Anodal transcranial direct current stimulation over the left temporal pole restores normal visual evoked potential habituation in interictal migraineurs

Migraine is a neurological disorder that is characterized by recurrent clinical attacks separated by variable-length headache-free intervals. Although the pathogenesis of migraine is far from completely understood, clinical neurophysiology and neuroimaging studies in recent decades have disclosed subtle functional and morphological abnormalities that manifest during the interictal phase and distinguish migraineurs from normal healthy subjects [1‐3]. Among the various subcortical and cortical areas implicated in migraine pathophysiology, emerging evidence highlights the temporal pole (TP) as a key neural substrate. In humans, the TP serves as a multimodal neural hub that receives and integrates various sensory modalities including olfactory, auditory, taste, and visual inputs. Moreover, the TP participates in the ventral visual stream (VVS) for visual information processing [4‐6]. During an olfactory task, interictal migraineurs exhibited significantly higher brain glucose metabolism in the left TP compared to control subjects [7]. Moreover, BOLD signal in the TP in response to noxious stimulation was reduced in interictal patients compared to patients who were actively experiencing a migraine [8, 9]. In resting-state MRI studies comparing interictal migraineurs to healthy control subjects, decreased grey matter density was observed in the left TP [10] and the left TP exhibited decreased connectivity with components of the default-mode network [11]. Finally, the TP was implicated as an important area for differentiating patients with migraine from healthy control subjects in a cross-sectional brain MRI investigation [12]. Taken together, these findings suggest that the TP is both intricately related to the pathophysiology of migraine and sensitive to the cyclical recurrence of migraine attacks.

Transcranial direct current stimulation (tDCS) is a non-invasive technique for neuromodulation in humans that affects cortical excitability in a polarity-specific manner [13, 14]. Anodal polarization increases the excitability of cortical areas below electrodes, whereas cathodal polarization typically decreases cortical excitability [15]. A number of tDCS studies in different pain disorders [16, 17] have demonstrated that tDCS is well-tolerated by patients [18]. Anodal tDCS proved effective over either the motor cortex or the dorsolateral prefrontal cortex when used as prophylactic strategy both in episodic [19] and chronic [20, 21] migraine. Moreover, some studies reported that, in addition to the therapeutic effects, tDCS over the visual cortex also normalized interictal cortical hyperresponsivity in episodic migraine [22].

Nonetheless, to our knowledge, no study to date has targeted the TP for anodal tDCS in migraine, to enhance interictal temporal lobe activity and thereby interfere with an aspect of migraine pathophysiology. Thus, we examined whether anodal stimulation of the TP could restore normal function of the TP and thus physiological information processing in migraine. Moreover, given that the TP processes all kinds of sensorial information except for somatosensory information, we examined the habituation responses of evoked potentials to somatosensory stimuli (as a negative control) as well as visual stimuli.

The present study mainly revealed that a single session of anodal tDCS over the left temporal pole restored normal visual but not somatosensory habituation in interictal migraineurs.

Neurophysiological studies have shown that interictal migraineurs exhibit dysfunctional sensory information processing in the form of habituation deficits in response to various sensory inputs, including visual and somatosensory inputs [2]. Recent neuroimaging studies have revealed subtle microstructural alterations in the brains of patients with migraine in areas associated with the ictal-interictal cycle. Among these studies, some evidence highlights a pathophysiological role for the TP in migraine [7‐12].

The TP region encompasses the most anterior segment of the temporal lobe and receives extensive inputs from visual regions of the thalamus [30, 31]. Additionally, the TP is highly interconnected with the amygdala, hippocampus, superior temporal gyrus, hypothalamus, occipitobasal cortex, prefrontal regions, and insula, suggesting its participation in autonomic regulation, memory, and emotional processing [32, 33]. The TP is considered a multisensory associative cortex because it is also connected to the main sensory systems of the temporal lobe, including the visual, auditory, olfactory, and gustative systems, but not the somatosensory system [32, 34]. Indeed, neuroimaging studies have demonstrated subregional activation of the TP in response to specific sensory stimuli, with the ventromedial aspect of the TP having a predominant role in higher order visual information processing [34] as part of the VVS.

Our finding that anodal (excitatory) stimulation of the left TP restored physiological visual information processing but not somatosensory processing in interictal migraineurs is largely consistent with the abovementioned roles of the left TP in high-level multimodal perceptual processing. A selective effect of tDCS over the TP on visual information processing is probably related to the role of the TP in the VVS and its lack of participation in somatosensory elaboration. Interestingly, another study observed similar normalization of abnormal interictal VEP habituation in response to the application of tDCS over the occipital cortex in migraineurs [22]. This can be explained either by a direct interconnection between the TP and occipital cortex along the VVS or an indirect effect of the tDCS on brain structures that positively modulate both cortices.

The VVS is involved in visual recognition and in the assignment or retrieval of a given meaning for visual information [35]. After early activation of the occipital area, the complexity of representation of visual information increases as information flows to the anterior regions of the VVS, with the TP located at the end of the stream and sending backward facilitatory projections to the occipital cortex to optimize sensory processing (e.g., improve perception and learning) [35, 36]. Consistent with this evidence, we observed that the enhancement of TP activity with anodal tDCS improved VEP amplitude habituation, a basic form of learning [37], without affecting initial baseline excitability (reflected by non-significant changes in 1st block VEP amplitudes). In habituation paradigms, early and late responses can behave differently as a result of regulation by different mechanisms; according to the dual-process theory, increasing responsiveness (sensitization) competes with decreasing responsiveness (habituation) to determine final behavioural outcomes. Facilitation occurs at the beginning of the stimulus session and accounts for an initial temporary increase in response amplitude, whereas habituation occurs throughout the recording session and accounts for delayed decreases in responsiveness [38]. Therefore, our results regarding the selective effect of anodal tDCS on delayed habituation in migraineurs appear to be in line with the putative mechanism of tDCS; that is, the ability of tDCS to affect the potentiation of long-term learning processes and synaptic plasticity underlying learning and memory [39]. Alternatively, it has been shown that anodal tDCS exerts modulatory effects on thalamo-cortical circuits by increasing functional coupling between the thalamus and cortex [17, 40]. These experimental observations are of particular interest in migraine because independent research groups have previously reported reduced functional [41, 42] and morphological [43, 44] thalamic integrity coupled with decreased intracortical inhibition during visual stimulation in migraineurs [45, 46]. We thus can hypothesize that an alternative mechanism of action for anodal tDCS in the present study is increased thalamo-cortical activity, which in turn increased delayed inhibitory mechanisms to restore normal VEP habituation.

Irrespective of the mechanism, the observation that tDCS over the left TP is able to restore normal VEP habituation in interictal migraineurs leads to hypothesize that together with the visual, motor, and dorsolateral prefrontal cortices [19, 20], the TP could represent a novel target for tDCS as a prophylactic strategy for treating migraine [47].

This study had some limitations. For example, we only stimulated the left TP, such that we cannot know whether anodal tDCS of the right TP would have yielded similar results. Several studies have shown divergent functional roles of the left and right TP, where the right TP is more involved in elaborating socio-emotional implications of multisensory perceptual stimuli [48] while the left TP is mostly implicated in perceptual decoding, semantic processing, and conceptualization [34]. Nonetheless, both the left and right TPs are joined via the interior white commissure to advance multimodal perceptual analysis [32], such that the relevance of the right TP cannot be discounted. Furthermore, the positioning method we used is accurate, although not as accurate as neuronavigation-based techniques, which are unfortunately only available for neurosurgical procedures in our clinic. Another shortcoming of the present study is the lack of inclusion of a healthy control group undergoing the same stimulations, although this would not add anything to the results of the study because the healthy subjects usually already habituate normally at the baseline, i.e. we cannot normalize the already normal information processing.

In conclusion, anodal but not sham tDCS selectively enhanced visual but not somatosensory habituation in interictal migraineurs probably by restoring normal inhibitory activity of the left TP. We propose that this effect can be explained by either a direct interference with short- and long-term synaptic plasticity mechanisms or an indirect potentiation of the thalamo-cortical circuit. Further studies are needed to determine whether TP stimulation also normalizes the habituation response to other sensory inputs, such as auditory and nociceptive inputs. Regardless of the underlying cellular and molecular mechanisms of our observed effect, we propose that the TP should be considered as a key site of involvement in the pathophysiology of migraine and as a potential therapeutic target. Clinical studies are needed to clarify whether repeated sessions of anodal tDCS improve TP function and connectivity in patients with migraine to ultimately reduce the number and severity of migraine attacks.