Assessing the synergistic effectiveness of intermittent theta burst stimulation and the vestibular ocular reflex rehabilitation protocol in the treatment of Mal de Debarquement Syndrome: a randomised controlled trial

Mal de Debarquement Syndrome (MdDS) is a rare central vestibular disorder characterised by a constant sensation of non-spinning vertigo (i.e. rocking, swaying, and bobbing) [1‐3]. These motion sensations are commonly accompanied by a range of symptoms such as imbalance, ‘brain fog’, visual induced dizziness, sensitivity to light and sounds, anxiety, depression, and migraine [4‐9]. MdDS is typically triggered by exposure to passive motion, like that experienced on a boat, airplane, or motor vehicle. While some degree of non-spinning vertigo is common and transient following such experiences (termed Mal de Debarquement), a diagnosis of the chronic form, MdDS, requires symptoms to persist for at least one month [3, 5, 10]. Not all onsets of MdDS can be attributed to a passive motion event, and a small subset of patients associate the onset of their symptoms to non-motion events, such as intense stress, sickness, or childbirth, and some individuals cannot identify any triggering event at all. The differing onset types have been classified as motion-triggered (MT) and non-motion-triggered (NMT) MdDS, respectively [3‐5, 11, 12]. The MdDS clinical population demonstrates a female/male ratio of between 8:2 and 9:1 [4, 5, 8, 10, 13]. In females, the age of MdDS onset is commonly between 40 and 50 years, whereas male exhibits greater variability in age of onset [4, 5, 8, 10, 13]. Previously, poor diagnostic criteria and lack of understanding of the condition in the medical community resulted in a high rate of misdiagnosed or undiagnosed patients [5]. Though more recent diagnostic guidelines have become available [3], diagnosis is still complicated by an overlapping of symptoms with other vestibular pathologies, such as persistent postural perceptual dizziness [14] and vestibular migraine [15]. The underlying pathogenesis of MdDS is unknown, which limits treatment options. Currently, standard care for people with MdDS is benzodiazapines and other anti-anxiety or anti-depressant medications, such as selective serotonin reuptake inhibitors [4, 8, 16]. Unfortunately, the effectiveness of these medications is mixed, from no reduction to only moderate reduction of symptoms, and none are considered to be curative [16]. Another drawback of these types of medications is their potential for addiction [8] and the development of tolerance [17], limiting prolonged usage. Other non-specific treatments, such as standard vestibular rehabilitation, chiropractic treatments, vestibular suppressants, and counselling, have been trialed with no substantial benefits [12, 16, 18, 19].

Given the limited utility and negative sequalea associated with pharmacological interventions, increasing emphasis has been placed upon the need for non-pharmacological strategies to target the symptoms of MdDS [8, 16, 17]. One intervention that has shown promise is MdDS-specific vestibular rehabilitation [20, 21], based on recalibrating the vestibular ocular reflex (VOR), a neurological reflex which maintains ocular stability by generating compensatory eye movements during head movement [22]. It has been proposed that MdDS is the result of VOR maladaptation involving the central integrative mechanism in the vestibular system, the velocity storage (VS), and a multisensory element that modulates the time constant of the VOR with respect to that of semicircular canal afferents [21]. Evidence of VOR maladaptation in humans has been demonstrated in NASA space flight experiments [23], where participants developed oscillating vertical nystagmus on head movement following prolonged exposure to a slowly rotating room [23]. Yakushin and colleagues confirmed that MdDS patients similarly have longer VOR time constants compared to age-matched controls, suggesting VOR maladaptation [24]. To target this phenomenon, Dai and colleagues pioneered the VOR rehabilitation protocol [21], with the aim to recalibrate the VOR by exposing the patient to full-field horizontal or vertical optokinetic (OKN) stimuli (stripes) (moving in the opposite direction of the patient’s rotation or “gravitational pull”) coupled with passive head movements, in order to induce changes in the VS [21]. In their experiments, 24 individuals with MdDS were treated with a 5-day unstandardised VOR rehabilitation protocol. While improvements were observed, approximately 1 in 3 people did not have complete or substantial recovery 1 year post-treatment [21]. Similar findings have been reported in subsequent research [20, 25]. The later studies also reported that the VOR rehabilitation protocol had a better response rate in those with MT MdDS compared to those with NMT MdDS. Additionally, despite improvements in non-spinning vertigo perception, remission was rare and residual symptoms, such as high visual sensitivity, migraine-like symptoms, and brain fog, were common post-treatment [20, 25]. This suggests that there is a need for studies exploring avenues to augment the effectiveness of the VOR rehabilitation protocol.

Another non-pharmacological strategy to target the symptoms of MdDS is neuromodulation, which is based off the theory that MdDS is a disorder of neuroplasticity [10]. It was theorised that areas of the brain responsible for unconscious balance control develop an internal representation of the external environment, i.e. the persistent background oscillations of the passive motion experience [10]. An exploratory functional brain-imaging study on MdDS patients measuring brain glucose metabolism found that MdDS patients displayed hypermetabolism in the left entorhinal cortex (EC) and amygdala and displayed hypometabolism in the left prefrontal (including the dorsolateral prefrontal cortex [DLPFC]) and temporal cortex, along with the right amygdala [26]. Then, functional connectivity measurements revealed increased connectivity between the EC and sensory processing areas located in the parietal and occipital lobes and decreased connectivity with the prefrontal/premotor cortex (frontal eye field, middle frontal gyrus). MdDS subjects also exhibited reduced connectivity between homologous structures in the prefrontal cortex [26]. The changes identified in the EC are of significant interest given its key role in mediating hippocampal-neocortical communication [27‐29], and spatial mapping, navigation, and cognition [30]. Another study, on individuals with transient (i.e. non-chronic) Mal de Debarquement, also measured brain glucose metabolism and found hypermetabolism in the super occipital gyrus, superior frontal gyrus, and superior and inferior parietal lobules, and hypometabolism in various cerebellar structures (i.e. inferior semi-lunar lobule, nodule, uvula and tonsil) [31]. These findings differ from the hypermetabolism and hypometabolism identified in patients with the chronic form of MdDS, though this may suggest that metabolic activity in the various parts of the brain change over time in people with MdDS, particularly between the transient experience and the chronic form. A morphometry study [32] identified duration of illness-dependent grey matter volume changes in visual-vestibular processing areas, default mode network structures, salience network structures, somatosensory network structures, and a structure within the central executive network (the DLPFC) [32]. Based on hypometabolism and grey matter changes identified in the DLPFC of people with MdDS [26, 32], this area has been the most common cortical target during neuromodulation investigations. Repetitive transcranial magnetic stimulation (rTMS), a form of non-invasive brain stimulation, targeting DLPFC has demonstrated promising short-term improvements among individuals with MdDS [33‐38]. Studies of resting state functional magnetic resonance imaging and electroencephalography in people with MdDS have indicated that DLPFC rTMS decreases functional connectivity between the left EC and the precuneus, right inferior parietal lobule, and the contralateral EC [39, 40], which are part of the posterior default mode network. This reduction in connectivity has been correlated with improved MdDS symptoms [41]. Further, the DLPFC has direct projections to main cortical oculomotor areas and several areas in the posterior parietal cortex responsible for gaze stability and oculomotor control [42, 43]. Spatial information received by the posterior parietal lobes also projects to the DLPFC, making it an important area for cognitive control over spatial information processing. This is pertinent to people with MdDS who, along with rocking dizziness, experience poor attention and significant intolerance to visual motion [33]. Additionally, it has been hypothesised that the DLPFC has indirect projections to the vestibular nuclei [43], which are known to be an integral part of the reflex arc of the VOR [44]. Given the potential link between the VOR and DLPFC, this represents a promising means by which to enhance outcomes associated with the VOR rehabilitation protocol. Prior stimulation of DLPFC may therefore enhance the tolerability and effectiveness of subsequently delivered VOR protocols. Stimulation of the left DLPFC has also been shown to effectively treat symptoms of anxiety and depression [45‐47], which commonly develops in people with MdDS [3‐5].

While valuable, traditional rTMS is time-consuming. This is a pertinent point when treating clinical populations and given that rTMS of DLPFC has been associated with increased procedural discomfort compared to stimulation of other cortical sites [48]. Recently, theta burst stimulation (TBS) protocols have been developed, which have been shown to produce similar clinical effects to rTMS but at a fraction of the time required [49]. A previous TBS study targeting various brain areas in those with MdDS suggests that the treatment has promise, but demonstrates only modest improvements when used in isolation [50]. However, it is known that non-invasive brain stimulation alters cortical excitability and has the potential to enhance synaptic plasticity and promote synaptogenesis, providing a strong neurophysiological rationale for its use as a means of enhancing responsiveness to subsequent treatments [51]. Thus, TBS may act synergistically with subsequently delivered VOR protocols to promote neuroplastic changes and enhance patient outcomes.

The aim of this study was to evaluate the effectiveness of the VOR protocol with and without iTBS pre-treatment on objective and subjective outcomes in people with MdDS. We hypothesised that participants who receive a pre-treatment of active iTBS prior to the VOR rehabilitation protocol would demonstrate greater improvements in balance, a greater reduction of symptoms, and significant improvements in mental health scores, compared with those receiving sham iTBS prior to the VOR rehabilitation protocol.

This study was the first to explore the synergistic effects of iTBS and the VOR rehabilitation protocol in people with MdDS. Despite significant improvements in subjective and objectives outcomes over time, there were no differences between the Active and Sham iTBS Groups. In both groups, MdDS symptom rating, mental health scores and disability perception significantly decreased, and effects were maintained up to the 16 weeks post-treatment follow-up. Both Sham and Active Groups demonstrated significant improvements in static posturography, with ~ 58–83% reductions across all outcomes by the end of treatment week. The results of this study suggests that iTBS of the DLPFC does not enhance outcomes beyond that achieved using the VOR protocol in the treatment of people with MdDS.

Over the last decade, various forms of non-invasive brain stimulation have been trialled on the MdDS population, such as rTMS [33‐37], TBS [50], Transcranial Alternating Current Stimulation (tACS) [79‐81], and Transcranial Direct Current Stimulation (tDCS) [57], and have demonstrated positive results. This is the first study that has trialled iTBS in this population and was chosen as the pre-treatment due to its potential to enhance synaptic plasticity, promote synaptogenesis, and facilitate synaptic connections within cortical tissue [82‐85], with the aim to enhance the brain’s responsiveness for subsequent treatments. Further, iTBS over DLPFC has previously been shown to enhance postural control among other population groups [86]. iTBS also requires less session time compared to rTMS protocols [49]. Cha and colleagues [50] have conducted the only other study utilising TBS in MdDS, where continuous TBS (cTBS) was administered over the occipital cortices, cerebellar vermi, and lateral cerebellar hemispheres of 26 patients. The participants then had the freedom to continue receiving cTBS over the brain targets of their own preference, which they felt were most effective in reducing their oscillating vertigo. After the first session, eleven participants chose the occipital cortex, nine chose the cerebellar vermis, one chose lateral cerebellar hemisphere, and five chose none. After 10–12 sessions of 1200 pulses over the target of choice, it was concluded that cTBS over either the occipital cortex or cerebellar vermis was effective in reducing subjective perception of oscillating vertigo acutely, improving mental health scores and reducing perceptions of disability. Improvements in objective balance were reported across all groups. While valuable, it is difficult to determine the mechanisms underlying the effects observed in the previous study given that multiple sites were stimulated and there was the potential for compound effects between the initial TBS session and the participant’s selected target site. When combined with the findings of the present study, it is plausible that no single optimal cortical target site exists for MdDS, but rather, may vary between patients depending on their predominant motion symptom experience, triggers, or underlying pathogenesis [3, 5, 6, 13].

There are various potential reasons as to why iTBS over DLPFC did not augment the effectiveness of the VOR protocol. First, previous work demonstrates that the effects of non-invasive brain stimulation are cumulative, increasing with repeated sessions [87]. Therefore, while ecologically valid and more feasible clinically, the 4-day protocol employed in the present study may not have been sufficient to induce observable changes beyond those achieved with the VOR rehabilitation protocol alone. However, this remains speculative and it should be noted that a recent paper identified no cumulative effects of TBS over DLPFC on cortical excitability [88]. Second, the current study utilised an excitatory iTBS protocol, based on the positive effects observed in people with MdDS after excitatory rTMS over left DLPFC [33, 34]. iTBS over DLPFC has been shown to improve postural control in other population groups [86], though has not been trialled in people with MdDS. The goal of the protocol used in this study was to decrease functional connectivity between the entorhinal cortex and the posterior default mode network, in accordance with the findings of Yuan and colleagues [41], where positive outcomes were associated with a decrease in functional connectivity after excitatory rTMS over the left DLPFC. However, the DLPFC has far reaching connectivity with other cortical and subcortical sites, which may elicit varying or competing effects via indirect stimulation. Indeed, iTBS over remote and interconnected cortical sites has been shown to enhance functional connectivity in the default mode network, which may worsen MdDS symptoms [89]. Though incompletely explored, there is also the potential that the effectiveness of neuromodulation protocols varies depending upon patient presentation. For example, while TBS has been shown to be comparable or superior to rTMS for depression [49, 90], studies have suggested that high-frequency rTMS is superior to TBS for neuropathic pain [91]. Finally, there is a substantial body of literature suggesting that rTMS [92, 93] and iTBS [94, 95] induce variable effects on cortical excitability, with ‘excitatory’ and ‘inhibitory’ protocol labels being a misnomer. Inter-individual variability in cortical responses may therefore have ‘washed out’ effects during group-level analyses. Exploration of participant-specific responses in larger samples would be an interesting and important avenue for future research.

Recently, Mucci and colleagues postulated that MdDS originates from the persistence of an adaptive internal model that functions to cancel sinusoidal disturbances of body position experienced aboard a vehicle in motion [96]. It was proposed that the internal model is a bilateral oscillator, consisting of a system of loops, involving glutamatergic and GABAergic pathways between the cerebellar cortex and the vestibular nuclei in the brainstem. This vestibulo-cerebellar oscillator then becomes noxiously permanent in those with some sort of predisposing factor (i.e. immunoendocrine condition or disruption). A computational analysis of this proposed loop was investigated by Burlando et al. [97] and showed that parameter changes, typically induced by synaptic plasticity, increased the system’s tendency to oscillate. The results of this study may suggest that iTBS over the left DLPFC is not effective in disrupting the system of loops that are theorised to exist between the vestibular nuclei and the cerebellum, whereas cTBS over the cerebellum may be able to affect the noxious oscillator as evidenced by the positive effects of this approach observed in previous research [50].

Regardless of the result that iTBS of the left DLPFC does not enhance outcomes beyond that achieved using the VOR protocol in the treatment of people with MdDS, our study further validates the VOR rehabilitation protocol as an effective treatment option. The exact mechanism by which the VOR rehabilitation protocol produces beneficial outcomes in MdDS patients is not fully understood. In light of the recent vestibulo-cerebellar oscillator theory, it may be possible that aspects of the treatment influence these noxious loops between the cerebellum and the brainstem. The visuomotor functions of the cerebellum include control of the VOR and optokinetic reflexive eye movements, and smooth pursuit [98], all of which are activated in the VOR rehabilitation protocol. In addition, the passive movement of the vestibular apparatus via head movement could lead to an increase in peripheral afferent signals arriving at the vestibular nuclei, into the loop, potentially disrupting or weakening it. Though it is still not considered a cure, the VOR rehabilitation protocol has the capacity to improve subjective and objective outcomes in people with MdDS up to sixteen weeks post-treatment. These results are in line with the findings of Dai [21, 25], and Mucci [20], whereby ~ 70% of patients demonstrated significant improvements in objective and subjective outcomes after the VOR treatment. Dai [25] and Mucci [20] both reported that people with MT MdDS responded better to the treatment than those with NMT MdDS, and a differing underlying pathological mechanism between the onset types has been proposed [3, 5, 6, 13]. Given the small sample size of this study, the comparison between MT and NMT participants was not made. In addition to the unequal response rates, remission is rare and residual symptoms remain. This highlights that though the VOR rehabilitation protocol is the most effective treatment for MdDS, there is a need to further explore how the protocol can be optimised as part of MdDS management.

The main limitation to this study was not collecting objective posturography data at the 16-week follow-up timepoint. Given the geographical location of the participants, obtaining this data was not possible, but may have provided further insight into any delayed or long-term effects. This is a pertinent consideration given that the effects of non-invasive brain stimulation are often delayed beyond the treatment session itself [87]. Another limitation of the study was that our sample was exploratory and inter-individual variability was not analysed. Due to the inability to administer an effective sham version of the VOR rehabilitation protocol, a third group, i.e. active iTBS + sham VOR protocol, was not included in this study. This may be viewed as a limitation, though a sham VOR rehabilitation protocol would be difficult to blind. Another limitation is that some of the participants received their diagnosis from a Primary Care Physician, as opposed to a vestibular specialist. Given that MdDS is relatively uncommon and not investigated widely amongst Primary Care Physicians, this consideration may have influenced the population recruited (though also enabled larger samples to be recruited). However, most participants were reviewed by specialists, and all were required to be reporting symptoms consistent with MdDS diagnosis guidelines [3]. Finally, participant-specific MRI images were not employed during neuronavigation to localise DLPFC for stimulation. While the BeamF3 heuristic utilised has been shown to produce a very close approximation of the scalp site used for MRI-guided stimulation [99], there is the potential that participant-specific MRI data may have further enhanced target site localisation, and the use of this approach represents an important consideration for further research.

This study demonstrates that a pre-treatment of iTBS of the left DLPFC does not enhance subjective or objective outcomes beyond that achieved using the VOR rehabilitation protocol in the treatment of people with MdDS. These findings further support the effectiveness of the VOR rehabilitation protocol in reducing MdDS symptoms. Though pretreatment of iTBS did not affect these improvements, further research into TBS efficacy is warranted, given the accumulating evidence in its ability to alter central processes in a non-invasive, non-pharmacological way, reduce MdDS symptoms when other areas of the brain are targeted and its potential to influence noxious loops between the cerebellum and the brainstem, which are theorised to play a vital role in MdDS.