Differences in Static and Dynamic Resting-State Functional Connectivity between Migraineurs with and without Photophobia, without Phonophobia or Osmophobia

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Migraine, the second leading cause of disability, predominantly affects women, and its underlying neural mechanisms are complex and heterogeneous. While there is evidence linking migraine to photophobia, the circuit mechanisms behind this association remain unknown. The manuscript titled "Functional Connectivity Specific to Photophobia in Migraineurs: A Static and Dynamic Resting-State Functional Magnetic Resonance Imaging Study" employs fMRI brain imaging techniques to investigate neural activity in patients experiencing only photophobia. With a sample of 30 patients, the authors scanned various cortical and cerebellar areas to identify potential brain regions associated with photophobia in migraineurs. The authors identified 10 connectivity pairs in patients with photophobia under static resting state conditions and six under dynamic resting-state conditions. Additionally, they found low connectivity between the cerebellar and temporal regions in patients with photophobia.

 While this paper represents an important contribution to our understanding of the circuit mechanisms underlying migraine with this unique sensory modality, several improvements are necessary before publication:

 

 Introduction: Provide an introduction explaining how fMRI imaging of brain activity can be translated into functional connectivity. Define static resting state and dynamic state and highlight the differences. This will help readers who are not experts in functional connectivity to understand the rationale behind the experimental design.

 

Additional methodological information: Include a detailed description of how the T score was calculated.

Figure annotation: Provide anatomical brain region annotations for better understanding of connectivity. Explain the color code used in the figures. Additionally, include a color-coded scale bar to visualize the differences between different connectivity pairs.

Method section clarification: In line 54, it is stated that all participants are women, while in the results section (line 101), it is mentioned that 73% of patients were female. Please double-check and ensure consistency in reporting.

Author Response

Reviewer 1

 

General comments:

Migraine, the second leading cause of disability, predominantly affects women, and its underlying neural mechanisms are complex and heterogeneous. While there is evidence linking migraine to photophobia, the circuit mechanisms behind this association remain unknown. The manuscript titled "Functional Connectivity Specific to Photophobia in Migraineurs: A Static and Dynamic Resting-State Functional Magnetic Resonance Imaging Study" employs fMRI brain imaging techniques to investigate neural activity in patients experiencing only photophobia. With a sample of 30 patients, the authors scanned various cortical and cerebellar areas to identify potential brain regions associated with photophobia in migraineurs. The authors identified 10 connectivity pairs in patients with photophobia under static resting state conditions and six under dynamic resting-state conditions. Additionally, they found low connectivity between the cerebellar and temporal regions in patients with photophobia.

 

 While this paper represents an important contribution to our understanding of the circuit mechanisms underlying migraine with this unique sensory modality, several improvements are necessary before publication:

 

 

Comment 1:

 Introduction: Provide an introduction explaining how fMRI imaging of brain activity can be translated into functional connectivity. Define static resting state and dynamic state and highlight the differences. This will help readers who are not experts in functional connectivity to understand the rationale behind the experimental design.

Response: Thank you for your helpful comments. We have now added the following information regarding fMRI to the Introduction section.

“Functional connectivity describes the temporal correlation between spatially separated neurophysiological events, often measured using techniques such as functional magnetic resonance imaging (fMRI), electroencephalography (EEG), or magnetoencephalography (MEG) [3, 4]. Functional connectivity using MRI is often assessed by examining the synchronization of blood-oxygen level-dependent signals in different brain regions [3, 4]. There are two types of functional connectivity: static and dynamic [5]. Static connectivity assesses average connectivity patterns over a fixed period. Functional connectivity was assumed to be stable during the observation period. Conversely, dynamic connectivity examines how connectivity patterns change over time, recognizing that functional connectivity may fluctuate with cognitive state or task.”

 

Comment 2

Additional methodological information: Include a detailed description of how the T score was calculated.

Response: Thank you for pointing this out. As you suggested, I have now included references to previous studies and have added a brief description of how to calculate the T-values.

“In brief, we describe the T value of the T-value. Static functional connectivity analysis was conducted through ROI-to-ROI analysis using the CONN toolbox [5, 6]. In this approach, the target voxel BOLD time series S(x, t) was replaced by the target ROI time series Rj(t). The resulting ROI-to-ROI correlation matrices illustrate the functional connectivity levels between each ROI pair. The ROI-to-ROI correlation was determined using the Fisher-transformed bivariate correlation coefficient between the BOLD time series of the two ROIs. Dynamic independent component analyses were performed for the dynamic FC using the CONN toolbox [5, 6]. These analyses investigated the temporal modulation properties of the ROI-to-ROI connectivity matrix to uncover circuits with similar modulated connections. Dynamic independent component analysis (ICA) matrices quantified the expression of different modulatory circuits and the connectivity change rate between ROI pairs, as indicated by the connectivity strength and sign variations co-varying with a specific component/circuit time series. Group-level modulatory components Gamma_l(i, j) were computed through several steps. First, a simplified generalized context-dependent psychophysiological interaction (gPPI) model was used to estimate the group-level modulatory components Gamma_l(i,j). These components were rotated using fastICA with a hyperbolic tangent contrast function. The ICA mixing matrix W was inverted to derive the dynamic IC/circuit time series. Finally, the group-level modulatory components were back-projected onto the subject-level components gamma_nk(i,j) via a series of standard first-level gPPI models, incorporating the estimated dynamic independent component/circuit time series h(t) as gPPI psychological factors.”

 

 

Comment 3:

Figure annotation: Provide anatomical brain region annotations for better understanding of connectivity. Explain the color code used in the figures. Additionally, include a color-coded scale bar to visualize the differences between different connectivity pairs.

Response: Thank you for your valuable comment. We have now enlarged the figure to better display the connectivity and have further added a list of abbreviations for the anatomical brain regions to the description. Due to software limitations, we are unable to provide color-coded scale bars to visualize the differences between different connectivity pairs; however, the meaning of the red and blue coloring has now been described.

 

Comment 4:

Method section clarification: In line 54, it is stated that all participants are women, while in the results section (line 101), it is mentioned that 73% of patients were female. Please double-check and ensure consistency in reporting.

Reply: Thank you for pointing out that "all participants are women" in line 51 was incorrect; this mistake has now been corrected.

Reviewer 2 Report

Comments and Suggestions for Authors

In this paper, the authors explored differences in brain connectivity between migraine patients with photophobia (sensitivity to light) and those without photophobia, phonophobia (sensitivity to sound), or osmophobia (sensitivity to smells). Using 3T functional MRI during the interictal phase the authors analyzed both static and dynamic resting-state functional connectivity across various brain regions.The authors identified ten connectivity differences in static resting-state functional connectivity in patients with photophobia and six differences in dynamic resting-state functional connectivity. Both analyses indicated that migraineurs with photophobia had significantly lower connectivity between the cerebellar hemisphere and the temporal region compared to those without photophobia. The authors concluded that reduced connectivity between the cerebellar hemisphere and the temporal region is specific to migraine patients with photophobia.

 

To improve the manuscript, please address these following points.

 

Ensure that potential confounding variables, such as medication use and duration of migraine history, are controlled or accounted for in the analysis.

 

Comparing the findings more explicitly with existing literature will help highlight how this study extends or challenges previous research.

 

Discuss the theoretical implications of reduced connectivity between the cerebellar hemisphere and the temporal region in the context of migraine pathophysiology and photophobia mechanisms.

 

Discussion on limitations should be expanded. For instance, elaborate on how the small sample size might have impacted the results and the specific challenges in recruiting during the ictal phase. Conducting a power analysis to determine the appropriate sample size needed to detect significant effects will be useful and including this information in the discussion section will strengthen the manuscript.

Author Response

Reviewer 2

 

General comments:

In this paper, the authors explored differences in brain connectivity between migraine patients with photophobia (sensitivity to light) and those without photophobia, phonophobia (sensitivity to sound), or osmophobia (sensitivity to smells). Using 3T functional MRI during the interictal phase the authors analyzed both static and dynamic resting-state functional connectivity across various brain regions. The authors identified ten connectivity differences in static resting-state functional connectivity in patients with photophobia and six differences in dynamic resting-state functional connectivity. Both analyses indicated that migraineurs with photophobia had significantly lower connectivity between the cerebellar hemisphere and the temporal region compared to those without photophobia. The authors concluded that reduced connectivity between the cerebellar hemisphere and the temporal region is specific to migraine patients with photophobia.

 

To improve the manuscript, please address these following points.

 

 

Comment 1:

Ensure that potential confounding variables, such as medication use and duration of migraine history, are controlled or accounted for in the analysis.

Response: The potential confounding variables you mentioned (history of MOH, duration of illness, and BMI) were not significantly different between the two groups. This information has now been added to Section 3.1.

 

Comment 2:

Comparing the findings more explicitly with existing literature will help highlight how this study extends or challenges previous research.

Response: Thank you for pointing this out. The following text has now been added to the Discussion section:

“Static resting-state functional connectivity analysis revealed 18 significant connectivity pairs in patients with photophobia, mainly involving the cerebellar hemispheres and regions such as the temporal occipital fusiform cortex. Dynamic resting-state analysis further revealed 16 significant connectivity pairs, primarily between the cerebellar hemisphere and other brain regions [5].”

 

Comment 3:

Discuss the theoretical implications of reduced connectivity between the cerebellar hemisphere and the temporal region in the context of migraine pathophysiology and photophobia mechanisms.

Reply: In the original manuscript, we discussed the cerebellum and migraine pathophysiology and photophobia mechanisms, but not those of the temporal region. The following text has now been added to address this:

“Migraineurs with photophobia exhibited reduced functional connectivity between the cerebellar hemispheres and temporal lobes, including the middle temporal gyrus for static and temporal fusiform cortex for dynamic connectivity. The middle temporal gyrus can be divided into four regions: anterior (aMTG), middle (mMTG), posterior (pMTG), and sulcus (sMTG). The aMTG is predominantly associated with the default mode network, sound perception, and semantic retrieval; the mMTG is mainly associated with semantic memory and semantic control networks; the pMTG is part of the traditional sensory linguistic areas; and the sMTG is associated with gaze direction decoding and intelligible speech [18]. The temporal fusiform cortex is considered a key structure for functionally specialized computations of high-level vision such as face perception, object recognition, and reading [19]. Reduced functional connectivity of the temporal fusiform cortex may cause photophobia due to functionally specialized computations of high-level vision. Regarding the altered functional connectivity of middle temporal gyrus, it is not clear which functions are directly related to photosensitivity and may be an indirect effect.”

 

 

 

Comment 4:

Discussion on limitations should be expanded. For instance, elaborate on how the small sample size might have impacted the results and the specific challenges in recruiting during the ictal phase. Conducting a power analysis to determine the appropriate sample size needed to detect significant effects will be useful and including this information in the discussion section will strengthen the manuscript.

Response: A discussion on how the small sample size affected the results, how the sample size was determined, and an explanation of the specific challenges in recruitment during the seizure period has been added/revised.

“These limitations unfortunately diminish the statistical robustness of the study, frequently necessitating suboptimal methods to establish significance and constraining the broader applicability of the findings [4]. This sample size limitation has also been observed in other functional MRI studies [10, 12, 14].”

“Additionally, no method has yet been established to determine the appropriate sample size required for fMRI studies; however, this may be resolved in the future when it becomes possible to model different sample size scenarios using computer simulations and assess the statistical power in each scenario.”

" It is difficult to predict precisely when a migraine attack will occur. Therefore, previous studies captured images for 30 consecutive days to obtain data during the attack period. However, scanning a large number of patients for 30 consecutive days is challenging, and finding a solution is not expected to be easy."

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have addressed all the comments.