Tinnitus is a common disorder characterized by the phantom auditory perception of sounds without external or internal auditory stimuli. Reported prevalence of tinnitus ranges from 19.7% to 25.3% in nation-wide studies [1,2]. For some subjects, tinnitus is disabling. It can result in sleep deprivation, loss of concentration, psychological distress, and depression [3,4]. Therefore, this symptom has forced clinicians to attempt to establish protocols for relieving tinnitus. There are various therapeutic modalities in the treatment of chronic tinnitus, including medications (antihistamines, sedatives, antiepileptics, antidepressants, antipsychotics, and vasodilators), psychotherapy, tinnitus retraining therapy, transcranial magnetic stimulation, and transcutaneous electrical stimulation. However, none of these symptomatic treatments has resulted in significant or lasting improvement of tinnitus.

Low-level light therapy (LLLT) was proposed as a therapeutic procedure for tinnitus over 20 years ago [5-7]. In contrast to high power lasers used to cut or destroy tissues, LLLT uses low-energy-lasers or light-emitting diodes to stimulate or inhibit cellular function [8]. Although the exact mechanism involved in the effect of LLLT on tinnitus is not clearly understood yet, it has been assumed to be a photochemical and photophysical stimulator of the mitochondria in hair cells [9,10]. LLLT acts by increasing blood microcirculation through sympathetic neural inhibition [11], increasing oxygen supply to hypoxic cell [12], prompting an increase in cell proliferation [13], enhancing adenosine triphosphate synthesis in mitochondria [14,15], and releasing various growth factors [16].

However, there are still some controversies concerning the effectiveness of LLLT in treating tinnitus as only some studies have shown positive results [9,17-21]. This discrepancy might be caused by inconsistencies in several technical parameters. First, as higher wavelength laser would deliver a larger amount of irradiance through greater penetration, different degrees of laser light transmission to the cochlea might have caused different therapeutic outcomes. Earlier studies evaluating LLLT with a wavelength of 650 nm have reported no significant reduction of symptoms in chronic tinnitus [6,7]. However, an infrared laser, especially at wavelengths around 800 nm [22], has a lower absorption of water, which enables a greater amount of energy to more deeply penetrate target tissues. Thus, it can transfer energy to hair cells and auditory nerves without being absorbed by the lymphatic fluid in the cochlea. Recent animal studies have suggested that a LLLT with a wavelength of 810-830 nm might promote hair cell survival following gentamicin damage in the cochlea and reduce salicylate-induced tinnitus [23]. Second, delivery route might also influence results of LLLT. LLLT delivered to the cochlea via the mastoid process is expected to be greatly absorbed by temporal bone. However, trans-tympanic delivery of LLLT shows more irradiation penetration since a less bony structures hinders irradiation [24]. Finally, different laser parameters (e.g., laser type, wavelength, output power), irradiation route (e.g., mastoid, external auditory meatus), and treatment schedules (e.g., frequency of treatments) might have caused different results [25]. In addition, inconsistencies of patient selection and measurement between studies further hinder authors and clinicians from making comparisons among studies. Considering these issues, the aim of the present study was to narratively review the current literature on LLLT to explore the true effect of LLLT.

A search was performed in PubMed database. It was restricted to the English language. Selected keywords were “low-level laser” OR “photobiomodulation” OR “phototherapy” AND “tinnitus”. The author also performed hand search of references of the selected studies to identify other possible relevant studies. Articles included should necessarily be presented with full access to the text. I verified those articles that presented titles and summaries that approached the subject of this research as well as methodology, results, and relevance for tis practical application.

Study characteristics

Ten studies with 588 patients have compared effects of LLLT with a placebo on patients with complaints of tinnitus. Characteristics of these studies are presented in Table 1. Seven studies used a low laser device with an intensity of 5 mW and a wavelength of 650 nm [5-7,19,26-28]. One study used the device with an intensity of 60 mW and a wavelength of 810 nm [29]. One study used the device with an intensity of 50 mW and a wavelength of 830 nm [30]. One study used the device with an intensity of 100 mW and a wavelength of 830 nm [21]. Nine studies applied light on the tympanic membrane through an auditory ear canal [5-7,19,21,26,27,29,30]. Only one study applied light to the ear via mastoid bone [28]. All but one study enrolled patients with chronic tinnitus lasting for at least three months [5-7,19,21,26,27,29,30]. One study did not report the duration of tinnitus [28]. Five studies included patients with cochlear tinnitus [7,21,27-29], whereas three studies included patients with idiopathic tinnitus [6,27,30]. The other two studies included patients with both tinnitus [5,19]. All studies used a sham laser as placebo intervention [5-7,19,21,26-30]. Five studies irradiated laser unilaterally [5-7,21,30], while other studies did not further report the laterality [19,26-29]. More detailed information regarding the sample size, laser modalities, and protocol of LLLT are presented in Table 1.

Table 1 . Characteristics of included studies

Study (yr)	Sample size	Comorbidity; duration of tinnitus (mon)	Treatment laterality	Method of LLLT	Delivery method
Choi et al. (2019) [21]	38	SNHL; >3	Unilateral	100 mW; 830 nm; 20 min/d for 10 sessions; 5 times/wk TINI device (WONTECH, Republic of Korea)	Trans-meatal
Dehkordi et al. (2015) [6]	66	Idiopathic; 48 for treatment group and 34 for control group	Unilateral	5 mW; 650 nm; 10 min/d for 20 sessions; 5 times/wk Tinnimed® laser device (HT International, Germany)	Trans-meatal
Mirvakili et al. (2014) [26]	120	SNHL; >12	Not reported	5 mW; 650 nm; 20 min/d for 20 sessions; 3 times/wk TINNImed laser device (Switzerland)	Trans-meatal
Ngao et al. (2014) [27]	43	Idiopathic; >6	Not reported	5 mW; 650 nm; 20 min/d for 70 sessions; 7 times/wk MedicLaser® + Tinnitool® (DisMark GmbH, Switzerland)	Trans-meatal
Mollasadeghi et al. (2013) [28]	82	SNHL (NIHL); 22 (average)	Not reported	5 mW; 650 nm; 20 min/d for 20 sessions; every other day TINNImed laser device (Switzerland)	Trans-mastoid
Teggi et al. (2009) [7]	54	SNHL; 26 for both groups	Unilateral	5 mW; 650 nm; 20 min/d for 90 sessions; once daily New-developed TCL-system with four different diode lasers	Trans-meatal
Cuda and De Caria (2008) [19]	46	Mixed (84.8% SNHL; 15.2% idiopathic); >36	Not reported	5 mW; 650 nm; 20 min/d for 90 sessions; once daily TinniTool EarLaser (DisMark GmbH)	Trans-meatal
Gungor et al. (2008) [5]	45 (66a))	Mixed (54% SNHL; 45% idiopathic); 96 (average)	Unilateral	5 mW; 650 nm; 15 min/d for 7 sessions; once daily Tinnimed® laser device (HT International)	Trans-meatal
Nakashima et al. (2002) [29]	45 (68a))	SNHL; not reported	Not reported	60 mW; 810 nm; 6 min/d for 4 sessions; once a week SOFT LASERY JQ 310 (Minato Medical Science, Japan)	Trans-meatal
Mirz et al. (1999) [30]	49	Idiopathic; 70 for treatment group and 62 for control group	Unilateral	50 mW; 830 nm; 15 min/time for 15 sessions; 5 consecutive days per a week Uni-laser 310P, type 201.000, 3B	Trans-meatal

LLLT, low-level light therapy; SNHL, sensorineural hearing loss; NIHL, noise-induced hearing loss; TCL, transmeatal cochlear laser.

^a)Used ears as the study subjects.

Outcomes and adverse events

Table 2 shows outcomes and adverse events. Tinnitus Handicap Inventory (THI) scores were measured after LLLT in seven studies [7,19,21,26-28,30]. Nine studies evaluated the severity of tinnitus according to rating scale scores [5-7,21,26-30]. One study evaluated a number of psychosocial questionnaires such as Beck Depression Inventory and State-Trait Anxiety Inventory [30]. Six studies revealed no statistically significant differences in outcomes between the LLLT group and the sham group [6,7,21,27,29,30]. However, four studies concluded that LLLT was effective in improving tinnitus [5,19,26,28]. All four studies showing positive results used the device at a wavelength of 650 nm. Of these studies, two enrolled patients with sensorineural hearing loss (SNHL) [26,28] and the other two enrolled patients with mixed tinnitus involving SNHL and idiopathic tinnitus [5,19]. Adverse events were reported qualitatively in three studies [5,21,29]. Only one patient developed a sudden onset hearing loss and one patient developed dizziness during the course of LLLT [29].

Table 2 . List of outcomes and adverse events

Study (yr)	Evaluation method	Evaluation timing	Result	Reported adverse event
Choi et al. (2019) [21]	NRS of loudness, duration, and annoyance, THI, Psychoacoustical matches of tinnitus loudness, and MML	Immediately, 2 weeks after treatment	No statistically significant differences between two groups	No AEs observed
Dehkordi et al. (2015) [6]	NRS of loudness, TSI, Psychoacoustical matches of tinnitus loudness	Immediately	No statistically significant differences between two groups	Not reported
Mirvakili et al. (2014) [26]	VAS of severity, THI	Immediately, 3 months after treatment	Short-term treatment of LLLT was effective, but its impact may be reduced over the time	Not reported
Ngao et al. (2014) [27]	VAS of annoyance, sleep disruption, depression, concentration, loudness, and pitch, THI	Immediately	Not significantly superior to placebo effect in improving tinnitus	Not reported
Mollasadeghi et al. (2013) [28]	VAS of loudness, THI, Psychoacoustical matches of tinnitus loudness	Immediately, 3 months after treatment	Short-term treatment of LLLT was effective, but an effect that was weakened after 3 months follow-up	Not reported
Teggi et al. (2009) [7]	VAS of loudness, THI, Psychoacoustical matches of pitch, loudness, and MML	Immediately	No efficacy as a therapeutic measure for tinnitus	Not reported
Cuda and De Caria (2008) [19]	THI	Immediately	The THI scores more significantly improved in the group receiving LLLT	Not reported
Gungor et al. (2008) [5]	VRS of loudness, duration, and annoyance	2 weeks after treatment	The loudness, duration, and degree of annoyance of tinnitus were improved in LLLT group. No significant improvement was observed in the placebo laser group	No AEs observed
Nakashima et al. (2002) [29]	VRS of loudness, duration, and annoyance	1 week after treatment	No significant difference was observed between two groups	Sudden hearing loss and dizziness
Mirz et al. (1999) [30]	VAS of loudness, annoyance, and attention involved, THI, TCSQ, BDI, STAI	Immediately	No statistically significant differences between two groups	Not reported

NRS, numeric rating scale; THI, Tinnitus Handicap Inventory; MML, minimum masking levels; TSI, Tinnitus Severity Index; VAS, visual analogue scale; VRS, verbal rating scale; TCSQ, Tinnitus Coping Style Questionnaire; BDI, Beck Depression Inventory; STAI, State-Trait Anxiety Inventory; LLLT, low-level light therapy; AE, adverse effect.

Results of efficacy of LLLT for patients with tinnitus were inconsistent. A previously reported meta-analysis has concluded that the pooled effect estimate of five control-case studies does not show a significant difference in THI score after treatment [31]. They failed to demonstrate a significant difference in THI score between the LLLT group and the control group in the subgroup analysis [31].

Many studies have demonstrated a positive effect of LLLT on tinnitus [9,17-21,32]. However, when compared to controls, it is hard to say that LLLT has a favorable effect on tinnitus regardless of the measurement method or laser parameters. Evidence from many studies has suggested that most forms of subjective tinnitus result from a loss of inhibition secondary to cochlear damage in central auditory structures [33,34]. This loss of inhibition can disrupt the normal synchronized neural activity constrained by feed-forward inhibition to acoustic features of stimulus. Thus, neural networks in the central auditory pathway might have been aberrantly increased to compensate the under-activity of brain regions related to cochlear damage. Although therapeutic effects of LLLT on tinnitus are still under investigation, LLLT may compensate for sensory deprivation in the auditory system. Montazeri et al. (2017) [32] have demonstrated that LLLT can increase compound action potential amplitude without changing otoacoustic emissions. This result may support the fact that LLLT can alleviate tinnitus by reducing neural network activity rather than healing cochlear damage. Patients enrolled for case-control studies might not be effective or appropriate enough. Among included studies, the study of Mollasadeghi et al. (2013) [28] was the only study that enrolled patients with chronic tinnitus caused by noise-induced haring loss. A noise trauma can result in injury to hair cells in the inner ear and degeneration of the auditory nerve. As discussed earlier, patients who sustained injury from noise trauma to the cochlea might benefit the most from irradiation by compensating sensory deprivation in auditory system. Thus, further studies are required to support the effect of LLLT on tinnitus mainly caused by a specific inner ear disease or injury.

I evaluated the efficacy of LLLT in patients with tinnitus. Results still showed inconsistent effect of LLLT. Further large-scale studies are needed to evaluate the efficacy of this therapy on tinnitus mainly caused by abnormal neural networks in the central auditory pathway.

None.

All work was done by JEC.

Ji Eun Choi is an editorial board member of the journal but was not involved in the review process of this manuscript. Otherwise, there is no conflict of interest to declare.

None.

None.