Introduction

Music is a powerful stimulus affecting human perceptual, cognitive and motor brain functions, including movement, language, memory, attention, and emotions [1]. Musical activities in a clinical setting are highly motivational, as they activate the dopaminergic mesocorticolimbic reward circuitry, allowing for their reiteration over time, which leads to learning and ultimately to brain plasticity [2]. Indeed, a recent meta-analysis of 84 neuroimaging studies using structural magnetic resonance imaging (MRI) and functional MRI (fMRI) evidenced plastic changes in the physiology and anatomy of primary and non-primary sensory areas of the supratemporal lobe as well as cortical and subcortical motor circuits in frontostriatal regions as a consequence of years of musical practice in healthy musicians [3].

The potential role of music for the neurorehabilitation in the realm of severe acquired brain injury (ABI), such as disorders of consciousness (DOC), is still largely unexplored. ABI includes a wide range of conditions caused by traumatic events (e.g., traumatic brain injury), by vascular event (e.g., ischemic or haemorrhagic stroke), or by cerebral anoxia, toxic or metabolic insults (see [4] for a review). These neurological conditions can result in mild to severe impairments of motor and cognitive functions as well as in emotional and neuropsychiatric symptoms, which in turn may lead to a significant level of disabilities and poor functional outcome. Various rehabilitation techniques and treatments have been developed for the neuropsychological deficits associated with ABI (e.g., aphasia, neglect, attention and memory deficits, etc.), ranging from behavioural to cognitive interventions [5]. However, current evidence on rehabilitation techniques of ABI highlights relevant methodological and measurement inconsistencies, mainly due to the heterogeneity of this group of patients. Furthermore, rehabilitation interventions are often very complex and target multiple outcomes, which can be difficult to evaluate independently.

In fact, ABI could result in DOC, which, in turn, based on the observable behavioural features and their inferred relationship to level of consciousness, is clinically manifested as coma, vegetative state (VS), or minimally conscious state (MCS) [6]. Coma is a condition clinically defined as the complete loss of spontaneous or stimulus-induced arousal [7]. This condition is a self-limiting state that typically resolves within 2 weeks and transitions into either VS or MCS [6]. VS is a condition of wakeful unconsciousness, which is diagnosed when spontaneous eye-opening remerges, although the absence of any discernible evidence of language comprehension, verbal or gestural communication, or reproducible purposeful behavioural responses to visual, auditory, tactile or noxious stimuli persist [8]. Lastly, MCS is a condition of severely altered consciousness characterised by minimal yet definite behavioural evidence of self or environmental awareness. It is a condition that usually results from the improvement in consciousness after coma or VS [6].

For the diagnosis of DOC type, the neurologist is typically aided by electroencephalography (EEG), a widely available routinely-used neurophysiological technique that records the synchronised post-synaptic electrical activity of neurons. In DOC patients, EEG signal recorded during rest is clearly atypical; it is indeed characterised by a generalised slowing in the theta or delta ranges and it presents other patterns including epileptiform activity, burst-suppression and alpha-coma. EEG is highly functional for the differentiation of DOC types: for patients without preserved consciousness, EEG shows predominant low, mostly delta, frequency, combined with a marginalization of the higher frequencies, both in terms of the global power of brain activity and in functional connectivity patterns; for MCS patients the opposite EEG pattern is seen: higher frequency bands are preserved both in global power and in functional long-distance connections [9].

Additional to the analysis of oscillatory patterns in the resting-state EEG, also the evoked responses to passive stimulation, typically auditory, have been studied in DOC patients. Among these responses, the one that has biomarker features, meaning it also has some predictive power in relation to the awakening outcome of the patient, is N100. According to a predominant theory of brain function (the Predictive Coding Theory or PCT; [10]), the N100 tackles the predictive process of anticipating the sensory environment based on learning its properties by extracting the events’ statistical probabilities and continuously updating its internal model. In the auditory system, a decreased N1 reveals a memory process which is further modulated by the precision of the context: if a context is uncertain, such as when a listener has never heard a song before, the N1s to all sounds of that song are enhanced as compared to when the tones of the song are learned [10].

Given the clinical and neurophysiological features of DOC patients, music may represent a useful means that could function as an alternative communication channel for multiple reasons. Music represents a nonverbal auditory stimulus with a marked emotional salience (e.g., [11, 12]); as such, it may be regarded as an optimal stimulus in contexts where cognition is severely compromised and stimuli with personal meaning may produce relevant behavioural changes [13–15]. This might be particularly relevant as the most successful rehabilitation interventions are those that are tailored to the individual [16]. Moreover, music-based interventions are based on integrity of the hearing system, which is usually intact in DOC conditions, despite a frequently impaired visual system [17]. Second, music listening may be seen as a kind of enriched sensory stimulation and, hence, a subtype of cognitive rehabilitation strategies; these strategies are based on the idea that an enriched environment benefits brain plasticity and improves recovery of injured brains [18]. Some studies [19, 20] showed as music affect neural networks, accelerating brain plasticity and avoiding sensory deprivation. Multisensory stimulation treatment, based on auditory, visual, tactile, olfactory and gustatory stimuli, can improve the level of consciousness, above all when the stimuli are of high emotional relevance. This aspect is very important also for music stimulation that ought to be used as rehabilitation treatment in DOC.

Neuropsychological studies have shown that music increases arousal and attentions in DOC patients, with significant cortical and Autonomic Nervous System (ANS) activity [21]. Specifically, music accelerates brain plasticity by increasing the activity in frontal, temporal, parietal and subcortical processes, with an important effect on the recovery process [22]. Moreover, in DOC patients, the emotional content of salient music leads to an activation of limbic structures [23, 24], as well as an increase in eye blinks during the exposure to preferred songs [21]. The impact of music on DOC patients is also visible in their EEG responses. O’Kelly et al. [21] investigated the cognitive correlates of music perception in DOC patients and found that passive exposure to preferred songs increased EEG amplitude in frontal midline theta and alpha in both VS and MCS cohorts. These findings resonate with what has been observed even in healthy subjects: the spectral power of the frontal midline theta is increased during listening to preferred pleasant music in contrast to unpleasant music [25]. The salience of reward circuits in DOC is confirmed by the fact that this condition is often associated with widespread impairment of the dopaminergic system, as well as a reduced level of dopamine in the cerebrospinal fluid [26]. Therefore, interventions aimed at stimulating dopamine release may be particularly appropriate for DOC. In fact, a context of preferred music is deemed to be more appropriate than a neutral context to the expression of residual cognitive functioning in patients with DOC [27]. In sum, music-based interventions have consistently been associated with behavioural and neural improvements among DOC patients (for a review, see [28]).

A further group of rehabilitation strategies for brain-lesioned patients, including DOC patients, consists of neuromodulation aimed to alter the physiology of the neural circuits responsible for the behaviour under investigation. Among several neuromodulation approaches, there are pharmacological interventions (e.g., amantadine hydrochloride) [29], deep brain stimulation (for a review see [30]), and Non-Invasive Brain Stimulation techniques (NIBS) [31]. NIBS exists in various forms, although one of the most adopted in clinical settings is transcranial electrical stimulation (tES). This technique consists of delivering a small (1–2 mA) current via two electrodes placed on the scalp. According to the stimulation parameters, tES affects neural excitability and has plastic effects (mediated by NMDA-receptors) [32]. These currents, cathodal (-) or anodal (+), generate an electric field that can modulate spontaneous neural activity by interfering with the membrane potential of the underlying neuronal compartments by inducing, respectively, hyperpolarization (inhibiting effect) or depolarization (facilitating effect) [33]. Importantly, the induced changes become more stable and long-lasting (i.e., aftereffects) when stimulation is repeated several times [34]. Key targets for stimulation in the neurocognitive domain are the dorsolateral prefrontal cortex (DLPFC), especially in working memory, attention, and executive functions, and the posterior parietal cortex (PPC), which plays a critical role in visual spatial perception and attentional processes [35]. Several tES studies show improvements in clinical (e.g., GCS, CRS-R, cognitive, motor) and surrogate (i.e., EEG, TMS, fMRI) outcomes of patients with DOC [36–38], for a review see Kuo et al. [39], although this effect is more evident among MCS patients rather than among unresponsive wakefulness syndrome patients (VS/UWS) [40]. NIBS is, thus, considered a useful tool in the rehabilitation of clinical populations such as patients with DOC [41]; for instance, tES has been used as a potential adjuvant treatment to improve symptoms in patients with hemispatial neglect after stroke [42]. However, there is an increasing understanding that patients do not homogeneously respond to NIBS]. This might be ascribed to the fact that the methodology adopted in tES studies is characterised by a large variability as concerns stimulation parameters, participant characteristics as well as the characteristics of associated rehabilitation treatment.

With this study we propose a novel rehabilitation protocol that combines music-based intervention and NIBS in neurological patients aimed to (i) assess the residual neuroplastic processes existing in DOC state, and (ii) evaluate the putative modulation of these processes by a non-pharmacological 2-weeks intervention combining electric (tES) and auditory (music) stimulation. The musical stimulation will consist of musical pieces that vary in familiarity and complexity (from pieces chosen from the favourite ones for the patient to novel pieces following classical tonal style up to atonal pieces). We hypothesize that tES, altering the physiology of neural circuits, can increase the plastic effect of music stimulation and thus increase the responsiveness of the patient. As far as we know, there are no studies in the literature that have evaluated the effectiveness of the association of these treatments in DOC patients Furthermore, an additional novel aspect will be the use of autobiographical music, hence emotionally relevant to the patients.

In addition, we also want to ascertain the influence of this rehabilitation protocol on caregiver’s burden and psychological distress, since acquired brain injury has an important impact on physical, psychological, social and also financial conditions of caregivers; their entire life is modified in order to take care of the patient, causing a strong emotional burden, with high levels of psychophysiological distress, depression, anxiety and a more complex symptomatology called prolonged grief disorders [43, 44]. In this regard, an aim of this study is to evaluate the caregiver’s psychological burden both before and post rehabilitation treatment in order to assess the effect of music stimulation and brain stimulation also on caregiver’s emotional condition.

The outcome measures will be clinical, namely based on the neurologist’s observations of clinical improvement and responsiveness, measured through specific scales as well as neurophysiologically. For the latter, continuous EEG is expected to display a decrease of power spectra of slow frequency bands (delta-theta) and an increase in power for faster frequency bands (alpha, beta, gamma). We also expect to evaluate residual memory and learning processes in DOC patients by measuring a higher EEG N100 component amplitude after compared to before treatment. Also, amplitudes of the later-stage auditory P2 and latencies of the N100 and P2 will be explored. More broadly, the present intervention combining music exposure with electric neuromodulation is expected to produce increased evoked responses indexed in EEG measures.

Material and methods

Experimental design and procedures

The experimental paradigm is one of the main strengths of the study, as a randomized cross-over design will be used in this trial, including a treatment condition with electric brain stimulation combined with music listening and a placebo condition with sham brain stimulation combined with noise listening. The randomization of the participants to be included in each group will be done using the software “Research Randomizer” (www.randomizer.org). The randomization procedure will be performed by an independent researcher unrelated to the study. The randomization file will be saved in a password-protected online database. This independent researcher will enrol participants and will assign the patients to intervention groups. Both patients and examiner will be fully blinded.

Assessment times of outcome measures will be conducted at admission (T₀), after the first weeks of post-intervention or post-placebo (T₁), and after the second two weeks of post-intervention or post-placebo (T₂) (Fig 1). Patients will be randomized into three groups depending on the order of the stimulation or placebo conditions:

1. Stimulation-Placebo: Patients receiving tES combined with music stimulation for 2 weeks, 1 week of wash out, then sham stimulation combined with noise (placebo) for other 2 weeks.
2. Placebo-stimulation (combined tES and music): Patients receiving sham stimulation and noise (placebo) for the first 2 weeks, then 1 week wash out, and finally tES and music stimulation for other 2 weeks.
3. Music-placebo: Patients receiving music stimulation and sham stimulation for 2 weeks, 1 week of wash out, then sham stimulation and noise (placebo) for another 2 weeks.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Study flow diagram displaying interventions and assessments.

https://doi.org/10.1371/journal.pone.0304642.g001

For the treatment condition, music stimulation, lasting overall around 30 minutes, will be provided by headphones for two weeks in a row every day (except weekends) twice a day: in the morning and in the afternoon. In the morning only, this music-listening treatment will be preceded by brain stimulation with tES applied in the hemisphere contralateral to the lesion (position of electrodes adjusted according to clinical condition of the patient). The placebo condition will have the same temporal course of the treatment condition, but, instead of music, patients will listen to white noise for 30 minutes in the morning, and pink noise for 30 minutes in the afternoon and only during the morning sessions, sham (current induction for 10 minutes ramp up and 10 minutes ramp down) in the hemisphere contralateral to the lesion will be delivered. Between the two conditions, one week of washout is planned.

Patients who will develop relevant clinical condition, such as respiratory problems, during the period of protocol, will be excluded from the study.

Discussion

With this trial we present a novel clinical intervention combining music exposure with NIBS in DOC patients, also assessing the impact of this rehabilitation protocol on caregiver’s burden and psychological distress. Both music and NIBS have been largely proven effective in the realm of brain lesions. Nonetheless, this is the first study to propose a clinical protocol including a combined effect of these two intervention strategies among post-coma patients. Further strengths of the current protocol are the sample size and the use of a variety of objective measures, including neurophysiological ones such as EEG, to show the cognitive and functional improvement of DOC patients.

In relation to EEG recorded during music stimulation, our intervention is expected to produce two outcomes. The first consists of increased evoked responses amplitudes to sounds reflecting a residual plasticity of the auditory cortex in DOC patients after the music-based intervention combined with electric stimulation. Second, the continuous EEG is expected to display a decrease of power spectra of slow frequency bands and an increase in power for faster frequency bands. Overall, our ambitious aim of inducing and evaluating beneficial effects of music and brain stimulation on psychological and neurophysiological measures would represent a novel rehabilitation approach, not yet explored in DOC patients.

The experimental paradigm we designed is one of the major strengths of the study. Indeed, as RCTs provide the most reliable evidence on the effectiveness of interventions, a pivotal condition especially in clinical research, this protocol significantly contributes to shed light on the most appropriate characteristics for DOC-targeted interventions. As such, this study may help pave the way towards a more generalised and replicable protocol on DOC rehabilitation, as well as highlight the potential role of music on brain injury rehabilitation. In this sense, as our experimental design includes both preferred and non-preferred musical stimuli, this study will allow us to entangle the beneficial effects of music ascribable to its acoustic characteristics from those arising from its autobiographical content, yet unexplored in prior studies on DOC [27, 46]. A deeper understanding of musically-induced positive emotions in DOC patients might be central in sharpening cognitive stimulation processes and personal memory activation. This is particularly relevant for the clinical population we are considering as DOC can be related to damage of the dopaminergic system [26]. In other words, as rewarding music can activate the dopaminergic system via changes in the limbic system (i.e., [73]), the association between music stimulation and cognitive improvements in DOC might be mediated by musically-related positive emotions.

Another strength of this trial refers to its implications on the caregiver. Taking care of patients with DOC has a high impact on a caregiver’s social, economic and emotional condition. High levels of depression, anxiety and prolonged grief were reported both at admission both at follow-up in caregivers of DOC patients [74]. However, most music interventions studies have focused on caregivers of patients affected by dementia (for a review see [75]). As far as we know, there are no studies assessing the effects of music stimulation on DOC patients’ caregivers’ burden. We aim to fill this gap considering that music may be an efficient additional caring tool for caregivers for several reasons. First, in our study caregivers actively contribute to the intervention by suggesting patients’ musical preferences to the neuropsychologist and the experimental team. This direct involvement is likely to promote caregivers’ empowerment and increased control, as seen in previous studies (i.e., [76, 77]). Second, music-based interventions are generally associated with fostered feelings of inner strength, personal growth, increased resilience, social confidence and self-esteem amongst caregivers [76]. Therefore, our rehabilitation protocol could be valid and effective in relieving caregiver’s emotional symptoms.

We also acknowledge some limitations that should be considered for the interpretation of the results. First, since our experimental paradigm provides for tailored salient emotional stimuli (namely preferred music), individual differences on music reward sensitivity might influence the elicited responses. Indeed, prior evidence suggested that individual music reward sensitivity affects both physiological (skin conductance) and neural (mainly dopaminergic mesolimbic reward system) responses (see [78, 79]). A second potential limitation is related to the possible overestimation of the treatment effect size, due to the lack of studies performing this multimodal intervention on DOC patients. Accordingly, in the event of preliminary analyses revealing an effect size significantly lower than predicted, we will consider an extension of the population size. Lastly, given the nature of our target population and intervention, which involves specialised healthcare professionals and specifical equipment, we are unable to blind participants and assessors. All in all, this protocol has a high clinical applicability, it is easy to offer, and it is very motivating for the patients and the caregivers. We, therefore, believe that our findings may help to open new paths in the development of new specific DOC rehabilitation approaches.

Supporting information

Acknowledgments

The authors have no financial or personal relationships with other people and organizations that could inappropriately influence their work.