Quantitative pupillometry as a marker of autonomic dysregulation in vestibular migraine

2.1 Participants

The study cohort initially comprised 74 participants, among whom 43 patients were diagnosed with VM according to the ICHD-3 criteria [5], and 31 healthy controls without a history of migraine or vestibular disorders. Three participants in the study group and one in the control group were excluded because of inadequate cooperation during vestibular testing. Thus, the final analysis consisted of 70 participants, comprising 40 patients with VM and 30 healthy controls.

The VM cohort consisted of patients with episodic VM diagnosed according to the ICHD-3 criteria. Among these patients, 4 (10.0%) had migraine with aura, whereas 36 (90.0%) had migraine without aura. The attack frequency ranged from 4 to 6 episodes per month, with a mean ± standard deviation (SD) of 4.9 ± 0.7 episodes per month. None of the participants met the diagnostic criteria for chronic migraine. These clinical characteristics were considered when interpreting interictal pupillometric responses.

The exclusion criteria included ocular disease, neurological or metabolic disorders, use of medications that could affect pupillary or autonomic function, and abnormal vestibular test results. All participants underwent a comprehensive otoneurological examination to exclude alternative causes of vertigo.

This study was approved by the Bakircay University Non-Interventional Clinical Research Ethics Committee (Approval No. 2116), and written informed consent was obtained from all participants at the time of hospitalization.

2.2 Inclusion criteria

Participants were eligible for inclusion if they met the following criteria: (1) age between 18 and 65 years; (2) normal ophthalmologic findings, including best-corrected visual acuity and normal anterior and posterior segment examinations; (3) a normal otoneurological evaluation, with no evidence of peripheral or central vestibular pathology other than VM; (4) for the VM group, fulfillment of the ICHD-3, diagnostic criteria for VM; (5) for the control group, no history of migraine, vertigo, or vestibular disorders; (6) no use of medications that could affect autonomic, neurological, or pupillary responses; and (7) provided written informed consent and could complete the full pupillometric protocol. Participant recruitment and data collection were conducted between October 2025 and November 2025.

All patients with VM were evaluated during the interictal phase, which was defined as the absence of migraine or vestibular symptoms for at least 72 hours before testing. This criterion was applied to minimize the influence of acute autonomic changes during attacks on pupillometric measurements. Nevertheless, although participants were assessed during an interictal period defined in this way, it cannot be completely excluded that some individuals were evaluated during the premonitory phase of migraine, which may precede headache onset by up to 48 hours [29].

Although the inclusion criteria allowed participants aged 18–65 years, the final sample included individuals within a narrower age range.

2.3 Ophthalmic examination

All participants underwent a comprehensive ophthalmic examination, which included best-corrected visual acuity assessment, slit-lamp biomicroscopy, and evaluation of both the anterior and posterior segments.

2.4 Pupillometric procedure

Non-contact pupillometric measurements were obtained using an automated quantitative pupillometry system (MonPack One, Metrovision, Pérenchies, France), and all examinations were performed by the same experienced clinician (Fig. 1).

Fig. 1.Representative pupillometry recordings obtained under different illumination conditions. (A) scotopic, (B) mesopic, (C) low photopic, and (D) high photopic illumination. (E) Representative pupillary light reflex response curve. BI: Binocular; OD: Oculus Dexter; OS: Oculus Sinister.

Bilateral pupillary measurements were recorded using an infrared pupillometer under five standardized illumination conditions: scotopic, mesopic, low photopic, high photopic, and resting light. Before testing, all subjects underwent dark adaptation in order to standardize baseline pupil diameter.

For each condition, dynamic pupillary parameters were recorded, including constriction amplitude, latency, duration, and velocity, as well as dilation latency, duration, and velocity. For statistical analysis, measurements obtained from both eyes were averaged.

The five illumination conditions were defined as follows: scotopic testing was performed under complete darkness to assess pupil responses when rod photoreceptors are predominantly active; mesopic testing evaluated pupil behavior under intermediate lighting conditions, in which both rod and cone photoreceptors contribute; low-photopic testing measured pupil responses under low-light conditions dominated mainly by cone activity; high-photopic testing examined pupillary constriction under bright illumination, when cone photoreceptors are fully activated; and resting light conditions were used to determine the natural baseline pupil diameter in the absence of a direct light stimulus.

2.5 Statistical analysis

The statistical analysis of the data was performed using the Statistical Package for the Social Sciences (SPSS) version 27 (IBM Corp., Armonk, NY, USA). Descriptive statistics were expressed as frequency, percentage, mean, SD, median, and minimum–maximum values. The relationships between categorical variables were analyzed using the Pearson chi-square test.

The normality of continuous variables was assessed using the Shapiro-Wilk test. Skewness and kurtosis values were also examined, and data distribution was visually evaluated using graphical methods. According to Tabachnick and Fidell, skewness and kurtosis coefficients between −1.5 and +1.5 are considered acceptable for normal distribution [30]. In the present study, all variables met this criterion.

For comparisons between groups, the independent-samples t-test was used for normally distributed variables, whereas the Mann-Whitney U test was used for variables that were not normally distributed. Within-group comparisons were performed using the paired-samples t-test or the Wilcoxon signed-rank test, as appropriate. A p-value of < 0.05 was considered statistically significant in all analyses.

The required sample size for the study was determined using G*Power (version 3.1.9.7, G*Power, Heinrich Heine University Düsseldorf, Düsseldorf, NRW, Germany). The effect size was estimated from previously published reference studies and calculated as Cohen’s d = 0.86. According to the power analysis, with a significance level of α = 0.05, a statistical power of 90% (1 − β), and an effect size of d = 0.86, a minimum of 30 participants per group was required. Therefore, the study was planned with a minimum sample size of 60 participants.

In addition to p-values, effect sizes were calculated to quantify the magnitude of the observed differences. Cohen’s d was used for comparisons between independent groups. Effect sizes were interpreted according to the thresholds recommended for pain research, defined as small (0.10), medium (0.30), and large (0.70) [31].

4.1 Pupillometric and autonomic findings

In the present study, no significant interictal differences in pupillometric parameters were observed between patients with VM and healthy controls. However, autonomic dysfunction in migraine is known to be variable and phase-dependent [32, 33], and interictal pupillometry may have limited sensitivity for detecting subtle central alterations. Consistent with this interpretation, the observed differences were associated with predominantly small effect sizes, suggesting that they are unlikely to reflect clinically meaningful autonomic alterations in VM during the interictal phase.

The photopic pupillary asymmetries observed in some participants should, therefore, be interpreted with caution. Although mild anisocoria may fall within the range of physiological variability [34], persistent asymmetries may still suggest the possibility of asymmetric central autonomic modulation. Nevertheless, pupillometry does not allow localization of neural activity to specific brainstem nuclei [35], and therefore, our findings cannot be attributed to lateralized activity in structures such as the locus coeruleus or the periaqueductal gray.

Taken together, these findings suggest that interictal pupillometry provides limited, but potentially complementary, information regarding autonomic regulation in VM. Its role as a standalone biomarker remains uncertain. Future studies integrating multimodal autonomic assessments with functional neuroimaging may help determine whether these subtle pupillary variations reflect central network-level modulation rather than peripheral influences or physiological variability.

4.2 Light-dependent modulation and brainstem mechanisms

In bright light, patients with VM exhibited a noticeable right-to-left asymmetry in pupil responses. In a similar context, Zou et al. [36] reported that patients with VM showed increased photophobia and greater sensitivity to visual triggers, both of which were significantly associated with more severe vertigo symptoms.

Their study further demonstrated that photophobia during VM attacks was positively correlated with interictal photosensitivity and visually triggered dizziness, thereby highlighting the contribution of visual hypersensitivity to VM and supporting the need for individualized management and preventive strategies [36].

This pattern of modulation likely reflects transient, centrally mediated hypersensitivity within the Locus Coeruleus–Periaqueductal Gray Mater (LC–PAG) network rather than fixed structural autonomic dysfunction. The increased reactivity observed under bright light may indicate that VM is associated with a functional, stimulus-dependent imbalance in sympathetic–parasympathetic integration, which is consistent with the broader concept of central sensory hypersensitivity in migraine. In addition, pupil dynamics may serve as an indirect marker of LC-mediated coordination between cognitive and autonomic activity across distributed neural networks [37].

4.3 Comparison with previous studies and broader interpretation

While previous studies have reported subtle interictal autonomic or pupillary asymmetries in certain migraine subtypes, our findings indicate only minimal interictal differences in patients with VM, suggesting that pupillometric changes may be episodic, stimulus-dependent, or subtype-specific rather than consistently detectable across all migraine phenotypes.

Kavuncu et al. [38] reported significant inter-eye differences in scotopic pupil diameter, together with shortened constriction latency in patients with migraine with aura, suggesting a shift in pupillary balance toward parasympathetic predominance.

Similarly, Harle et al. [39] demonstrated interictal interocular differences in pupillary latency in patients with migraine, indicating that subtle autonomic asymmetry may persist even between attacks.

Against this background, our findings suggest that the light-dependent and lateralized pupillary asymmetry observed in VM may reflect a residual but adaptive form of central autonomic modulation rather than fixed structural impairment. This interpretation is consistent with the framework of LC–PAG-mediated hypersensitivity within brainstem autonomic circuits and further supports the concept of brainstem-centered functional dysregulation in VM. These findings suggest that VM is not characterized by fixed or degenerative autonomic deficits, but rather by reversible and context-dependent modulation of central autonomic control. The subtle and light-dependent asymmetry identified in the present study may therefore represent a physiological expression of the brainstem’s dynamic regulation of sensory and autonomic processes. In this context, pupillometry may serve as a useful and noninvasive approach for detecting phase-dependent autonomic alterations and for exploring brainstem network involvement in VM.

Together, these pupillometric findings further highlight the possible contribution of brainstem autonomic nuclei and their thalamocortical connections to the pathophysiology of VM. To better interpret these results, it is important to consider how autonomic dysregulation interacts with vestibular, trigeminal, and cortical circuits in producing the complex clinical manifestations of VM.

4.4 Autonomic dysregulation in vestibular migraine

The ANS maintains physiological homeostasis through a dynamic balance between sympathetic and parasympathetic activity, and disruption of this equilibrium in migraine may result in sympathetic hyperactivity or parasympathetic hypoactivity, particularly during attacks [40].

Autonomic regulation within interconnected vestibular–trigeminal–brainstem networks represents an important modulatory component of VM pathophysiology [41, 42]. While the ANS contributes to clinical manifestations such as nausea, vertigo, and light sensitivity, current evidence suggests that these changes reflect phase-dependent modulation of central sensory networks rather than primary structural autonomic dysfunction [40].

Functional imaging studies have demonstrated altered connectivity among the hypothalamus, thalamus, brainstem nuclei, and vestibular cortex, regions that collectively mediate sensory and autonomic integration. This network-level dysregulation likely contributes to symptoms such as dizziness, nausea, cardiovascular instability, and photophobia commonly observed in VM [40].

At the brainstem level, reciprocal connections among the vestibular nuclei, nucleus tractus solitarius, dorsal motor nucleus of the vagus, and locus coeruleus form a vestibulo-autonomic network that enables bidirectional modulation between vestibular input and autonomic output. Activation of the trigeminovascular system, together with the release of vasoactive neuropeptides such as calcitonin gene-related peptide (CGRP) and substance P, induces neurogenic inflammation and vasodilation [14], which may transiently influence vestibulo-ocular reflex control and autonomic tone, thereby contributing to clinical symptoms such as vertigo, motion intolerance, and lightheadedness. Beyond these autonomic circuits, the vestibular nuclei themselves may undergo sensitization under the influence of migraine-related brainstem regions, with additional modulation provided by inhibitory cerebellar feedback from the nodulus and uvula, where canal–otolith integration occurs [43].

Although CGRP and substance P are well-established mediators in migraine pathophysiology, their direct effects on vestibulo-ocular reflex (VOR) control have not been experimentally established. CGRP primarily modulates nociceptive transmission and vascular tone [22]. Accordingly, our findings do not support a direct mechanistic link between these neuropeptides and VOR disruption.

The thalamus acts as a central relay that integrates vestibular and nociceptive inputs. Its higher-order nuclei, including the posterior, lateral posterior, and lateral dorsal nuclei, project to cortical regions involved in multisensory processing [26, 44]. Functional imaging studies have demonstrated increased thalamic activation during both vestibular stimulation and migraine episodes [26, 45], supporting the concept of dynamic, phase-dependent central sensory dysregulation rather than fixed structural sensitization.

At the cortical level, patients with migraine exhibit increased excitability, which may be influenced by modulatory input from the locus coeruleus and dorsal raphe nuclei [46, 47]. In addition, abnormal cortical interactions between visual and vestibular networks have been demonstrated in VM, contributing to heightened motion sensitivity and increased susceptibility to motion sickness, even during interictal periods [48, 49].

Collectively, these findings indicate that VM reflects multilevel central dysregulation involving brainstem autonomic nuclei, thalamic relay structures, and cortical sensory networks. Taken together, these mechanisms support the interpretation of VM as a functional and dynamic disorder characterized by transient hypersensitivity within the vestibulo-autonomic axis rather than structural damage. This integrated framework also links the pupillometric evidence of brainstem asymmetry with broader processes of central sensitization across vestibular and nociceptive networks.

4.5 Comparison with previous research and physiological interpretation

Earlier studies in migraine without vestibular involvement have demonstrated reduced baseline pupil diameter and shortened constriction latency, reflecting an accelerated pupillary light response, findings that have been interpreted as reflecting relative parasympathetic predominance [38].

In contrast, our results are consistent with those of Gufoni and Casani [50], who reported no significant interictal pupillary asymmetry, suggesting that autonomic dysfunction in VM may be episodic and phase-dependent. The stronger intragroup correlations observed in VM patients may further indicate central reorganization within the vestibulo-autonomic network rather than peripheral impairment, implying that pupillary changes may represent adaptive responses to fluctuating cortical excitability.

Previous pupillometric studies in migraine have yielded heterogeneous findings depending on the subtype and timing of assessment. Harle et al. [39] demonstrated increased interocular differences in pupillary light reflex latency and a mild imbalance between sympathetic and parasympathetic activity that persisted during the interictal phase, suggesting subtle autonomic asymmetry even outside headache periods. Kavuncu et al. [38], in contrast, reported a shift in pupillary balance toward parasympathetic predominance in patients with migraine with aura, reflected by prolonged constriction latency and reduced constriction velocity.

In our present study, no overall interictal differences were observed in static or dynamic pupillary parameters between patients with VM and healthy controls; however, a significant right-to-left asymmetry was detected under high-photopic illumination. This pattern differs from the more generalized parasympathetic predominance described in migraine with aura and instead suggests a stimulus-dependent, lateralized modulation of central autonomic pathways, particularly within midbrain networks, such as the locus coeruleus and periaqueductal gray, which are involved in light sensitivity and pain processing.

Taken together, these findings suggest that migraine-related autonomic dysfunction is functional and context-dependent rather than structural, with distinct patterns across migraine subtypes. Whereas migraine with aura may be associated with a more sustained parasympathetic bias, VM appears to involve dynamic, light-dependent asymmetry consistent with transient, centrally mediated hypersensitivity within the vestibulo-autonomic network. Compared with previous clinical studies that assessed pupil diameter under limited conditions, the present study evaluated both static and dynamic pupillary responses across multiple illumination levels, which represents a methodological strength and allows for a more comprehensive characterization of autonomic function and light-dependent modulation in VM.

Our present study also investigated interictal pupillary dynamics in patients with VM and compared them with those of healthy controls. The absence of significant differences in baseline pupillometric measures suggests that interictal autonomic regulation may appear largely preserved; however, autonomic dysfunction in migraine is known to be variable, phase-dependent, and non-linear, and interictal pupillometry may, therefore, have limited sensitivity for detecting subtle central disturbances [32, 33]. Accordingly, the absence of statistically significant differences should not be interpreted as definitive evidence of normal autonomic function in all patients with VM.

Interictal pupillometry may have limited sensitivity for detecting subtle disturbances of central autonomic regulation. Although minor pupillary asymmetries were observed, these findings may reflect physiological variability rather than true pathological alterations. At present, the clinical utility of pupillometry as a diagnostic tool or early biomarker in VM remains uncertain, particularly because data on its sensitivity, specificity, and reproducibility in clinical settings are still lacking. Therefore, it would be premature to regard pupillometry as a reliable method for identifying early markers of the disorder. Future studies incorporating larger sample sizes, standardized protocols, and multimodal autonomic and neurophysiological assessments are needed to clarify its potential value and to further elucidate the complex network-level mechanisms underlying VM. Collectively, these findings support the concept that VM is primarily a disorder of dynamic network dysfunction rather than fixed structural autonomic impairment.

4.6 Clinical implications

From a clinical perspective, these findings suggest that pupillometry may serve as a non-invasive and objective tool for evaluating autonomic function in VM. Although no persistent interictal abnormalities were identified, the subtle, light-dependent asymmetries observed in patients with VM may indicate context-related alterations in central autonomic regulation. Assessment of pupillary responses under different illumination conditions may, therefore, contribute to a better understanding of the autonomic mechanisms associated with vestibular and migraine symptoms. Future studies should incorporate ictal-phase evaluations, heart rate variability, and functional neuroimaging in order to clarify the temporal dynamics of autonomic activity in VM and to further assess the potential diagnostic and prognostic relevance of pupillometry.