S3 Guideline: Chronic Tinnitus

Tinnitus counselling [84‐89] is a fundamental part of the therapy of a person with chronic tinnitus (for details on the implementation, see Appendix 2).

The aim of counselling is a clarifying psychoeducational explanation and the presentation of strategies for dealing with a benign disease to reduce fears or exaggerated expectations of a cure. It is the basis for constructive habituation mechanisms and serves to avoid negative, self-reinforcing cycles in tinnitus [44, 90, 91].

The statement that there are no therapeutic options is false. The patient should be encouraged to deal with his or her ear noise in an informed way through counselling. The task of the first consulting physician is thus to advise the patient about his/her possible individual aetiopathogenesis (based on anamnesis and findings), prognosis, tinnitus-increasing factors (hyperacusis, hearing ability) or avoidance of harmful influences (e.g. noise).

If the doctor does not have the time resources for detailed counselling, they will refer the patient to a doctor or psychological psychotherapist who is qualified in tinnitus therapy. It is important for the patient, and additionally to prevent somatisation tendencies, that the treatment is not finished and that they can receive advice from the attending doctor at any time. The doctor’s role should also include counselling on alternative or new methods of treatment. This must be done based on the latest scientific knowledge. The prognostic content of counselling should be that the ringing in the ears can very often be changed by employing a suitable, non-drug therapy via a gradual reduction of the tinnitus distress and that the doctor can carry out this therapy (or refer the patient to a doctor or psychological psychotherapist in appropriately qualified practices and facilities who are qualified in suitable tinnitus therapy). As a result of successful therapy, the suffering caused by tinnitus is reduced.

Reference can be made to various manualised interventions with or without consideration of hearing loss or concomitant psychological disorders. Self-help (Sect. 4.1.12) has also been shown to be effective with guidance, although specific interventions seem to be more effective.

4.1.2.1 Hearing aids

For the effectiveness of hearing aids in tinnitus therapy, there are only studies with moderate or weak evidence. However, this is mainly because there are practically no studies that examine the effectiveness of hearing aids alone for the treatment of chronic tinnitus. Therefore, evidence cannot be obtained, even though there are certainly indications from clinical experience that hearing aids promote tinnitus suppression and habituation. However, this requires intensive counselling and support of the patients, as a U.S. study with 133 hearing-impaired tinnitus patients proves. Hearing aid acceptance is primarily lower in the case of severely distressing tinnitus [92].

The benefit of hearing aids may be higher for low and medium tinnitus frequencies (up to 6 kHz) than for high-frequency tinnitus [93]. Other, more recent studies, however, show good effects even with isolated high-frequency hearing loss and high-frequency tinnitus [94]. In this study, 114 tinnitus patients were randomised into three groups and fitted with different types of hearing aids. Since no further counselling or education took place, a pure hearing aid effect could be demonstrated, which led to a significant improvement in the THI after 3 and 6 months. However, there were no differences between different hearing aid types or fitting strategies.

Other studies investigated differences between individual device types and found good effect sizes but no device-specific differences [95, 96]. A Swedish study investigated the effect of hearing aids on tinnitus patients, with a total of 100 patients included in the study, 50 of whom had hearing loss with tinnitus and 50 of whom had hearing loss without tinnitus. Of these 100 participants, data from 92 could finally be analysed, 46 from each group. For the patients with tinnitus and hearing loss, a significant improvement in the THI was achieved via the hearing aid application. Both groups, including the non-tinnitus patients, also improved in tests that measured cognitive functions (Reading-Span Test and Hearing-in-Noise Test; [97]).

Overall, however, there is a lack of convincing studies and meta-analyses demonstrating the effectiveness of hearing aids alone, for the systematic reasons outlined above. Accordingly, Hoare et al. [98] conclude in their Cochrane Review that a recommendation for the use of hearing aids for the indication of tinnitus cannot be made because of the poor methodology of the studies. In an update of this Cochrane Review in 2018 [99], this assessment is maintained, but overall, after evaluating studies that compared noisers (noise generators) and hearing aids, the effectiveness of hearing aids is generally confirmed.

If hearing aids are fitted, this is generally done according to the applicable national hearing aid guidelines. In individual cases of isolated high-frequency hearing loss and high-frequency tinnitus, a hearing aid fitting can be useful even without the presence of hearing loss corresponding to these guidelines [94, 100‐103].

4.1.2.2 Noise generators or noisers

Noise CDs or noise generators are also discussed for tinnitus suppression, often in combination with a hearing aid [104]. A Cochrane meta-analysis evaluated six studies with a total of 553 participants and found that no improvement in tinnitus could be measured by external sounds alone or their amplification (by hearing aids). The analyzed studies stated that sound therapy was supportive. However, a clear determination of the evidence was not possible because of the multimodal therapy approaches [105]. In a review of a total of 89 studies, the high proportion of bias, caused by the commercial interests of the manufacturers and study sponsors, but also the weak methodology and often unclear definition of the outcome parameters, was criticised [106].

A Cochrane analysis from 2018 evaluated eight studies with a total of 590 participants on the effectiveness of noise therapy, mediated either by hearing aids or by sound generators. The authors criticised the fact that there was virtually no blinding and a high risk of bias in all studies. In particular, the comparison of the use of hearing aids and noisers showed no significant effects due to noiser use. Evidence of measurable superiority of sound therapy or noiser treatment over placebo or education and counselling could not be found for any device tested. Overall, the quality of the studies was weak because no second or subsequent effects on depression or anxiety were recorded. The general quality of life was also not taken into account in the studies. The authors conclude that this Cochrane analysis does not provide any evidence that noise or sound therapy for tinnitus is superior to general counselling or placebo treatment. The authors further conclude that in future studies a much better methodology with blinding and randomization and especially with the recording of other secondary outcome parameters must indeed be applied [99]. Furthermore, long-term effects of noise treatment and thus possible damage to the auditory pathway due to constant sound stimulation have not yet been investigated and thus not recorded [34, 107‐109]. In studies on classic TRT, in which noisers are regularly used, data predominate with the statement that TRT produces the same effects as other counselling and habituation therapies even without the use of noisers ([110]; see Sect. 4.1.—TRT).

4.1.2.3 Cochlear implant

Patients with profound hearing loss or deafness may be indicated for a CI to improve their hearing. If these patients had tinnitus, an improvement is retrospectively observed more often than not [111‐117]. This applies to unilateral CI implantation for homolateral [118], but also for contralateral [118] and bilateral [116, 119, 120] tinnitus. A second implant can further improve the quality of life, also concerning tinnitus [121]. Unilaterally deaf patients with very distressing tinnitus who were fitted with a CI also showed comparable results pro- and retrospectively [122‐124]. The effect on the phantom sound was also independent of the tinnitus quality. Narrowband noise, tonal or even polyphonic tinnitus behaved in the same way [118].

Prospective observations of the course of tinnitus in connection with hearing loss-induced CI implantations also come to the same conclusion, although the number of study participants is still small overall [113, 119, 125]. In a prospective study of 174 CI users, 71.8% had tinnitus before surgery; in 20% it disappeared 6 months after surgery, and in 51.2% it improved [125]. Furthermore, improvements are also shown when stress processing and coping strategies are additionally assessed [116]. Exacerbations of tinnitus are possible, but remain as exceptions. In addition, there is a single non-controlled pilot study [126] with 21 CI implantations for the indication of unilateral tinnitus, from which the patients benefited significantly, according to the authors.

In individual cases with chronic tinnitus, in which the CI had only occasionally or not at all influenced the tinnitus, special electrical stimuli (biphasic with a fixed stimulation rate of 100–200 or 5000 St/s at a pleasant volume) were used. Some of the patients responded positively to at least one of the types of stimulation tested, i.e. the tinnitus was partially suppressed [127]. It was also pointed out that an electrode that is fully inserted (full length) is superior to a partial insertion [128]. However, systematic studies are lacking. At present, this procedure can still be described as experimental.

Studies on CI indications alone with the indication of tinnitus without hearing loss do not exist. Tinnitus with simultaneous occurrence of a CI-relevant—also unilateral—hearing loss can increase the indication for a CI, but cannot be the sole indication [111, 114, 115, 119, 125]. Recent studies also come to the same conclusions [129‐133].

Intracochlear electrical stimulation involves delivering current to the auditory nerve via a CI electrode. The primary goal is to rehabilitate hearing in hearing-impaired patients. Some of these patients are naturally also affected by chronic tinnitus so that the effect of electrostimulation on tinnitus can be investigated in patients with a cochlear implant and tinnitus.

The 23 studies reviewed here are clinical longitudinal case studies without randomisation, and consequently, they had to be assigned to a low level of evidence (level of evidence 4). Furthermore, as a rule, on-off examinations in the period up to 1 year postoperatively were indicated, so that no statement on the long-term effect is possible. All patients were adults who had undergone CI surgery due to postlingual progressive deafness.

Various standardised questionnaires (TF, THI, TFI) were generally used to assess the outcome of tinnitus in the course before and after surgery. Some studies also recorded stress and depression indicators in addition to tinnitus. In 21 of 25 studies, a statistically significant positive effect of the CI could be determined based on the scores regarding tinnitus perception as well as the co-incidences of anxiety disorders and depression.

Among the 25 studies, there were three studies in which a comparative group was included in the observation.

In the study by Seo et al. [134], a comparison was made between tinnitus patients with a CI (n = 16) and a control group with an active middle ear implant (AMEI, n = 11) before and after surgery. As a result, surgery significantly improved THI 6 months postoperatively in 91% of patients in the group with AMEI and 56% in the group with CI. The study was only assigned a level of evidence of grade 4 due to the lack of randomisation.

Two studies could be assigned to evidence level 1b. In the first study [135], a group of 38 postlingually deaf adults were randomised. While one group received bilateral simultaneous implantation, the second group of 19 patients received sequential bilateral surgery within 2 years. Among the 38 patients, 16 had chronic tinnitus. The Tinnitus Handicap Inventory (THI) and the Tinnitus Questionnaire (TQ) were used to assess the change in subjective tinnitus perception before and after CI fitting. As a result, the preoperative tinnitus prevalence was 42%. A tendency towards a positive effect was found in both the simultaneously and sequentially operated group based on the questionnaires. There was no difference between simultaneous and sequential surgical strategies. In the group with sequentially operated patients, five persons experienced a complete suppression of tinnitus after implantation of the second side. At 2 years after unilateral implantation, two patients developed tinnitus that did not exist before the operation. This disappeared in both patients after implantation of the second side. The total follow-up time was 3 years.

In a second study [136], patients with tinnitus were also prospectively fitted with a CI either unilaterally (n = 19) or simultaneously bilaterally (n = 19). The evaluation was based on the THI, the TQ and a visual analogue scale. The prevalence of tinnitus preoperatively was 42%. After 1 year, the score in the TQ was reduced to 71.4% and in the THI to 80%. In six patients who had no tinnitus before the operation, tinnitus was induced postoperatively.

In summary, there is usually a positive effect on preoperatively existing tinnitus after a CI. Conversely, however, in rare cases, a new occurrence of tinnitus after implantation is also possible, whereby it cannot be determined with certainty whether the effect is due to the implantation or occurs spontaneously. The positive effect of a CI on tinnitus seems to be independent of age, i.e. patients over 80 years of age can also benefit from CIs for tinnitus and hearing loss.

4.1.2.4 Hearing therapy

A meta-analysis presented in 2010 shows weak evidence that auditory therapeutic approaches such as hearing therapy or audiotherapy are effective for hearing losses that can often be detected in tinnitus patients, including unilateral hearing loss as well as for primary central or psychogenic hearing losses [137].

Hearing therapy (audiotherapy) can be manualized [138]. This involves targeted exercises to improve central auditory processing skills such as directional hearing, focusing and differentiation in noise, with and without hearing aids, and specifically overhearing of tinnitus. Furthermore, auditory therapy improves the acceptance of hearing aids and can thus promote the tinnitus situation [139]. Auditory discrimination training (ADT), which requires tinnitus patients to do frequency discrimination exercises, also improves tinnitus distress [140‐142]. A systematic review examined 10 studies on the effectiveness of general auditory perception training. A significant improvement in tinnitus distress was found, but all studies had low evidence due to methodological weaknesses [137, 143]. Since this review, only two relevant studies have been added: Zarenoe et al. [144] investigated the effects of additional motivational training for hearing aid fitting in tinnitus patients with hearing loss. A total of 50 patients (40–82 years old) were randomised, half of whom received the motivational training and the other half were fitted with hearing aids only. The success of the intervention was monitored with the THI. Tinnitus distress decreased significantly in both groups, but more markedly in the training group. Tugumia et al. [145] compared auditory (n = 6) and visual training (n = 6) in tinnitus patients, but could not determine any significant differences concerning tinnitus distress between the groups. However, auditory training slightly improved the THI scores, while visual training slightly worsened them.

Acknowledged, psycho-physiologically based therapy methods with the goal of a tinnitus habituation form an important basis for the treatment of patients with chronic tinnitus today. They are applied on an outpatient and inpatient basis. However, it is a prerequisite that the patients are suitable for such procedures and that they can, and even must, accept the therapy and the model on which it is based.

In the case of cognitive habituation, the phantom sound still exists, but it is perceived less or is no longer perceived without active attention to the tinnitus. Habituation is the result of a specific, cognitive learning process of the brain [146]. Therefore, it is also the purpose of therapy to bring the brain from a stressful tinnitus perception as far as possible to the tinnitus habituation. This neurophysiological learning process can be described as desensitisation [147] or as habituation [41, 148]. Habituation is the goal of therapy, but an improvement in symptoms—even without reaching the final goal of complete habituation—can already represent a great gain for the person affected.

A suitable therapy method with proven effectiveness in controlled studies is structured tinnitus-specific CBT [149‐154]. It aims at an improved, more indifferent handling of the tinnitus, at best again habituation.

According to a recent Cochrane Review [155] based on 28 relevant studies with a total of 2733 participants, there is (low grade) evidence that this therapy reduces the negative influence of tinnitus on quality of life. It has very few side effects and can also reduce the accompanying symptoms of depression and anxiety [155]. In this review, the effects of different forms of CBT were investigated. All participants had tinnitus for at least 3 months and their average age ranged from 43 to 70 years. The duration of CBT ranged from 3 to 22 weeks and it was mostly conducted as an outpatient intervention in clinics or online.

In some of the studies, CBT was compared only with waiting list control groups; in others, comparisons were made with active interventions (psychoeducation, auditory stimulation, TRT). The literature on behavioural therapy was systematically and meta-analytically summarised for tinnitus. The results were as follows:

A total of 14 studies compared CBT with no intervention/waiting list control. These studies showed that the tinnitus distress (measured with tinnitus questionnaires) was significantly reduced at the end of treatment with CBT compared to the control group (standardized mean difference [SMD]: −0.56, 95% confidence interval [CI]: −0.83–−0.30; 10 studies; 537 participants; low certainty of evidence). Converted into a score on the THI (range 0–100), this corresponds to a 10.91 point lower score in the CBT group, with the minimum clinically relevant reduction (MCID) for this scale being 7 points.

Possible adverse effects were recorded in seven of these studies: CBT probably leads to no or only minor adverse effects. Six studies reported no adverse effects and in one study one participant worsened under the CBT condition (risk ratio [RR]: 3.00; 95% CI: 0.13–69.87).

Concerning secondary endpoints, CBT may lead to a slight reduction in comorbid anxiety or depressive symptoms.

Three studies compared CBT with acoustic interventions. Indexed by the THI, CBT is likely to reduce tinnitus distress compared with acoustic intervention as measured by the THI (range 0–100; mean difference [MD]: −5.65; 95% CI: −9.79–−1.50; 3 studies; 444 participants; moderate certainty of evidence).

The evidence suggests that CBT can slightly reduce depression compared to auditory interventions, but there are no relevant differences in anxiety symptoms or health-related quality of life. No adverse effects were reported for either intervention.

One study (42 participants) compared CBT with TRT (including bilateral sound generators according to the TRT protocol). It was found that CBT was better at reducing tinnitus distress (measured by THI) compared to TRT (MD: −15.79; 95% CI: −27.91–−3.67; low certainty of evidence). In three participants, the values worsened during the study: once in the CBT (n = 22) and twice in the TRT group (n = 20; RR: 0.45; 95% CI: 0.04–4.64).

In total, 16 studies compared CBT with another active control (e.g. relaxation, information, internet-based discussion forums). Overall, CBT can reduce tinnitus distress compared to other active treatments (SMD: −0.30; 95% CI: −0.55–−0.05; 12 studies; 966 participants; low certainty of evidence). Converted to a THI score, this corresponds to a 5.84 point lower score in the CBT group than in the other active control group. Adverse effects were systematically recorded in a few studies only. In one study, it was reported that tinnitus worsened in three participants: once in the CBT group and twice in the information-only group.

All results given refer only to the time point at the end of treatment. There is no evidence of efficacy for further follow-up time points (6 months and 12 months after the end of treatment; [155]).

Other, older studies document significant reductions in depression scores in randomised controlled trials (RCTs), presented in a meta-analysis (SMD: 0.37; 95% CI: 0.15–0.59; I² = 0%; [156]) as well as in single RCTs [157‐161].

In uncontrolled cohort studies, Brüggemann et al. [162] and Seydel et al. [163, 164] were able to demonstrate a reduction in tinnitus distress for outpatient and also inpatient therapeutic methods. Schaaf et al. [165] stress high effect sizes for inpatient tinnitus therapies.

One problem with clinical implementation is availability. The possibility of outpatient tinnitus-specific behavioural therapy exists only rarely. Here, new possibilities arise through the approval of online-based offers. In several studies, online-based behavioural therapy approaches for tinnitus were found to be as effective as classic behavioural therapy in face-to-face form, but other studies emphasise the superiority of direct face-to-face psychotherapy. Before internet-based therapy, the patient should be seen by the psychotherapist at the beginning to assess comorbidities. When considering internet-based behavioural therapies, special consideration should be given to the following: which interventions were used, the number of sessions, elements of the therapy and whether a therapist was present.

Guided self-help, as well as cognitive-oriented internet programmes (ICVT), can be alternative psychotherapeutic applications of CBT for tinnitus [160, 166]. The effect sizes are a little less smaller than with regular face-to-face CBT, but on the other hand, the so-called guided self-help, as misleadingly described by the authors, is a much more cost-effective psychotherapy than face-to-face interventions [167].

For the evaluation of internet-based behavioural therapy and tinnitus-related programmes, a systematic review and meta-analysis evaluate studies from 1990 to 2018. A total of 25 studies were included, six of which were dedicated to the treatment of tinnitus patients. Especially concerning anxiety and depression, only small effects were found in these studies. The authors conclude that internet-based CBT could be an alternative to traditional face-to-face therapy, but that the current studies cannot yet provide sufficient data certainty.

An RCT with a 2-months follow-up period investigated the effectiveness of internet-based behavioural therapy. In total, 72 patients underwent this audiologist-controlled behavioural therapy model and were compared with 73 patients who were only interviewed once a week (monitoring control group). After 2 months, the monitoring group also received internet-based behavioural therapy. Therapy success was measured with the TFI. In addition, accompanying symptoms such as insomnia, anxiety, depression and hearing loss were recorded. After the internet intervention, the reduction in tinnitus stress was significantly greater than in the control group (51% compared to 5%). This effect was seen as early as 4 weeks after the start of the therapy. In addition, the therapy also improved sleep problems, depression, hyperacusis and cognitive impairment, as well as leading to a better quality of life. Overall, the effects were followed up and remained stable 2 months after the intervention [168]. Another study by this research group randomly compared 46 patients who participated in a guided 6‑week internet-based behavioural therapy with 46 patients who received face-to-face CBT. The treatment effect (within group effect) was large in both groups (27 and 32 points reduction in TFI, respectively); the difference between the two groups (between group effect) was small, leading the authors to conclude that both interventions are similarly effective [169].

A meta-analysis examined the effect of internet-based CBT on quality of life, depression and tinnitus-related anxiety. A total of 12 RCTs with 1144 patients were analysed, each comparing psychological interventions with a waiting list. A second network meta-analysis examined 19 studies with 1543 patients in which different behavioural therapy approaches were compared with each other (“head-to-head”). Concerning tinnitus distress, face-to-face behavioural therapy, i.e. in direct contact with a psychologist, had statistically significantly the greatest potential for improvement: Tinnitus improved in 75%, depression in 83% and anxiety in 87%. Since this study compared several forms of behavioural therapy, the authors emphasise that all forms of CBT are effective therapy for tinnitus [170].

Another study compared internet-based behavioural therapy for tinnitus treatment with personalised behavioural therapy (N = 43). In this study, special emphasis was placed on finding out which patients were more open to internet-based offers than others. What was particularly striking here was that the patients who were open to such therapy offers also benefited more from internet therapy, but were also less burdened overall. The authors conclude that internet-based behavioural therapy can be an alternative to modern therapy approaches for open patients. However, this therapy requires independent work and a high level of self-motivation [171].

In addition to frequently and sufficiently evaluated behavioural therapy methods, psychodynamic interventions are also used in tinnitus therapy. This happens especially within the framework of so-called multimodal therapy approaches. The condition (as for all psychotherapy methods) is that the patient and therapist must be convinced of the approach and the procedure.

Scientifically, it remains unsatisfactory that representatives of psychodynamic therapies have so far not presented any evaluated studies on the proof of evidence for tinnitus beyond casuistics.

Indirectly, the benefit of a psychodynamic approach that included cognitive–behavioural elements could be proven in an inpatient quality evaluation. From 1994 to 2007, 37 tinnitus patients were treated in an inpatient therapy setting in closed groups by the same psychotherapist in individual and group therapy over 4–8 weeks with an integrative psychodynamic behavioural therapy approach with a focus on tinnitus and hearing symptoms. This proved more effective, with an effect size of 0.93, than a comparison group treated with a serotonin reuptake inhibitor alone [172]. However, controlled RCTs and meta-analyses on the proof of efficacy regarding psychodynamic methods for the treatment of chronic tinnitus are not available.

The core of TRT is an acoustic therapy with frequency-unmodulated noise, which was developed and introduced in the Anglo-American area based on the neurophysiological model [173‐175].

TRT is a habituation technique that reduces the auditory, emotional and autonomic impact of the tinnitus noise and thus reduces the stress response to the tinnitus stimulus. It integrates three to five intervention steps including a detailed tinnitus history, auditory distraction from the tinnitus by broadband noise via a tinnitus instrument and psychological counselling.

Furthermore, TRT is a specific implementation of general tinnitus habituation therapy that uses direct counselling to reduce negative tinnitus-related reactions and the strength of the tinnitus signal [176]. As a result of tinnitus, reactions of stress, anxiety, panic attacks or loss of quality of life (fight, flight or freeze) occur. Without negative association, the fight–flight response to the tinnitus is suppressed. The main goal of TRT is to achieve habituation to the tinnitus by retraining auditory, limbic and autonomic processing in the brain [176]. This means that due to the high plasticity of the central nervous system, it is possible to reduce the response to repeated stimulation with neutral sound stimuli and by counselling [177].

In contrast to this, a working group set up in 1996 by the ADANO of the DGHNO-KHC took on the task of adapting TRT for German conditions and defining quality requirements. In the recommendation published in 2000 [178], it was suggested that sound therapy be expanded to include CBT interventions (TRT according to ADANO) and be carried out in a team by an ENT doctor, a licensed psychotherapist (doctor or psychologist) in cooperation with a hearing aid acoustician [65, 85, 86, 146, 179‐181].

This form of therapy is more appropriately termed ‘tinnitus management therapy’.

Recent studies [182, 183] show that TRT with noise device therapy is not superior to treatment with placebo noise generators, although other standard treatments show success here. The specifically defined counselling sequence seems to be of particular importance in differentiating the TRT protocol from other forms of therapy for tinnitus. The effect sizes are greater for patients with decompensated tinnitus. In addition, the presence of hearing loss and hyperacusis must be given special consideration.

Studies showing that sound therapy, which goes beyond CBT in the context of TRT according to ADANO, results in an additional benefit for the patient are not available and a direct comparison of the effectiveness between TRT and CBT [184] cannot yet be made definitively due to heterogeneous outcome variables. Further studies must provide clarity here. There is no convincing proof of the effectiveness of TRT according to the evidence-based criteria required here [105]. Recent studies on ‘classic TRT’, i.e. without psychotherapeutic interventions, do not show better efficacy than other habituation therapies with counselling as an essential component. If the therapy is carried out over a longer period, the results are stable even after 18 months [182, 185, 186].

General use of tones, auditory scenes and broadband or narrowband noise in the range of the tinnitus frequency has been tested and sold in many approaches and application forms for tinnitus treatment. Noisers as apparative applications, CD or other sound carriers, and more recently smartphone-based applications have been used. Highly sophisticated methods (e.g. external biostatistician, involvement of a centre for clinical studies) and external quality management were not used.

Effectiveness could not be proven for any of the methods, or studies on this topic were not initiated at all.

4.1.5.1 Music therapy

Initial clinical studies are available for three different specific forms of music therapy. For tinnitus-centred music therapy (TIM), in which the applied music is therapeutically changed within the tinnitus frequency, an application observation is available for 158 patients with acute and 18 patients with chronic tinnitus [187].

Argstatter et al. [188‐191] and Grapp et al. [192] published studies on music therapy according to the Heidelberg Concept. However, a larger and statistically carefully planned study is missing to make a recommendation [193]. Furthermore, the fact that (Heidelberg) music therapy also works with behavioural therapy units and relaxation therapy is not mentioned in the studies and consequently not taken into account in the results. More recent studies are not available.

4.1.5.2 Tailor-made notched music therapy

The working group of Pantev reports on person-specific filtered music applications (with recess—notch—of the tinnitus frequency, so-called tailor-made notched music therapy). The basic idea is that peripheral hearing loss leads to a reduced lateral inhibition in the range of the affected frequency, which should result in a cortical reorganisation. This maladaptive adaptation is supposed to be reversed by listening to music with an appropriate notch filter (usually half an octave). Initial evidence for this has been reported in 39 and 24 patients with chronic tinnitus, respectively, although only one study included a control population and was pseudorandomized and double-blind [194, 195]. Changes were observed in the perceived loudness of tinnitus and/or tinnitus distress. In the study by Teismann et al. [195], changes were only reported in patients with a tinnitus frequency of ≤ 8 kHz and partly only in follow-up measurements. In a subsequent study, 32 patients were divided into three groups. Two groups received either anodal (N = 10) or cathodal (N = 11) electrostimulation in combination with TMNMT. Overall, 11 patients received TMNMT and sham stimulation. All three groups reported an improvement in THQ (p = 0.04) when comparing baseline with post-treatment. The effect was still significant 1 month after the end of treatment. There was no modulation of the effect by electrostimulation [196].

These initial study results prompted a large-scale placebo-controlled trial of TMNMT. Stein et al. [197] published these results, in which 100 patients were randomly treated with a verum alienation (i.e. analogous to the previously determined tinnitus frequency) and 100 patients with a placebo alienation. The primary outcome measure (Tinnitus Questionnaire) did not show the predicted effect of treatment, but patients in the verum group reported a reduction in perceived loudness using a visual analogue scale (this effect was found only in the no-dropout analysis—F [1, 81] = 4.075; p = 0.047— and not in an intention-to-treat analysis). In a Chinese study, 43 patients with chronic idiopathic tinnitus were assigned to either a group receiving TMNMT or masking treatment (sound masking). The authors reported significant or highly significant changes in perceived loudness (visual analogue scale) and the Tinnitus Handicap Inventory ([198]; the publication is available in Chinese; only the abstract is available in English). Methodological aspects remain very unclear.

Furthermore, TMNMT was also studied in combination with Ginkgo biloba [199]: 26 patients were treated with the combination for 3 months. The THI score was reduced from 33.9 to 23.1 (p = 0.03), especially regarding the emotional component (p = 0.02). Due to the lack of control groups, it remains unclear in this study whether the reported effects are due to TMNMT, Ginkgo biloba or their interaction.

In the studies presented here, a reduction in tinnitus stress or a reduction in perceived loudness or a reduction of both is reported. It remains unclear where the differences come from. In the methodologically strongest study [197], only a reduction in loudness was reported, but not on the primary outcome variable (tinnitus distress). Furthermore, the patients studied suffered only from mild or moderate tinnitus.

The principle of TMNMT has now been implemented in hearing aids and tested on 20 patients with tonal tinnitus. The patients were randomly divided into a test group (N = 10) and a control group (N = 10) [200]. The test group received hearing aids with a notch filter, the control group received hearing aids without a notch filter. The outcome variable was the TQ at baseline and after 3 months. The two groups were statistically analysed independently of each other, which represents a methodological shortcoming. After 3 months, an improvement in TQ was shown for the test group (Cohen’s d: 0.84), but not for the control group (no statistics reported). The publication describes itself as a proof-of-concept and should be understood as such and nothing more.

There is a systematic review of internet and smartphone applications for the treatment of tinnitus [167], but only one of the studies reported above was included in the analysis [199]. This review also critically discusses the aforementioned study concerning the lack of a control group, but emphasises the improvement achieved especially in patients with tinnitus of a duration of fewer than 3 months (… ‘the improvement was good with this therapy form and is found to be more beneficial in individuals whose tinnitus is of recent onset of fewer than 3 months.’), who are then not classified as chronic according to all general criteria.

TMNMT treatment costs have even been reimbursed by some health insurances as part of a pilot project, although efficacy has not been proven in large-scale studies with external quality assurance. A survey of the 457 physicians who use this therapy also showed, with a response rate of 25.6%, that a significant improvement neither in tinnitus distress nor loudness could be achieved through notched music therapy [201].

In addition to TMNMT, another method was presented in which music was altered based on individual tinnitus characteristics. Li et al. [202] investigated the effect of listening to spectrally altered music on tinnitus. They studied 15 patients with chronic tinnitus who were asked to listen to this music for about 2 h a day for 1 year. The control group comprised 19 patients who listened to unchanged music. Changes in THI in the test group, but not in the control group, were found 3, 6 and 12 months after the start of therapy with medium effect sizes, as well as in a questionnaire for the assessment of anxiety symptoms 6 months after the start.

4.1.5.3 Sound therapy

Not only music but tones or complex sequences of tones and auditory scenes have been studied in various ways for the treatment of tinnitus. Durai and Searchfield [203] compared the influence of listening to natural sounds with static broadband noise in a cross-over design. Patients were randomly assigned to two groups in which the order of the listening condition was counterbalanced. The two phases lasted 8 weeks each, whereby the individual duration varied greatly. Overall, various effects were measured over time for both conditions, but as there was no untreated control group, these effects are difficult to interpret. The TFI as the primary outcome variable showed a greater reduction in the broadband noise condition, but the tinnitus loudness matching worsened. Overall, nature noise was expected to lead to improvement, but the opposite was true. In another study, broadband noise was compared as a control condition against a variant of sound therapy called ‘harmonic sound therapy’ [204]. In this therapy, sounds with narrowband noise around the first and fourth subharmonic are presented around the tinnitus frequency. A total of 18 patients were randomly assigned in a cross-over design with two arms and received the respective therapy conditions for 2 h a day for 3 months. Again, when the study arms were compared independently, several effects were seen in various variables over the time of treatment and when the arms were compared independently. When the study arms were compared directly, there was only a reduction in perceived loudness and THI. The analysis strategy has statistical flaws as not all comparison cells are included in the analysis.

Broadband noise compared to sine tones was investigated by Li et al. [205]. Overall, 14 patients each were randomly assigned to one of the two conditions and were to use the stimulation for 3 months, three times a day for 30 min. Both THI and loudness showed a reduction during therapy up to 12 months afterwards.

Simonetti et al. [206] used fractal tones, i.e. harmonic and melodic tones and tone sequences that are, however, not predictable. These tones were to be listened to for 8 h a day for 6 months. There was no control group and only six patients were reported. The small sample group showed a change in THI. Due to the small sample size and the lack of a control group, the data are difficult to interpret.

As another auditory stimulation procedure, Munro and colleagues [207] used ocean noise presented binaurally with ‘binaural beats’. The perception of binaural beats occurs when two stimuli with small frequency differences are presented simultaneously in dichotic listening. This creates the perception of a modulation. In 20 patients, the effects of this 10-min stimulation were compared with those in 20 patients with stimulation with unmodified ocean noise. No differences were found between the stimulation conditions.

Noise stimulation was also studied during sleep in tinnitus patients [208]. In total, 58 patients were randomly divided into three groups receiving either an individualized tinnitus sound or a non-individualized sound, both conditions via in-ear headphones, or a sound via a bedside device. The application was to be every night for 3 months. The application of the two conditions with in-ear headphones reduced the scores for tinnitus distress compared to the third group. The individualized noise resulted in a reduction in tinnitus loudness compared to the other two noise conditions.

Tyler et al. [209] investigated the effect of masking tinnitus compared to no masking in a cross-over design in 18 patients. The data were not statistically analysed and the description of the effects remains descriptive.

Masking of tinnitus in patients with and without hearing loss was studied by Rocha and Mondelli [210]. In each case, 15 patients were assigned to a non-randomised group and received tinnitus masking (in patients without hearing loss) or masking with a hearing aid (in patients with hearing loss) for 6 months. The authors report an improvement in THI and VAS in both groups, which cannot be assessed conclusively as crucial statistical parameters are not reported. Comparison of the groups is problematic because of confounding variables.

Sereda, Davies and Hall [211] report on a descriptive level in a feasibility study of eight patients that hearing aids with tinnitus-specific masking are applicable and user-friendly.

4.1.5.4 Acoustic neuromodulation

Acoustic reset neuromodulation is a procedure that aims to achieve tinnitus reduction by presenting certain tones in the range of the individual tinnitus frequency. A first clinical pilot study on 63 patients (unilaterally blinded, randomised, placebo-controlled) showed statistically significant improvements. However, the placebo group of 5 patients was very small and their tinnitus was in a different frequency range [212]. Results of phase III studies are not available or were not allowed to be published at the request of the study sponsors [213]. An uncontrolled cohort study suggested efficacy, but also described isolated tinnitus exacerbations [214]. A systematic review concluded that the available evidence is insufficient to recommend the procedure in routine clinical practice [215]. Placebo-controlled studies are lacking [216].

Treatment of acute tinnitus follows the treatment recommendations for acute sudden hearing loss. For the treatment of acute hearing loss, corresponding guidelines (AWMF 017/2914 currently being updated [82]; guideline USA [83]) recommend systemic or intratympanic steroid treatment with, however, moderate evidence. Therefore, acute tinnitus with hearing loss or following hearing loss should be treated appropriately. If the tinnitus occurs acutely without measurable hearing loss, the standard cortisone therapy recommended for hearing loss is not recommended.

Drug therapy for chronic tinnitus:

A review from the United States essentially refers to two types of pharmacotherapies for tinnitus. The first is being researched in the hope of directly eliminating tinnitus. The second group of drugs has been developed to treat or at least alleviate possible comorbidities and thereby also improve the quality of life of tinnitus patients. While there are practically no promising approaches for the first category to eliminate tinnitus from perception, there are numerous medications that are suitable for the treatment of comorbidities; usually modern antidepressants [217]. These are used because individual psychosomatic factors can play a decisive role concerning the actual distress of the tinnitus [4].

Therapeutic approaches such as intratympanic steroid treatment do not affect chronic tinnitus [218, 219] if it does not occur in conjunction with acute hearing loss. A temporary increase in tinnitus intensity and tinnitus distress is not considered new-onset tinnitus but should be considered and treated as a fluctuation in the subjective perception of chronic tinnitus [220].

Many classes of drugs have been used or tried for chronic tinnitus, including various antiarrhythmics, anticonvulsants, anxiolytics, glutamate receptor antagonists, antidepressants, muscle relaxants and others [221], with little evidence of greater benefit than harm [222]. Experimental studies, some of them phase II or even phase III, with NMDA receptor antagonists or AMPA receptor antagonists had also shown promising approaches, but ultimately a breakthrough was not apparent, which is why these types of drugs did not reach market maturity [223].

Meta-analyses and RCTs with usable results and numerous non-usable and comparable publications are available. For some preparations, there are indications of possible efficacy in individual randomised clinical trials, but there are no replicated positive results from randomised clinical trials with sufficient evidence or positive results from meta-analyses for a single preparation. Accordingly, neither the European Medicines Agency (EMA) nor the Food and Drug Administration (FDA) has approved any preparation for the treatment of tinnitus [224]. The tinnitus guideline from Great Britain, which was updated in 2020, also made a recommendation against drug treatments, especially the drug betahistine [225].

For betahistine, Ginkgo biloba and antidepressants, Cochrane meta-analyses show no evidence for the efficacy of the respective preparations in chronic tinnitus.

4.1.6.1 Betahistine

A Cochrane Review of the use of betahistine in the treatment of tinnitus states that in England, for example, more than 100,000 prescriptions for betahistine are written out for tinnitus each month, especially by general practitioners but also by specialists. In the Cochrane Review, five studies with a total of 305 participants were evaluated. Differences in tinnitus distress and the corresponding questionnaires after treatment with betahistine or placebo were not found, nor were accompanying symptoms such as depression influenced differently by betahistine than by placebo. Thus, there is no evidence for betahistine in the treatment of chronic tinnitus [226].

4.1.6.2 Antidepressants

The Cochrane Review of antidepressants for tinnitus [227] identified six RCTs (610 patients) on this topic. Only one study was considered to be of high quality. This study compared the effect of paroxetine (SSRI—serotonin reuptake inhibitor) with placebo and showed no significant difference in effect between the groups. No effect was shown for trazodone (serotonin antagonist and reuptake inhibitor) and only a small effect was shown for tricyclic antidepressants, but this may be due to methodological problems in these studies. Side effects were frequently reported, including sedation, sexual dysfunction and dry mouth. Nevertheless, antidepressants are often used successfully in the treatment of depression and anxiety; however, not to improve tinnitus, but to treat the accompanying depression and/or anxiety symptomatology or a distressing sleep disorder.

4.1.6.3 Benzodiazepines

The systematic review by Jufas and Wood [228] on the use of benzodiazepines for tinnitus included six clinical trials that investigated the use of diazepam, oxazepam and clonazepam. There were mixed results across the studies and methodological issues, and thus this limits the assessment of effect. In summary, it was concluded that the use of benzodiazepines for subjective tinnitus does not have a robust evidence base and that these drugs should only be used as bridging drugs with strict indications, e.g. in the context of the initiation of antidepressant therapy, because of their considerable side effects (especially the short-term development of dependence).

4.1.6.4 Ginkgo biloba

Ginkgo biloba is the most commonly used herbal supplement for tinnitus. Older systematic reviews included four RCTs on Ginkgo biloba for tinnitus as a primary complaint [229]. A second systematic review included five RCTs, with most studies having low methodological rigour and conflicts of interest [230]. The results were positive for Ginkgo, but the authors stated that a clear conclusion on efficacy was not possible. A meta-analysis pooled data from six RCTs and concluded that Ginkgo has no benefit over placebo [231]. Ginkgo biloba can interact with other blood thinners to cause severe bleeding, which may increase the risk of bleeding in patients with underlying coagulation disorders [232].

A review from Norway as an update of a Cochrane Review [229] evaluated new randomised and placebo-controlled studies on the efficacy of Ginkgo biloba with a total of over 6000 patients: Evidence was found neither for efficacy in cognitive deficits, in dementia, in apoplexy, in claudicatio nor in tinnitus. Rather, there are—albeit mild—side effects such as dizziness, stomach complaints or allergic reactions, sometimes also an increased bleeding tendency [233]. As early as 2001, the efficacy of Ginkgo extract for tinnitus was investigated in the British Medical Journal (BMJ) in a double-blind and placebo-controlled study with a very large patient population (1121 participants). Ginkgo led to the same (small) improvement in tinnitus penetrance and intensity as a placebo [234].

Also in connection with other indications, Ginkgo extracts only showed an effect under very special circumstances. In a systematic review by Spiegel et al. [235], positive effects of EGb 761 (Ginkgo extract) compared to placebo were demonstrated in dementia in combination with tinnitus. The basic assumption for this analysis is that Ginkgo extract has a positive effect on tinnitus and vertigo symptoms, which has already been proven elsewhere. Five studies were systematically included that fulfilled the inclusion criteria: a diagnosis of dementia according to generally accepted criteria, long-term treatment of at least 20 weeks as well as simultaneous measurement of parameters of other indication areas such as the simultaneous occurrence of tinnitus and dizziness and finally an evaluation before and after the treatment. For this, five studies were identified that found EGb 761 to be significantly superior to placebo for both tinnitus and dizziness. The authors conclude that EGb 761 is also effective for neurosensory symptoms associated with dementia. Why numerous other randomised and placebo-controlled studies investigating (and failing to confirm) Ginkgo effects on tinnitus were not included in the analysis is not answered in the paper.

A meta-analysis of three reviews, on the other hand, found no influence on the severity of tinnitus and no improvement in tinnitus intensity and quality of life after administration of Ginkgo biloba [236].

The methodological comparability of the studies available so far is not given with different administration of the drug compared to placebo and different outcome variables.

4.1.6.5 Zinc

A Cochrane analysis evaluated RCTs assessing zinc against placebo in the treatment of tinnitus in adults. Three studies with a total number of 209 participants were reviewed. Improvement in tinnitus distress was assessed with questionnaires (Tinnitus Handicap Questionnaire). On the whole, there was no significant difference in any of the studies compared to placebo, neither in terms of primary treatment success nor in terms of secondary success rates. However, no side effects were described or found. None of the studies could find a significant improvement in quality of life or even an improvement in depression and anxiety in patients treated with zinc [237].

4.1.6.6 Melatonin

Melatonin is a hormone secreted by the pineal gland and is involved in the regulation of the sleep–wake cycle.

Three RCTs with a total of 193 participants have studied melatonin for the treatment of tinnitus and each has shown benefits with the greatest improvement in patients with severe tinnitus and insomnia [238]. However, given the small total number of patients studied and the methodological limitations, including the lack of a placebo group in the largest study, these results should be interpreted with caution. Although another study showed a potential benefit for patients with concurrent sleep disturbance due to tinnitus, this study lacked randomisation, blinding or placebo control [239]. Only one study reported potential side effects of melatonin, including bad dreams and fatigue [240]. A review evaluated studies investigating the treatment of tinnitus patients with melatonin. In five studies, no evidence was found for therapeutic success concerning tinnitus, but at least sleep disturbances improved [241].

4.1.6.7 Oxytocin

In a pilot study, the treatment of chronic ringing in the ears with the hormone oxytocin was investigated. Oxytocin is a hormone that, among other things, helps to regulate signal transmission in the brain and is stimulated by high doses of oestrogen. It mainly regulates contractions during birth and the production of breast milk and sperm. Social factors are also thought to be controlled by regulation in the amygdala, such as empathy and trust, but also attention to acoustic stimuli.

The study tested whether attention to chronic ear noises can also be influenced by the hormone: 15 patients with chronic tinnitus were given oxytocin as a nasal spray twice a day for 10 weeks. Five of these patients reported an improvement in their ringing in the ears—it became quieter. A second double-blind, placebo-controlled study was then conducted with 17 tinnitus patients who received only a single dose (oxytocin or placebo). Here, the patients who received the hormone reported a minimal improvement in their tinnitus 30 min to 24 h after hormone administration. Overall, there was only a very small difference compared to the placebo. In the long-term treatment (15 patients over 10 weeks), the reduction in tinnitus distress and loudness was measurable, but also not very strong and not placebo-controlled [242].

4.1.6.8 Steroids

Controlled studies on the treatment of chronic tinnitus with systemic steroid administration are not available. In a corresponding search, only 2 studies were found that used corticosteroids intratympanically, both without significant effect. One study randomised 70 adult tinnitus patients to intratympanic treatment with either methylprednisolone or saline. However, the severity of tinnitus did not change significantly in any group [218]. The other study also randomised intratympanic treatment (n = 36) with dexamethasone or saline, also without significant differences [243]. For a more recent study, 107 patients aged 20–77 years with idiopathic chronic tinnitus were randomised. One half received six intratympanic injections of dexamethasone, a total of twice a week for 3 weeks. The control group received saline injections. Treatment outcome was measured with the Tinnitus Handicap Inventory (THI), before treatment, after 1 week, after 1 month and after 6 months. In this study, the verum-treated group showed an improvement in THI after 6 months compared to the control group, although only just marginally significant, while hearing thresholds did not improve with this treatment [219]. Other studies with a similar study design had not found any effects in the case of long-term tinnitus, but at best in the case of acute tinnitus.

4.1.6.9 Medicinal influence on neurotransmission

4.1.6.10 Cannabis

In recent years, the use of cannabinoids has been discussed for practically all chronic diseases, and the first studies exist regarding the evidence of cannabis treatment in ENT medicine. A review examined 79 publications but discussed all possible ENT indications such as blepharospasm, alleviation of radiation side effects and psychological treatment of a cancer diagnosis. According to these authors, real evidence for a meaningful cannabis treatment is not yet given; this would have to be provided by further research [245].

One paper directly discusses the effect of cannabinoids on tinnitus [246]. It essentially refers to animal experiments that show that effects on the dorsal cochlear nucleus are to be expected in the brain through cannabis medication and thus hyperactivity and consequently an intensification of the tinnitus could be achieved. A positive or alleviating effect on ringing in the ears and tinnitus is therefore not to be expected from cannabis, at least according to the animal studies currently available.

In a recent review, connections between cannabis effects and an influence on the immune response of the auditory system are presented and a potentially positive effect for tinnitus therapy is postulated, without, however, being able to draw on studies for this [247].

Based on the aforementioned data, no pharmaceutical can be recommended for the treatment of chronic tinnitus. However, if psychiatric comorbidities (e.g. anxiety disorders, depression) exist in connection with tinnitus, these should be treated. Regarding the type of treatment, reference is made to the corresponding guidelines (S3 guideline for the treatment of unipolar depression; S3 guideline for anxiety disorders).

Repetitive transcranial magnetic stimulation (rTMS) is a procedure that enables non-invasive influencing of neuronal excitability of superficially located brain areas. For about 20 years, rTMS has been investigated for the treatment of tinnitus. There are numerous randomised controlled studies with a total of over 1000 participants. In some cases, different stimulation protocols were investigated [248].

Only some of these studies showed significant efficacy of rTMS. Systemic reviews and meta-analyses all concluded that the treatment has few side effects and is safe. The study data on efficacy are heterogeneous.

A methodically good and comprehensive randomised and placebo-controlled study evaluates the therapeutic effect of 1‑Hz low-frequency repetitive transcranial magnetic stimulation for tinnitus treatment. The authors, who have many years of study experience with this type of treatment, compared the efficacy of a 2-weeks treatment with 10 sessions and 2000 stimuli each on the left temporo-parietal cortex against a comparative treatment with a sham coil. In total, 153 patients were enrolled in the study; 75 received the real treatment, and 78 the sham treatment. After the treatment, the tinnitus distress remained identical in all patients, and the accompanying symptoms such as quality of life and depression, which were also examined, showed no significant difference between the two treatment groups. Although the therapy was well tolerated overall, it did not show a therapeutic effect; the follow-up period was 6 months [249].

Systematic reviews including a Cochrane analysis concluded that short-term treatment effects are detectable, but longer-lasting effects are not proven [249, 250]. The number of studies is limited, and long-term effects (including side effects) have not been sufficiently investigated.

In a recent meta-analysis on different forms of neurostimulation, the effects of 28 randomised controlled studies on rTMS were systematically analysed [251]. It showed that active rTMS significantly reduced tinnitus distress at the end of treatment compared to sham treatment (effect size: −0.45; CI = −0.66; −0.24; p < 0.0001; [252]). Longer-term effects of treatment were evaluated in some studies for the period between 1 week and 6 months after treatment. These showed sustained improvements (effect size: −0.42; CI = −0.68; −0.15; p = 0.0024). The effects were most pronounced with stimulation of the left auditory cortex, and women responded better to treatment than men did.

A Chinese meta-analysis evaluated 22 studies with a total of 1228 patients in which rTMS was compared with sham treatment. The mean difference in THI reduction between active rTMS and sham was −7.92 (CI = −14.18; −1.66) 1 week after treatment, −8.52 (CI = −12.49; −4.55) 1 month after treatment and −6.53 (CI = −11.4; −1.86) 6 months after treatment. However, not all studies (only two) were included in the 6‑month follow-up, especially those that found no improvements [253].

Since the minimum clinically relevant reduction (MCID) for the THI scale is 7 points [254], the effects observed in the follow-up period of 6 months compared to sham treatment are consistently classified as clinically relevant.

However, the studies examined in this meta-analysis were also very heterogeneous; the results were not significant in 13 studies.

In a recent meta-analysis that also evaluated the same studies, Dong et al. [255] concluded that rTMS did not provide a significant improvement compared to sham treatment. Ten RCTs with 567 participants were analysed. Neither short-term (p = 0.72), nor medium-term (p = 0.41), nor long-term (p = 0.14) showed significant improvements for tinnitus distress in either the THI or the TQ.

No serious side effects were reported in any of the studies. Headaches were listed as the most frequent side effects of active rTMS. The frequency of side effects did not differ significantly between active and sham rTMS, although the difference was not significant [253].

However, it must be taken into account that these data are based only on RCTs of rTMS for tinnitus with about 1000 patients, and not all of these trials systematically reported side effects.

In summary, the efficacy of repetitive transcranial magnetic stimulation is questionable.

4.1.8.1 Transcranial electrical stimulation

In transcranial direct current stimulation (tDCS), a low current (0.5–2 mA) is applied to the cortex via the scalp. Depending on the polarity, this results in an increase or decrease in cortical excitability in the stimulated areas. Fregni et al. [256] were the first to propose this for tinnitus treatment. A review [257] included 17 studies, but only two were randomised and controlled. The meta-analysis concluded that there was insufficient evidence that tDCS was effective against tinnitus. Further RCTs of tDCS with different forms of stimulation were requested. Since then, further studies have been published, which certify that the method is safe, but that the effect on tinnitus is small to non-existent.

A Swiss study investigated the effect of transcranial direct electrical stimulation (tDCS) in a double-blind and placebo-controlled manner on 42 patients [258]. The cathode was applied via the auditory cortex, the anode was applied prefrontal. Although there were no side effects, there were no effects on tinnitus.

In another study, 40 patients were treated for 6 months either with anodal tDCS or with a so-called sham stimulation, i.e. placebo. Afterwards, all were fitted with hearing aids for 6 months. After 3 months of hearing aid use, all improved in tinnitus distress, regardless of whether they had previously received real or sham stimulation [259].

A Japanese research group investigated the influence of tDCS on connectivity [260]. Since the connectivity between the left and right auditory cortex seems to be less pronounced in tinnitus patients than in normal-hearing patients, nine tinnitus patients were compared with nine control participants. After tDCS treatment, the connectivity between the auditory cortex and somatosensory and motor brain areas decreased in the tinnitus patients, while this connection remained strong in the control group.

A total of 22 patients with tinnitus existing for more than 6 months were divided into two groups in a study; 11 were stimulated anodally with 2 mA for 20 min over left temporo-parietal areas in five sessions, and 11 patients had sham stimulation. Tinnitus loudness in VAS, THI and stress level were assessed. No significant differences were found between the anodal and sham stimulation groups. The tinnitus distress of some patients worsened or the tinnitus changed; long-term effects were not found [261].

One study compared tDCS with sham stimulation at the shoulder in 24 patients. No significant differences were found in the TFI between the two groups, not even concerning tinnitus intensity [262].

In another study, 35 patients with chronic tinnitus were treated with tDCS for 10 days for 20 min each, either left temporal or bifrontal stimulation or only with sham. All three groups improved in THI scores, but differences between groups were not found [263].

In a double-blind placebo-controlled study, 25 patients were treated with tDCS in 10 sessions, and 15 patients received sham treatment. After 1 month, various data were evaluated, in particular THI and VAS scores. Some side effects were also accurately recorded, especially itching, but without any serious side effects. However, the paper only describes the data collected concerning future study protocols; specific results were not reported [264].

A meta-analysis summarised the results of 32 RCTs and found a positive effect, especially for cathodal transcranial direct current treatment of the cortex, concerning the severity of tinnitus and the improvement of quality of life. However, measuring instruments were not described and outcome parameters were not evaluated reliably; long-term observations were also not available. Four studies investigated tDCS, and two compared tDCS with tRNS (transcranial random noise stimulation). Only in two (older) studies was the outcome with tDCS better than sham stimulation [265].

Out of a total of 85 studies on tDCS, 34 were evaluated in another review [266]: Here too, however, mainly the very different forms of application and study designs were described; an evaluation of the concrete therapy results of the individual studies was not undertaken. It is emphasised in this review that tDCS is often mixed with other methods such as various forms of acoustic stimulation in particular, which makes an overall evaluation difficult.

4.1.8.2 Vagus nerve stimulation

Vagus nerve stimulation is thought to stimulate cholinergic basal nuclei, which are assumed to be responsible for extensive changes in cortical organisation. This is hypothesized to improve learning effects, as has been found in animal studies. Tinnitus patients should be acoustically stimulated to ‘unlearn’ their tinnitus. Stimulation of the vagus nerve should then support this learning [267]. Vagus nerve stimulation can be done invasively by an implanted vagus nerve stimulator or non-invasively as transcutaneous vagus nerve stimulation by electrostimulation of the external auditory canal. Through these mechanisms, stimulation of the vagus nerve, coupled with tonal stimuli, should enable the treatment of tinnitus. Experimental studies could prove the safety of the procedure, both for direct and transcutaneous stimulation [268‐272].

While the first studies implanted a stimulation electrode on the neck, in a less invasive procedure an electrode was positioned transcutaneously on the auricle and stimulation of the vagus nerve was attempted for 30 min with a pulse rate of 25 Hz and an amplitude of 1–10 mA. During this time, the patients listened to music filtered at the tinnitus frequency. A total of 30 patients were treated; after 10 sessions, 50% reported relief from tinnitus. Side effects did not occur [273].

In 2014, an electrode was implanted cervically on the vagus for a study of tinnitus patients in the United States. The patients listened to music daily and were stimulated regularly. In 2017, this study was completed and 16 patients were treated in this way for 1 h daily for 1 year. They were compared with 14 patients who had also received implants, but for whom the therapy had only begun 6 weeks later. In these first weeks, the improvement of the tinnitus distress of the therapy group was 10% higher than in the control group; overall, the tinnitus distress of the treated group improved by 50%, that of the other group by 28%. The interpretation of the results classifies the improvement as significant, measured with the THI and other questionnaires. Tinnitus loudness was measured audiometrically. In addition, the Beck Depression Inventory (BDI) and an anxiety questionnaire were used. There were no significant changes according to the tables and graphs in any of the measured values. In the follow-up, subgroups were examined and it was found that the patients who did not have high-frequency tinnitus and noise-induced tinnitus seemed to benefit more. Finally, the authors state that vagus nerve stimulation paired with tones may be effective for subgroups of tinnitus patients, but this would need to be clarified in larger studies [274].

It has also been suggested that vagus stimulation could be supported by simultaneous administration of a muscarinic receptor type 1 to improve learning effects [275]. It has also been investigated whether cervical implanted vagus stimulation paired with tones alters speech understanding and influences voice and hearing functions. However, the study of seven tinnitus patients did not find any side effects of vagus stimulation in this regard [276].

A review from 2020 summarises all areas of application of invasive and transcutaneous vagus stimulation. The main areas of application are depression and epilepsy. Only a few studies can be claimed for tinnitus; results are not recorded and referenced [277].

4.1.8.3 Bimodal acoustic and electrical stimulation

The bimodal acoustic and electrical stimulation is supposed to simultaneously stimulate the auditory pathway and the trigeminal pathways thereby inducing plastic changes in the brain. This is based on animal experiments, for example, as described by Markovitz et al. [278].

In this therapy method, different acoustic stimuli adapted to the patient’s hearing ability are simultaneously given via headphones, and trigger points on the cheek and neck are electrically stimulated. This is intended to activate touch-sensitive nerves in the area of the auditory pathway. Since these parts of the auditory pathway are decisive for the amplification of tinnitus, which was established in animal experiments, success is also assumed in humans. In the pilot study after an animal study, 20 patients were then treated, in whom the tinnitus loudness and annoyance were reduced after 28 days. However, the effect only lasted 30 days. Treatment success is seen in a post hoc analysis, especially for patients with somatotinnitus [279, 280].

Another method combines the presentation of tones with electrostimulation of the tongue [281]. Again, only a pilot study or study protocol has been published to date, but this does not include a placebo group [282]. Although these seem to document general tolerability, valid and above all placebo-controlled studies are not yet available, but have been announced. The new study will include 192 patients who will be treated in four arms and followed up for 1 year. This study will also include an acoustic-stimulation-only arm without tongue stimulation [283]. In 2020, Conlon et al. [284] published the results of the study presented in 2017, unfortunately without placebo control. For this, 326 patients were recruited in two centres and randomly assigned to three equally sized stimulation groups each in a double-blind fashion. These groups differed in stimulus parameters and the sounds paired with them, whereby the first group used high-frequency sounds with synchronously applied electrical stimulation of the tongue, the second group used comparable stimuli without synchronisation, and the third group used acoustic stimulation with low-frequency sounds that were not synchronised with electrical tongue stimulation. Treatment was given for 12 weeks with 2 × 30 min of application daily. After treatment, the scores of all three groups improved significantly, the findings were measured with THI and TFI and checked again after 12 months, where they remained stable. The data of 20% of the participants did not improve with the bimodal stimulation. The evaluation does not differentiate which different hearing losses were present and especially why the values of all groups improved, although the stimulation parameters of the third group were relatively unsuitable for the postulate of neuroplastic changes and thus at least suggest a placebo effect. Side effects were described but were not severe. Furthermore, there is a considerable bias in the study (industry sponsorship bias, 11 of 13 authors had a conflict of interest in this respect)

4.1.8.4 Invasive electrical stimulation

Invasive forms of tinnitus treatment are so far solely experimental and include vagus nerve stimulation with an implantable electrode, constant electrical stimulation of the nervus vestibulocochlearis, extradurally implanted electrodes for brain stimulation and neural stimulators for deep brain stimulation. These are invasive methods that are not suitable for widespread and general use. Research in this area is limited to a few cases, and the pathophysiological mechanisms on which such therapies are based are neither sufficiently explored nor understood [285‐288]. Controlled studies and meta-analyses are lacking; in addition, considerable side effects have been described in some of the few cases.

A review paper shows possible therapeutic perspectives through deep brain stimulation (DBS), in which an electrical pulse generator (brain pacemaker) is implanted in brain structures. The therapy has proven itself in the symptomatic treatment of otherwise therapy-resistant Parkinson’s disease and should now be examined concerning its possible use in tinnitus. This is based on considerations that in tinnitus patients, stimulation zones have been found in very different areas of the brain, which would then have to be controlled individually [289].

Anatomically and surgically targeted implantation of electrodes seems feasible if it would be therapeutically useful for tinnitus [290].

In a double-blind, placebo-controlled and randomised cross-over study, an electrode was implanted in the auditory cortex under general anaesthesia in nine patients with severely distressing unilateral tinnitus. This electrode was connected to a stimulator inserted in the pectoralis region. Patients were biphasically stimulated for 4 months, then randomised into two groups with either placebo or true stimulation. After a wash-out phase, cross-over stimulation was performed again. Therapy success was measured with a questionnaire (structured tinnitus interview—STI). In one patient the stimulator had to be explanted due to severe psychological decompensation and in three patients it was explanted after the end of the study, while five patients were still stimulated for another 3 years. During the open phase, the condition of five patients improved and that of two patients worsened. In the subsequent controlled phase, improvements were also seen, but in both the placebo and the directly stimulated group. No side effects were seen in any of the patients, in particular no hearing changes. The authors conclude from this study that direct electrostimulation of the auditory cortex is associated with a significant placebo effect, simply because of the surgery. However, there is no therapeutic effect on tinnitus distress when placebo effects are taken into account [291].

Six patients were stimulated with deep brain stimulation (DBS) at different positions of the caudate nucleus and were examined concerning a change in tinnitus. The control was carried out utilising functional magnetic resonance imaging, and they were compared with 14 patients matched according to TFI. A reduction in tinnitus loudness was achieved by stimulating five positions of the caudate nucleus, but not at 15 other positions. However, the changes only lasted for the duration of the stimulation [292].

In a similar study, six patients with therapy-resistant tinnitus had an electrode implanted in both caudate nuclei and were continuously stimulated for 24 weeks. One patient had to be explanted because of suicide risk; in the others, THI and TFI scores were evaluated. In three patients, the TFI improved by > 13 points, in four the THI by > 20 points; they were classified as responders. Side effects occurred but were reversible: postoperative pain, tinnitus worsening in all six participants, headache in five and milder side effects in two to three participants. There was no follow-up [293].

A review examined 21 studies since 2005 that invasively treated patients with severe and refractory tinnitus with various forms of neuromodulation. The studies were all of low quality due to small numbers of cases, lack of control groups and imprecise definition of outcome parameters. Although isolated successes have been reported, there is no evidence to support the use of invasive stimulation techniques as an alternative treatment option for chronic tinnitus [294].

4.1.8.5 Intracochlear electrical stimulation

See Cochlear implant (Sect. 4.1.2.3).

4.1.8.6 Transcutaneous electrical nerve stimulation

In TENS treatment, interference currents with a frequency of 100–300 Hz are usually applied directly to the auricle or on the mastoid. The electrodes are applied like a band-aid or held in place during treatment. The treatment devices, usually used for orthopaedic complaints, can be purchased by the patient. Scientific studies have been published only sporadically, so far without scientific evidence.

In 2014, Lee et al. [295] studied 65 patients with chronic tinnitus, 45 of whom received TENS treatment and 20 of whom only received sham treatment. Therapy success was measured with the THI and VAS. Subjective improvement was experienced by 62.2%, especially those patients who had low-frequency tinnitus and mild hearing loss, but relief lasted only 1 month.

In 2019, Li et al. [296] treated 46 patients with acute tinnitus: 23 with TENS, and 23 with sham treatment. Therapy success was measured with the TQ and THI; there was a just significant improvement of p < 0.01 after 4 weeks compared to placebo treatment.

Tutar et al. [297] treated 60 patients with chronic tinnitus in ten TENS sessions every 4 days. The patients were divided into three groups: 20 patients each were treated on one ear, both ears, or had only sham treatment. The treatment was carried out with a frequency of 200 Hz and 10–30 mA. The values of both treatment groups improved significantly (p < 0.05) compared to the non-treatment. But the values of the control group also improved, albeit less. The authors therefore also emphasise a placebo effect.

Feasibility studies, reliable therapy control studies or meta-analyses with larger numbers of cases are not available for TENS treatment.

A meta-analysis of this treatment assessed 17 studies with 1215 participants, but only four studies could be evaluated for a meta-analysis. There were significant improvements in THI scores and VAS values. The complaints of 40% of the patients improved for at least 3 months [298].

4.1.8.7 Low-level laser therapy

In so-called low-level laser therapy (LLLT—formerly: ‘soft laser’), which has been marketed since 1985, the ear, mastoid or auditory canal is irradiated with low intensity (approx. 100 mW) and a wavelength of usually 830 nm for 15–20 min over several days. Work on anatomical specimens in the past has shown that this irradiation cannot reach the inner ear or even higher structures of the auditory pathway. There has been no scientific evidence, but occasionally new studies have been presented that use this form of therapy.

In 2015, Dehkordi et al. [299] concluded that after treatment of 66 tinnitus patients, of whom 33 received real and 33 received sham irradiation, no differences were found between the groups; in both groups the suffering from tinnitus improved slightly (p < 0.589 in TSI).

Choi et al. [300] also saw no significant improvement in 2019 after treatment of 38 patients (19 real, 19 sham): no significant improvements and no placebo effects.

A recent systematic review [77] lists three RCTs and records positive results for the application of physiotherapeutic/manual techniques in the cervical spine, including cervical mobilization, myofascial techniques and osteopathy. However, the methodological quality of the included studies is low. Another systematic review [301] includes two RCTs on cervical spine treatments. Cervical spine therapy can record positive effects concerning the severity of tinnitus and includes physiotherapeutic exercises and trigger point treatment. The methodological quality of the RCTs included is again low. In addition, a current RCT [302] with 61 participants concludes that a combination of physiotherapeutic and manual therapy (exercises for craniocervical and temporomandibular joints as well as myofascial techniques) together with patient education offers a significantly better outcome concerning tinnitus severity than physiotherapy with education alone.

Several nutritional supplements have been used for tinnitus, including lipoflavonoids, garlic, homeopathy, traditional Chinese–Korean herbal medicine, honeybee larvae, and various vitamins and minerals. Evidence of the efficacy of these therapies for tinnitus is not available [232]. A more recent review also finds that there is no evidence of dietary supplements on tinnitus. In particular, a reduction of salt or caffeine had neither a positive nor a negative effect on tinnitus perception [303]. In a review by Wegner et al. [226], five studies with approx. 300 tinnitus patients each were examined concerning betahistine administration versus placebo (vitamin B₁₂), and no effect (−0.16) on tinnitus loudness and impairment was found after 12 weeks.

Concerning the effect of acupuncture in patients with chronic tinnitus, no recommendation can be made due to the existence of qualitatively insufficient studies; a benefit cannot be demonstrated, and minimal harm is assumed.

A systematic review of acupuncture for the treatment of tinnitus included nine RCTs with a total of 431 participants [304]. This systematic review highlighted the heterogeneity of study designs as well as their methodological limitations using the Cochrane bias risk assessment tool. There were variations in study design of the types of acupuncture interventions, frequency, intensity and duration of treatments, selection of other control groups, as well as variability with blinding and selection of outcome measures, many of which were not validated [303]. The authors concluded that the small number of RCTs of acupuncture for the treatment of tinnitus with small sample sizes and methodological issues is insufficient to conclude effectiveness.

Recent reviews (18 studies with 580 tinnitus patients [305] and five studies with 182 tinnitus patients [306]) show a weak effect of (electro)acupuncture compared to the various comparison groups. However, the methodological limitations described above still apply, and thus no reliable conclusions on the effectiveness of acupuncture are possible.

Despite some studies on the significance and functionality of self-help (SH) for tinnitus, there is little evidence or no knowledge of whether and how SH has a specific effect on tinnitus. Presumably, this is because the largest part of SH is carried out voluntarily or autonomously out of one’s pocket in the form of membership fees and indirect costs such as the time spent personally.

Since nothing can be sold or earned here, no lobby would encourage studies on the evidence of SH.

What is SH? If we want to be consistent with the term, in the true sense of the word, SH is when those affected help each other: Mostly this happens in face-to-face SH groups (SHG) whose meetings are more or less structured with a fixed core and changing participants or who contact each other in internet forums.

However, the term SH moves away from the idea of SH when working through the tinnitus issue oneself with the help of a ‘self-help book’ or ‘internet-based self-help’ or increasingly with tinnitus apps misleadingly counted as the original SH.

If minimal-contact interventions with experts are integrated into the latter forms of intervention, it is a matter of treatment/therapy and no longer of SH.

In healthcare reforms, the stronger involvement of patients in their healthcare is a common theme. Self-help is commonly understood as seeing tinnitus sufferers as active partners in their healthcare, taking responsibility for their well-being. It is an integral part of the UK Department of Health tinnitus management guidelines [225], the US tinnitus practice guideline [307], the American Academy of Otolaryngology (AAO; http://​otojournal.​org) and the American Tinnitus Association ([308]; ATA; McGinnis 2001; www.​ata.​org).

Despite many publications on the importance and functionality of SHG activity, there is little evidence on whether and how SH specifically affects tinnitus [308, 309].

Only recently, a study on the effect of SH was published [310]. At the time of the survey, 800 of the 13,000 members of the non-profit SH organisation Deutsche Tinnitus-Liga (DTL) were active in SH groups. In a cross-sectional study, the tinnitus distress (Mini-TQ; [64]) was assessed in 1108 affected persons (mean 61 years, 60% men) using valid items of the Structured Tinnitus Interview (STI; [310]) on tinnitus knowledge (TK), tinnitus coping and quality of life (QoL; WHOQOL-BREF 1998). As a result, the regression analyses show highly significant correlations of community SH activity with TK, coping as well as health system orientation and self-confidence, but no significant differences concerning general QoL.

Recently, SH activity was shown as a gradient in the four groups of current SHG participation (n = 217), previous SHG participation (n = 118), DTL membership without SHG participation (n = 641) and no references to SHG or DTL (n = 132). The former and latter groups thus show the greatest differences and most clearly demonstrate the benefits of SHG participation [310]. When controlled logistic regressions are determined for age, gender, education, equivalised income and tinnitus distress, odds ratios of 6.94 for TW, 3.83 for knowledge of help options, and 7.75 for self-confidence in knowing more about tinnitus than most doctors are found between these two groups. The other two groups are in between with corresponding and also significant gradations. Despite all the methodological limitations of this cross-sectional study, it seems more likely that tinnitus-related knowledge and other benefits are a result of SHG participation than vice versa.

Smartphone apps as guided SH? A plethora of apps for tinnitus can be found, which in the broadest sense of the development of digital coping forms can be attributed to Internet programmes and thus also categorised as guided SH [312]. The advantage of the apps is that they are installed on the smartphone and, like SH books, can be conveniently used anytime and anywhere, partly offline and partly online. In a review, Mehdi et al. [313] identified and evaluated 34 systematically recorded apps for tinnitus. Although all of them were considered to have a certain technical functional value, they all lacked scientific evidence. Nagaraj and Prabhu [167] came to similar conclusions. In the course of the current expansion of the Digitalisation Act of the Federal Republic of Germany, the Federal Institute for Drugs and Medical Devices (BfArM) is attempting, according to § 139e of the German Social Law Book V (SGB V), to bring the exploding variety of health apps into regulated channels by the authorities (Digital Health Applications, DiGA, https://​www.​bfarm.​de/​DE/​Medizinprodukte/​DVG/​_​node.​html).

In the field of health services, the spread of digitalisation (telemedicine; [314]) has resulted in increasingly inconsistent terminology, the keywords of which provoke considerable miscommunication for SH. Thus, under laboratory conditions of research, partly inconsistent word combinations such as ‘self-help via internet’, ‘internet-based self-help treatment’, ‘internet-guided self-help’, ‘internet-based therapeutic self-help intervention’ and ‘internet-based self-help training’ etc. have been developed and evaluated for various psychotherapy procedures, which, in contradiction to their taxonomy, lack any reference to SH [315, 316]. A correct taxonomy, e.g. CBT, would be ‘Healthcare interventions delivered over the internet’ [317] or ‘internet-delivered CBT’ [318].

However, if there is no consensus on terminology among researchers, it is very difficult for guideline writers, and not possible at all for media, to report correctly on SH versus therapy and to convey an accurate message to the public. A common and clearly defined terminology would undoubtedly help to navigate this [316].

Implications for practice: Service providers should inform their patients about SH options analogous to the recommendations mentioned at the beginning and encourage them very early on to consider an SHG as a possible option for their self-management [307].

To avoid blatant misunderstandings, internet-based therapy programmes including apps with or without expert support should under no circumstances be terminologically declared as SH.