associated with photobiomodulation therapy (PBMT) for hearing loss in human and animal studies.


| INTRODUCTION 1.| Background and epidemiology
Hearing impairment is one of the most common medical conditions, 1 affecting approximately 466 million people worldwide. 2Impact can adversely affect employment, communication and social interaction.This can lead to manifold psychosocial burdens, 3 as well as significant detrimental economic impacts to the individual and wider society. 4mmon rehabilitation options include hearing aids, and surgery such as cochlear implantation, although operative trauma can potentially exacerbate existing hearing loss.Current research suggests there may be scope for photobiomodulation therapy (PBMT) to mitigate ototoxic trauma associated with surgery, traumatic noise exposure or from chemical trauma by ototoxic medications such as platinum-based chemotherapy.
Photobiomodulation therapy (PBMT), also referred to as 'lowlevel laser light (LLLT)' and 'near-infrared light', is a non-invasive therapy that uses light energy to enhance or modulate the activities of specific cells to improve or change the function of body tissues.[7][8]

| PBMT mechanism of action
Cell damage within the inner ear is a complex combination of inflammatory and oxidative stress.The mechanism of PBMT on neural-cell recovery and regeneration is yet to be clarified.The prevailing theory focuses on mitochondrial cytochrome c oxidase, a key protein in cellular metabolism and repair, and one of three major proteins in the human body responding to near infrared wavelength. 9These proteins absorb nearinfrared wavelength energy and then modulate biochemical reactions within cells.Cytochrome c oxidase is a large transmembrane protein complex in the mitochondrial electron transport chain that consists of five protein complexes that together produce adenosine 5-triphosphate (ATP). 10This theory is supported by research showing that PBMT enhances ATP production. 11Increased ATP production may lead to enhanced cell metabolism, promoting the damage-repair process.
Several studies have shown that PBMT can reduce inflammation within inner ear cells in vitro. 12,13Clinical studies on the use of PBMT to protect against hearing loss, tinnitus, and vestibular dysfunction have been published.To the best of the author's knowledge, there are presently no systematic reviews synthesising the effect of PBMT on hearing loss.

| Objectives
This review is to assess the application of PBMT in the treatment of hearing loss, examining evidence from both animal and human studies.

Population:
Humans or animals

Intervention: Photobiomodulation therapy
Comparator: There is no formal comparator or control.
Comparators were expected to vary according to the study type.Comparators may include other methods of hearing preservation, for example, administration of drugs via systemic or local routes.

Outcomes:
Pre-and post-PBMT audiometric outcomes, evidence of inflammation e.g., fibrosis (impedance values, histology), measures of neural responses (evoked compound action potentials, auditory brainstem evoked potentials), spiral ganglion neuron density and/or number, neo-osteogenesis, and hair cell count.
Adverse events associated with PBMT.

| METHODS
The study protocol was registered in the PROSPERO prospective database of systematic reviews (CRD42020210574 and CRD42020212259) and was created according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta Analyses) guidelines. 14

| Study inclusion criteria
All human experimental study designs were eligible, including casecontrol, case series, cohort, randomised controlled trials.Animal studies (live, in vitro) were eligible if they included at least one quantitative outcome measure.There were no restrictions placed on the follow-up length or the study duration.Only studies with full texts available were included.Studies with insufficient reporting data on pre-and post-intervention audiometric outcomes and those that did not assess the effect of PBMT on hearing outcomes were excluded.

Key points
• Photobiomodulation therapy (PBMT) is a potential noninvasive treatment option for hearing loss with minimal adverse effects.
• To date, only one human trial has assessed the use of PBMT in the treatment of hearing loss.
• This study showed no statistically significant improvement in hearing outcomes.
• Animal studies utilising longer wavelength PBMT for longer duration of therapy demonstrated improvements in audiological outcomes.
• Further human trials are required to determine the efficacy and safety of PBMT to treat hearing loss.

| Search strategy
The initial search was conducted in September 2020 and repeated in July 2022 to include relevant papers published since the original search.The following electronic databases were searched: MEDLINE, EMBASE, CENTRAL, ClinicalTrials.govand Web of Science including Web of Science Core collection.No limit was placed on language or publication year.Search strategy terms used are summarised in the Appendix S2.Hand-searching reference lists of the included relevant systematic reviews and citation searches were conducted to identify additional studies missed from the electronic database searches.and corroborated by a third (Jameel Muzaffar).

| Data extraction
A standardised Microsoft Excel sheet was used for data extraction from the included studies.This was designed and piloted prior to the data extraction phase.The data of interest comprised of study characteristics (study design, location, duration), primary and secondary outcome data and operative adverse events.Missing data were sought, where possible, by email contact with study authors.
Any discrepancies were identified and resolved through discussion within the author team.

| Risk of bias quality assessment
Two reviewers (Yasmin Nikookam/Nawal Zia) independently assessed the methodological quality of the included studies.Animal studies were assessed using the SYRCLE tool. 15Human studies were assessed using the Oxford Centre for Evidence Based Medicine (OCEBM) grading system, and the Cochrane Risk of Bias 2 (RoB2) tool for randomised control trials. 16,17Any disagreements were resolved through discussion between the two authors (Yasmin Nikookam/ Nawal Zia), and where necessary, consultation with the third review author (Jameel Muzaffar).The above process was followed for the second search conducted by two authors (Andrew Lotfallah/Jameel Muzaffar).

A flowchart detailing study selection according to the Preferred
Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines is included in Figure 1 (See Supplementary Materials).
Sixteen studies were conducted on animal models, published between 2012 and 2021.From these, 10 were non-randomised controlled trials, 4 were in vitro studies, one ex vivo study, and one randomised-controlled trial.Three studies used mouse models, ten used rat models, two used a guinea pig model, and one used a gerbil model.The wavelength used was classified in all studies: 11 used 808 nm, 3 used 830 nm, 1 used 908 nm, 1 used 810 nm.The duration of PBMT ranged from 4 to 60 min in a session.The follow-up duration of PBMT in animals ranged from 5 to 28 days.Table 1 summarises study characteristics for both human and animal studies.

| Quality of studies
The methodological quality of included studies was modest, mainly consisting of non-randomised controlled studies (n = 10).All animal studies (n = 16) were prospective, and all studies had a minimum of five animals, or two cell lines which underwent PBMT.
The single human study 7 comprised a randomised, double-blind controlled prospective study that had 30 subjects.This study was OCEBM grade I. Participants were randomly allocated to treatment, placebo and control groups with adjustment to ensure similar baseline characteristics, although the method of randomisation was not described.Double blinding of subjects and researchers administering placebo and treatment laser therapy was also implemented to limit potential bias.
Heterogeneity of audiological, PBMT duration, power, and wavelength outcomes within and between human (n = 1) and animal (n = 16) studies precluded a meta-analysis.Within the human study, the limitations were the reporting of complications following PBMT and the follow-up duration of patients who had PBMT.Quality assessment of the human studies is summarised in Table 4. Within the animal studies there were limitations in post-treatment observation duration of animals receiving PBMT, audiological data prior to PBMT delivery, age of subjects, housing of animals and PBMT technique.
Quality assessment of animal studies is summarised in Table 5.

| Audiological outcomes
Audiological outcomes in humans are summarised in Table 2. Audiological outcome measures were assessed using pure-tone audiometry, speech perception, and transient evoked otoacoustic emissions.
Pre-PBMT hearing status was reported, and all patients recruited had moderate to severe, non-fluctuating hearing loss, with an average pure-tone audiometric score of 39.6 dB HL (SD 15.3 dB). 7e causes of hearing loss were not stated.The PBMT treatment protocol was based on a pilot study conducted by HearingMed (unpublished) showing an improvement of word recognition scores following PBMT. 7diological outcomes in animals are summarised in Table 3  Not stated?Severe?
PBMT administration details: Followed a protocol that included 7 steps.The laser was placed in 4 different positions: 1).Temporomandibular joint, anterior to the external auditory meatus of the ear, 2 inches from the surface of the skin.
2).Midline of the cervical spine, 3 inches from the skin surface.
3) Top of the head, 2 inches from the skin surface.À1.8 to
Following reported as a range:

5-40 min
Mean age of subjects:

PBMT administration details:
The laser module was placed at the outer ear canal at exactly the angle that allowed a total coverage of the cochlea with the laser beam of approximately 7 mm diameter.
NOTE: PBM was administered before NIHL was induced.

Audiological data:
ABR recordings 2 days before and 2 weeks after noise exposure.

ABR-threshold shift
Groups 3-7 showed a lower group mean hearing loss following the noise exposure compared to group 2.

Cochlear hair cells
No loss of inner hair cells could be detected bilaterally for any of the groups.Group 2: a significant decrease in outer hair cell counts was found for both ears compared to group 1.

Good outcomes.
A single NIR pre-treatment induces a very effective protection of cochlear structures from noise exposure.Pre-exposure of
[ The morphology in the PBM/NAC group showed less hair cell loss than the other groups.
Good outcome: PBM/NAC combination therapy may prevent hearing loss more effectively than PBM alone.

Note:
Animals were also divided into three groups for immunohistochemistry of iNOS: naive Induced hearing loss-exposed to 1 octave band noise centered at 4 kHz for 5 h (121 dB sound pressure level) in a ventilated sound exposure chamber.

| Photobiomodulation therapy outcomes
Twelve studies summarised all three of wavelength used, duration of PBMT, and follow-up period.Wavelength size of 808 nm was used most frequently (n = 11).PBMT was effective in hearing enhancing recovery, reducing loss of hair cells and reducing the immune inflammatory response in the included animal studies.One study found preemptive PBMT therapy before noise-induced hearing loss (NIHL) was effective at hearing preservation. 18One study found that pretreatment using PBMT prior to cochlear implantation resulted in better hearing outcomes. 29Another study found a PBMT power of 165 versus 110 mW/cm 2 resulted in a greater preservation of hearing. 309][20][21][22][23][24][25][26][27][28][29][30][31] Direct comparison of differing ototoxic agents was not conducted in any of the studies included.
One study assessed the effect of PBMT in combination with antioxidants N acetyl-L-cysteine (NAC) or acetyl-L-carnitine (ALCAR). 20is study showed conflicting results on the effectiveness of PBMT, in that the laser in combination with an antioxidant was not superior to the antioxidant alone when comparing ABRs.However, the hair cell count illustrated that PBMT in addition to an antioxidant was superior to the antioxidant alone.
Overall, there was a trend toward benefit from PBMT in animal studies, irrespective of delivery method, wavelength and power used or the animal species treated.][25][26][27]29 In the single human study included, no statistically significant effect of PBMT on hearing was detected. 7Notably, the PBMT wavelength was shorter, and the power used lower compared to all animal studies.No study reported any intra-or post-PBMT adverse events or deaths.

| DISCUSSION
This systematic review is the first to report outcomes of PBMT on hearing loss.Generally, animal subjects showed improvements in hearing preservation outcomes whereas no statistically significant improvement was detected in the single human study.

| PBMT versus placebo
The only human results from Goodman et al.'s study found none of the three measures of hearing (audiometric thresholds, speech recognition test, or otoacoustic emissions) showed statistically significant difference between the treatment, placebo, or control groups. 7Therefore, this study concluded that PBMT was not effective in recovering hearing loss.However, the sample size was small (n = 30) and comparison with animal studies is difficult as the delivery characteristics differed.In the human study, PBMT was delivered in seven cycles across different anatomical locations, including the temporomandibular joint, spine, ear and top of the head.The duration of therapy ranged from 15 to 60 s.Overall duration of therapy was relatively very short, lasting approximately 4 min in total with low power (7.5 mW).
In contrast, animal studies that produced positive results from PBMT, typically irradiated through the tympanic membrane for 60 min.Moreover, the wavelength used by Goodman et al. was 532 nm constantly, with 635 nm in a pulsed manner. 7Prior evidence suggests that the wavelength with the highest potency of biomodulation is 800-830 nm. 27This is likely due to the absorption spectrum of cytochrome c oxidase which has a peak of 830 nm, and the penetration of light through tissue is superior at 830 nm than 532 nm. 32Hence, there were duration, power, anatomical site and wavelength differences between the human and animal studies.
Goodman et al.'s findings contrast with the 16 animal studies that assessed the effect of PBMT for a range of ototoxic insults.9][30][31] The effect of PBMT in reducing T A B L E 5 SYRCLE risk of bias assessment.

| PBMT versus systemic therapy
One animal study evaluated the added effect of PBMT in addition to systemic antioxidant therapy (NAC and ALCAR). 20This study found combination therapy of NAC and PBMT to significantly improve NIHL when compared to a previous study that assessed the effect of PBMT alone.It appears that NAC may have a role in accelerating hearing improvement with PBMT (improved on the 6th day of irradiation), when compared to PBMT alone (improved on the 10th day of irradiation). 20,26Additionally, authors found the antioxidant NAC to be superior to ALCAR, but NAC and ALCAR alone was not superior to combination therapy with PBMT. 20

| PBMT positioning and characteristics
In contrast to most animal studies which irradiated in close proximity to the tympanic membrane, Goodman et al.'s human study utilised external irradiation (temporomandibular joint, cervical spine, top of head and external auditory meatus).They state that a transmeatal approach would have resulted in greater penetration but suggest that it would be less practical. 7e positioning of PBMT for optimal delivery varied across studies.Beyer et al. found that irradiation of the mastoid leads to therapeutically insufficient light doses when compared to irradiation through the tympanic membrane. 33 An excess amount of laser irradiation may lead to destruction rather than promotion. 34Subsequently, determining optimal PBMT parameters is vital and must be balanced against safe PBMT delivery.
There are peaks in the typical responsive wavelengths for cytochrome c oxidase (670 and 830 nm). 9 Cytochrome c oxidase mediates photobiomodulation in the far-red and near-infrared range.Although cytochrome c oxidase also absorbs strongly at wavelengths less than 630 nm, this is within the visible light range and has a lower rate of tissue penetrance than wavelengths in the near-infrared range. 9Penetrance of lasers through the tympanic membrane and other tissue structures of the inner otic capsule is superior in the near infrared range (780-1100 nm). 25 Therefore, wavelengths must be carefully selected according to PBMT delivery method and structures the light must pass through to reach the cochlea.Additionally, it is important to ascertain whether shorter, concentrated delivery of PBMT induces a more significant effect on hearing loss when compared to a more prolonged delivery at lower concentration.

| PBMT as a pre-treatment therapy
Basta et al. assessed the effect of PBMT prior to noise exposure.They found a single pre-treatment dose of PBMT induced statistically significant protection of cochlear structures based on ABR recording and histological analysis. 18Pre-exposure of 10 min was the optimal dosing. 18These findings suggest the scope for the implementation of PBMT as a pre-treatment therapy prior to procedures known to cause inner ear damage and hearing loss, such as cochlear implantation.
More research is required to ascertain the true benefits of PBMT as a pre-treatment in hearing preservation.

| Future of PBMT
Overall, results suggest that PBMT could be an effective method for hearing preservation.However, almost all studies to date have been conducted on animal models with only one conducted on human subjects.Most studies assessed outcomes over a short duration of time, with the longest period of follow up being 28 days.This precludes comment on the long-term effects PBMT may have on hearing loss, or whether further courses are needed to maintain benefit.Including a longer follow-up period would enable researchers to assess the longer-term effects and complications of PBMT.Thus, enabling researchers to determine the suitability of using PBMT in further human trials.

| CONCLUSION
Though the evidence base is far from comprehensive, hearing outcomes following PBMT appear to be superior to non-PBMT in animal studies.PBMT theoretically enables a non-invasive mode of delivering therapy which may enable audiological function to be preserved and maintained following injury to inner ear structures.The low-risk profile and promising data from animal models suggest PBMT warrants further investigation as an intervention to prevent or treat hearing loss.Further research should focus on optimising PBMT light delivery and dosing for the inner ear to inform future human trials.

AUTHOR CONTRIBUTIONS
Manohar Bance, Jameel Muzaffar and Peter Kullar conceived the paper and supervised the work undertaken.Searches and data extraction were performed by Yasmin Nikookam, Nawal Zia, Andrew Lotfallah and Jennifer Davis-Manders.All authors reviewed the study protocol and were involved in drafting and providing critical edits to the manuscript.All authors agree to be accountable for all aspects of the work presented.
Initial searches were performed by Yasmin Nikookam and verified by Nawal Zia.The two reviewers independently screened titles and abstracts of the studies from the database search inclusion and to identify duplicates.Full texts were reviewed independently against the inclusion and exclusion criteria.Disagreements at the abstract and full-text screening stages were discussed within the author team (Yasmin Nikookam/Nawal Zia) and where applicable, with a third reviewer (Jameel Muzaffar), and consensus was reached in determining eligible studies.The second search was conducted in the same manner by two authors (Jennifer Davis-Manders/Andrew Lotfallah) Correlates to not stated.a Correlates to number of cell wells used.NIKOOKAM ET AL.T A B L E 2 Primary outcomes in human studies.

5 )= 5
We also divided the animals into three groups for immunohistochemistry of iNOS: naïve (n = 5), non-treatment (n Beyer et al. used a PBMT power and dose similar to Goodman et al. (1 mW, 593 and 633 nm vs. 7.5 mW, 532 and 635 nm in Goodman et al).7 Therefore, for optimum dosimetry, evaluation of light transmission factors for chosen irradiation modalities is necessary.The externally applied light dose needs to be calculated according to the tonotopy of the cochlea, as different anatomical landmarks and mediums will transduce different frequencies; this is relevant when considering the anatomical location of the PBMT probe.

Identification of studies via databases and registers Identification Screening Included
Change Pure-Tone Audiometry:* Calculated as change in PTA ( posttest minus pretest), negative values indicate improvement in thresholds.Following expressed as a range: in the audiometric HFA.Change calculated as posttest minus pretest; negative values indicate improvement in thresholds.Following expressed as a range: T A B L E 4 Cochrane risk of bias 2 tool.Green = low risk of bias; red = high risk of bias; yellow = unclear risk of bias. 1. Was the allocation sequence random?Yes.2.Was the allocation sequence concealed until participants were enrolled and assigned to interventions?Yes.3.Did baseline differences between intervention groups suggest a problem with the randomisation processs?No. 4. Were participants aware of their assigned intervention during the trial?No. 5. Were carers and trial personnel aware of participants' assigned intervention during the trial?Yes and No. 6. Deviations that arose because of the trial context?No. 7. Was an appropriate analysis used to estimate the effect of assignment to intervention?Yes.8.Participants aware of intervention?No.9.Personnel Aware of intervention?Yes and No. 10.Balanced non-protocol interventions?Yes.11.Failures in implementation affecting outcome?No.12.Non-adherence affecting outcome?No.13.Outcome data for all participants?Yes.14.Method of measuring the outcome inappropriate?No.15.Measurement or ascertainment of outcome differ between groups?No.16.Outcome assessors aware of intervention received?Unsure.17.Could assessment have been influenced by knowledge of intervention?No.18.Results selected from multiple outcome measurements and multiple analyses of the data?No.19.Trial analysed in accordance with a pre-specified plan?Yes.