Abstract
Background
Structural imaging offers insight into migraine pathogenesis. Magnetic resonance imaging (MRI) morphometry plays a crucial role in identifying these alterations, yet the clinical significance remains debated. While gray matter volume and cortical curvature are commonly analyzed, cortical thickness offers a more direct measure of cytoarchitectural differences and neuroplastic changes in migraine. Advanced structural MRI techniques, including surface-based morphometry and voxel-based morphometry, have provided insights into cortical thickness alterations in migraine. These methods enable high-resolution assessment of brain morphometry, revealing dynamic changes associated with migraine phases and treatment.
Methods
This narrative review synthesizes findings from cortical thickness studies, focusing on methodological approaches, variations in imaging sequences and study designs, including cross-sectional and longitudinal studies.
Results
Studies using surface-based morphometry (i.e. SBM) and voxel-based morphometry (i.e. VBM) have reported inconsistent findings. Increased thickness is frequently observed in pain-processing regions, such as the somatosensory cortex, insula and anterior cingulate cortex reflecting hyperexcitability or maladaptive neuroplasticity. by contrast, cortical thinning has been noted in regions such as the orbitofrontal cortex, posterior cingulate cortex and visual cortex, suggesting neuronal loss or impaired cortical integrity. Differences between episodic and chronic migraine further highlight progressive structural changes associated with disease burden. Emerging evidence also suggests that preventive treatments, including calcitonin gene-related peptide monoclonal antibodies and botulinum toxin A, may reverse some of these cortical alterations, particularly in treatment responders.
Conclusions
Cortical thickness analysis provides valuable insights into migraine pathophysiology, offering a potential biomarker for disease progression and treatment response. However, inconsistencies across studies highlight the need for standardized MRI protocols and larger longitudinal investigations to clarify the clinical relevance of cortical thickness changes in migraine.
This is a visual representation of the abstract.
Introduction
Migraine is a neurological disorder characterized by recurrent headache attacks and dynamic brain change across different disease phases. The transient headache pain episodes have been associated with long-term structural alterations in the cortex, which may reflect neuronal plasticity and disease progression. These changes vary across migraine phases and may be influenced by preventive treatments. Understanding the relationship between cortical thickness and migraine progression could provide insights into pathophysiology, disease burden and therapeutic effects (1–3).
While the pathophysiology of migraine remains incompletely understood, advanced magnetic resonance imaging (MRI) has contributed to our understanding of the disorder. Structural MRI specifically provides robust and high-resolution measurements of the brain morphometry with high test-retest reliability. The most widely used structural MRI measurement is cortical thickness assessment. Cortical thickness is determined by the size and density of neuronal and glial cells in the cortex, neuropil density and even cortical blood volume (4). When the brain is exposed to a repetitive activity, the thickness of the cortex may change in task-related areas (5). Likewise, repetitive nociceptive stimulation may induce plastic adaptations in migraine, which may be reflected in the cortical thickness (6). Likewise, transient changes in the cortical blood volume may contribute to the cortical thickness changes in migraine (e.g. during attacks) (7,8).
Studies on cortical thickness may provide valuable insights to the underlying pathophysiology of migraine. In the present review, we provide an overview of the major cortical thickness findings in migraine, how they have improved our understanding of the disease, and their implications for migraine treatment.
Structural MRI methods
The two main methods of measuring cortical structure are surface-based morphometry (SBM) and voxel-based morphometry (VBM) (9). Both rely on three dimensional anatomical MRI images of the brain, which are composed of small cubes called voxels, typically 1 × 1 × 1 mm in size, which determines scan resolution.
In VBM, each voxel is assigned a probability of belonging to the grey or white matter. Using this segmentation, VBM calculates the volume of grey matter in the cortex. Some voxels, contain both grey and white matter due to their size. While VBM only partially considers this, SBM provides a method of taking it into account. The SBM approach uses the probabilities of voxels of belonging to grey or white matter to draw surfaces that delineate the cortex. This method draws two surfaces: one at the boundary between white and gray matter, and another at the boundary between gray matter and cerebrospinal fluid. The distance between these two surfaces is the cortical thickness in SBM. Thus, VBM estimates only grey matter volume, whereas SBM estimates cortical (i.e. grey matter) volume, thickness, area and gyrification (10).
The advantage of SBM is that the cortical thickness is measured at a high and subvoxel resolution, providing more precise measurement. Thus, SBM is better suited to detect small scale changes in cortical structure. However, SBM analysis is often more time consuming than VBM due to the need for manual corrections of errors in the segmentation. By contrast, VBM provides a comprehensive analysis of the whole brain, as opposed to primarily the cortex for SBM, and can detect subtle structural changes in grey and white matter volumes. However, VBM is more highly sensitive to methodological choices, including smoothing kernel size, statistical thresholds and registrations parameters, all of which can significantly impact the results. The selection of smoothing kernels (e.g. 4 mm vs. 8 mm full-width at half maximum (FWHM)) influences spatial resolution and the ability to detect localized cortical alterations, which larger kernels enhancing sensitivity but reducing spatial specificity (11,12). Moreover, individual differences in brain folding patterns can reduce alignment accuracy in VBM (13). Similarly, statistical thresholding, such as uncorrected p-value vs. family-wise error (FWE) or false discovery rate corrections can lead to variability in findings. Studies with less stringent thresholds (e.g. p < 0.05 uncorrected) may report more widespread cortical alterations, while those applying strict corrections (e.g. p > 0.05 FWE corrected) may yield more conservative results. These methodological differences contribute to the inconsistencies observed across migraine neuroimaging studies, highlighting the importance of standardized statistical approaches to improve reproducibility (14–17). In conclusion, both VBM and SBM can be used to evaluate of the cortical structure. Given that SBM provides more precise cortical thickness estimates, it is generally preferable for migraine research, particularly when investigating regional changes. However, VBM remains useful for exploring global grey matter differences and detecting additional white matter alterations. Future studies should directly compare these methods in migraine populations to optimize methodological choices.
Anatomical sequences
High resolution anatomical imaging is usually acquired with T1-weighted sequences. The most used sequences are: (i) the magnetization-prepared rapid acquisition gradient-echo (i.e. MPRAGE); (ii) fast spoiled gradient echo (FSPGR); and (iii) fast low angle shot (i.e. FLASH). The different sequences have slight differences in sensitivity, influencing cortical thickness measurements. Application of MRI scanners with different magnetic field strengths (3.0-Tesla vs. 1.5-Tesla) may yield varying sensitivity for grey and white matter segmentation (18). Studies comparing cortical thickness across different field strengths suggest that higher field strength yield more precise segmentation but can also introduce greater susceptibility for artifacts (19). This improvement in imaging quality at higher field strength is also supported by the findings from one study, where standard MRI imaging was used to assess cortical thickness asymmetries, suggesting that higher resolution imaging can yield more informative data (20). Use of different MRI sequence parameters such as repetition time (TR), echo time (TE) and flip angle can also affect the cortical thickness measurements. While longer TR and TE values can increase the signal-to-noise ratio, this may also induce the risk of motion artifacts due to longer scan time (21). Additionally, one study discussed the preprocessing of anatomical brain images and highlighted the importance of using a standardized neuroimaging pipeline to harmonize cortical thickness measurements across scanners and sites (22). Standardization of MRI sequences and parameters would allow for better comparisons of cortical thickness studies in migraine and ensure consistency and validity of results. The standardization might aid in improving our understanding of migraine and potentially facilitating its clinical applications.
Design of studies of the cortical thickness
Cortical thickness is mainly investigated in cross-sectional or longitudinal study designs, each with distinct strengths and limitation, influencing the interpretations of findings.
Cross-sectional studies compare cortical thickness between patients with migraine and healthy controls at a single tie point (23,24). They are valuable for identifying structural differences but cannot determine whether these differences are pre-existing trait, consequences of migraine or transient changes. As a result, cross-sectional studies may overestimate or misinterpret associations, especially when analyzing factors that fluctuate, such migraine frequency, medication use or disease progression. Their inability to track individual changes over time limits their utility in assessing the effects of treatment or migraine transitions.
Longitudinal studies, in contrast, provide a dynamic perspective by tracking cortical thickness changes over time. Repeated scans allow researchers to examine alterations related to different migraine phases, such as the headache phase and interictal phase (8), and evaluate the impact of preventive treatments (25). However, longitudinal studies are resource-intensive, requiring standardized imaging protocols, long-term patient retention and control for confounding factors such as lifestyle changes are medications effects. High attrition rates and variability in migraine patterns may introduce bias, potentially limiting statistical power.
Additionally, within-group analysis assesses whether clinical variables, such as monthly migraine days, correlate with cortical thickness, helping to refine patient-specific interpretation (26). Within-subject designs, which compare interhemispheric differences in unilateral migraine, can minimize interindividual variability but may still be affected by confounders like attack frequency or laterality (14).
The main approaches to compare the cortical thickness data are whole-brain and region-of-interest (ROI) based analysis. The whole-brain analysis compares the entire cortex to identify clusters that differ significantly between groups, often employing voxel-wise comparisons corrected for multiple comparisons. The most used correction methods include the false discovery rate (FDR) and FWE rate, which reduce the risk of false positives which reduce the statistical power and maintains it at an acceptable level (24,27). By contrast, ROI analysis focuses on predefined cortical regions based on prior evidence or hypotheses, reducing the number of statistical comparisons but potentially overlooking novel findings and reducing detectability of changes in subregions of ROIs. Additionally, studies utilizing non-parametric permutations tests and general linear models help to improve robustness, particularly in studies with smaller sample sizes (16). Standardizing these methodological approaches improve comparability across migraine neuroimaging studies.
Cortical thickness studies in migraine
The underlying biological mechanisms behind cortical thickness findings in individuals with migraine are not fully understood. Some of the cortical thickness alterations might represent a trait that predisposes to the development of migraine (15,28), while others might be a consequence of the repeated occurrence of migraine attacks (6,29). Evidence shows that cortical thickness alterations observed in individuals with migraine are not static and can change between the ictal and interictal phases of migraine and be remodeled over the years (8,30). It is crucial to consider that cortical thickness studies in migraine should be interpreted within the context of cortical thickness research performed on various chronic pain conditions. Several studies have been conducted in patients with different types of headaches, such as cluster headache (31), persistent headache after COVID-19 (32) or post-traumatic headache (33), as well as those with other chronic painful conditions (34,35). These studies have identified cortical alterations within specific nociceptive regions. These regions are also evident in migraine patients (27), including its subtypes. The presence of shared cortical thickness abnormalities among individuals with chronic pain raises the question of whether the observed changes in migraine patients are unique to migraine or are common to other chronic pain disorders, potentially linked to the recurrent experience of pain. Thus, one must exercise caution when interpreting the results obtained by VBM and SBM studies in migraine.
Exploring the mechanisms behind cortical thickness
Cortical thickness alterations in migraine result from structural, functional and neurobiological changes in the brain. Key contributing factors include cortical spreading depression (CSD), neuroplasticity, genetic influences, and the cumulative impact of migraine frequency and duration. CSD, a key mechanism in migraine pathophysiology, particularly in migraine with aura, disrupts brain function and metabolism (36), which may lead to transient cortical thickness changes. One study suggests that patients with frequent migraine exhibit increased cortical thickness in specific regions, likely as an adaptive response to heightened sensory processing demands, involving neurochemical changes and neuroplasticity (37). Neuroplasticity also plays a crucial role in cortical thickness variation. Changes in chronic migraine has been proposed to reflect focal dysplasia, reactive gliosis and activity-dependent plasticity due to repeated attack (38). One study found that reduced cortical thickness in the posterior cingulate correlates with migraine improvement, suggesting that treatment may reverse some structural changes (25). These findings highlight the influence of headache frequency and duration on cortical structure. One study found that catechol-O-methyltransferase polymorphisms may drive structural and functional brain changes, interacting with environmental factors to alter cortical thickness and morphology, indicating that genetic factors further contribute to structural changes in migraine (39). Additionally, studies indicate that chronic migraine patients may exhibit increased MRI-based brain age, suggesting cumulative structural damage from recurrent attacks (40). This implies that migraine-related changes may affect not only pain processing, but also long-term brain health. In summary, cortical thickness variability in migraine arises from a complex interplay of CSD, neuroplasticity, genetic predisposition and migraine frequency. Advances in neuroimaging and genetic research will continue to enhance our understanding of these mechanisms and their relevance to migraine management.
Factors influencing cortical thickness
Cortical thickness in migraine varies by subtype, age, comorbidities and chronicity. Migraine with aura is linked to distinct anatomical differences, including increased gray matter volume and cortical hyperresponsiveness, likely driven by CSD (41,42). Age also influences cortical thickness, with structural differences seen between younger and older individuals with migraine (43,44). Chronic migraine further impacts cortical structure, potentially leading to atrophy, focal dysplasia, and reactive gliosis due to repeated attacks and neuroplastic changes (38). Comorbidities such as depression and anxiety add complexity because they are linked to structural brain changes, including altered cortical thickness (8,45). Depression, for example, has been associated with increased cortical thickness in specific regions, possibly as a compensatory response to chronic pain (8). In summary, cortical thickness variability in migraine is influenced by multiple factors, highlighting the need for tailored research and clinical approaches to better understand and manage the condition.
Imaging the interictal phase
Migraine can be divided into two phases: the interictal phase, which occurs between attacks and can last for days or weeks, and the ictal phase, which encompasses the prodromal, headache, and postdromal stages of an attack. Most studies examining cortical thickness in patients with migraine have focused on the interictal phase.
Findings on cortical thickness in patients with migraine are mixed (Table 1). Some studies report no significant differences in cortical thickness between patients with migraine and healthy controls, nor associations between cortical thickness and migraine frequency or duration (46). However, a multicenter study involving 131 patients with migraine (with and without aura) found thinner cortex in several regions, including the bilateral central sulcus (motor and somatosensory areas), left middle frontal gyrus, left visual cortices and the right occipitotemporal gyrus, compared to 115 healthy controls (23). Notably, this study also identified negative correlations between attack frequency, migraine duration and cortical thickness in the left middle frontal gyrus and left central sulcus, suggesting that recurrent attacks might induce plastic changes in pain-related areas. These findings align with the hyperexcitability often seen in patients with migraine, including allodynia. Also, the involvement of the visual cortices and occipitotemporal gyrus indicate that repeated CSD exposure over time leads to long-term alterations.
Demographics, clinical characteristics, and significant finding in cortical thickness studies.
aMigraine with aura; bmigraine with complex aura; cmigraine with interictal photosensitivity; dmigraine without interictal photosensitivity; emigraine without aura; ftreatment responders; gtreatment non-responders; hepisodic migraine; ichronic migraine; jictal state.
ACC = anterior cingulate cortex; AI = anterior insula; CM = chronic migraine; CT = cortical thickness; DMC = dorsomedial cortex; F = female; FG = fusiform gyrus; HC = healthy controls; IC = isthmus cingulate; IFG = inferior frontal gyrus; IPC = inferior parietal cortex; M = male; MA = migraine with aura; MO = migraine without aura; MFS = middle frontal sulcus; MTC = middle temporal cortex; NS = not specified; OCC = occipital cortex; ParsOp = pars opercularis; PC = precentral cortex; PCC = posterior cingulate cortex; PCG = postcentral gyrus; PFC = prefrontal cortex; POC = postcentral cortex; SBM = surface-based morphometry = SFC = superior frontal cortex; SFG = superior frontal gyrus; SSC = somatosensory cortex; STG = superior temporal gyrus; STS = superior temporal sulcus; TE = echo time; TOC = temporo-occipital incisure; TR = repetition time; VBM = voxel-based morphometry; WML = white matter lesions.
Other studies have reported conflicting results. For example, some observed increased cortical thickness in regions such as the insula, superior frontal gyrus, paracentral gyrus, temporal pole and precuneus (25), whereas others reported reduced cortical thickness in areas including the inferior frontal gyrus, posterior cingulate, bilateral cuneus and middle temporal gyrus, comprisng regions associated with the default mode network (52).
Another study including 57 patients with episodic migraine, 57 patients with chronic migraine and 52 healthy controls observed decreased cortical thickness in both episodic and chronic migraine patients, with respect to healthy controls in the left temporal gyrus and right fusiform gyrus and increased cortical thickness in the right inferior temporal gyrus in healthy controls and patients with chronic migraine, with respect to episodic migraine (27). These variabilities suggest that migraine-related brain changes are not uniform but rather dynamic and influenced by multiple factors with both compensatory and maladaptive changes.
Variability in findings may also stem from differences in study designs, sample sizes and methodological choices, such as variations in statistical thresholds, smoothing kernel sizes and multiple comparison correction methods. Studies employing different spatial smoothing kernels (e.g. 4 mm vs. 8 mm FWHM) may yield conflicting results because smaller kernels preserve finer anatomical details, whereas larger kernels enhance signal detection at the cost of spatial specificity. Additionally, differences in statistical thresholding, such as uncorrected voxel-wise p-values vs. cluster-based FWE or FDR corrections, further contribute to discrepancies. For example, one study comparing 63 patients with migraine to 18 healthy controls found increased cortical thickness in the left temporo-occipital incisure and middle frontal cortex, along with reduced cortical thickness in the left superior frontal and precentral cortices, using a stringent statistical threshold (p < 0.01, cluster extent of 100 mm²) (24). However, this study did not explore the role of migraine frequency as a contributing factor.
Cortical thickness alterations in specific regions, such as the prefrontal cortex, may contribute to changes in sensory perception and processing, while increased cortical thickness in the temporal-occipital transitional zone could indicate dysfunction in visual processing (24). This dysfunction might relate to photophobia, although this has not been directly investigated. Similarly, increased cortical thickness in the lingual gyrus and pericalcarine area in patients with interictal photosensitivity suggests altered processing of visual stimuli. These changes might also involve the isthmus cingulate, an area linked to emotional processing (48).
Age-related changes in cortical thickness have also been studied. Normally, cortical thickness decreases with age, but some studies suggest that migraine may alter this trajectory. For example, decreased age-related thinning was observed in the bilateral postcentral cortex, right fusiform, and right temporal pole in patients with episodic migraine (17). Another study reported that female patients with migraine showed no insular thinning with age compared to healthy controls (52). In young individuals (aged 6–18 years), reduced cortical thickness was observed in those over 12 years old in regions such as the superior and middle frontal gyri, pre- and post-central cortex, paracentral lobule, and posterior insula (49). These findings suggest that cortical changes in migraine may be dynamic and influenced by factors such as attack frequency, raising the possibility that preventive treatments could modulate these changes over time (Figure 1).
Cortical changes in migraine across phases and treatment impact.
In summary, studies investigating cortical thickness during the interictal phase reveal complex and sometimes contradictory findings. Increased cortical thickness in specific regions has been linked to migraine characteristics such as photosensitivity, while age-related changes in cortical thickness suggest that alterations may occur plastically rather than as fixed traits. Further research is needed to clarify the relationship between cortical thickness, migraine characteristics and potential treatment effects.
Imaging in patients with migraine with aura
Studies investigating cortical thickness in migraine with aura (MA) have often focused on the visual cortex. The earliest study in this field examined 12 individuals with MA with visual aura found bilateral increases in cortical thickness in the V3A and MT+ areas (53). However, the study did not report the time since the last visual aura episode, a factor that may influence cortical measurements.
Subsequent studies have reported mixed findings. One study comparing 166 individuals with MA (including visual, sensory, aphasia and motor aura) to 137 healthy controls found increased cortical thickness in the visual cortex, specifically in visual areas V2 and V3A (15). This finding suggests that structural brain alterations may occur in individuals with MA, potentially due to repeated episodes of cortical spreading depolarization (a mechanism underlying aura), which could lead to increased neuronal density in these regions. Alternatively, the increased cortical thickness might reflect a predisposition to aura attacks (15). However, this study reported a five-year gap between the average age of the cohort at the time of the scan (48 years) and the reported age at the last aura attack (43 years), which may influence the interpretation of findings.
By contrast, other studies have not identified consistent cortical thickness changes in the visual cortex. One study comparing 28 individuals with visual MA to 28 healthy controls reported no cortical thickness differences in V3A or MT+ (46). Similarly, a multicenter study involving 38 MA patients and 115 healthy controls found cortical thinning in areas V1 and V2 (23). Of note, the study did not specify the subtype aura experienced by patients with migraine.
Cortical thickness changes in regions beyond the occipital lobe have also been explored. One study of 20 individuals with side-locked visual MA reported increased cortical thickness in the contralateral inferior frontal gyrus compared to the headache side (14). Another study of 60 individuals comprising both visual and sensory MA patients found no differences in the visual cortex but noted slightly thinner somatosensory cortex compared to healthy controls (51).
A study investigating patients with pure visual migraine aura using SBM identified additional findings. Patients with pure visual aura exhibited increased cortical thickness in high-level visual-information-processing areas (e.g. lingual gyrus and Rolandic operculum), whereas those with complex auras (patients reporting in addition to visual aura, unilateral paresthesia and/or language symptoms) showed cortical thinning in temporal, frontal, insular, postcentral and visual areas (47). However, the study did not explore associations between these findings and the frequency or duration of aura episodes, which could influence structural changes.
In summary, studies of cortical thickness in MA suggest that structural changes may occur in both the visual cortex and other brain regions. While some studies report increased cortical thickness in areas associated with visual processing, others find cortical thinning or no differences. Variability in results may stem from differences in sample sizes, methodologies and statistical thresholds. Future large-scale studies are needed to better understand the relationship between cortical thickness changes in MA and their association with clinical symptoms.
Imaging during the headache phase
Only one study assessed cortical thickness changes in patients with migraine during the headache phase of migraine (8). The study scanned 15 individuals with migraine without aura during both ictal and interictal phase of spontaneous migraine attacks. The main findings were a reduction of volume and thickness of pain-related cortical areas, the precentral and calcarine cortices, as well as decreased cortical thickness of the temporal pole during attacks compared to the interictal state. The study did not find an association between the side of the migraine headache and reduced cortical thickness. However, it showed a trend toward reduction of cortical thickness ipsilaterally to the pain side during attacks and an increase contralaterally to the pain side. The mean time between the ictal and interictal scans was 30 days, with the shortest interval of 12 days. These findings suggest rapid and short-lasting cortical thickness changes that might be a consequence of hypoperfusion, cell shrinking or a fast plastic adaptation to non-specific stimuli (7). While the study did not include a healthy control group, the reduced cortical thickness might be a consequence of attacks since the thickness of the temporal pole and pericalcarine cortex are suggested to be involved in migraine pathophysiology (54,55).
In summary, the only study investigating cortical thickness changes during the headache phase of migraine suggests dynamic, reversible cortical alterations in pain-related areas. These findings highlight the potential for rapid cortical responses to migraine attacks, but further studies with larger cohorts and healthy control groups are needed to validate these results and better understand the underlying mechanisms.
Impact of treatment on cortical thickness
An important question is whether altered cortical thickness in patients with migraine can reverse with effective preventive treatment. If such changes were reversible, they might serve as a potential biomarker for treatment response. Notably, these alterations could occur even with treatments that primarily target peripheral mechanisms, such as botulinum toxin A or calcitonin gene-related peptide (CGRP) monoclonal antibodies because these therapies modify sensory input that eventually reaches the central nervous system (57).
So far, few studies have explored the treatment effect on cortical thickness (Table 2). One study reported that the posterior cingulate cortex became thinner after prophylactic treatment in 19 patients with migraine, compared to pre-treatment scans (25). The posterior cingulate cortex is involved in pain perception, and this finding suggests possible plastic improvement in pain-related cortical changes. However, the posterior cingulate cortex was not initially thicker in untreated patients with migraine compared to 19 matched healthy controls. Additionally, the study's small sample size and lack of detail about the types of preventive treatments used limit the generalizability of these findings. Further research, particularly with specific preventive medications like CGRP-targeted therapies, may provide clearer insights into these mechanisms.
Pharmacological preventive treatments for migraine.
aTreatment responders; btreatment non-responders; cchronic migraine with medication-overuse headache; dchronic migraine.
ACC = anterior cingulate cortex; AI = anterior insula; BoNT-A = botulinum neurotoxin type A; CT = cortical thickness; DMC = dorsomedial cortex; IPC = inferior parietal cortex; NS = not specified; ParsOp = pars opercularis; PCC = posterior cingulate cortex; PFC = prefrontal cortex; SFG = superior frontal gyrus; SSC = somatosensory cortex; STG = superior temporal gyrus.
A recent study examined cortical changes after three months of treatment with the anti-CGRP monoclonal antibody galcanezumab in 36 patients with migraine. Responders (≥50% reduction in monthly migraine days) showed decreased cortical thickness in regions including the somatosensory cortex and anterior cingulate cortex, whereas non-responders exhibited reductions in other areas, such as the dorsomedial cortex. These changes were interpreted as reflecting recovery from maladaptive neural activity, with reduced nociceptive input leading to neural remodeling. However, the study suggested that longer treatment durations might be necessary to achieve more substantial cortical normalization, indicating that cortical plasticity might require extended intervention periods before stabilizing (50,56).
Another potential application is using baseline cortical thickness differences to predict treatment response. One study examined this possibility in 23 patients with chronic migraine who underwent MRI before receiving botulinum toxin A as a preventive treatment (50). The study found that patients who later responded to botulinum toxin A had increased cortical thickness in regions including the right primary somatosensory cortex, anterior insula, left superior temporal gyrus and inferior parietal gyrus (pars opercularis) compared to non-responders. While these findings are preliminary, they highlight the potential for cortical thickness differences to serve as predictors of treatment response and warrant further investigation.
Another study assessed whether the baseline MRI scan could predict the response to topiramate in a series of 57 patients, among which 42 tolerated the drug and were further evaluated. Cortical thickness was not observed as response predictor, but grey matter area values in the left cuneus curvature, left superior temporal gyrus and right insula, as well as the right inferior parietal cortex volume, were associated with the probability of achieving a 50% response (59).
Furthermore, one study investigated the morphometric effect after nasal-bupivacaine sphenopalatine ganglion blockade in 12 patients with chronic migraine and medication-overuse headache (58). The study reported decreased thickness six weeks after twice weekly treatment in the left temporal pole and lateral occipital-temporal sulcus cortex.
In summary, studies suggest that preventive treatments, including CGRP monoclonal antibodies and botulinum toxin A, may induce changes in cortical thickness in patients with migraine. These changes could reflect neural plasticity and recovery from maladaptive neural activity, particularly in responders. Although cortical thickness changes in regions such as the somatosensory cortex, anterior cingulate cortex, and anterior insula have been observed in treatment responders, alternative MRI markers, including grey matter area, cortical surface area and subcortical volume, may provide more reliable biomarkers for predicting treatment efficacy. Baseline cortical thickness differences might also help predict treatment responses, although the variability across studies suggests that cortical thickness is unlikely to become a robust individual predictor. Nevertheless, these investigations provide valuable insights into migraine pathophysiology and hold potential for advancing treatment strategies.
Expert opinion on future directions
It is our opinion that cortical thickness has untapped potential that still needs to be explored. Studies investigating cortical thickness in migraine have reported varying and sometimes contradictory results, showing both increased and decreased cortical thickness in different brain areas. At present, no reliable cortical thickness biomarker exists for the migraine brain. A biomarker could provide significant benefits, including (i) a greater understanding of migraine pathogenesis; (ii) the ability to monitor treatment efficacy; and (iii) an index of the transition from episodic to chronic migraine.
Additionally, measuring clinical outcomes and identifying the mechanisms that drive migraine progression could facilitate the development of preventive therapeutic interventions.
Further research is necessary to enhance our understanding of cortical thickness in migraine pathogenesis and the effects of available treatments, further research is necessary. Future investigations should focus on the following aspects:
Baseline and event-based imaging
Individuals with migraine should be scanned during a migraine-free state, including the hours subsequent to the imaging acquisition, and re-examined during migraine development to better understand the pathophysiological events involved.
Large-scale studies
Studies with larger sample sizes are essential to ensure sufficient statistical power for detecting correlations between cortical thickness and clinical outcomes, such as migraine pain severity, frequency and duration.
Longitudinal follow-up studies
Longitudinal studies of treatment-naïve migraine cohorts should investigate changes in cortical thickness post-treatment, with a specific focus on the reversibility of cortical thickness to baseline levels compared to healthy controls, ensuring that the same imaging machines are employed and the acquisition scheme is consistent.
Biomarker identification
Longitudinal studies should aim to identify characteristics of cortical thickness changes associated with a good response to preventive treatment (e.g. pre-treatment prediction of treatment response). Comparing changes in responders versus non-responders could help identify biomarkers of treatment efficacy. Other causes of chronic pain should be controlled.
In summary, while cortical thickness is a valuable metric, other brain morphology parameters may offer additional or complementary insights into migraine pathophysiology and treatment response. Expert opinions in the field suggest that gray matter volume, cortical surface area, cortical curvature, subcortical structures and white matter structure should also be considered in biomarker research.
Conclusions and future perspectives
Cortical thickness studies reveal dynamic, phase-specific changes in migraine, with thinning and thickening observed interictally, ictal, and in migraine with aura. Treatment response studies suggest neuroplastic reversibility, highlighting cortical thickness as a potential biomarker. However, inconsistencies necessitate longitudinal, well-powered studies to clarify its role in migraine pathophysiology and therapeutic monitoring.
Clinical implications
Inconsistent increases and decreases in cortical thickness are observed in pain-related and visual processing regions.
Changes in cortical thickness may be reversed by preventive treatment suggesting neuroplastic remodeling.
Future standardized studies are needed to enhance our understanding of cortical thickness and identify a potential biomarker.
Footnotes
Author contributions
FMA initiated the study. All authors were responsible for drafting sections of the manuscript. All authors contributed with critical revision of the manuscript and approved the final manuscript.
Declaration of conflicting interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: SY, MMAK, NZ and RHC have no conflict of interest to report. BAC reports person fees from Pfizer, outside the submitted work. RM reports personal fees from AbbVie, Eli Lilly, Lundbeck, Pfizer, Teva and Biomedia, outside the submitted work. DGA has received honoraria for speaker/advisory board/consultancy/research support from Novartis, Teva, Lilly, Pfizer, AbbVie, Lundbeck and Organon. DGA has received grants from the International Headache Society, and competitive research funds from Carlos III Health Research Institute and Regional Health Administration, SACYL. DGA is a board member of the Spanish Society of Neurology and the European Union of Medical Specialist (UEMS). GC serves as Associate Editor for The Journal of Headache and Pain, Cephalalgia, Cephalalgia Reports, BMC Neurology (Pain section), Frontiers in Neurology (Neurotechnology section) and Frontiers in Human Neuroscience (Brain imaging and Stimulation section). PPR has received honoraria as a consultant and speaker from AbbVie, Amgen, Eli Lilly, Lundbeck, Medscape, Novartis, Pfizer and Teva Pharmaceuticals. PPR research group has received research grants from Instituto Salud Carlos III, EraNet Neuron, AbbVie, Novartis, Teva Pharmaceuticals, AGAUR, FEDER RIS3CAT, and has received funding for clinical trials from AbbVie, Amgen, Biohaven, Eli Lilly, Lundbeck, Novartis, Pfizer and Teva Pharmaceuticals. FMA has received Honoria or personal fees from Pfizer, Teva, Novartis, Lundbeck, AbbVie and Eli Lilly for lecturing or participating in advisory boards; is principal investigator for phase IV trials sponsored by Novartis and by Teva; serves as president of Danish Headache Society and board member of the European Headache Federation; serves as associate editor for Acta Neurologica Scandinavica, Frontiers in Neurology, Frontiers in Pain Research and Headache Medicine.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Faisal Mohammad Amin was supported by Lundbeck Foundation Ascending Investigator Grant (R396-2022-301). The funding source played no role in this study.
