Fig. 11.1
Tinnitus networks. Brain networks involved in phantom perception. Increased activity in the auditory cortex (brown) as a consequence of auditory deprivation is necessary, but not sufficient for tinnitus perception. The stimulus becomes consciously aware if auditory activity is connected to a larger coactivated awareness network involving subgenual (sgACC) and dorsal anterior cingulate cortex (dACC), posterior cingulate cortex (PCC), precuneus, parietal cortex, and frontal cortex (blue). Salience to the phantom percept is reflected by activation of dACC and anterior insula (yellow). Tinnitus annoyance is reflected by coactivation of a nonspecific distress network consisting of the anterior cingulate cortex (sgACC and dACC), anterior insula, and amygdala (red). Memory mechanisms involving the parahippocampal area, amygdala, and hippocampus (green) play a role in the persistence of the phantom percept (Modified from (De Ridder et al. 2011); Copyright 2011 National Academy of Sciences, U.S.A.)
11.3 Tinnitus Measurement
As tinnitus is a purely subjective phenomenon, measurement of treatment outcome is not trivial. Tinnitus loudness can be either assessed by psychoacoustic measurements (loudness matching or minimal masking level) or by visual analogue or numeric rating scales. The impact of tinnitus on quality of life is usually assessed by validated questionnaires (Zeman et al. 2014). As psychoacoustic measures of tinnitus loudness have shown only limited test-retest reliability (Henry and Meikle 2000), tinnitus loudness assessment by visual analogue scales or numeric rating scales may provide more useful information (Adamchic et al. 2012). Validated questionnaires are the recommended primary outcome measurement for clinical trials (Langguth et al. 2007). However, there exist several validated questionnaires which assess similar but not identical constructs (Milerova et al. 2013). Even if the scores of different questionnaires correlate with each other (Zeman et al. 2012), comparability across studies using different questionnaires is impaired.
11.4 Rationale for the Application of rTMS in Tinnitus
As mentioned in the introduction, tinnitus is related to altered activity of cortical networks involving also central auditory areas. Since rTMS has the ability to focally modulate cortical activity, it has been assumed that it can interfere with the tinnitus-related abnormal neural network activity and thereby influence the perception of tinnitus.
In a recent study, stimulation sites thought to be most effective in various neurological diseases were found to represent different nodes within the same brain network as defined by resting-state functional connectivity MRI (Fox et al. 2014). Based on this observation, one would expect that tinnitus can be modulated by targeting nodes of tinnitus-related abnormal cortical networks. Indeed, single sessions of rTMS over the temporal or temporoparietal cortex but also over the frontal and parietal cortex have been shown to reduce tinnitus transiently in a subgroup of tinnitus patients (for an overview, see (Langguth and De Ridder 2013)). With the goal to produce longer-lasting modulation of tinnitus-related cortical activity, repeated applications of rTMS have been investigated as a potential treatment for some forms of tinnitus. Thus, in summary, analogous to what has been proposed for implanted electrodes overlying the auditory cortex in tinnitus, only those patients who exhibit good functional connectivity between the stimulation target and the putative tinnitus network are likely to respond to neuromodulatory approaches (De Ridder and Vanneste 2014).
11.5 Clinical Effects of rTMS in Tinnitus
Based on the notion that tinnitus is related to auditory cortex hyperactivity, low-frequency rTMS has been applied with the aim to reduce tinnitus by reducing auditory cortex hyperactivity. Since this approach was first proposed (Eichhammer et al. 2003; Langguth et al. 2003), it has been investigated in an increasing number of studies applying low-frequency rTMS in long trains of 1200–2000 pulses repeatedly over 5–10 days (Table 11.1). Beneficial effects of low-frequency rTMS have been confirmed by many (Anders et al. 2010; Khedr et al. 2008, 2009; Plewnia et al. 2007b; Marcondes et al. 2010; Smith et al. 2007; Rossi et al. 2007) but not all further controlled studies (Piccirillo et al. 2013; Langguth et al. 2014; Hoekstra et al. 2013). Moreover, the degree of improvement and the duration of treatment effects varied across studies, probably due to differences in study design, stimulation parameters, and selection criteria of the participants.
Table 11.1
Effects of repeated sessions of rTMS in tinnitus patients
Articles | Number of patients | Target, coil type (placement) | Control condition | Stimulation frequency and intensity | Number of pulses/session and number of sessions | Results | Class of the study |
|---|---|---|---|---|---|---|---|
Kleinjung et al. (2005) | 14 | Auditory cortex activation area in PET, F8c (FDG-PET-guided navigation) | Sham coil | 1 Hz, 110 % RMT | 2000 pulses, 5 sessions | Significant tinnitus reduction (prolonged effect up to 6 months) | III |
Rossi et al. (2007) | 16 | Left TPC, F8c (navigation and 10–20 EEG system) | Tilted coil combined with electrical skin stimulation | 1 Hz, 120 % RMT | 1200 pulses, 5 sessions | Significant tinnitus reduction (no prolonged effect) | III |
66 (active: 16, 17, 17; control: 16) | Left TPC, F8c (10–20 EEG system) | Stimulation of nonauditory cortical areas | 1/10/25 Hz, 100 % RMT | 1500 pulses, 10 sessions | Significant tinnitus reduction for all active conditions (prolonged effect up to 12 months); less efficacious for tinnitus with longer duration | III | |
Anders et al. (2010) | 42 (active: 22; control: 20) | Auditory cortex, F8c (10–20 EEG system) | Tilted coil | 1 Hz, 110 % RMT | 1500 pulses, 10 sessions | Significant tinnitus reduction (not initially, but at 3–6 months after the stimulation) | II |
Marcondes et al. (2010) | 19 (active: 10; control: 9) | Left superior temporal cortex, F8c (10–20 EEG system) | Sham coil | 1 Hz, 110 % RMT | 1020 pulses, 5 sessions | Significant tinnitus reduction (prolonged effect up to 6 months); effect correlated to a reduced activity of inferior temporal cortices in SPECT | III |
Mennemeier et al. (2011) | 21 | Auditory cortex activation area in PET, F8c (FDG-PET-guided navigation) | Sham coil combined with electrical skin stimulation | 1 Hz, 110 % RMT | 1800 pulses, 5 sessions | Significant tinnitus reduction (43 % responders, 33 % improvement); no correlation with activity changes in PET | II |
Piccirillo et al. (2011) | 14 | Left TPC, F8c (navigation and 10–20 EEG system) | Sham coil | 1 Hz, 110 % RMT | 1500 pulses, 10 sessions | Nonsignificant tinnitus reduction | III |
Chung et al. (2012) | 22 (active: 12; control: 10) | Left auditory cortex, F8c (navigation) | Sham coil | cTBS, 80 % RMT | 900 pulses, 10 sessions | Significant tinnitus reduction; more efficacious on emotional component of tinnitus | III |
Plewnia et al. (2012) | 48 (active: 16, 16; control: 16) | Bilateral temporal cortex or TPC, F8c | Active stimulation behind the mastoid | cTBS, 80 % RMT | 900 pulses, 20 sessions | Nonsignificant tinnitus reduction | III |
Hoekstra et al. (2013) | 50 (active: 25; control: 25) | Bilateral primary auditory cortex, F8c (navigation) | Sham coil | 1 Hz, 110 % RMT | 4000 pulses (2000 left, 2000 right), 5 sessions | Nonsignificant tinnitus reduction | I |
Lee et al. (2013) | 15 | Left temporal cortex, F8c (10–20 EEG system) | Tilted coil | 1 Hz, 100 % RMT | 1200 pulses, 10 sessions | Significant tinnitus reduction, negatively correlated to the duration of tinnitus | III |
Piccirillo et al. (2013) | 14 | Left temporoparietal junction, F8c | Sham coil | 1 Hz, 110 % RMT | 20 sessions | Nonsignificant tinnitus reduction | III |
Bilici et al. (2015) | 75 (active 30, 15; control 30) | Left TPC, Cc | Sham coil | 1/10 Hz, 110 % RMT | 900 pulses (1 Hz) or 600 pulses (10 Hz), 10 sessions | Significant tinnitus reduction for all active conditions, less pronounced in combination with paroxetine | III |
Langguth et al. (2014) | 185 (active: 47, 48, 46; control: 44) | PET-guided temporal cortex, left temporal cortex, combined left temporal + prefrontal cortices, F8c (navigation and 10–20 EEG system) | Sham coil | 1 Hz (temporal cortex), 20 Hz (prefrontal cortex), 110 % RMT | 2000 or 4000 pulses, 10 sessions | Significant tinnitus reduction for all three active conditions, but no statistical significant difference in comparison to sham; better effects on a descriptive level for combined frontal and temporal rTMS | I |
11.6 Duration of Treatment Effects
While some studies demonstrated effects that outlasted the stimulation period for several months (Khedr et al. 2008, 2009; Marcondes et al. 2010) up to 4 years (Burger et al. 2011), others were not able to achieve long-lasting effects (Plewnia et al. 2007b; Rossi et al. 2007). One case report (Mennemeier et al. 2008) and a case series (Langguth et al. 2008b) suggest that patients who respond once to rTMS treatment also experience further positive effects from a second series of rTMS, but controlled studies investigating maintenance therapy are lacking.
11.7 Stimulation Frequency
Currently, it is also still unclear, whether low-frequency rTMS is the optimal stimulation frequency. Two studies demonstrated that 10 Hz and 25 Hz rTMS are at least as efficient as 1 Hz for tinnitus treatment (Khedr et al. 2008, 2009, 2010). High-frequency priming stimulation, which enhanced effects of low-frequency rTMS in a preclinical study (Iyer et al. 2003), has failed to enhance the therapeutic efficacy of low-frequency rTMS for the treatment of tinnitus (Langguth et al. 2008a). Also theta-burst stimulation has been investigated with conflicting results. In one study, ten sessions of continuous theta-burst TMS over the auditory cortex have reduced tinnitus loudness and tinnitus impairment (Chung et al. 2012). In contrast, bilateral continuous theta-burst over 4 weeks had no superior effect on tinnitus as compared to sham stimulation (Plewnia et al. 2012)
11.8 Stimulation Target
The optimal target for stimulation and the best method for coil positioning are still a matter of debate (Langguth et al. 2010). Various neuroimaging methods reveal slightly different areas of abnormal neuronal activity in tinnitus, and accordingly different targets have been chosen for stimulation. Based on FDG-PET data that reveal increased neuronal activation predominantly of the left auditory cortex independent of tinnitus laterality (Arnold et al. 1996), this area has been chosen as treatment target in many studies. Whereas a first study revealed a relationship between PET activation in the auditory cortex and treatment outcome (Langguth et al. 2006), this finding could not be confirmed in a larger sample (Schecklmann et al. 2013). A recent study performing FDG-PET before and after treatment found no relationship between activation changes in the stimulated area and clinical outcome, questioning the use of FDG-PET for identification of the optimal treatment target.
Other imaging studies identified abnormalities predominantly in temporoparietal areas (Plewnia et al. 2007a). Based on fMRI (Smits et al. 2007) and MEG studies (Llinas et al. 1999; Muhlnickel et al. 1998; Weisz et al. 2007b), the primary involvement of the auditory cortex contralateral to the perceived tinnitus has been hypothesized (De Ridder 2010). A recent study confirmed this notion by demonstrating that rTMS over temporoparietal areas is more efficient when applied contralaterally to the perceived tinnitus than ipsilaterally (Khedr et al. 2010). However, this is somewhat contradictory to another recent finding that shows lower efficacy of left temporal rTMS in right-sided tinnitus as compared to left-sided tinnitus (Frank et al. 2010).
Pathophysiological concepts and neuroimaging findings are stressing the relevance of nonauditory areas in tinnitus (De Ridder et al. 2014). Therefore, stimulation protocols have been extended to the frontal cortex. In one pilot study, 32 patients received either low-frequency temporal rTMS or a combination of high-frequency prefrontal and low-frequency temporal rTMS (Kleinjung et al. 2008). Directly after therapy, there was an improvement of the tinnitus questionnaire score for both groups, but there were no differences between groups. Evaluation after 3 months revealed a remarkable advantage for combined prefrontal and temporal rTMS treatment. A pilot study demonstrated similarly a tendency toward increased efficacy when 1 Hz left temporal rTMS was preceded by 1 Hz right prefrontal rTMS (Kreuzer et al. 2011). These data indicate that modulation of both frontal and temporal cortex activity might represent a promising enhancement strategy for improving TMS effects in tinnitus patients.
It is known from animal experiments that neuronal plasticity can be enhanced by dopaminergic receptor activation (Bao et al. 2001). However, in pilot studies, the administration of neither 100 mg of levodopa nor 150 mg bupropion before rTMS was successful in enhancing rTMS effects in tinnitus patients (Kleinjung et al. 2009, 2011).
There is some evidence from several studies that the clinical characteristics of patients may affect the therapeutic outcome of rTMS in tinnitus patients. Several studies reported that patients who had their tinnitus for a shorter duration may have better treatment outcomes (Khedr et al. 2008; Kleinjung et al. 2007). However, when larger samples were analyzed, this effect could neither be confirmed nor other robust predictors for treatment outcome could be identified (Frank et al. 2010; Lehner et al. 2012).
11.9 Neurobiological Mechanisms of rTMS Effects in Tinnitus
The mechanisms by which rTMS exerts its clinical effects on tinnitus are still incompletely understood. The concept that 1 Hz rTMS reduces tinnitus by inducing long-term depression (LTD)-like effects on increased neuronal activity in the auditory cortex has been challenged by the findings that (1) treatment outcome of 1 Hz rTMS is worse in patients with more pronounced auditory hyperactivity (Langguth et al. 2006) and that (2) both low- and high-frequency rTMS over the temporoparietal cortex exert beneficial effects on tinnitus (Khedr et al. 2008, 2010).
In line with these findings, a recent investigation in healthy controls has demonstrated that both low- and high-frequency rTMS over the temporal cortex reduce auditory cortex excitability as measured with the auditory-evoked P50 amplitude (Nathou et al. 2014)
FDG-PET scans before and after rTMS were not successful for identifying the neuronal correlates of rTMS-induced tinnitus reduction (Mennemeier et al. 2011). In particular, no relationship between the treatment-related change of metabolic activation of the auditory cortex and clinical effects could be detected (Mennemeier et al. 2011).
A study which investigated the effects of auditory cortex stimulation in healthy controls with voxel-based morphometry found alterations in the temporal cortex and in the thalamus, suggesting that temporal rTMS may influence thalamocortical processing (May et al. 2007).
The exact cortical region in which temporal rTMS exerts clinical effects in tinnitus patients is still a matter of debate (Langguth et al. 2010). It has been argued that the primary auditory cortex is difficult to reach by TMS, since it is located far from the brain surface in the Sylvian fissure in the lateromedial direction. Furthermore, following the tonotopic organization of the primary auditory cortex, the representation of low frequencies is located more lateral, whereas the representation of high frequencies is more medial. Thus, one would expect better outcomes in patients with low-frequency tinnitus since the related abnormalities in the auditory cortex are expected to be more lateral and should therefore be better reached by rTMS. However, such a relationship could not be demonstrated (Frank et al. 2010). It has been proposed that rTMS might exert direct effects on the superficial secondary auditory cortex which then further propagate to the primary auditory cortex, analogous to what has been described for electrical stimulation of the secondary auditory cortex in tinnitus. A recent study which used MEG to record auditory-evoked potentials suggests that rTMS induces changes in both primary and secondary auditory cortex activity (Lorenz et al. 2010). The auditory steady-state response, which is supposed to be generated in the primary auditory cortex, was more consistently influenced by rTMS, and its changes also correlated with perceptual changes (Lorenz et al. 2010). Also a very recent study which investigated the effects of paired associative auditory and cortical stimulation (Schecklmann et al. 2011) does not provide clear evidence where exactly temporal TMS interferes with auditory processing.
11.10 Methodological Considerations
Both tinnitus perception and distress are known to be susceptible to placebo effects (Dobie 1999). Therefore, evaluation of treatment efficacy requires adequate methodology for the control of nonspecific effects. Different kinds of sham treatments have been suggested as control conditions. In addition to the sham coil system, which mimics the sound of the active coil without generating a magnetic field, an angulation of an active coil tilted 45° or 90° to the skull surface or a stimulation of nonauditory brain areas has been described (see Table 11.1). Finding an optimal control condition for treatment studies is also difficult because of limitations in blinding of patients and operators to different stimulus conditions and due to the fact that TMS itself results in auditory and somatosensory stimulation in addition to the cortical effect. Indeed, a very recent study provides empirical support for the relevance of a double mechanism consisting of a direct cortical modulating effect and an indirect effect via somatosensory-auditory interactions mediated through trigeminal and C2 nerve activation (Vanneste et al. 2011a). As a possible approach for differentiating the two effects, the use of a control condition involving electrical stimulation of the facial nerve has been proposed (Mennemeier et al. 2009; Rossi et al. 2007). Similarly, also interactions between the acoustic artifact of the coil and auditory cortical stimulation may be relevant (Schecklmann et al. 2011).
11.11 Safety Aspects
Even if rTMS is a safe technique (Wassermann 1998; Rossi et al. 2009), some precautions need to be met, mainly due to the theoretical risk of triggering a seizure (though extremely improbable with LF rTMS) or especially of inducing auditory changes because of the noisiness of rTMS at high intensities. The potential harm to hearing function has to be particularly considered in the treatment of tinnitus, since many tinnitus patients suffer from hearing loss. Actually, rTMS has recently been reported to transiently decrease the amplitude of the otoacoustic emissions, reflecting active cochlear effects (Tringali et al. 2012). Despite the absence of recognized auditory toxicity (Schonfeldt-Lecuona et al. 2012), some patients with tinnitus may complain of a worsening of hyperacusis and painful hypersensitivity to noises after rTMS therapy (Rossi et al. 2009). One recent study in tinnitus patients did not show any deterioration in hearing function after a treatment series of 20 sessions of theta-burst stimulation (Schraven et al. 2013). A clinically relevant side effect is the risk of worsening of tinnitus, which has been reported in several studies for a small subgroup of patients. However, little is known whether the worsening of tinnitus, reported in these patients after treatment, is only transient or longer lasting.
Conclusion
In summary, there are an increasing number of studies investigating rTMS for the treatment of tinnitus. Though encouraging, results must still be considered as preliminary due to small sample sizes, methodological heterogeneity, high interindividual variability, and limited knowledge about the duration of therapeutic effects. Replication in multicenter trials with many patients and long-term follow-up are required before firm conclusions can be drawn (Landgrebe et al. 2008). Further clinical research is also needed to get a clear definition of subgroups of tinnitus patients which benefit most from rTMS and how their medical histories, their comorbidities, and their medication may affect the outcome. Better understanding of the pathophysiology of the different forms of tinnitus and the neurobiological effects of rTMS will be critical for optimizing or even individualizing treatment protocols.
A few years ago, a Cochrane meta-analysis of rTMS for the treatment of tinnitus (Meng et al. 2011), which only included randomized controlled studies with parallel groups (Anders et al. 2010; Marcondes et al. 2010; Khedr et al. 2008), came to the conclusion that there is currently limited evidence for efficacy and that further studies are needed before firm conclusions can be drawn. Recently published evidence-based guidelines concluded that “LF (1 Hz) rTMS unilaterally applied to temporal or temporoparietal cortical areas can interact with an abnormal hyperactivity of auditory cortices that may constitute the neural correlate of tinnitus perception. Literature data showed that this type of rTMS protocol has a possible therapeutic efficacy in this clinical condition. The efficacy of active rTMS is superior to placebo in the treatment of subjective tinnitus, but the effects are usually partial and transient at clinical level” (Lefaucheur et al. 2014).
Related posts:
Basic Principles of rTMS in Motor Recovery After Stroke
Therapeutic Applications of rTMS for Aphasia After Stroke
rTMS in Dysphagia After Stroke
Clinical Applications of rTMS in Parkinson’s Disease
Neurophysiology of rTMS: Important Caveats When Interpreting the Results of Therapeutic Interventions
rTMS in Visual Hemineglect After Stroke
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
