Review Reports - Susceptibility Weighted Imaging as a Biomarker for Cortical Spreading Depression

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study explores the potential of susceptibility-weighted imaging (SWI) in MRI as a biomarker for cortical spreading depression (CSD), the phenomenon underlying migraine aura. Findings revealed a posterior-to-anterior gradient of SWI changes aligning with symptom progression, supporting SWI as a marker for CSD. Even regions without reported symptoms showed changes, suggesting silent CSD. SWI offers a practical and widely available imaging tool for understanding and diagnosing migraine aura. However, limitations include the retrospective design and variability in imaging protocols.

Manuscript is interesting and findings are novel. It's written well. A few minor comments:

1. Please include gender, race and age details in TableS1 if possible. Please describe if these are not relevant and why.

2. In discussion authors point to differences in histological composition of posterior regions i.e. atleast in monkey this region has more neurons. It will be useful to describe if there are other changes for example glial cell numbers, pericytes, or type of neurons etc .

3. And, if anything is known about neural circuitry, vascular systems that might contribute to this? This question arises because of discussion on oxyhemoglobin.

4. Please add more details in legends of Figure 1. Legends should clearly state what is the inference from these images.

5. Authors conclude MRI with SWI might be able to detect CSD. Please add a few lines on what more will be needed to completely establish this.

Author Response

Reviewer 1

“Manuscript is interesting and findings are novel. It's written well. “

Response: We thank the reviewer for the favorable evaluation.

“ Please include gender, race and age details in TableS1 if possible. Please describe if these are not relevant and why.”

Response: Unfortunately, we did not acquire a consent to publish of identifiable data from the included patients, and therefore we cannot publish further identificatory details due to local regulations.

“In discussion authors point to differences in histological composition of posterior regions i.e. atleast in monkey this region has more neurons. It will be useful to describe if there are other changes for example glial cell numbers, pericytes, or type of neurons etc “

Response: We agree with the reviewer that besides neurons, other cell types might play a role in the spreading of the CSD waves. We therefore added in the first part of the discussion the following: “ One hypothesis explaining the emergence of CSD in the posterior regions is likely related to the histological composition of the posterior regions. In human and monkey it was shown that the posterior regions have a higher density of neurons compared to the anterior regions. Since the CSD is a neuronal phenomenon that is inhibited by astroglial cells, which have a higher density in the anterior regions of the cortex, it is more likely that the CSD originates in the posterior regions.^{5,6,11,16,17,18} Furthermore, in females with migraine aura the cortical thickness in the occipital regions was higher than in females without migraine.¹⁹”

Rockel AJ, Hiorns RW, Powell TP. The basic uniformity in structure of the neocortex. Brain. 1980 Jun;103(2):221-44.

Collins CE, Airey DC, Young NA, Leitch DB, Kaas JH. Neuron densities vary across and within cortical areas in primates. Proc Natl Acad Sci U S A. 2010 Sep 7;107(36):15927-32.

Gaist D, Hougaard A, Garde E, et al. Migraine with visual aura associated with thicker visual cortex. Brain. 2018 Mar 1;141(3):776-785.

“3. And, if anything is known about neural circuitry, vascular systems that might contribute to this? This question arises because of discussion on oxyhemoglobin.”

Response: We are not aware of any specific aspects of neuronal circuitry specific to the occipital lobe relevant to the cortical spreading depression. However, there are peculiarities of the vasoreactivity in migraine patients compared to controls. We therefore added in the first part of the discussion the following text: “In addition, the vasomotor autoregulation was impaired in patients with migraine aura, possibly explaining the increase in deoxyhemoglobine during an attack.^20-30”

²⁰ Reinhard M, Schork J, Allignol A, Weiller C, Kaube H. Cerebellar and cerebral autoregulation in migraine. Stroke. 2012 Apr;43(4):987-93.

²¹Bäcker M, Sander D, Hammes MG, Funke D, Deppe M, Conrad B, Tölle TR. Altered cerebrovascular response pattern in interictal migraine during visual stimulation. Cephalalgia. 2001 Jun;21(5):611-6.

²²Lee MJ, Cho S, Woo SY, Chung CS. Paradoxical association between age and cerebrovascular reactivity in migraine: A cross-sectional study. J Neurol Sci. 2019 Mar 15;398:204-209.

²³Rosengarten B, Sperner J, Görgen-Pauly U, Kaps M. Cerebrovascular reactivity in adolescents with migraine and tension-type headache during headache-free interval and attack. Headache. 2003 May;43(5):458-63.

²⁴Jiménez Caballero PE, Muñoz Escudero F. Peripheral endothelial function and arterial stiffness in patients with chronic migraine: a case-control study. J Headache Pain. 2013 Jan 31;14(1):8.

²⁵Vernieri F, Moro L, Altamura C, Palazzo P, Antonelli Incalzi R, Rossini PM, Pedone C. Patients with migraine with aura have increased flow mediated dilation. BMC Neurol. 2010 Mar 10;10:18.

²⁶Lauritzen M. Long-lasting reduction of cortical blood flow of the brain after spreading depression with preserved autoregulation and impaired CO2 response. J Cereb Blood Flow Metab. 1984 Dec;4(4):546-54.

²⁷González-Quintanilla V, Toriello M, Palacio E, González-Gay MA, Castillo J, et al. Systemic and cerebral endothelial dysfunction in chronic migraine. A case-control study with an active comparator. Cephalalgia. 2016 May;36(6):552-60.

²⁸Marouf, W., Hetzel, A., Reinhard, M. et al. Cerebral ultrasound perfusion imaging in a migraine attack with prolonged aura.J Neurol 255, 599–600 (2008).

²⁹ Napoli R, Guardasole V, Zarra E, Matarazzo M, D'Anna C, Saccà F, Affuso F, Cittadini A, Carrieri PB, Saccà L. Vascular smooth muscle cell dysfunction in patients with migraine. Neurology. 2009 Jun 16;72(24):2111-4.

³⁰ Vernieri F, Tibuzzi F, Pasqualetti P, Altamura C, Palazzo P, Rossini PM, Silvestrini M. Increased cerebral vasomotor reactivity in migraine with aura: an autoregulation disorder? A transcranial Doppler and near-infrared spectroscopy study. Cephalalgia. 2008 Jul;28(7):689-95.

“4. Please add more details in legends of Figure 1. Legends should clearly state what is the inference from these images. “

Response: Point well taken. We now added details to the figure in the manuscript.

“5. Authors conclude MRI with SWI might be able to detect CSD. Please add a few lines on what more will be needed to completely establish this.”

Response: Given the limitations of our study, especially the retrospective design and the low number of included patients, we propose that SWI should used in further prospective larger studies using MRI with a higher field strength, to validate our findings. We now added this to the manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors

Susceptibility Weighted Imaging as a Biomarker for Cortical Spreading Depression

The paper by Scutelnic et al. explores the use of susceptibility weighted imaging (SWI) in magnetic resonance imaging (MRI) as a potential biomarker for detecting cortical spreading depression (CSD), hypothesized to be the pathophysiological correlate of migraine aura. The authors aimed to assess whether SWI changes in patients with migraine aura correspond to the clinical presentation, focusing on the topological distribution of these changes.

Authors found a posterior to anterior gradient of SWI changes during acute migraine aura, especially when visual symptoms were present. Moreover, all patients with visual disturbances showed SWI changes in the posterior region and the distribution of SWI changes corresponded anatomically to the typical progression of symptoms in migraine aura. Some regions showed PFV despite patients not reporting corresponding symptoms, suggesting possible silent CSD or assessment bias. Based on their findings, authors conclude that MRI with SWI might be useful in detecting traces of CSD, supporting its potential as an imaging marker for migraine aura.

While the study is conducted with rigor and limitations are addressed, in my opinion some changes must be made before considering publication.

· Although all the necessary concepts are there, some are taken for granted and not discussed enough, assuming that the reader has all the necessary notions to be prepared to understand the manuscript.

· The Method section is well written and deeply explain how the study was carried out. I would suggest some changes:

o Lines 55-56: the sentence “Excluded were patients not fulfilling the inclusion criteria” is in my opinion redundant. If patients do not meet inclusion criteria they cannot be enrolled; exclusion criteria should give different reasons for why patients cannot be considered.

o Line 65: it’s not clear if and how the research group collaborated: “a board certified neuroradiologist (NS)” carried out all the MRI evaluations but multiple “experienced board-certified neurologists” was in charge of migraine diagnosis. Did the neurologists helped the neuroradiologist?

o The “NS” abbreviation in line 65 doesn’t seem useful: I didn’t find it again the manuscript.

· Results: although the tables seem comprehensive, more detailed explanations of the results in the text would make for a better understanding. For example, in line 102 authors talk about “the 12 patients with speech disturbance” without a previous explaining of their presence.

· The discussion section is well arranged in my opinion. However, it’s peculiar in my opinion that 11 out of 50 lines (22%) are dedicated to the study limitations.

o Adding to the previous comment, authors didn’t mention the little sample size (18 patients) as a potential limit. The small sample size could be even more confounding considering all the other limits.

· The reference list is very short (only 15 papers are cited) and most of them seem quite old (6, 40%, are more than five years old). I would suggest authors to improve and updated the reference section if possible.

Overall, the hypothesis made by the authors sound very interesting. However, in my opinion the manuscript should be revised and refined before acceptance.

Comments for author File: Comments.pdf

Author Response

Reviewer 2

“Although all the necessary concepts are there, some are taken for granted and not discussed enough, assuming that the reader has all the necessary notions to be prepared to understand the manuscript.”

Response: We thank the reviewer for this remark. We attempted to keep the introductory part as succinct as possible. However, we now did expand the discussion on the possible reasons as to why the CSD originates in the occipital lobe and added multiple references, many of which are more recent: “ One hypothesis explaining the emergence of CSD in the posterior regions is likely related to the histological composition of the posterior regions. In human and monkey it was shown that the posterior regions have a higher density of neurons compared to the anterior regions. Since the CSD is a neuronal phenomenon that is inhibited by astroglial cells, which have a higher density in the anterior regions of the cortex, it is more likely that the CSD originates in the posterior regions.^{5,6,11,16,17,18} Furthermore, in females with migraine aura the cortical thickness in the occipital regions was higher than in females without migraine.¹⁹“

Rockel AJ, Hiorns RW, Powell TP. The basic uniformity in structure of the neocortex. Brain. 1980 Jun;103(2):221-44.

Collins CE, Airey DC, Young NA, Leitch DB, Kaas JH. Neuron densities vary across and within cortical areas in primates. Proc Natl Acad Sci U S A. 2010 Sep 7;107(36):15927-32.

Gaist D, Hougaard A, Garde E, et al. Migraine with visual aura associated with thicker visual cortex. Brain. 2018 Mar 1;141(3):776-785.

“In addition, the vasomotor autoregulation was impaired in patients with migraine aura, possibly explaining the increase in deoxyhemoglobine during an attack.^20-30”

²⁰ Reinhard M, Schork J, Allignol A, Weiller C, Kaube H. Cerebellar and cerebral autoregulation in migraine. Stroke. 2012 Apr;43(4):987-93.

²²Lee MJ, Cho S, Woo SY, Chung CS. Paradoxical association between age and cerebrovascular reactivity in migraine: A cross-sectional study. J Neurol Sci. 2019 Mar 15;398:204-209.

²⁴Jiménez Caballero PE, Muñoz Escudero F. Peripheral endothelial function and arterial stiffness in patients with chronic migraine: a case-control study. J Headache Pain. 2013 Jan 31;14(1):8.

²⁵Vernieri F, Moro L, Altamura C, Palazzo P, Antonelli Incalzi R, Rossini PM, Pedone C. Patients with migraine with aura have increased flow mediated dilation. BMC Neurol. 2010 Mar 10;10:18.

²⁸Marouf, W., Hetzel, A., Reinhard, M. et al. Cerebral ultrasound perfusion imaging in a migraine attack with prolonged aura.J Neurol 255, 599–600 (2008).

“The Method section is well written and deeply explain how the study was carried out.”

Response: We thank the reviewer for this comment.

“I would suggest some changes: Lines 55-56: the sentence “Excluded were patients not fulfilling the inclusion criteria” is in my opinion redundant. If patients do not meet inclusion criteria they cannot be enrolled; exclusion criteria should give dierent reasons for why patients cannot be considered.”

Response: As requested, we deleted that part.

“Line 65: it’s not clear if and how the research group collaborated: “a board certified neuroradiologist (NS)” carried out all the MRI evaluations but multiple “experienced board-certified neurologists” was in charge of migraine diagnosis. Did the neurologists helped the neuroradiologist?”

Response: We thank the reviewer for pointing this out. Since this is a retrospective study based on patients who were investigated and treated at the emergency department, multiple neurologists were involved. Indeed, at the time of the MRI examination, also multiple neuroradiologists evaluated the MRI images in the emergency setting. Because the primary aim in the emergency setting was to exclude stroke, little or no consideration was given to SWI changes. Therefore, at a later timepoint, one neuroradiologist (Dr. Nedelina Slavova, “NS”) evaluated the MRI images for SWI changes. This new evaluation of the MRI images was done for study purposes only. We added these aspects to the method section: “The MRI scans were specifically evaluated for SWI changes at a later timepoint than the presentation at the emergency department.”

“The “NS” abbreviation in line 65 doesn’t seem useful: I didn’t find it again the manuscript.”

Response: Thank you for this remark. We respectfully believe that the specification “NS” is important, given that the evaluation of the MRI images is a central part of our paper. We must declare which neuroradiologist (i.e. Dr. Nedelina Slavova, “NS”) assessed the MRI images for SWI changes. Therefore, if the reviewer and editor agree, we would like to maintain it in the manuscript.

“Results: although the tables seem comprehensive, more detailed explanations of the results in the text would make for a better understanding. For example, in line 102 authors talk about “the 12 patients with speech disturbance” without a previous explaining of their presence.”

Response: Point well taken. We now added the following sentence to the results section: “Overall, 18/18 (100%) patients had visual symptoms, 10/18 (56%) sensory symptoms, 12/18 (67%) speech disturbance, and 3/18 (17%) motor symptoms.”

“The discussion section is well arranged in my opinion. However, it’s peculiar in my opinion that 11 out of 50 lines (22%) are dedicated to the study limitations.”

Response: We agree that we discuss the limitations of our study extensively, which are inherent to a hypothesis generating study as ours. We believe, that despite promising results, the interpretation must be made with caution in the context of possible limitations. However, we now expanded our discussion by adding possible explanations on why the CSD begins posteriorly (see our comment above).

“Adding to the previous comment, authors didn’t mention the little sample size (18 patients) as a potential limit. The small sample size could be even more confounding considering all the other limits.”

Response: We agree. We now added this as an additional limitation.

“The reference list is very short (only 15 papers are cited) and most of them seem quite old (6, 40%, are more than five years old). I would suggest authors to improve and updated the reference section if possible.”

Response: We thank the reviewer for this observation. We now expanded the reference list to 30 papers, many of which are more recently published.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors,

your manuscript talks about a topic of scientific interest with potential daily clinical implications.

I ask you to answer the following questions:

- In the introduction, could you better specify the role of oxyhemoglobin in migraine attacks?

- In the materials and methods, could you describe whether in the inclusion criteria patients were strictly subjected to MRI within a short onset of symptoms?

- In the materials and methods, could you specify whether among the exclusion criteria there were patients with particular pre-existing pathological conditions?

- In the discussion could you specify the role that this neuroimaging can have in clinical practice from the point of view of the patient's prognosis? Can an early neuroradiological diagnosis somehow influence the therapeutic choice by facilitating the management of symptoms?

Author Response

Reviewer 3

“Dear Authors,

your manuscript talks about a topic of scientific interest with potential daily clinical implications.

I ask you to answer the following questions:

- In the introduction, could you better specify the role of oxyhemoglobin in migraine attacks?”

Response: We thank the reviewer for raising this point. We believe that the increase of oxyhemoglobin is a consequence of the CSD. We are not aware of any active role of oxyhemoglobin during a migraine attack with aura. However, little is known about this topic, and further research is necessary to address this question. We now added to the introductory part: “However, whether oxyhemoglobin has a specific role during a migraine attack or is simply a consequence of CSD is not known.”

“- In the materials and methods, could you describe whether in the inclusion criteria patients were strictly subjected to MRI within a short onset of symptoms?”

Response: We included patients presenting at the emergency department due to acute neurological symptoms. There was therefore a selection of patients with emergency MRI done shortly after begin of focal neurological symptoms. The maximum time frame from symptom onset to imaging was chosen to be eight hours, due to previous data showing persistent SWI changes eight hours after symptom onset. Because we already highlight these aspects in the methods section, we would prefer, if the reviewer and the editor agree, not to change the manuscript in this regard.

“- In the materials and methods, could you specify whether among the exclusion criteria there were patients with particular pre-existing pathological conditions?”

Response: We thank you for pointing this out. We did not exclude patients with other pre-existing pathological conditions. The co-morbidities are given in a supplementary table. We added: “Patients with other co-morbidities (eg psychiatric, vascular risk factors) were not excluded.”

“- In the discussion could you specify the role that this neuroimaging can have in clinical practice from the point of view of the patient's prognosis? Can an early neuroradiological diagnosis somehow influence the therapeutic choice by facilitating the management of symptoms?”

Response: We thank the reviewer for raising this important point. Since we did not perform a case-control study by including patients with other differential diagnoses, we cannot assess the usefulness of SWI changes in differentiating between a migraine aura and other diagnoses. However, we believe that in a specific clinical context, the SWI changes might help in reinforcing the clinical suspicion of migraine aura. Further studies are needed with a case-control design to address this important clinical question. We added these aspects to the discussion: “The value of SWI changes in differentiating between a migraine aura and other differential diagnoses (e.g. transient ischemic attacks) could not be addressed in this study. Therefore future case-control adequately powered studies are necessary to address this question.”

Reviewer 4 Report

Comments and Suggestions for Authors

The application of susceptibility-weighted imaging (SWI) as a biomarker for cortical spreading depression (CSD) is an area of growing research interest.

The authors present a single-center retrospective cohort study to have a double-check on how CSD can be detected, which has been igniting research in the field of migraine with aura for so many years.

Apart from the selection of patients, the methodology and technique which are unexceptionable I find the conclusions astounding: the results point to a definite marker of CSD in aura. In fact they found a posterior to anterior gradient of SWI changes during acute migraine aura when visual symptoms were present.

This corroborates this technique as a confirmatory method of aura.

I would just like to point out that the reported literature lacks some references that need to be discussed:

doi: 10.1186/s10194-023-01704-z.

doi: 10.1186/s10194-024-01889-x.

doi: 10.1186/s10194-024-01827-x.

This research finally underscores SWI as potential to uncover biomarkers for cortical spreading depression in aura migraine. Further work is needed to validate findings.

Author Response

“The application of susceptibility-weighted imaging (SWI) as a biomarker for cortical spreading depression (CSD) is an area of growing research interest.

This corroborates this technique as a confirmatory method of aura.”

Response: We thank the reviewer for these remarks.

“I would just like to point out that the reported literature lacks some references that need to be discussed:

doi: 10.1186/s10194-023-01704-z.

doi: 10.1186/s10194-024-01889-x.

doi: 10.1186/s10194-024-01827-x.”

Response: As requested, we now briefly discuss the references in the manuscript: “In contrast to more sophisticated laboratory³¹ but also imaging methods, such as BOLD and machine learning algorithms, MRI-SWI is widely available.^32,33”

³¹Dalkara T, Kaya Z, Erdener ŞE. Unraveling the interplay of neuroinflammatory signaling between parenchymal and meningeal cells in migraine headache. J Headache Pain. 2024 Jul 31;25(1):124.

³²Mitrović K, Savić AM, Radojičić A, Daković M, Petrušić I. Machine learning approach for Migraine Aura Complexity Score prediction based on magnetic resonance imaging data. J Headache Pain. 2023 Dec 18;24(1):169.

³³Raggi A, Leonardi M, Arruda M, et al. Hallmarks of primary headache: part 1 - migraine. J Headache Pain. 2024 Oct 31;25(1):189.

“This research finally underscores SWI as potential to uncover biomarkers for cortical spreading depression in aura migraine. Further work is needed to validate findings.”