Migraine type-dependent patterns of brain activation after facial and intranasal trigeminal stimulation

doi:10.21203/rs.3.rs-1857511/v1

Download PDF

Research Article

Migraine type-dependent patterns of brain activation after facial and intranasal trigeminal stimulation

https://doi.org/10.21203/rs.3.rs-1857511/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

In migraine, the trigeminal nerve is intimately involved in the pathophysiology of the disease. We hypothesized that alterations in the sensory trigeminal activation in migraine would be reflected by EEG-derived event-related potentials (ERP). We aimed to investigate differences in the temporal and spatial processing of trigeminal stimuli between interictal migraine patients and healthy subjects.

ERP to trigeminal stimuli were recorded at 128-channels to allow localization of their cortical sources with high temporal resolution. Seventeen patients with episodic migraine without aura, with episodic migraine with aura, and 17 healthy subjects participated in the study. The first branch of the trigeminal nerve was stimulated using intranasal chemical (CO₂), cutaneous electrical, and cutaneous mechanical (air puff) stimuli. Analyses were performed with regard to micro-state segmentation, ERP source localization, and correlation with the patients’ clinical characteristics.

Topographical assessments of EEG configurations were associated with the pathological condition. The source analysis revealed altered trigeminal-sensory response patterns in the precuneus, temporal pole, and cerebellum for both migraine groups during the interictal phase. The estimated current source density was positively correlated with migraine disease duration, indicating brain functional and structural changes as a consequence of the disease. Hyperactivity of the cerebellar posterior lobe was observed as a specific trigeminal response of migraine patients with aura.

In conclusion, our results suggest the presence of brain changes accompanying the advancement of migraine as an expression of dysfunctional central pain processing. Hence, we identified EEG patterns in response to mechano-/chemosensory stimuli that can serve as biomarkers of migraine.

Headache

Clustering Analysis

EEG

Pain

Odor

Nose

The trigeminal nerve is intimately involved in the pathophysiology of migraine. While the ophthalmic and maxillary branches of the nerve are purely sensory, the mandibular branch has sensory and motor functions. A variety of sensory inputs from the scalp, face, nose, mouth, and meninges are processed by the trigeminal pathway providing tactile, proprioceptive, and nociceptive as well as chemosensory information. Nociceptive afferents in the nasal cavity have free nerve endings located within the lining of the nasal vestibule and nasal chambers. When stimulated by chemical substances, they lead to sensations like stinging, burning, warmth, or cold (Doty 1978; Reeh 2001), which are significant components of the extensive somatosensory representation of the craniofacial region in the central nervous system.

The exact mechanisms underlying migraine pain are still not fully understood, mainly because of the absence of any identifiable brain pathology in established imaging and EEG techniques. It is assumed that migraine headaches originate from trigeminal nerve terminals in meninges due to neurogenic inflammation (Goadsby 2007; Messlinger 2020; Moskowitz 1993; Coppola 2020).

Data on functional imaging show that the posterior part of the hypothalamus is significantly activated in the pre-ictal period of the migraine attack and a role its onset has been suggested (Schulte 2017; Maniyar 2014). Furthermore, changes in hypothalamic-brainstem connectivity are thought to drive the attacks (Schulte 20016). Observing spontaneous migraine attacks in a patient by magnetic resonance imaging revealed altered functional coupling between the hypothalamus and a) spinal trigeminal nuclei and b) the dorsal rostral pons pre-ictally. Hence, a top-down activation of trigeminal structures seems likely. As a result, peripheral and central trigeminal nerve endings release pro-nociceptive neurotransmitters, i.e., Calcitonin Gene-Related Peptide (CGRP) (Goadsby 2017). Peripheral and central sensitization results in the transduction of nociceptive stimuli, which then may clinically manifest as migraine headache.

Although, the trigeminal system plays a major role in migraine pathophysiology, investigations of trigeminal sensory pathways in migraines using multichannel EEG are lacking. Previous experimental approaches relying on blink reflex (Sandrini 2002; Brooks 2013) and/or classical pain-related event-related potential (ERP) recordings (Sohn 2016; Valeriani 2003) pointing towards abnormal temporal processing of external stimuli and dysfunctional sequential recruitment of neuronal networks involved in pain-processing in migraine, possibly the pathophysiological background of the disease (de Tommaso 2014). Psychophysical tests in patients with migraine showed a mechanical hypersensitivity in trigeminal and non-trigeminal regions, most pronounced in relation to a migraine attack (Scholten-Peeters 2020).

We hypothesized that alterations in the trigeminal activation in migraine would be reflected by EEG recordings based on a sufficient temporal and spatial resolution. The present study aimed to investigate whether there are differences in the processing of trigeminal stimuli between migraine patients with (PMA) and without aura (PM) and healthy subjects (N) expressed as different brain response patterns. Specifically, responses were studied during the interictal period with regard to three types of stimuli: (a) intranasal chemical nociceptive stimulation with gaseous CO2, (b) cutaneous electrical stimulation, and (c) cutaneous mechanical stimulation using air puffs. To determine whether there were differences in trigeminal pain-processing at different brain levels, we used the recording of ERPs obtained with 128 channel EEG. This allowed us to examine the spatial distribution of the ERP sources inside the cortices.

Participants

Fifty-one right-handed subjects participated in the study, including 17 PM patients (mean age ± sd = 40.1 ± 13.1 y), 17 PMA patients (mean age ± sd = 30.1 ± 10.3 y) and 17 N subjects (mean age ± sd = 37.8 ± 11.2 y). Patients were diagnosed according to the International Headache Society (IHS) criteria of the International Classification of Headache Disorders (Headache Classification Committee ICHD-III 2018) for “migraine without aura” and “typical aura with migraine headache.” Exclusion criteria included a previous history of olfactory disorders due to sinonasal disease, head trauma, neuropsychiatric disease (except mild depression) and neurological disease (e.g., Parkinson’s disease, Alzheimer’s disease). The study protocol was approved by the Ethics Board of the Faculty of Medicine at the TU Dresden (protocol number EK 58022015). Detailed information about the experiment was given to all participants, and informed written consent was obtained. All aspects of the study were performed following the Declaration of Helsinki.

Questionnaires

Migraine-related disability was assessed using the migraine disability score (MIDAS; sum score > 20 corresponds to high migraine-related disability) (Stewart 2001) and the Headache Impact Test (HIT-6; sum score > 60 corresponds to high migraine-related disability) (Kosinski 2003). To assess clinically relevant depression, patients completed the Beck Depression Inventory (Beck 1996).

Experimental procedure

The study was divided into two sessions, separated by at least five days. During first visit, participants underwent olfactory and trigeminal testing and training for the experimental condition to familiarize them with the study procedures. Olfactory function was tested using the „Sniffin' Sticks” identification test (Hummel 1997). Trigeminal testing was performed using a lateralization task (Frasnelli 2009). Only subjects with a normal range of olfactory/trigeminal function were eligible to participate in the study. Further, the Beck Depression Inventory-II (BDI) (Beck, 1996) was administered to assess self-reported symptoms of depression and to exclude moderate or severe depression in participants. Then eligible subjects were instructed through a training session simulating the actual experiment by randomly receiving a short trial of the three stimulus conditions. They also learned to perform an simple computer game based on tracking a white square with the mouse pointer, introduced to maintain stable concentration levels and avoid unwanted ocular movements that might produce EEG artifacts.

The EEG with 128 channels (Ag–AgCl active-electrodes, BioSemi, Amsterdam, NL) was recorded during the second visit. The first branch of the trigeminal nerve was stimulated in three different ways; described below. Electrodes were mounted on a standard 10/10 System head cap (BioSemi cap) and connected to the EEG amplifier (BioSemi Active Two AD-box). The outputs of all the AD converters were digitally multiplexed and sent via a single optical fiber to a USB2 Receiver (BioSemi) connected to a PC, where the data were registered and stored through acquisition software (BioSemi ActiveView 605). The sampling frequency of the signal was set at 512 Hz. An electrode gel (Signa gel—Parker laboratories, Inc. Fairfield, New Jersey, USA) was applied to each electrode tip to maximize the conductivity and reduce the impedance. Four pairs of flat-type, active electrodes were used to record the vertical electrooculogram (EOG), the horizontal EOG, both earlobes, and both mastoids. The olfactometer trigger was registered synchronously to the EEG signal. The whole acquisition lasted 120 minutes, with a break in between. The order of the conditions was randomized across the subjects. Acoustic white noise was played through headphones to cover the background noise. Participants were asked to rate intensity (0, not perceived to 100, extremely strong) and pleasantness (-5, very unpleasant to + 5, very pleasant, 0 neutral) on a visual analog scale during the experiment.

Stimulus presentation

The three stimuli, intranasal nociceptive chemical, cutaneous electrical, and cutaneous mechanical, were delivered to areas innervated by the first trigeminal branch. For chemical nociceptive stimulation, gaseous CO2 was chosen because it does not produce a concomitant odorous sensation which might interfere with pain sensation. CO2 was presented to the right nostril, while both electrical and mechanical stimuli were presented to the right forehead.

The chemical stimulation was presented using a computer-controlled air-dilution olfactometer (OM6B; Burghart, Wedel, Germany) with a constant flow of odorless, humidified air of controlled temperature (7.2 l/min with an 80% relative humidity and 36°C). The electrical stimulus (500µs duration, 200Vmax) was delivered through Ag–AgCl electrodes (COP10S-SEI; EMG s.r.l., Cittadella, Italy) using a constant current stimulator (Digitimer DS3; Digitimer Ltd., Welwyn Garden City, Hertfordshire, UK). The electrode was attached to the skin on the right eyebrow, innervated by the trigeminal nerve’s ophthalmic branch, with a ground electrode on the left forearm. However, due to the presence of large EEG artifacts, the electrical condition was excluded from the EEG analysis. The mechanical stimulus was an air puff of 250ms duration, generated by the olfactometer. The opening of the Teflon tubing, through which the puff was delivered, had a diameter of 2 mm; the stimulator outlet was positioned approximately at 5 mm distance from the skin surface pointing towards the right forehead, airflow was 7.2 l/min. This stimulus was irritating but not painful.

Stimulation parameters

The setup of the stimulus condition was based on a pilot study aiming to match the stimuli in intensity. For this, 15 individuals were exposed to chemical, electrical, and mechanical conditions. They were asked to evaluate the conditions on a scale from 0 (not perceived) to 100 (extremely strong) in intensity. Initially, the chemical condition were split into two levels; CO2_low (44% v/v CO2) and CO2_high (56% v/v CO2). As regards the electrical stimulation, we first set the electrical threshold (minimum level of electrical stimulus detection), then increased the relative experimental condition by 20% of this value. The mean ratings and standard deviation for the group in each condition are reported in supplemental digital content- (SDC-)Table 1.

Table 1

Clinical and demographic data of episodic migraine patients with aura
Age	Sex	BDI	Aura	Midas	HIT-6	years with migraine	migraine days last 3 month
32	F	37	y	23	63
18	F	24	seeing stars, silverish flickery scotoma central, concentric visual field narrowing, hypaesthesia arm, leg, aphasia	46	66	10	24
28	F	24	flickery lights temporal, dysesthesia arm	160	65	5	75
58	F	18	y	112	55	38	21
51	F	24	flickery scotoma temporal	71	63	13	15
24	F	1	y	1	69
27	F	2	dysesthesia arm, leg	21	54	5	12
28	f	1	y	7	57
39	F	16	y	12	62
31	F	12	y	160	65
26	M	1	y	3	53
26	F	5	y	3	55
22	F	5	y	8	53
24	F	1	y	5	54
26	F	0	y	5	56
27	M	7	y	19	63
25	F	9	y	50	64
Age	Sex	BDI	Aura	Midas	HIT-6	years with migraine	migraine days last 3 month
y = yes

A Friedman test revealed a statistically significant difference across the 4 intensity ratings (χ²(3, n = 15) = 20.3, p < 0.001). Higher median (Md) intensity values were observed for CO2_high (Md = 52) compared to CO2_low (Md = 31), electrical condition (Md = 32), and mechanical condition (Md = 33). A posthoc test (Wilcoxon) with Bonferroni correction showed no significant differences between CO2_low vs. electrical condition nor for CO2_low vs. mechanical or electrical condition vs. Puff (p > 0.16).

Therefore, as the final set of experimental conditions, we chose 1) cutaneous mechanical stimulation using air puffs, 2) cutaneous electrical stimulation (20% higher than the individual threshold), 3) intranasal chemical stimulation with gaseous CO2 (44% v/v CO2). The three stimuli were presented at intervals of 20s (range: 18-22s) with 50 repetitions each.

Data analysis

ERP were evaluated using the Cartool software (Brunet 2011). The pre-processing steps included filtering (low pass 15 Hz, high pass 0.1 Hz, and notch 50 Hz), baseline correction, and manual artefact rejection. For each subject, a grand average (GA) was computed for each condition. The single-subject GA were entered in the group grand mean (GM) for a final total of 9 GM (3 groups X 3 conditions).

Functional microstates, as defined by Lehmann and colleagues (Lehmann 1987), were computed simultaneously for the three groups in each condition using a microstate segmentation algorithm (Katayama 2007; Pascual-Marqui 1995). Topographical atomize and agglomerative hierarchical clustering (T-AAHC) analysis was used to identify the stable map voltage topography (microstates). The best clustering was individuated by maximizing a meta-criterion, based on the median of all optimal numbers of clusters defined by 6 criteria as described in (Brechet 2019). Cluster maps correlating more than 80% were merged, and segments less or equal to 24 ms were rejected. The synthetic maps were then visually compared to the time course topographical maps obtained by the grand mean for each condition Fig. 2 (i) / 5 (i). Segmentation of the GM can identify similarities and differences in the time course of the post-stimulus brain process. The identified micro-states, a set of template maps that best represent a certain epoch segment, were fitted back to the single subject GA data using a procedure based on a spatial correlation of the cluster maps (Michel 1999). Ultimately, the fitting procedure generated parameters of interest in the relevant maps that pinpoint specific brain process characteristics to the group/condition. The corresponding anatomical brain structures responsible for the voltage scalp map distribution were identified using source localization techniques based on autoregressive average (LAURA) algorithm (Grave 2004)using 6000 source points distributed in the gray matter and integrated into Cartool software (Brunet 2011). The inverse solution matrix was multiplied by the GA of each subject to obtain the results of the inverse solution, which represents the estimated current source density (eCSD). Coordinates of the localized sources are reported in MNI- (Montreal Neurological Institute) space. The analysis of psychophysical, track performance, and demographical data was carried out using IBM SPSS Statistics v.23 (SPSS Inc., Chicago, Ill., USA).

The eCSD of critical brain activities for each subject was extracted in a spatial region of interest (sROI) defined by the solution point in the maximum activity and the six cartesian nearest neighbors and mediated in the timeframe defined by the fitting back procedure.

Demographics and Migraine Screening

The three groups were homogenous in gender (χ²(2, n = 51) = 0.30, p = 0.86), and age (χ²(2, n = 51) = 4.47, p = 0.11) with a relative mean age ± s.d. N = 38 ± 11, PM = 40 ± 13, PMA = 30 ± 10 .

Of 17 patients with migraine without aura, 5 showed pronounced migraine prodromes, 4 were on prophylactic migraine medication, and 9 on nonmedication prophylactic therapy (Table 3). In addition, 6 of the patients complained about tension type headache or facial pain, and 8 patients about pain in other body regions. 11 patients took acute migraine medication and 6 consumed analgesics for pain conditions other than migraine. A psychological diagnosis was seen in 3; concomitant somatic diseases in 4 patients. Of 17 patients with migraine with aura, 2 showed pronounced migraine prodromes, only one was on prophylactic migraine medication and 2 on non-medication prophylactic therapy. Four of the patients complained about tension type headache or facial pain, and 3 patients about pain in other body regions. Six patients were on prescribed acute migraine medication, and 1 consumed analgesic for pain conditions other than migraine. A psychological diagnosis was seen in 4, and concomitant somatic diseases in 4 patients. In each patient group, 8/17 (47%) presented a Migraine Disability Index (MIDAS) of ≥ 21 points, expressing a rather strong negative impact of migraine on daily life activities. Headache Impact Test (HIT-6) value of ≥ 60 points corresponding to the severe migraine-related burden on daily life was seen in 7/17 (41%) and 9/17 (53%) of patients with migraine without aura and migraine with aura, respectively. For further details, see Tables 1 and 2.

Table 2

Clinical and demographic data with Episodic migraine Patients without aura.
Age	Sex	BDI	Midas	HIT-6	years with migraine	migraine days last 3 month
26	F	3	12	64	10	15
43	F	13	70	61	20	6
50	F	13	85	53	30	45
54	F	2	5	54	30	5
57	F	4	94	67	35	35
23	F	4	2	65
21	F	9	25	59
26	M	0	2	50
55	F	8	24	51	43	6
55	F	6	45	57	30	27
44	M	14	6	62
50	F	5	12	51
24	M	4	14	56
46	F	1	30	51	37	24
40	F	13	9	49	20	15
43	F	16	19	66	19	30
25	F	14	10	73
Age	Sex	BDI	Midas	HIT-6	years with migraine	migraine days last 3 month

Table 3

Statistical test results for the variables extracted after the fitting back of the segmentation to the single subject grand average for the mechanosensory condition. p, a p-value of the statistical test, H, the value of the Kruskal-Wallis test, **/* indicates p < 0.025/0.05, § indicates a value at the threshold supposes a trend of significance.
tf	Map	mean duration	GEV	first onset	max GFP
		p (H)
1	5	0.1810 (3.419)	0.1738 (3.500)	0.3455 (2.125)	--
	6	0.7565 (0.5581)	0.806 (0.4314)	0.1126 (4.369)	--
	9	0.0033 (11.40) **	0.0036 (11.28) **	0.2441 (2.820)	0.0199 (7.836)**
2	6	0.975 (0.05070)	0.7132 (0.6760)	--	--
	7	0.0471 (6.110) *	0.0225 (7.591) **	--	--
	8	0.9988 (0.002374)	0.9852 (0.02975)	--	--
	10	0.3467 (2.119)	0.3898 (1.884)	--	--
	12	0.0596 (5.639)^§	0.0982 (4.641)	--	--
3	9	0.9402 (0.1233)	0.9071 (0.1951)	0.5935 (1.044)	0.3902 (1.882)
4	12	0.0051 (10.56) **	0.0485 (6.054) *	--	--
4	13	0.0438 (6.256) *	0.0283 (7.131) *	--	--

To examine possible differences between the three groups, we performed a Kruskal-Wallis test with the dependent screening variables of odor identification (OI), trigeminal lateralization (TL), and BDI and with group as a categorical variable. The test revealed no statistically significant differences for odor identification, but for trigeminal lateralization (χ²(2, n = 51) = 8.51, p = 0.014) and for the BDI (χ²(2, n = 51) = 8.18, p = 0.017). BDI posthoc analyses (Mann-Whitney) with Bonferroni correction (p < 0.017) showed significant differences between N- and PM- group (p = 0.005) (when the Bonferroni constraint was released, the test became sensitive also between N- and PMA-group (p = 0.031)). Trigeminal lateralization test post-hoc analyses (Mann-Whitney) with Bonferroni correction (p < 0.017) revealed significant differences between N- and PMA- groups (p = 0.006). ((when the Bonferroni constraint was released, the test became sensitive also between N- and PM-group (p = 0.029)). Data are illustrated in the boxplot in Fig. 1 (a).

Figure 1 approx. here

Intensity-, pleasantness-, and task-performance ratings during the EEG session

The data are illustrated in Fig. 1, (b), (c), and (d).

Psychophysics and performance statistical results within the group

Statistical analysis (Friedman test) revealed that the N group rated intensity (χ²(2, n = 15) = 6.15, p < 0.046) and pleasantness (χ²(2, n = 15) = 10.5, p < 0.005) differently among the three stimulus conditions. The post-hoc analysis revealed that N rated the intensity of the CO2 higher than the intensity of electric and less pleasant than both electric and the air-puff. The PM group presented with significantly different ratings between the three conditions both in intensity (χ²(2, n = 16) = 26.0, p < 0.001) and hedonics (χ²(2, n = 16) = 27.9, p < 0.001), where post-hoc analysis revealed that PM rated the CO2 and the air puff higher in intensity, and CO2 pleasantness was lower than electrical and air puff higher than electrical stimulation. The PMA-group presented significantly different ratings of intensity (χ²(2, n = 14) = 15,527, p < .000) and hedonics (χ²(2, n = 14) = 10.145, p < .006) (Fig. 1, (b) and (c)). Post-hoc analysis showed that the electrical stimulus intensity was rated lower than the CO2 and the air puff. In contrast, the CO2 pleasantness was rated lower than the electrical and the air puff. All post-hoc tests were carried out with Wilcoxon with Bonferroni threshold correction (p < 0.017). No statistically significant difference was detected for mean performance between stimulus conditions within the group (Fig. 1d).

Psychophysics and performance statistical results between groups

Kruskal-Wallis Test on the categorical variable group (N, PM, PMA) and the dependent variables of electrical threshold stimulus acquisition, mean intensity (CO2, electrical stimulus, puff), mean hedonics (CO2, electrical stimulus, puff), and mean performance (CO2, electrical stimulus, puff) was performed. Results revealed a statistically significant difference for the following variables: mean intensity rating of the CO2 stimulus (χ²(2, n = 51) = 11.4, p = 0.003; post-hoc PM > PMA & PM > N), mean hedonic rating of the CO2 stimulus (χ²(2, n = 51) = 12.8, p = 0.002, post-hoc PM < PMA), mean tracking task performance index during EEG acquisition following the CO2 stimulation (χ²(2, n = 51) = 9.11, p = 0.011 (PMA better performer)), mean tracking task performance index during EEG acquisition following the electrical stimulation (χ²(2, n = 48) = 7.52, p = 0.02 (PMA better performer)) and mean tracking task performance index during EEG acquisition following the mechanical stimulation (χ²(2, n = 47) = 7.40 p = 0.025 (PMA better performer)). No statistically significant differences were detected for mean electrical thresholds or intensity, and pleasantness of the stimulation applied.

Eeg Results

Grand Mean

A grand mean (GM) for each group and condition was obtained on an epoched time range of 1.2s; the relative figure is reported in the supplemental digital content (SDC-Figure 2.).

The GM for each condition/group is also illustrated as red (positive) – blue (negative) voltage topography in Fig. 2(i) for PUFF and Fig. 3(i) for CO2. The corresponding scalp topographies, over 60ms time average, are reported on top of each diagram and illustrate the brain activity time course of the 1.2 s post-stimulus onset.

Microstate segmentation and Source analysis

Mechanosensory condition

The segmentation, obtained from the 128-channel grand averages of the three groups, identified a total of 18 microstates, explaining 81.9% of the global variance (Fig. 2(iii) and (iv)). Identified stable topographies (microstates) were labelled with numbers from 1 to 18. For the control group (N) ten micro-states (namely 1,4,5,6,9,12,10,13,15,16) represented the processing of the mechanosensory stimulus after the onset. For the PM group the epoch was segmented in eight micro-states, namely 3,5,6,9,12,8,14,17 and for the PMA group in nine micro-states, namely 2,9,11,5,6,7,10,12,18. Microstates 5,6 and 9 are common to the three groups within the same time frame of appearance (tf1: from 78ms to 412ms and tf2: from c.a. 572ms to 792ms) and same order (respectively in tf1 5, 6, and 9 and in tf2 9 ). Groups N and PM share map 12 from c.a. 417ms to 472ms (tf3), while N and PMA share map 10 from ca. 466ms to 607ms (tf4). Finally, PM and PMA share map 12 from c.a.732ms to 880ms (tf5). Interestingly, the initial time range of the epoch, from 0ms to 123ms, and the final part of the epoch, from 466ms till the end, are represented by different microstates across groups suggesting distinct stimulus processing in the three groups. Comparing to the original data, microstate analysis clearly reflects it (Fig. 2 (I)) with the identified microstates (Fig. 2 (iii), and (iv)). Topographies in Fig. 2(i) similar to the microstates in Fig. 2(iii) and (iv) were highlighted with the same color frame.

Figure 2 approx. here

Fitting back to the single subject EEG-grand average. The fitted back procedure was based on the microstates included in the epoch segment that contained shared microstates. This corresponds to our segmentation of the mechanosensory condition to a time frame from 76- 1200ms and includes microstates 5, 6,7,8, 9, 10, 12, and 13. For them, we extracted the variables ‘mean segment duration’ (Duration) and ‘global explained variance’ (GEV) for each subject. Both are indicative of how much the corresponding microstate is present in the relative single subject epoch segment. In addition, we also extracted the ‘first onset’ (Fonset) of appearance as an effect of the disease condition (map 5, 6, 9 ) and the maximum of the global field power (max GFP) in map 9. Based on the segmentation on the GM, we have defined time frames (tf) of interest where to extract the variable at the single-subject level, these were tf1, from 76 to 428ms, for Duration and GEV, and Fonset of map 5, 6, and 9. In addition, in this time frame, we also extracted the maximal Global Field Power (max GFP) that reflects the strength of the signal for the specific condition. In tf2, from 416 to 609ms, we extracted Duration and GEV of maps 6, 7,8, 10, and 12. In tf3, from 566 to 793ms Duration, GEV, first onset, and max GFP on the second appearance of map 9 and in tf4, from 729 to 1200ms Duration, GEV of the second appearance of map 12, and the first appearance of map 13 were extracted.

A non-parametric Kruskal-Wallis one-way analysis of variance (one variable x 3 groups) was applied on the variable extracted from the fit back on the single subject above specified. The detailed statistical results are reported in Table 3. Mean values and SE of the variables are illustrated in Fig. 3 along with the time frame (tf) in the analysis.

Figure 3 approx. here

Results showed that map 9 in tf1 had a significant different mean duration, GEV, and max GFP across the groups. Post hoc tests revealed that Duration, GEV, and max GFP were significantly bigger for the PMA condition with respect to N and PM.

Map 7 within tf2 had a significantly longer duration and bigger GEV in the PMA group than N and PM groups.

Finally, map 12 was significantly longer and explained more variance for both group PM and PMA with respect to N, and map 13 had a longer duration and a bigger GEV for N with respect to PM and PMA.

These results suggested that a longer duration of map 9 and map 7 and higher signal strength for map 9 are specifically observed in PMA. Map 12 is a unique topography associated with the migraine condition in both aura and non-aura symptoms. Finally, map 13 represents an exclusive topography characteristic of the normal condition.

Source analysis for the mechanosensory condition. The source analysis of map 9 revealed maximal activity in the right cerebellum, lower activity in the left cerebellum, and the left temporal inferior and superior pole.

The source analysis of map 7 pointed to a maximum activity in the left and right Cerebellum and the left and right gyrus rectus.

Map 12 is generated by a source in the right cerebellum and the left inferior frontal triangular area. Finally, map 13 is generated by a source in the right postcentral gyrus and left temporal pole. Brain activities associated with the discussed maps are illustrated in Fig. 4.

Figure 4 approx. here

Chemosensory condition

The segmentation on the chemosensory condition Grand Means for the three groups identified a total of 18 stable topographies, explaining 80.3% of the global variance (Fig. 5(iii) and (iv)). The micro-states, originally named with numbers in crescent order, were renamed with alphabetic letters from a to t to avoid confusion with the maps identified in the mechanosensory segmentation. In detail, the chemosensory stimulus processing was described by eleven micro-states for the control (N), namely a, b, d, e, f, h, i, m, o, s, and t; six microstates for the PM group, namely c, f, g, p, q, and t and by height microstates for the PMA group, namely a, f, c, h, l, n, r, and t. The three groups shared map f from c.a.78ms to 287ms, although with notable different duration and onset, and map t from c.a. 636ms till the end of the epoch. N and PMA groups shared microstate a from the beginning of the epoch to c.a. 46ms and microstates h from c.a. 236ms to 441ms. Finally, PM and PMA groups shared map c that appears in different time frames, at the beginning of the epoch for PM and around 107ms for PMA. The micro-states analysis clearly reflects the original data, as is visible from Fig. 5(iii), (iv), and (i), where we also highlight the similar topographies with the same color code.

Figure 5 approx. here

Fitting back to the single subject EEG-grand average. To test the consistency of the group analysis at the single-subject level, we fitted back to the single-subject the shared micro-states among the groups, namely: a, c, f, h, and t, highlighted in Fig. 5(iii). For each micro-state, we extracted the following variables: Duration, GEV, and the first onset within a specific time frame (TF) of interest defined as TF0, within 0 to 223ms for map a and c, TF1 within 80 to 359ms, and TF2 361 to 559ms for map f, TF3, within 639 to 1199ms for map t and TF4 between 119 and 225ms, for map h. We additionally extracted the first onset of map f on TF1 and the first onset of map c on TF0.

A non-parametric Kruskal-Wallis one-way analysis of variance among the three groups was used to test statistical differences on the defined variables. Details of the statistical results are reported in Table 4; mean and SE across the groups are reported in Fig. 6 for each variable. The results showed a significant difference in the mean duration of map c within the TF0. The post hoc test showed that map c was shorter with a smaller GEV for the N-group, confirming the segmentation. Also, map f within the TF2 was shorter with a smaller GEV for the N-group. Figure 6 illustrates the mean values and SE of the tested variables. This suggests that a longer map c in the initial part of the epoch and a longer map f in TF2 indicates a migraine condition.

Table 4

Statistical test results for the variables extracted after the fitting back of the segmentation to the single subject grand average for the chemosensory condition. p, p-value of the statistical test, H, the value of the Kruskal-Wallis test, **/* indicates p < 0.025/0.05. § indicates a value that supposes a trend of significance.
Map – TF	mean duration	GEV	first onset
Map – TF	p (K-W)
a - TF0	0.8803 (0.255)	0.8405 (0.348)	-
c - TF0	0.0298 (7.026)*	0.0471 (6.109)*	0.1991 (3.228)
h - TF4	0.4700 (1.510)	0.6571 (1.134)	-
f - TF1	0.4675 (1.521)	0.4675 (1.521)	0.8945 (0.223)
f - TF2	0.0490 (6.032)*	0.0516 (5.927)*	-
t - TF3	0.0655 (5.452)	0.2652 (2.655)	-

Figure 6 approx. here

Source analysis for the chemosensory condition: The source analysis of map c revealed activity in the right precuneus, left temporal Mid area, and the frontal superior gyrus. The source analysis of map f indicated activity in the temporal pole mid-left, the cerebellum Crus 1 left, and the precentral gyrus right. Brain areas involved in the maps discussed above are illustrated in Fig. 7.

Figure 7 approx. here

Table 5 summarizes the maps of interest, the correspondent maximum of brain activities in MNI coordinates, and the relative brain areas nomenclature.

Table 5

Source localization of the relevant brain topographies for migraine condition, migraine with aura and control.
		MNI (mm)			Brain area
	Map	x	y	z	Brain lobe	Brain sub-area	Brodnmann area
general migraine	c	15.11	-57.41	39.28	Parietal lobe	Precuneous R	7
		-63.45	-3.02	-15.11	Temporal lobe	Superior temporal gyrus L	22/21
		15.11	81.58	-3.02	Frontal lobe	Frontal pole R	10
	f	-51.36	9.06	-33.23	Temporal lobe	Temporal pole L	21/20
		-21.15	-63.45	-33.23	Cerebellum	Posterio lobe	Crus 1
		51.36	-9.06	45.32	Frontal lobe	Precentral gyrus R	4
	12	9.06	-33.23	-57.41	Cerebellum	Posterio lobe R	9
	12	-51.36	27.19	15.11	Frontal lobe	Inferior frontal gyrus opercular part.	45/46
migraine with aura	7	-33.23	-45.32	-57.41	Cerebellum	Posterior lobe L	8
		45.32	-45.32	-45.32	Cerebellum	Posterior lobe R	Crus 2
		3.02	63.45	-21.15	Frontal lobe	Fusiform gyrus R	11
		-9.06	69.49	-21.15	Frontal lobe	Lingual gyrus L	11
	9	15	-63	-39	Cerebellum	Posterior lobe L	8
		-15	-63	-39	Cerebellum	Posterior lobe L	8
		-51.36	15.11	-39.28	Temporal lobe	Inferior temporal gyrus L	38/20
Control	13	57.41	-9.06	39.28	Parietal lobe	Postcentral gyrus R	3
Control	13	-45.32	21.15	-39.28	Temporal lobe	Temporal Pole L	20

Correlation analysis with clinical data

General migraine condition.

Our results showed a relationship between the general migraine disorder (with and without aura) and map c and f within the chemosensory condition and map 12 within the mechanosensory condition. To test the relationship between the eCSD signal of the corresponding elicited brain activities and the clinical data available for the PM and PMA groups, namely BDI, MIDAS, HIT-6, ‘Years with migraine’ and ‘migraine days last 3 months’ we performed a Spearman rho correlation analysis. There was a strong positive correlation between eCSD in the right precuneus, left temporal pole, and right cerebellum and the variable ‘Years with migraine’, with respective values r = 0.61, p = 0.039; r = 0.60, p = 0.043; and r = 0.62, p = 0.34, indicating a high level of eCSD associated with longer years of persistent migraine disease.

Further, medium to low correlation have been found for patient reported measures as HIT-6 (left temporal pole: r=-0.43(p = 0.02); right cerebellum: r=-0.35(p = 0.05)). Since other patient reported measures as Midas did not show any correlation, we view these results with caution, also due to subjectivity and inter-individual variability in a rather small group of patients.

Details of the correlations are reported in Fig. 8 and Table 6.

Table 6

Spearman rho Correlation analysis and significance between the eCSD for the max brain activity in map c (precuneus), f (temporal pole) and 12 (cerebellum) and the clinical scores BDI, MIDAS, HIT-6, Years with migraine, Migraine days last three months.
	Spearman r	BDI	MIDAS	HIT-6	Years with migraine	Migraine days last 3 months
Precuneus	r	-0.03	-0.15	-0.17	0.61	-0.03
Precuneus	P	0.88	0.44	0.38	0.04 *	0.91
Temporal Pole	r	-0.16	0.11	-0.43	0.60	-0.18
Temporal Pole	P	0.41	0.55	0.02 *	0.04 *	0.57
Cerebellum	r	-0.22	0.01	-0.36	0.62	-0.21
Cerebellum	P	0.26	0.94	0.05	0.03 *	0.51

Figure 8 approx. here

Migraine condition with aura.

Segmentation analysis and the fitting back to the single subjects of the variable defining the microstates have indicated map 7 and the strength and duration of map 9 as specific topographies for the migraine condition with aura. To confirm these results, we have computed the results of the inverse solution and the related eCSD in the sources for each group. A Kruskal-Wallis test revealed a statistically significant difference in eCSD signal extracted from the ROI with maximum brain activity sources of map 9 (right cerebellar cortex) across the three different groups, N, PM, and PMA, H = 8.496, p = 0.0143. The PMA group recorded a higher mean eCSD of 1.163 than the other two groups N and PM, which recorded a mean eCSD of 0.9348 and 0.9215. The post-hoc test confirmed the significant differences with an alpha level < 0.05 and multiple comparison correction. This confirms an increase in the strength of the eCSD signal in the right cerebellar cortex during the tf1 for the PMA group.

Data are illustrated in the box plot of Fig. 9.

Figure 9 approx. here

Our study revealed characteristic EEG topographies for migraine patients with or without aura and healthy controls. Map c, map f, and map 12 in the late part of the epoch (tf4) were specific for the general migraine conditions (with and without aura). The fit corroborated the identification to the single-subject data that, coherently with the group analysis, explained in a statistically significant way the data variance. The relative source analysis indicated a maximum of activities in brain areas respectively located in the precuneus right (BA7), temporal pole left (BA21/22), and the cerebellum in the right posterior lobe (area IX). eCSDs, extracted in the ROI centered in the maximum activity of the previously mentioned brain areas, positively correlated with the clinical score ‘years of migraine.’ This indicates an intensification of brain response in the above-mentioned areas with the duration of migraine.

Evidence also suggested that modulation in strength and duration of the signal for map 9 in tf1 and the relative eCSD in the left posterior lobe of the cerebellum (area VIII) is specific to migraine with aura condition.

Regarding the clinical and psychophysical data, there were no differences between the study groups in olfactory identification ability. Trigeminal sensitivity however, was significantly higher in migraine patients with and without aura than controls. The BDI score was significantly higher in both PM and PMA groups, indicating more depression-relevant symptoms in migraine patients than controls. Anxiety and depressive disorder are known comorbidities in migraines resulting in an additional disease burden (Alwhaibi 2020) [2]. Interestingly, tracking performance was significantly better in the PMA group.

EEG identification of general migraine condition

In our study, we were able to identify unique maps and relative brain areas related to the migraine condition. All the brain areas highlighted have been related to pain conditions.

The precuneus is part of the default mode network (DMN) and is involved pain sensitivity ( Goffaux 2014; Emerson 2014; Schwedt 2015) [24,19,62] and modulation of pain in both acute and chronic conditions (Emerson 2014)[19] and specifically in chronic migraine (Schwedt 2015)[62], involving overactivity and abnormal linkage with cortical networks (Zhang 2016; Tomasi 2011; Huang 2011; Griebe 2014; Schwedt 2015; Goffaux 2014) [77,72,29,27,61,24]. In a work that studied the functional connectivity (FC) of the precuneus in migraines, the authors observed an abnormal FC wihin the DMN (Zhang 2016) [77], suggesting this area playing a crucial role in dysfunctions commonly associated with migraine. Our data showed that: (1) precuneus was activated during map c within the chemosensory stimulation, (2) Map c was unique for the migraineurs group, with and without aura, and (3) the uniqueness was statistically confirmed at the single-subject level. Moreover, the eCSD from the ROI centered in the precuneus maximum activity was positively correlated with the variable ‘years with migraine’, confirming the crucial role of this specific area in the pathophysiology of the migraine.

The temporal pole (TP) participates in pain processing by mediating affective responses to painful stimuli and is a multisensory convergence zone responsible for processing visual, olfactory, and auditory stimuli (Demarquay 2008; Moulton 2011). The activity that we observed in this area is positively correlated with the duration of migraine. The TP is differentially structurally altered in chronic migraine (Schwedt 2015) [64]. Our result agrees with the TP hyperexcitability found by Moulton et al. (Moutlon 2011) in interictal migraine patients. The same area was found exacerbated during migraine attacks (Burstein 2010). Notably, in agreement with our results, the temporal pole has been shown to have atypical activation and atypical functional connectivity in several migraine fMRI studies (Maleki 2012a, 2012b; Tessitore 2013; Stankewitz 2011). Altogether this indicates that the trigeminal response in the temporal pole can differentiate migraine brain from healthy controls, and it can possibly have a role in migraine etiology.

The cerebellum has been linked to processing sensorimotor, cognitive, and affective information of pain-perception (Moulton 2011; Mehnert 2019; Ruscheweyh 2014) and pain in migraines (Carpenter 1961; Coombes 2016; Kros 2018; Mehnert 2019; Vincent 2007). A recent study has explored the alteration of gray matter volume and diffusion properties of the cerebellum in migraineurs. The findings showed microstructural changes that suggested a major contribution to dysfunction of the transmission and modulation of noxious information, trigeminal nociception, and conduction and integration of multimodal information in migraine (Qin 2019). Hyperactivity in the cerebellum in migraine patients in response to negative stimuli, similar to our chemosensory condition, has been linked to decreased inhibition of aversive processing that can act as trigger to aggravate headache in migraine patients (Wang 2017).

EEG changes in migraine patients with aura

The two subgroups of the disease, migraine with and without aura, are thought to differ in terms of pathophysiology (Zhang 2016). Our results have shown an increase of duration in time and strength of map9 in tf1 during the mechanosensory stimulation for the PMA group and a significant increase of the eCSD localized in the posterior lobe cerebellum area VIII (Figs. 3 and 6, Table 5). This is in line with the reported evidence for interictal cortical hyperexcitability in migraine, which is most pronounced in patients with aura (Brigo 2012). The characteristic modulation of the signal, individuated by the modulation in strength and duration of map 9, represent a unique signature of EEG-noxious response of the migraine with aura condition.

Up to now, several studies have linked clinical and pathophysiological evidence between general migraine and lesions in the cerebellum (Kruit 2006, 2005, 2004). From this point of view, our results on the increase of signal in the cerebellar cortex in migraine and especially in the PMA group corroborate previous results with the addition that we identified the areas (location and time tf4 after stimulus onset) in the cerebellum posterior lobe area IX which positively correlated with the duration of the symptomology for both migraine and migraine with aura.

Altered sensory processing in migraine

Different maps and relatively different brain areas are highlighted with different nociceptive trigeminal input, possibly due to the different valence of the stimuli. This agrees with the results of previous studies that indicate that the brains of migraineurs might also be interictally different in terms of how they respond to sensory stimulation (Boulloche 2010; Charles 2013; Coppola 2007; Russo 2012; Liu 2012) suggesting a permanent interictal alteration of the function and structures of the brain as a consequence of the migraine state.

To the best of our knowledge, our study is the first one involving a chemosensory trigeminal stimulation of patients in an interictal condition. Studies of intranasal ammonia-induced activation in the interictal period showed that migraineurs have regional brain activation that differs from controls (Aderjan 2010; Schwedt 2014; Stankewitz 2013, 2011a; Russo 2012a; Moulton 2011, 2008).

The air-puff condition can be compared to other cutaneous noxious stimuli. Several works have reported a hypersensitivity to touch, indicated as mechanical allodynia. This response is typical of up to two-thirds of migraine patients (Lipton 2008; Selby 1960) especially recognized in the cephalic/facial region and registered at the cerebral level using fMRI techniques. Our experiment observed an increase in the duration and intensity of the EEG signal for the PMA group within the air-puff condition in the temporal segment tf1 where map 9 was present (Fig. 2). The increase of duration and intensity of map 9, expressed by the GEV, were statistically significant also at the single-subject level. This can be explained by hyperactivity typically associated with migraines with aura. We also added the information that the relative hyperexcited brain area has its maximum in an area of the right cerebellum, area 8 (Fig. 9a)), with other local maxima in the left cerebellum and the inferior temporal gyrus. All this agrees with the idea that migraine may be considered a brain state of altered excitability, as pointed out also by previous works (Charles 2013; Coppola 2007).

Performance and trigeminal functions

Pain processing, such as pain memory and prior pain experience, affects neurocognitive aspects of the human brain (Wiech 2008). Expectation, vigilance, and reappraisal modulate the ability to perceive pain (Keltner 2006; Meyer 2002). We believe that the trigeminal hypersensitivity, and the increased performance, especially in the migraine with aura group in our study, is an index of this enhanced cognition of pain. The same explanation can be applied to observing activity in several cognitive areas for PM and PMA groups, such as frontal pole and inferior frontal gyrus, fusiform and lingual gyrus. Also, activity in the motor cortex (BA4) and cerebellum can be ascribed to the preparation of nocifensive reflexes initiated by the hypersensitivity to the trigeminal-stimulation (Ruehle 2006), and to movement suppression or movement evoked by the noxious stimulus itself (David 2002).

Electrical neuroimaging as diagnostic tool.

Finally, it is crucial to stress that common neuroimaging modalities, such as MRI and CT, do not show migraine-specific brain abnormalities, and no specific tests or biomarkers are available to diagnose the disease. An accurate diagnosis can be made only with recurring headaches and the associated symptoms, resulting in an often long and exhausting procedure for migraineurs. To delineate an effective and prompt treatment, a diagnostic tool would be then much desirable. The results of this work show that a simple mechanosensory/chemosensory experimental paradigm targeting the trigeminal nerve might pinpoint unique topographies and relative brain areas discriminating migraineurs from controls. This result is significant since migraine is not commonly associated with a structural abnormality detectable with standard MRI or CT procedure (Evans 2020).

All together, our results highlight the existence of functional differences between migraine patients with and without aura, where difference in response to presented stimuli may be used as a diagnostic tool warranting further studies.

Acknowledgments:

We are grateful to all patients and volunteers who diligently sat into the study. We thank Caterina Pacifico for the editing and proofreading the manuscript. None of the authors had a conflict of interest. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Aderjan D, Stankewitz A, May A (2010) Neuronal mechanisms during repetitive trigemino-nociceptive stimulation in migraine patients. Pain 151:97–103
Alwhaibi M, Alhawassi TM (2020) Humanistic and economic burden of depression and anxiety among adults with migraine: A systematic review. Depress Anxiety; 37:1146–1159
Beck AT, Steer RA, Brown GK (1996) Manual for the Beck depression inventory-II. Psychological Corporation, San Antonio, TX
Boulloche N, Denuelle M, Payoux P, Fabre N, Trotter Y, Geraud G (2010) Photophobia in migraine: an interictal PET study of cortical hyperexcitability and its modulation by pain. J Neurol Neurosurg Psychiatry 81:978–984
Brechet L, Brunet D, Birot G, Gruetter R, Michel CM, Jorge J (2019) Capturing the spatiotemporal dynamics of self-generated, task-initiated thoughts with EEG and fMRI, vol 194. Neuroimage, pp 82–92
Brigo F, Storti M, Nardone R, Fiaschi A, Bongiovanni LG, Tezzon F, Manganotti P (2012) Transcranial magnetic stimulation of visual cortex in migraine patients: a systematic review with meta-analysis. J Headache Pain 13:339–349
Brooks JB, Fragoso YD (2013) The blink reflex test does not show abnormalities in a large group of patients with chronic migraine. Arq Neuropsiquiatr 2013; 71: 862–865
Brunet D, Murray MM, Michel CM (2011) Spatiotemporal analysis of multichannel EEG: CARTOOL. Comput Intell Neurosci; 813870
Burstein R, Jakubowski M, Garcia-Nicas E, Kainz V, Bajwa Z, Hargreaves R, Becerra L, Borsook D (2010) Thalamic sensitization transforms localized pain into widespread allodynia. Ann Neurol 68:81–91
Carpenter MB, Hanna GR (1961) Fiber projections from the spinal trigeminal nucleus in the cat. J Comp Neurol 117:117–131
Charles A (2013) Migraine: a brain state. Curr Opin Neurol 26:235–239
Coombes SA, Misra (2016) G. Pain and motor processing in the human cerebellum. Pain 157:117–127
Coppola G, Parisi V, Di Renzo A, Pierelli F (2020) Cortical pain processing in migraine. J Neural Transm (Vienna) 127:551–566
Coppola G, Pierelli F, Schoenen J (2007) Is the cerebral cortex hyperexcitable or hyperresponsive in migraine? Cephalalgia. 27:1427–1439
Davis KD, Pope GE, Crawley AP, Mikulis DJ (2002) Neural correlates of prickle sensation: a percept-related fMRI study. Nat Neurosci 5:1121–1122
de Tommaso M, Ambrosini A, Brighina F, Coppola G, Perrotta A, Pierelli F, Sandrini G, Valeriani M, Marinazzo D, Stramaglia S, Schoenen J (2014) Altered processing of sensory stimuli in patients with migraine. Nat Rev Neurol 10:144–155
Demarquay G, Royet JP, Mick G, Ryvlin P (2008) Olfactory hypersensitivity in migraineurs: a H(2)(15)O-PET study. Cephalalgia 2008; 28: 1069–1080
Doty RL, Brugger WE, Jurs PC, Orndorff MA, Snyder PJ, Lowry LD (1978) Intranasal trigeminal stimulation from odorous volatiles: psychometric responses from anosmic and normal humans. Physiol Behav 1978; 20: 175–185
Emerson NM, Zeidan F, Lobanov OV, Hadsel MS, Martucci KT, Quevedo AS, Starr CJ, Nahman-Averbuch H, Weissman-Fogel I, Granovsky Y, Yarnitsky D, Coghill RC (2014) Pain sensitivity is inversely related to regional grey matter density in the brain. Pain 2014; 155: 566–573
Evans RW, Burch RC, Frishberg BM, Marmura MJ, Mechtler LL, Silberstein SD, Turner DP (2020) Neuroimaging for Migraine: The American Headache Society Systematic Review and Evidence-Based Guideline. Headache; 60: 318–336
Frasnelli J, Charbonneau G, Collignon O, Lepore F (2009) Odor localization and sniffing. Chem Senses 34:139–144
Goadsby PJ (2007) Recent advances in understanding migraine mechanisms, molecules and therapeutics. Trends Mol Med 13:39–44
Goadsby PJ, Holland PR, Martins-Oliveira M, Hoffmann J, Schankin C, Akerman S (2017) Pathophysiology of Migraine: A Disorder of Sensory Processing. Physiol Rev 97:553–622
Goffaux P, Girard-Tremblay L, Marchand S, Daigle K, Whittingstall K (2014) Individual differences in pain sensitivity vary as a function of precuneus reactivity. Brain Topogr 27:366–374
Granziera C, DaSilva AF, Snyder J, Tuch DS, Hadjikhani N (2006) Anatomical alterations of the visual motion processing network in migraine with and without aura. PLoS Med 3:e402
Grave de Peralta Menendez R, Murray MM, Michel CM, Martuzzi R, Gonzalez Andino SL (2004) Electrical neuroimaging based on biophysical constraints. Neuroimage; 21: 527–539
Griebe M, Flux F, Wolf ME, Hennerici MG, Szabo K (2014) Multimodal assessment of optokinetic visual stimulation response in migraine with aura. Headache 54:131–141
Headache, Headache Classification Committee of the International Headache Society (IHS) (2018) The International Classification of Headache Disorders, 3rd edition. Cephalalgia; 38: 1-211
Huang J, Zong X, Wilkins A, Jenkins B, Bozoki A, Cao Y (2011) fMRI evidence that precision ophthalmic tints reduce cortical hyperactivation in migraine. Cephalalgia 31:925–936
Hummel T, Sekinger B, Wolf SR, Pauli E, Kobal G (1997) 'Sniffin' sticks': olfactory performance assessed by the combined testing of odor identification, odor discrimination and olfactory threshold. Chem Senses 22:39–52
Katayama H, Gianotti LR, Isotani T, Faber PL, Sasada K, Kinoshita T, Lehmann D (2007) Classes of multichannel EEG microstates in light and deep hypnotic conditions. Brain Topogr 20:7–14
Keltner JR, Furst A, Fan C, Redfern R, Inglis B, Fields HL (2006) Isolating the modulatory effect of expectation on pain transmission: a functional magnetic resonance imaging study. J Neurosci 26:4437–4443
Kosinski M, Bayliss MS, Bjorner JB, Ware JE Jr, Garber WH, Batenhorst A, Cady R, Dahlof CG, Dowson A, Tepper S (2003) A six-item short-form survey for measuring headache impact: the HIT-6. Qual Life Res 12:963–974
Kros L, Angueyra Aristizabal CA, Khodakhah K (2018) Cerebellar involvement in migraine. Cephalalgia 38:1782–1791
Kruit MC, Launer LJ, Ferrari MD, van Buchem MA (2006) Brain stem and cerebellar hyperintense lesions in migraine. Stroke 37:1109–1112
Kruit MC, Launer LJ, Ferrari MD, van Buchem MA (2005) Infarcts in the posterior circulation territory in migraine. The population-based MRI CAMERA study, vol 128. Brain, pp 2068–2077
Kruit MC, van Buchem MA, Hofman PA, Bakkers JT, Terwindt GM, Ferrari MD, Launer LJ (2004) Migraine as a risk factor for subclinical brain lesions. JAMA 291:427–434
Lehmann D, Ozaki H, Pal I (1987) EEG alpha map series: brain micro-states by space-oriented adaptive segmentation. Electroencephalogr Clin Neurophysiol 67:271–288
Lipton RB, Bigal ME, Ashina S, Burstein R, Silberstein S, Reed ML, Serrano D, Stewart WF (2008) Cutaneous allodynia in the migraine population. Ann Neurol 2008; 63: 148–158
Liu J, Zhao L, Li G, Xiong S, Nan J, Li J, Yuan K, von Deneen KM, Liang F, Qin W, Tian J (2012) Hierarchical alteration of brain structural and functional networks in female migraine sufferers. PLoS One; 7:e51250
Maleki N, Becerra L, Brawn J, Bigal M, Burstein R, Borsook D (2012) Concurrent functional and structural cortical alterations in migraine. Cephalalgia 32:607–620
Maleki N, Linnman C, Brawn J, Burstein R, Becerra L, Borsook D (2012) Her versus his migraine: multiple sex differences in brain function and structure. Brain 135:2546–2559 -b
Maniyar FH, Sprenger T, Monteith T, Schankin C, Goadsby PJ (2014) Brain activations in the premonitory phase of nitroglycerin-triggered migraine attacks. Brain 2014; 137: 232–241
Mehnert J, May A (2019) Functional and structural alterations in the migraine cerebellum. J Cereb Blood Flow Metab 39:730–739
Messlinger K, Balcziak LK, Russo AF (2020) Cross-talk signaling in the trigeminal ganglion: role of neuropeptides and other mediators. J Neural Transm (Vienna) 127:431–444
Meyer GA (2002) The art of watching out: vigilance in women who have migraine headaches. Qual Health Res 12:1220–1234
Moskowitz MA, Macfarlane R (1993) Neurovascular and molecular mechanisms in migraine headaches. Cerebrovasc Brain Metab Rev 5:159–177
Moulton EA, Becerra L, Maleki N, Pendse G, Tully S, Hargreaves R, Burstein R, Borsook D (2011) Painful heat reveals hyperexcitability of the temporal pole in interictal and ictal migraine States. Cereb Cortex 21:435–448
Moulton EA, Burstein R, Tully S, Hargreaves R, Becerra L, Borsook D (2008) Interictal dysfunction of a brainstem descending modulatory center in migraine patients. PLoS ONE 3:e3799
Pascual-Marqui RD, Michel CM, Lehmann D (1995) Segmentation of brain electrical activity into microstates: model estimation and validation. IEEE Trans Biomed Eng 42:658–665
Qin Z, He XW, Zhang J, Xu S, Li GF, Su J, Shi YH, Ban S, Hu Y, Liu YS, Zhuang MT, Zhao R, Shen XL, Li J, Liu JR, Du X (2019) Structural changes of cerebellum and brainstem in migraine without aura. J Headache Pain 20:93
Reeh PW, Kress M (2001) Molecular physiology of proton transduction in nociceptors. Curr Opin Pharmacol 1:45–51
Ruehle BS, Handwerker HO, Lennerz JK, Ringler R, Forster C (2006) Brain activation during input from mechanoinsensitive versus polymodal C-nociceptors. J Neurosci 26:5492–5499
Ruscheweyh R, Kuhnel M, Filippopulos F, Blum B, Eggert T, Straube A (2014) Altered experimental pain perception after cerebellar infarction. Pain 155:1303–1312
Russo A, Tessitore A, Esposito F, Marcuccio L, Giordano A, Conforti R, Truini A, Paccone A, d'Onofrio F, Tedeschi G (2012) Pain processing in patients with migraine: an event-related fMRI study during trigeminal nociceptive stimulation. J Neurol 2012; 259: 1903–1912.-a
Russo A, Tessitore A, Giordano A, Corbo D, Marcuccio L, De Stefano M, Salemi F, Conforti R, Esposito F, Tedeschi G (2012) Executive resting-state network connectivity in migraine without aura. Cephalalgia 2012; 32: 1041–1048.-b
Sandrini G, Proietti Cecchini A, Milanov I, Tassorelli C, Buzzi MG, Nappi G (2002) Electrophysiological evidence for trigeminal neuron sensitization in patients with migraine. Neurosci Lett 317:135–138
Scholten-Peeters GGM, Coppieters MW, Durge TSC, Castien RF (2020) Fluctuations in local and widespread mechanical sensitivity throughout the migraine cycle: a prospective longitudinal study. J Headache Pain 21:16
Schulte LH, Allers A, May A (2017) Hypothalamus as a mediator of chronic migraine: Evidence from high-resolution fMRI, vol 88. Neurology, pp 2011–2016
Schulte LH, May A (2016) The migraine generator revisited: continuous scanning of the migraine cycle over 30 days and three spontaneous attacks. Brain 2016; 139: 1987–1993
Schwedt TJ, Chiang CC, Chong CD, Dodick DW (2015) Functional MRI of migraine. Lancet Neurol 2015; 14: 81–91
Schwedt TJ, Chong CD (2014) Correlations between brain cortical thickness and cutaneous pain thresholds are atypical in adults with migraine. PLoS One 2014; 9: e99791
Schwedt TJ, Chong CD, Chiang CC, Baxter L, Schlaggar BL, Dodick DW (2014) Enhanced pain-induced activity of pain-processing regions in a case-control study of episodic migraine. Cephalalgia 2014; 34: 947–958
Schwedt TJ, Chong CD, Wu T, Gaw N, Fu Y, Li J (2015) Accurate Classification of Chronic Migraine via Brain Magnetic Resonance Imaging. Headache 2015; 55: 762–777
Selby G, Lance JW (1960) Observations on 500 cases of migraine and allied vascular headache. J Neurol Neurosurg Psychiatry 23:23–32
Sohn JH, Kim CH, Choi HC (2016) Differences in central facilitation between episodic and chronic migraineurs in nociceptive-specific trigeminal pathways. J Headache Pain 2016; 17: 35
Stankewitz A, Aderjan D, Eippert F, May A (2011) Trigeminal nociceptive transmission in migraineurs predicts migraine attacks. J Neurosci 2011; 31: 1937–1943. -a
Stankewitz A, May A (2011) Increased limbic and brainstem activity during migraine attacks following olfactory stimulation. Neurology 2011; 77: 476–482. -b
Stankewitz A, Schulz E, May A (2013) Neuronal correlates of impaired habituation in response to repeated trigemino-nociceptive but not to olfactory input in migraineurs: an fMRI study. Cephalalgia 33:256–265
Stewart WF, Lipton RB, Dowson AJ, Sawyer J (2001) Development and testing of the Migraine Disability Assessment (MIDAS) Questionnaire to assess headache-related disability. Neurology 2001; 56: S20-28
Tessitore A, Russo A, Giordano A, Conte F, Corbo D, De Stefano M, Cirillo S, Cirillo M, Esposito F, Tedeschi G (2013) Disrupted default mode network connectivity in migraine without aura. J Headache Pain 14:89
Tomasi D, Volkow ND (2011) Association between functional connectivity hubs and brain networks. Cereb Cortex 21:2003–2013
Valeriani M, de Tommaso M, Restuccia D, Le Pera D, Guido M, Iannetti GD, Libro G, Truini A, Di Trapani G, Puca F, Tonali P, Cruccu G (2003) Reduced habituation to experimental pain in migraine patients: a CO(2) laser evoked potential study. Pain 2003; 105: 57–64
Vincent M, Hadjikhani N (2007) The cerebellum and migraine. Headache 2007; 47: 820–833
Wang M, Su J, Zhang J, Zhao Y, Yao Q, Zhang Q, Zhang H, Wang S, Li GF, Liu JR, Du X (2017) Visual cortex and cerebellum hyperactivation during negative emotion picture stimuli in migraine patients. Sci Rep 7:41919
Wiech K, Ploner M, Tracey I (2008) Neurocognitive aspects of pain perception. Trends Cogn Sci 2008; 12: 306–313
Zhang J, Su J, Wang M, Zhao Y, Yao Q, Zhang Q, Lu H, Zhang H, Wang S, Li GF, Wu YL, Liu FD, Shi YH, Li J, Liu JR, Du X (2016) Increased default mode network connectivity and increased regional homogeneity in migraineurs without aura. J Headache Pain 17:98

No competing interests reported.

SMFigure1GACO2PUFF.tif
SDC-Figure 1. Grand Mean butterfly plot for the three groups in the study and the two stimulus conditions. Butterfly plots of 128-channels Grand Average of the chemo -(CO2) and mechano- (PUFF) sensory evoked potentials evaluated for the control group (N), patients with migraine without aura (PM), and patients with migraine with aura (PMA). The recording scale is reported in microvolts on a total time range of 1.2 s.
SMTable1pilotintesity.docx
SDC-Table 1. Intensity ratings of the various stimuli: Fifteen subjects evaluated intensity in a pilot study for the different stimuli conditions.
SMTable2patientswithauraclinical.docx
SDC-Table2. Full clinical information of episodic migraine patients with aura
SMTable3patientswithoutauraclinical.docx
SDC-Table3. Full clinical information of episodic migraine patients without aura
SupplementalDigitalContent.docx

Download PDF

Editorial decision: Major revision
27 Sep, 2022
Reviews received at journal
06 Aug, 2022
Reviewers agreed at journal
16 Jul, 2022
Reviewers invited by journal
15 Jul, 2022
Editor assigned by journal
15 Jul, 2022
Submission checks completed at journal
15 Jul, 2022
First submitted to journal
14 Jul, 2022

You are reading this latest preprint version

Migraine type-dependent patterns of brain activation after facial and intranasal trigeminal stimulation

Status:

Version 1

Abstract

Figures

Introduction

Methods

Participants

Questionnaires

Experimental procedure

Stimulus presentation

Stimulation parameters

Data analysis

Results

Demographics and Migraine Screening

Intensity-, pleasantness-, and task-performance ratings during the EEG session

Psychophysics and performance statistical results within the group

Psychophysics and performance statistical results between groups

Eeg Results

Grand Mean

Microstate segmentation and Source analysis

Mechanosensory condition

Chemosensory condition

Correlation analysis with clinical data

Discussion

EEG identification of general migraine condition

EEG changes in migraine patients with aura

Altered sensory processing in migraine

Performance and trigeminal functions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1