Structural brain changes in patients with persistent headache after COVID-19 resolution

Vol.:(0123456789) 1 3 Jou nal o Neu ology h ps://doi.o g/10.1007/s00415-022-11398-z ORIGINAL COMMUNICATION S uc u al b ain changes inpa ien s wi hpe sis en headache a e COVID‑19 esolu ion Ál a oPlanchuelo‑Gómez1,2 · Da idGa cía‑Azo ín3,4 · ÁngelL.Gue e o3,4 · Ma ga i aRod íguez5 · San iagoAja‑Fe nández1 · Rod igodeLuis‑Ga cía1 Recei ed: 10 Augus 2022 / Re ised: 21 Sep embe 2022 / Accep ed: 22 Sep embe 2022 © The Au ho (s) 2022 Abs ac Headache is among he mos equen ly epo ed symp oms a e esolu ion o COVID-19. We assessed s uc u al b ain changes using T1- and di usion-weigh ed MRI p ocessed da a om 167 subjec s: 40 pa ien s who eco e ed om COVID- 19 bu su e ed om pe sis en headache wi hou p io his o y o headache (COV), 41 heal hy con ols, 43 pa ien s wi h episodic mig aine and 43 pa ien s wi h ch onic mig aine. To e alua e g ay ma e and whi e ma e changes, mo phome y pa ame e s and di usion enso imaging-based measu es we e employed, espec i ely. COV pa ien s showed signi ican lowe co ical g ay ma e olume and co ical hickness han heal hy subjec s (p < 0.05, alse disco e y a e co ec ed) in he in e io on al and he usi o m co ex. Lowe ac ional aniso opy and highe adial di usi i y (p < 0.05, amily-wise e o co ec ed) we e obse ed in COV pa ien s compa ed o con ols, mainly in he co pus callosum and le hemisphe e. COV pa ien s showed highe co ical olume and hickness han mig aine pa ien s in he cingula e and on al gy i, pa acen al lobule and supe io empo al sulcus, lowe olume in subco ical egions and lowe cu a u e in he p ecuneus and cuneus. Lowe di usion me ic alues in COV pa ien s compa ed o mig aine we e iden i ied p ominen ly in he igh hemisphe e. COV pa ien s p esen di e se changes in he whi e ma e and g ay ma e s uc u e. Whi e ma e changes seem o be associ- a ed wi h impai men o ibe bundles. Besides, he g ay ma e changes and o he whi e ma e modi ica ions such as axonal in eg i y loss seemed sub le and less p onounced han hose de ec ed in mig aine, showing ha pe sis en headache a e COVID-19 esolu ion could be an in e media e s a e be ween no mali y and mig aine. Keywo ds COVID-19· Headache· G ay ma e · Di usion enso imaging· Mig aine * Da id Ga cía-Azo ín [email p o ec ed] 1 Labo a o io de P ocesado de Imagen (LPI), Uni e sidad de Valladolid, 47011Valladolid, Spain 2 Ca di Uni e si y B ain Resea ch Imaging Cen e (CUBRIC), Ca di Uni e si y, Ca di CF244HQ, UK 3 Depa men o Neu ology, Headache Uni , Hospi al Clínico Uni e si a io de Valladolid, A enida Ramón y Cajal, 3, 47003Valladolid, Spain 4 Depa men o Medicine, Uni e sidad de Valladolid, 47005Valladolid, Spain 5 Depa men o Radiology, Hospi al Clínico Uni e si a io de Valladolid, 47003Valladolid, Spain Abb e ia ions AD Axial di usi i y ANCOVA Analysis o co a iance CM Ch onic mig aine COVID-19 Co ona i us disease 2019 COV G oup o pa ien s wi h pe sis en headache a e COVID-19 esolu ion DTI Di usion enso imaging DWI Di usion-weigh ed imaging EM Episodic mig aine FA F ac ional aniso opy FDR False disco e y a e FEW Family wise e o GM G ay ma e HC Heal hy con ols ICHD-3 In e na ional Classi ica ion o Headache Diso de s, 3 d edi ion MD Mean di usi i y MNI Mon eal Neu ological Ins i u e RD Radial di usi i y RT-PCR Re e se- ansc ip ase-polyme ase-chain- eac ion TBSS T ac -based spa ial s a is ics TE Echo ime Jou nal o Neu ology 1 3 TFCE Th eshold- ee clus e enhancemen TI In e sion ime TR Repe i ion ime WM Whi e ma e Backg ound Co ona i us disease 2019 (COVID-19) p esen s wi h a combina ion o espi a o y and sys emic symp oms, wi h headache being one o he mos p ominen ea u es. Head- ache is desc ibed by app oxima ely a qua e o pa ien s [1, 2], wi h a pheno ype ha combines ea u es o ension- ype headache and mig aine [3]. The p esence o headache has been associa ed wi h a mo e e icien immune esponse [4, 5] and a be e sho -and-mid- e m p ognosis [6–8]. The median du a ion o headache is a ound 2weeks [2], bu in a i h o pa ien s, headache becomes pe sis en and adop s a ch onic pa e n, pa icula ly i i pe sis s 2mon hs a e he acu e phase [9]. In o de o in es iga e on he biological unde pinnings o pe sis en headache a e COVID-19 esolu ion, magne ic esonance imaging (MRI) e eals i sel as an ideal echnique o assess he s uc u al changes in he b ain due o i s associ- a ed good con as o issue and non-in asi eness. MRI da a p ocessing allows o analyze quan i a i e p ope ies o he g ay ma e (GM) and whi e ma e (WM). To assess GM changes, mo phome y analysis om T1-weigh ed images allows he p ecise quan i ica ion o ea u es such as he ol- ume, co ical cu a u e, hickness and su ace a ea o di - e en co ical and subco ical GM a eas [10]. In headache diso de s, olume has been widely assessed [11], while a ea, cu a u e and hickness ha e been ba ely employed [12, 13]. On he o he hand, di usion MRI acquisi ions allow he quan i a i e desc ip ion o he WM p ope ies. Di u- sion enso imaging (DTI) is commonly employed o model he di usion MRI da a, and whi e ma e desc ip o s can be es ima ed om i . F ac ional aniso opy (FA), mean di usi - i y (MD), axial di usi i y (AD), and adial di usi i y (RD) a e common choices. These me ics ha e been ex ensi ely used o s udy he p esence o b ain changes in pa ien s wi h ch onic headache diso de s, especially mig aine [14–18]. A ew s udies ha e assessed b ain changes a e COVID- 19 esolu ion using MRI, none o hem ocused on pe sis en headache. In a s udy ca ied ou 3mon hs a e COVID-19 eco e y, pa ien s p esen ed highe global GM olume, and, wi h ega d o he WM, highe FA and lowe MD, RD, and AD in compa ison wi h heal hy con ols [19]. These ends e lec con a y pa e ns o wha usually has been associ- a ed wi h a ophy o de e io a ion. In con as , a la ge s udy wi h he UK Biobank composed o hund eds o longi udinal acquisi ions o heal hy con ols and pa ien s wi h p e ious COVID-19 in ec ion has epo ed lowe co ical hickness in he pa ien g oups [20]. In ano he s udy wi h a 1-yea ol- low-up, eco e ed pa ien s p esen ed a lowe olume ac ion o in acellula wa e han heal hy con ols, and dec eased FA was ound in pa ien s who we e admi ed o an in en- si e ca e uni compa ed o hose who we e no admi ed [21]. Finally, in a s udy assessing pa ien s who eco e ed om COVID-19 accompanied by anosmia o hyposmia, he au ho s ound lowe FA and highe RD in eco e ed pa ien s [22]. To da e, he e a e no s udies assessing s uc u al changes in pa ien s wi h pe sis en headache a e COVID-19 eso- lu ion (he eina e e e ed o as COV g oup). The main objec i e o his s udy is he e alua ion and cha ac e iza- ion o long- e m b ain s uc u al changes in COV pa ien s. In addi ion, a seconda y objec i e is he compa ison o hese changes wi h hose disco e ed in mig aine, he headache dis- o de ha has been mos ex ensi ely s udied using MRI da a. Ma e ials andme hods Pa icipan s An obse a ional s udy wi h a case–con ol design, nes ed in a p ospec i e coho [2, 9], was conduc ed o assess he po en ial s uc u al b ain changes o COV pa ien s. One he one hand, a g oup composed o heal hy con ols was used as a e e ence o a case–con ol s udy design. On he o he hand, o be e cha ac e ize he changes associa ed wi h headache, he COV pa ien s we e compa ed wi h a g oup composed o episodic mig aine (EM), and ano he g oup composed o ch onic mig aine (CM) pa ien s. Fo he COV pa ien s, he inclusion c i e ia we e adap ed om acu e-phase s udies [2]: (1) mic obiologically con- i med COVID-19 diagnosis based on a eal- ime e e se- ansc ip ase-polyme ase-chain- eac ion (RT-PCR) assay om espi a o y ac samples, o al e na i ely by he p es- ence o an i-SARS-CoV-2 IgM + IgA an ibodies, ollowing he Wo ld Heal h O ganiza ion p o ocols [23, 24]; (2) new- onse headache p esen ed du ing he acu e phase o COVID- 19, ul illing c i e ia o acu e headache a ibu ed o sys- emic i al in ec ion [25], ha was no be e accoun ed o ano he seconda y headache diso de acco ding o he In e - na ional Classi ica ion o Headache Diso de s, 3 d edi ion (ICHD-3) [25]; (3) age g ea e o equal o 18yea s old; and also he ollowing speci ic inclusion c i e ia: (4) pe sis ence o headache o a leas 3mon hs a e he acu e phase o COVID-19; (5) no esolu ion o he headache a he momen o MRI acquisi ion; (6) ag eemen o pa icipa e; (7) lack o use o d ugs wi h po en ial e ec on he cen al ne ous sys em (be e speci ied in exclusion c i e ia). The s udy was conduc ed in he Headache Uni o Hospi al Clínico Uni e si a io de Valladolid (Valladolid, Jou nal o Neu ology 1 3 Spain), a hi d-le el uni e si y public hospi al. All con- secu i e pa ien s ha we e in ec ed om COVID-19 in Valladolid Eas heal h a ea be ween Ma ch 8, 2020, and Ap il 11, 2020, we e sc eened (n = 580 hospi alized pa ien s and n = 1614 cases managed in an ou pa ien se - ing, ou a popula ion a isk o 162,431 inhabi an s) [2] and p ospec i ely ollowed up o a leas 9mon hs [9]. All hese pa ien s and hose ha we e e e ed o he head- ache ou pa ien clinic we e assessed o eligibili y. Fo he COV pa ien s included in he sample, di e se clinical ea u es o headache and he p esence o anosmia we e collec ed. All he pa icipan s we e aged be ween 18 and 60yea s. Rega ding he speci ic inclusion and exclusion c i e ia o he o he h ee g oups in his s udy (heal hy con ols, EM and CM), hey a e a ailable elsewhe e [18]. B ie ly, he diagnosis o pa ien s wi h EM and CM was based on he ICHD-3 c i e ia [25], and hese pa ien s p esen ed a s a- ble clinical si ua ion wi h a leas 1yea su e ing om mig aine. The exclusion c i e ia we e common in all ou g oups and we e: (1) dea h du ing ollow-up; (2) una ail- abili y o pa icipa e due o uns able medical condi ion; (3) p io his o y o cogni i e impai men o demen ia; (4) speech o language diso de s ha made headache e alu- a ion impossible; (5) consen wi hd awal o pa icipa e in he s udy; (6) use o he apeu ic o illici d ugs wi h po en ial e ec s on he ne ous sys em, including an i- dep essan s, ba bi u a es, neu olep ics, an iepilep ics, and opia es; (7) p io his o y o o he p ima y o seconda y headache diso de s, excluding in equen ension- ype headache (less han one a ack pe mon h); (8) p e ious his o y o mode a e- o-se e e c anio-ce ical auma; (9) p io his o y o neu ological o neu osu gical diso de s o he han COVID-19 (in he COV g oup); (10) su e ing om o he pain ul synd omes, o neu ological o psy- chia ic diso de s; (11) p egnancy o childbea ing; (12) condi ions ha con aindica ed an MRI acquisi ion; (13) claus ophobia; (14) p esence o unexpec ed ascula mal o ma ion in MRI ha could a ec he mo phologi- cal e alua ion; (15) any clinical condi ion ha a oided an accu a e desc ip ion o he headache pheno ype. Bo h mig aine pa ien s and COV pa ien s we e excluded i hey had used any p io p e en i e ea men be o e MRI acquisi ion. The pa ien s wi h mig aine we e ec ui ed a e hei i s isi o he a o emen ioned Headache Uni and, in case o p esc ibed p e en i e ea men , i began jus a e he MRI acquisi ion in a pe iod sho e han 1mon h. Da a and imaging om pa ien s wi h mig aine and con ols we e ob ained be o e he begin- ning o he COVID-19 pandemic. The E hics Re iew Boa d o Valladolid Eas heal h a ea app o ed his s udy (PI-GR-20-2017). MRI acquisi ion T1- and di usion-weigh ed images (DWI) we e collec ed o he ou g oups o subjec s. Fo he iden i ica ion o pos- sible abno mali ies, T2-weigh ed and FLAIR images we e addi ionally acqui ed om he COV g oup. All pa ien s we e scanned in he same MRI scanne , a Philips Achie a 3T MRI uni (Philips Heal hca e, Bes , The Ne he lands) wi h a 32-channel head coil. T1-weigh ed images we e acqui ed using a Tu bo Field Echo sequence wi h he ollowing pa ame e s: epe i ion ime (TR) = 8.1ms, echo ime (TE) = 3.7ms, lip angle = 8º, 256 × 256 ma ix size, spa ial esolu ion o 1 × 1 × 1 mm3 and 160 sagi al slices co e ing he whole b ain. Fo he di usion-weigh ed da a, TR = 9000 ms, TE = 86ms, lip angle = 90º, 61 di usion g adien o ien a- ions, one baseline olume, b- alue = 1000s/mm2, 128 × 128 ma ix size, spa ial esolu ion o 2 × 2 × 2 mm3 and 66 axial slices co e ing he whole b ain we e he employed pa ame e s. The pa ame e s o he acquisi ion o he T2-weigh ed images we e: Tu bo Spin Echo sequence, TR = 3000ms, TE = 80ms, lip angle = 90º, 560 × 560 ma ix size, spa ial esolu ion o 0.43 × 0.43 × 5 mm3, and 28 axial slices. Finally, 3D high- esolu ion FLAIR images we e ob ained using TR = 4800ms, TE = 308ms, in e sion ime (TI) = 1650ms, lip angle = 90º, 240 × 240 ma ix size, spa- ial esolu ion o 0.56 × 1.04 × 1.04 mm3, and 321 sagi al slices o co e he whole b ain. Image p ocessing The analysis was ocused on he assessmen o he GM and WM s uc u e. In he ollowing sec ions, he co esponding p ocessing s eps ollowed o he sepa a e analysis o each ype o issue a e desc ibed. G ay ma e mo phome y The au oma ic co ical pa cella ion pipeline om he F ee- Su e so wa e ( 6.0.0) was applied o he T1-weigh ed images. The s eps om his p ocedu e included he seg- men a ion o co ical and subco ical GM s uc u es and he calcula ion o geome ical desc ip o s o u he analysis. Speci ically, he p e ious p ocessing s eps o he pipeline we e skull s ipping, Talai ach ans o ma ion, segmen a ion o subco ical g ay and whi e ma e , including he bound- a y essella ion o bo h issue ypes, in ensi y no maliza ion, and su ace de o ma ion [26–29]. The pipeline was used o ex ac he mean co ical cu a u e, hickness, su ace a ea and GM olume o 68 (34 bila e al) co ical egions. Jou nal o Neu ology 1 3 In addi ion, GM olume was compu ed o se en bila e al egions and he bila e al ce ebellum (16 egions). This yields a o al o 288 mo phome ic ea u es o he GM. Di usion MRI p ocessing Fi s , DWIs we e p ep ocessed. The p ep ocessing s eps we e denoising ollowing he Ma chenko-Pas u P incipal Componen Analysis p ocedu e, Gibbs inging emo al, co - ec ion o eddy cu en s, mo ion and B1 ield inhomogenei- ies. These s eps we e ca ied ou wi h he “dwidenoise”, “m degibbs”, “dwi slp ep oc” and “dwibiasco ec ” ools om MR ix ( e sion 3.0.2) [30–36]. Once he p ep ocess- ing was inished, a whole b ain mask excluding he skull was ex ac ed using he “dwi2mask” ool om MR ix [37]. A e p ep ocessing, di usion enso i ing was pe - o med on he DWIs using leas squa es, and ac ional ani- so opy (FA), mean di usi i y (MD), and axial di usi i y (AD), we e ob ained as WM desc ip o s using he “d i i ” ool om FSL so wa e sui e ( e sion 5.0.9) [38]. An addi- ional desc ip o , he adial di usi i y (RD) was also calcu- la ed as he a e age o he second and hi d eigen alues om he di usion enso , also ob ained wi h “d i i ”. Fo he pos e io s a is ical analysis, he a o emen ioned di usion pa ame e s we e compa ed in di e se WM ac s. Speci ically, ac -based spa ial s a is ics (TBSS) was used as a p ocedu e o de ine a WM skele on whe e he alues om he di usion pa ame e s a e assessed [39]. B ie ly, he TBSS me hod consis s o a non-linea egis a ion o he FA images o a empla e in he Mon eal Neu ological Ins i u e (MNI) space wi h he FNIRT and FLIRT ools om FSL. The egis e ed images we e a e aged, and a mean FA skel- e on o he WM ac s was gene a ed using a FA alue o 0.2 as h eshold o dis inguish WM om GM. The TBSS p o- cess was epea ed o he non-FA pa ame e s using he FA egis a ion as e e ence. To iden i y speci ic WM egions, he JHU ICBM-DTI-81 Whi e-Ma e A las was used [40, 41]. Mo eo e , he minimum olume pe egion o conside s a is ically signi ican esul s was 30 mm3. S a is ical analysis The main objec i e o he analysis was he compa ison o he pa ame e s o he COV pa ien s wi h heal hy con ols (HC). To be e cha ac e ize he p ope ies o he COV g oup, hese pa ien s we e compa ed agains EM, and CM. The de ailed compa isons o GM and WM di usion desc ip o s be ween he wo mig aine g oups and HC is a ailable elsewhe e [12, 18, 42]. Two-by- wo g oup compa isons o age and sex we e ca - ied ou using he Mann–Whi ney U es and Fishe ’s exac es , espec i ely. Wi h ega d o he compa ison o he GM mo phome y pa ame e s be ween he COV and HC g oups, da a we e es ed o no mal dis ibu ion wi h he Shapi o–Wilk es . I his assump ion was no me , a Mann–Whi ney U es was applied. Fu he mo e, he homogenei y o a iance assump- ion was es ed by he Le ene es . I no mali y assump ion was me , a - es was used o compa e he alues be ween wo g oups, conside ing equal o di e en a iance depending on he esul s o he Le ene es . An analysis o co a iance (ANCOVA) was always used o compa e GM olume alues conside ing in ac anial olume as co a ia e o no in e es . Fo any seconda y compa ison wi h he EM o CM g oups, an ANCOVA was employed including age as co a ia e o no in e es . As seconda y assessmen , sex was also added as co a ia e o no in e es o he ANCOVA. In addi ion, o he main compa isons be ween COV and HC, esul s we e also co ec ed o age and sex ollowing he same ANCOVA p o- cedu e conduc ed in he compa ison be ween COV and he mig aine g oups. The Benjamini–Hochbe g False Disco e y Ra e (FDR) p ocedu e was applied o co ec he esul s o mul iple compa isons [43]. The h eshold o s a is ical sig- ni icance was se a p < 0.05. Rega ding he TBSS analysis o he di usion desc ip- o s, he “ andomise” ool om FSL wi h he h eshold- ee clus e enhancemen (TFCE) op ion was employed o de e - mine he di e ences be ween he g oups o in e es [44, 45]. B ie ly, his ool is used o pe o m a pe mu a ion es . We employed 5000 pe mu a ions and es ablished he s a is ical h eshold o s a is ical signi icance a p < 0.05 a e applying a amily-wise e o (FWE) co ec ion o mul iple compa i- sons. Fo he compa isons be ween he COV g oup and he EM and CM g oups, conside ing ha pa ien s om he i s g oup a e, on a e age, olde han pa ien s wi h mig aine, esul s we e age-co ec ed. As a seconda y analysis, in he compa ison be ween COV and he o he h ee g oups, esul s we e addi ionally co ec ed o sex. In addi ion, o assess he magni ude o he iden i ied di - e ences o he mo phome y and di usion pa ame e s, he Cohen’s d was compu ed o he compa isons wi h s a is i- cally signi ican di e ences. The alues employed o assess each pa ame e and speci ic egion we e hose included in he Desikan–Killiany a las o mo phome y, and he skel- e on and a ea o each egion acco ding o he JHU ICBM- DTI-81 a las o di usion pa ame e s. The pooled s anda d de ia ion was compu ed as he squa ed oo o he addi ion o he a iance o each assessed g oup mul iplied by he pe inen sample size minus one di ided by he o al sample size o he wo g oups minus wo. The inal Cohen’s d alue was he di e ence be ween he mean alue o he wo g oups di ided by he pooled s anda d de ia ion. To assess he ela ionship be ween clinical pa ame e s o COV pa ien s and g ay and whi e ma e desc ip o s, Spea man’s ank co ela ion coe icien was employed. The Jou nal o Neu ology 1 3 clinical a iables we e he disease du a ion and he head- ache equency. The alues o he g ay ma e mo phome y pa ame e s we e he same as hose used o he s a is ical compa isons be ween g oups. Fo he whi e ma e desc ip- o s, a ROI-based app oach was ca ied ou . Indi idual label maps o each subjec we e ex ac ed applying he in e se wa p ields o he egis a ion o he FA images desc ibed o he TBSS p ocedu e o he JHU ICBM-DTI-81 a las. Fo he g ay ma e and whi e ma e esul s, only he egions and pa ame e s wi h s a is ically signi ican di e ences be ween COV pa ien s and any o he g oup we e conside ed. Resul s In o al, 66 COV pa ien s we e ini ially ec ui ed. The inal sample was composed o 42 COV pa ien s and 43 HC bal- anced o age and sex. Median ime be ween onse o head- ache a ibu ed o COVID-19 and MRI acquisi ion in COV pa ien s was 10mon hs ( ange 3–20mon hs). In addi ion, 43 pa ien s wi h EM, and 43 pa ien s wi h CM we e conside ed. Two pa ien s o each o he i s wo g oups we e disca ded o he di usion MRI analysis due o egis a ion e o s. The s udy lowcha de ailing he sample a each s age is shown in Fig.1. The wo-by- wo compa isons o he demog aphic cha ac e is ics be ween he COV g oup and he o he h ee g oups a e a ailable in Table1. Fu he in o ma ion abou speci ic clinical cha ac e is ics ela ed o headache in COV pa ien s, i.e., pheno ype, loca ion, mean in ensi y, quali y, and symp oms du ing headache, oge he wi h he p esence o anosmia, is a ailable in Table2. G ay ma e mo phome y pa ame e s Compa ison be weenCOV andHC Compa ed o HC, COV pa ien s showed s a is ically sig- ni ican lowe GM olume in he bila e al pa s o bi alis (co ec ed p = 0.029 le ; co ec ed p = 0.033 igh ), and he igh usi o m gy us (co ec ed p = 0.033) and on al pole (co ec ed p = 0.033). Mo eo e , pa ien s also p e- sen ed lowe co ical hickness han HC in he igh pa s o bi alis (co ec ed p = 0.008). No s a is ically signi ican di e ences we e ound in co ical cu a u e o su ace Fig. 1 Flowcha o he s udy. The numbe o pa ien s wi h pe sis en headache a e COVID-19 esolu ion and heal hy con ols is di e en in he g ay ma e and whi e ma e assessmen s due o egis a ion e o s wi h he di usion images o una ailable di usion MRI acquisi ions Jou nal o Neu ology 1 3 a ea. The loca ion o hese indings is g aphically shown in Fig.2. A e he co ec ion o age and sex, hese esul s we e s a is ically signi ican , and, addi ionally, lowe co - ical hickness alues we e ound in COV compa ed o HC in he le os al an e io cingula e gy us (co ec ed p = 0.024). Compa ison be weenCOV andmig aine Rega ding he di e ences be ween COV and EM pa ien s a e co ec ing o age, signi ican ly highe co ical hick- ness (co ec ed p = 0.012) in he le pa acen al co ex, and lowe subco ical olume alues in he le accumbens (co ec ed p = 0.022) and he igh halamus (co ec ed Table 1 Clinical and demog aphic cha ac e is ics o pa ien s wi h pe sis en headache a e COVID-19 esolu ion, heal hy con ols (HC), epi- sodic mig aine (EM) and ch onic mig aine (CM) pa ien s † Fishe ’s exac es . ‡Mann–Whi ney U es . §Two- ailed, unpai ed S uden ’s - es (Welch es ). 1. COVID-19 headache s. HC. 2. COVID-19 headache s. EM. 3. COVID-19 headache s. CM. Da a a e exp essed as means ± SD. The pheno ype o he does no adhe e o mig aine o TTH ollowing ICHD-3 guidelines. TTH = ension- ype-headache COVID-19 headache (n = 42) HC (n = 43) EM (n = 43) CM (n = 43) S a is ical es Gende , male/ emale 11/31 (26/74%) 11/32 (26/74%) 9/34 (21/79%) 6/37 (14/86%) 1. p = 1† 2. p = 0.62† 3. p = 0.18† Age (yea s) 43.8 ± 10.2 41.8 ± 10.2 40.1 ± 6.4 41.3 ± 7.1 1. U = 1015, p = 0.33‡ 2. = 2.00, p = 0.049§ 3. U = 1110, p = 0.070‡ Disease du a ion 10.1 ± 3.4mon hs 16.0 ± 11.4yea 21.9 ± 10.2yea Headache equency (days/mon h) 30 ± 0 3.5 ± 1.9 22.8 ± 6.7 Di usion MRI sample n = 40 n = 41 n = 43 n = 43 Gende , male/ emale 11/29 (28/72%) 11/30 (27/73%) 9/34 (21/79%) 6/37 (14/86%) 1. p = 1† 2. p = 0.61† 3. p = 0.17† Age (yea s) 43.7 ± 10.3 41.0 ± 9.6 40.1 ± 6.4 41.3 ± 7.1 1. U = 959, p = 0.19‡ 2. = 1.90, p = 0.062§ 3. U = 1050, p = 0.084‡ Disease du a ion 10.2 ± 3.5mon hs 16.0 ± 11.4yea 21.9 ± 10.2yea Headache equency (days/mon h) 30 ± 0 3.5 ± 1.9 22.8 ± 6.7 Table 2 Speci ic clinical cha ac e is ics ela ed o headache o pa ien s wi h pe sis en headache a e COVID-19 esolu ion Mean in ensi y was exp essed as mean ± SD. The pheno ype o he does no adhe e o mig aine o TTH ol- lowing ICHD-3 guidelines. TTH = ension- ype-headache. Anosmia was no a cha ac e is ic di ec ly ela ed o headache, bu i was included due o i s ela ionship wi h COVID-19 Headache cha ac e is ic Full sample (n = 42) Di usion MRI sample (n = 40) Pheno ype, mig aine/TTH/o he 20/13/9 (47.6/31.0/21.4%) 20/12/8 (50/30/20%) Mean in ensi y (1–10) 7.3 ± 1.6 7.4 ± 1.6 Anosmia 22 (52.4%) 21 (52.5%) Loca ion Hemic anial 10 (23.8%) 9 (22.5%) Holoc anial 31 (73.8%) 30 (75%) Quali y P essing 30 (71.4%) 28 (70%) Th obbing 20 (47.6%) 20 (50%) Bu ning 1 (2.4%) 1 (2.5%) Symp oms du ing headache Pho o- and phonophobia 21 (50%) 20 (50%) Nausea o omi ing 16 (38.1%) 16 (40%) Agg a a ed by mo emen 22 (52.4%) 21 (52.5%) Jou nal o Neu ology 1 3 p = 0.025) we e obse ed in he COV g oup. No s a is ically signi ican di e ences we e iden i ied in co ical cu a u e o su ace a ea. The egions wi h s a is ically signi ican di e - ences we e he same a e addi ionally co ec ing he esul s o sex. Wi h espec o he compa ison be ween COV and CM pa ien s a e co ec ing o age, s a is ically signi ican di e ences we e ound o e e y pa ame e excep he su ace a ea. Fi s , lowe co ical cu a u e alues in he le cuneus (co ec ed p = 0.007) and he igh p ecuneus (co ec ed p = 0.028) we e obse ed in COV pa ien s. Also, highe co ical GM olume in he le caudal mid- dle on al gy us (co ec ed p = 0.012), pa acen al co ex (co ec ed p = 0.032) and pos e io cingula e gy us (co - ec ed p = 0.038) we e ound in COV pa ien s. Mo eo e , highe co ical hickness in he le banks o he supe- io empo al sulcus (co ec ed p = 0.007) and pa acen- al co ex (co ec ed p = 0.007) we e iden i ied in COV pa ien s. The egions wi h s a is ically signi ican di e - ences emained he same a e addi ionally co ec ing he esul s o sex. A summa y o hese esul s can be ound in Fig.3, and de ailed esul s conside ing all g oups a e a ailable in he Supplemen a y ma e ial (Tables S1–3). Rega ding he e ec size o he compa isons wi h signi ican di e - ences, all he alues we e in he ange o medium e ec size (0.5 ≤ d < 0.8), as shown in Fig.4. Fig. 2 G ay ma e egions ha p esen s a is ically signi ican changes be ween pa ien s wi h pe sis en headache a e COVID- 19 esolu ion (COV) and heal hy con ols (HC). A e alse disco - e y a e co ec ion, COV pa ien s showed lowe g ay ma e olume (GMV) and co ical hickness (CT) han HC. L le , R igh . These esul s we e also s a is ically signi ican a e co ec ing o age and sex. Addi ional esul s a e his co ec ion a e no shown Jou nal o Neu ology 1 3 Compa ison be weenpa ien s wi hmig aine andHC The esul s o he compa isons be ween CM, EM and HC, which a e no he ocus o his s udy, ha e been published elsewhe e, employing a highe sample size and balancing age and sex acco ding o he mig aine pa ien s’ cha ac e - is ics [12]. To be e unde s and he esul s o his esea ch, we he e summa ize hese esul s. B ie ly, co ical cu a u e was highe and co ical hickness, su ace a ea and co ical GM olume we e lowe in bo h mig aine g oups compa ed o HC. Cu a u e alues we e highe in mig aine pa ien s in egions such as he cingula e gy us, la e al occipi al co ex, he p ecuneus and he pa acen al co ex. Co ical hickness was lowe in pa ien s, wi h highe le el o signi icance, in he in e io empo al and usi o m gy i. GM olume and su ace a ea we e commonly lowe in pa ien s in he insula, he supe io empo al gy us, pa s iangula is and pa s o bi - alis, and addi ionally lowe a ea alues we e ound in he p ecuneus, cingula e gy us, sup ama ginal gy us and di e se on al, empo al and pa ie al gy i. Mo eo e , GM olume and widesp ead su ace a ea egions p esen ed lowe alues in CM compa ed o EM, in con as o co ical hickness, o Fig. 3 G ay ma e egions ha p esen s a is ically signi ican changes be ween pa ien s wi h pe sis en headache a e COVID-19 esolu ion (COV) and pa ien s wi h episodic mig aine (EM) and/o ch onic mig aine (CM). A e alse disco e y a e co ec ion, COV pa ien s showed highe co ical g ay ma e olume (GMV) and co - ical hickness (CT) han EM and CM, lowe subco ical GMV han EM, and lowe co ical cu a u e (CC) han CM. L le , R igh Jou nal o Neu ology 1 3 which highe alues we e iden i ied in CM in he in e io empo al gy us. Di usion pa ame e s inwhi e ma e Compa ison be weenCOV andHC S a is ically signi ican lowe FA was ound in COV com- pa ed o HC in 15 egions, and in 7 addi ional egions a e co ec ing o sex. These egions we e he in e nal and ex e nal capsule, he ce eb al peduncle, he co icospinal ac , he bila e al co ona adia a, he co pus callosum, he halamic adia ion, he supe io longi udinal asciculus, and he supe io on o-occipi al asciculus. In addi ion, highe RD alues we e ound in COV pa ien s in h ee egions om he le hemisphe e, and in eigh addi ional le egions a e co ec ing o sex: supe io longi udinal asciculus, co ona adia a, he ex e nal and in e nal capsule, he ce eb al pedun- cle, and he sagi al s a um. These FA and RD di e ences a e shown in de ail in Fig.5, showing he di e ences o each egion in Tables S4–7. Figu e6 shows a g aphical depic ion o hese WM changes. The dis ibu ion o he alues o he ou DTI pa ame e s o COV and HC g oups is shown o he h ee egions wi h RD signi ican di e ences and he le e o- len icula pa o in e nal capsule. These ou egions a e ep esen ed o obse e he dis ibu ion o he alues o all he RD di e ences and he egion wi h he highes e ec size in he FA compa ison, o be e app ecia e he FA di e ences be ween g oups. In o de o gain insigh abou he di ec ion and magni ude o hese di e ences, Fig.7 p esen s he p es- ence o absence o s a is ically signi ican changes o each WM egion, oge he wi h hei di ec ion and Cohen’s d. The e ec size o he iden i ied di e ences was e y low (d < 0.2) excep o compa isons in ol ing he in e nal capsule. Compa ison be weenCOV andmig aine Widesp ead s a is ically signi ican di e ences we e ound be ween he COV g oup and pa ien s wi h EM and CM conside ing he ou WM desc ip o s ha we e employed (Fig.7). Rega ding FA, COV pa ien s showed lowe alues han EM and CM in 15 (also in i e addi ional egions a e co - ec ion o sex) and 17 egions (also in one addi ional egion a e co ec ion o sex), espec i ely. Fo all FA compa i- sons, he main egions wi h di e ences we e he co pus cal- losum, bila e al co ona adia a, and le ex e nal and in e nal capsule, le ce eb al peduncle, and le halamic adia ion, inding addi ional di e ences in he le supe io longi udi- nal asciculus in he compa ison wi h EM. Wi h espec o he AD, COV pa ien s p esen ed s a is i- cally signi ican lowe alues in compa ison wi h EM in 39 egions (same numbe a e co ec ing o sex), in 14 egions compa ed o CM, mos ly in he igh hemisphe e, and in 8 ou o hese 14 egions a e co ec ing o sex. The main bila e al egions whe e di e ences agains CM and EM we e obse ed we e he ce ebella and ce eb al peduncles, he in e nal capsule, and he halamic adia ion. Mo eo e , di - e ences agains EM we e also ound in he co pus callosum, co ona adia a, co icospinal ac , supe io longi udinal as- ciculus, ex e nal capsule, and halamic adia ion. S a is ically signi ican lowe MD alues we e ound in COV pa ien s wi h espec o EM, in 25 egions (in h ee egions less a e co ec ion o sex), and CM, in 14 egions (in i e egions less a e co ec ion o sex). Mos Fig. 4 Cohen’s d alues o he compa isons wi h s a is ically sig- ni ican di e ences o g ay ma e mo phome y pa ame e s. Cells in ed e lec compa isons wi h lowe alues in pa ien s wi h pe sis en headache a e COVID-19, while cells in g een e lec he opposi e end. *Resul s ob ained exclusi ely when adding sex as a co a ia e Jou nal o Neu ology 1 3 COV pa ien s would p esen some p ope ies ela ed o a si ua ion simila bu milde han EM. A hypo hesis ha may explain his si ua ion is ha changes in he a o emen ioned di ec ion could be ela ed o he e ec o ime, conside ing ha pa ien s wi h mig aine ha e su e ed om headaches a longe ime compa ed o he COV pa ien s. Mo eo e , i is in e es ing o no e ha FA and RD di e ences be ween COV pa ien s and con ols we e ound in he le hemisphe e. On he o he hand, lowe MD and RD alues in COV pa ien s compa ed o mig aine pa ien s we e iden i ied in he igh hemisphe e, al hough highe RD alues we e also ound mos ly in ew le hemisphe e egions in COV pa ien s com- pa ed o CM only a e co ec ing he esul s o sex. Wi h ega d o he addi ional di e ences ound in he compa ison be ween pa ien g oups, and in line wi h p e i- ous s udies wi h he same coho o mig aine pa ien s, COV pa ien s showed simila AD di e ences compa ed o hose ound in CM wi h espec o EM, wi h widesp ead lowe AD alues [18, 42, 46]. Thus, COV pa ien s p esen ed a si ua ion simila o CM, wi h a possible in ensi ied com- ponen o he axial di usion. This e ec sugges s ha he axial pa ame e s could be highly ela ed o he headache e- quency. CM and COV pa ien s p esen a high headache e- quency, wi h 15 o mo e a acks pe mon h and almos e e y day, espec i ely [9]. In ac , in he sample employed in his s udy, all COV pa ien s su e ed om daily headache. On he con a y, pa ien s wi h EM, especially excluding pa ien s wi h high- equency EM (mo e han eigh headache days pe mon h), p esen ed opposi e alues in compa ison wi h he o he wo headache g oups. This e ec is in line wi h he la es mig aine s udies assessing AD in pa ien s wi h EM [16, 18, 59, 60]. Thus, in ela ion o he almos daily headache a acks o COV pa ien s, he pe manen ac i a- ion o he b ain can cause speci ic changes possibly associ- a ed wi h axonal damage o loss, o sho - e m demyelina- ion [48, 61–63]. Conside ing he di e ences be ween he headache g oups, we expec ed di e ences in AD be ween COV pa ien s and HC, bu we ob ained no s a is ically sig- ni ican esul s. The eason o his lack o di e ences may be associa ed wi h he highe age o con ols in his s udy, which may be a key ac o in he changes o di usion in he main o axonal di ec ion. Ne e heless, o conside he po en ial di e ences caused by he age di e ence, esul s we e co ec ed by age. In heal hy adul s, i has been epo ed ha AD signi ican ly changes wi h age. In egions such as he co ona adia a, which p esen ed s a is ically signi ican di e ences in COV pa ien s acco ding o ou esul s, AD has been epo ed o dec ease wi h age in adul s [64]. This esul seems o be in line wi h ou co ela ion esul s in COV pa ien s, as we ound a nega i e co ela ion be ween disease du a ion and mean AD alues in he pos e io co ona adia a. Howe e , he opposi e esul s ha e also been epo ed in GM and o he WM egions such as he o nix [64, 65]. GM esul s showed a simila se o di e ences in com- pa ison wi h he desc ibed RD and MD di e ences in WM, i.e., a si ua ion analogous bu milde han EM. In con as o a p e ious s udy ha iden i ied di e ences mo phome y pa ame e s be ween pa ien s wi h mig aine (EM and CM) and con ols in di e se egions, s a is ically signi ican di - e ences we e de ec ed in a small g oup o egions [12]. The numbe o egions wi h signi ican di e ences was pa - icula ly small be ween EM and COV pa ien s. Since COV pa ien s p esen a smalle deg ee o changes wi h espec o HC, he e ec o ime may be impo an in GM changes. No able limi a ions a e p esen in his s udy. Fi s , he e we e no baseline MRI acquisi ions o he COV pa ien s. This implies ha we we e unable o obse e he e ec s o he pe sis en headache be o e, du ing and a e he in ec ion o de e mine he changes associa ed wi h each pa hophysiological mechanism. To compensa e he lack o longi udinal da a, we compa ed he main g oup o in e es no only wi h heal hy con ols, bu also wi h he wo cu en ly dis inguished mig aine ypes, EM and CM. Howe e , pa ien s wi h mig aine, especially EM, a e younge han COV pa ien s. Despi e all compa isons wi h mig aine g oups we e co ec ed o age, his ac o could ha e in luenced he esul s. Mo eo e , due o he age di - e ence be ween he pa ien g oups, he con ols included in his s udy we e olde wi h espec o p e ious s udies whe e he a ge was he compa ison wi h pa ien s wi h mig aine [12, 18, 42]. The ema kable di e ence o age o he con ol g oups migh ha e al e ed he es ablished hypo hesis ega ding he di e ences o he di e se pa ame- e s be ween con ols and each pa ien g oup. We p e e ed o ollow he esul s om p e ious s udies wi h he same pa ien s o a oid any po en ial bias in he esul s caused o employing con ols no balanced o age and sex in compa ison wi h pa ien s wi h mig aine. Conside ing he di e se pheno ypes o headache in he pa ien s who eco - e ed om COVID-19, an assessmen o he di e ences be ween he mig aine and con ol g oups and each long- e m headache ype was no conduc ed. The eason was ha we p e e ed o p ese e a g oup wi h a ela i ely high sample size, as changes in headache diso de s ha e been epo ed o be sub le and he e o e a small sample size can be insu icien o iden i y s a is ically signi ican di e - ences. Due o he lack o in lamma ion bioma ke s a he di e se imepoin s o he ollow-up pe iod, we could no clea ly es ablish whe he changes a e s ongly associa ed wi h he COVID-19 in ec ion o mo e ela ed o headache. In addi ion, no T2-weigh ed o FLAIR images we e a ail- able in ei he he con ols o he pa ien s wi h mig aine. Conside ing he po en ial ela ionship be ween WM hype - in ensi ies and WM in eg i y, and also he iden i ica ion o hese indings in pa ien s wi h mig aine [66–68], we we e unable o assess a possible ac o ha migh in luence he Jou nal o Neu ology 1 3 esul s. Finally, since ec ui men o he COV g oup was pe o med ea ly in he COVID-19 pandemic e olu ion, no gene aliza ion can be made wi h ega d o he possible e ec s o o he COVID-19 a ian s such as del a o omi- c on, o he e ec o accina ion o ein ec ion. Conclusions Pa ien s wi h pe sis en headache a e COVID-19 esolu- ion showed di e se changes in GM and WM s uc u e. Changes in GM we e sub le and a ec ed an e io a eas, including he pa s o bi alis, he usi o m gy us and he on al pole. On he one hand, he obse ed WM changes we e di use, in ol ed mos o he WM ac s and seemed ela ed o he impai men o WM ibe bundles. On he o he hand, he WM changes p esen ed a si ua ion analo- gous bu milde han mig aine. Fu u e s udies a e needed o disc imina e long- e m and e ol ing changes associ- a ed wi h COVID-19 in ec ion and headache o assess he di e ences be ween long- e m headache pheno ypes a e COVID-19 eco e y. Supplemen a y In o ma ion The online e sion con ains supplemen- a y ma e ial a ailable a h ps:// doi. o g/ 10. 1007/ s00415- 022- 11398-z. Acknowledgemen s ÁP-G was unded by he Eu opean Union (Nex Gene a ionEU). Au ho con ibu ions ÁP-G: Analyzed and in e p e ed he da a; s a is- ical analysis; MRI p ocessing; c ea ion o igu es; d a ed he manu- sc ip ; e ised he manusc ip o in ellec ual con en . DG-A: S udy concep and design; analyzed and in e p e ed he da a; ec ui men o pa icipan s; d a ed he manusc ip ; e ised he manusc ip o in el- lec ual con en ; s udy supe ision. ÁLG: S udy concep and design; analyzed and in e p e ed he da a; ec ui men o pa icipan s; e ised he manusc ip o in ellec ual con en . MR: Analyzed and in e p e ed he da a; e ised he manusc ip o in ellec ual con en . SA-F: Ana- lyzed and in e p e ed he da a; c ea ion o igu es; e ised he manu- sc ip o in ellec ual con en . RL-G: Analyzed and in e p e ed he da a; ec ui men o pa icipan s; d a ed he manusc ip ; e ised he manusc ip o in ellec ual con en . Funding Open Access unding p o ided hanks o he CRUE-CSIC ag eemen wi h Sp inge Na u e. The s udy was unded by he Regional Heal h Adminis a ion—Ge encia Regional de Salud, Cas illa y Leon (GRS 2284/A/2020). A ailabili y o da a and ma e ials The da a ha suppo s he indings o his s udy a e a ailable om he co esponding au ho , upon eason- able eques . Decla a ions Con lic s o in e es The au ho s epo no compe ing in e es s. E hical app o al and consen o pa icipa e The E hics Re iew Boa d o Valladolid Eas heal h a ea app o ed his s udy (PI-GR-20-2017). All pa icipan s signed an in o med consen be o e hei pa icipa ion in he s udy. Open Access This a icle is licensed unde a C ea i e Commons A i- bu ion 4.0 In e na ional License, which pe mi s use, sha ing, adap a- ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons licence, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle a e included in he a icle's C ea i e Commons licence, unless indica ed o he wise in a c edi line o he ma e ial. I ma e ial is no included in he a icle's C ea i e Commons licence and you in ended use is no pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he copy igh holde . To iew a copy o his licence, isi h p:// c ea i eco mmons. o g/ licen ses/ by/4. 0/. Re e ences 1. Fe nández-de-las-Peñas C, Na a o-San ana M, Gómez-Mayo - domo V e al (2021) Headache as an acu e and pos -COVID-19 symp om in COVID-19 su i o s: a me a-analysis o he cu en li e a u e. Eu J Neu ol 28:3820–3825. h ps:// doi. o g/ 10. 1111/ ene. 15040 2. Ga cía-Azo ín D, Sie a Á, T igo J e al (2021) F equency and pheno ype o headache in co id-19: a s udy o 2194 pa ien s. Sci Rep 11:14674. h ps:// doi. o g/ 10. 1038/ s41598- 021- 94220-6 3. T igo López J, Ga cía-Azo ín D, Planchuelo-Gómez Á e al (2020) Pheno ypic cha ac e iza ion o acu e headache a ibu ed o SARS-CoV-2: an ICHD-3 alida ion s udy on 106 hospi al- ized pa ien s. Cephalalgia 40:1432–1442. h ps:// doi. o g/ 10. 1177/ 03331 02420 965146 4. T igo J, Ga cía-Azo ín D, Sie a-Mencía Á e al (2021) Cy okine and in e leukin p o ile in pa ien s wi h headache and COVID-19: a pilo , CASE-con ol, s udy on 104 pa ien s. J Headache Pain 22:51. h ps:// doi. o g/ 10. 1186/ s10194- 021- 01268-w 5. Planchuelo-Gómez Á, T igo J, de Luis-Ga cía R e al (2020) Deep pheno yping o headache in hospi alized COVID-19 pa ien s ia p incipal componen analysis. F on Neu ol 11:583870. h ps:// doi. o g/ 10. 3389/ neu . 2020. 583870 6. Ca onna E, Ball é A, Llau adó A e al (2020) Headache: a s ik- ing p od omal and pe sis en symp om, p edic i e o COVID-19 clinical e olu ion. Cephalalgia 40:1410–1421. h ps:// doi. o g/ 10. 1177/ 03331 02420 965157 7. Gonzalez-Ma inez A, Fanjul V, Ramos C e al (2021) Headache du ing SARS-CoV-2 in ec ion as an ea ly symp om associa ed wi h a mo e benign cou se o disease: a case–con ol s udy. Eu J Neu ol 28:3426–3436. h ps:// doi. o g/ 10. 1111/ ene. 14718 8. T igo J, Ga cía-Azo ín D, Planchuelo-Gómez Á e al (2020) Fac- o s associa ed wi h he p esence o headache in hospi alized COVID-19 pa ien s and impac on p ognosis: a e ospec i e coho s udy. J Headache Pain 21:94. h ps:// doi. o g/ 10. 1186/ s10194- 020- 01165-8 9. Ga cia-Azo in D, Layos-Rome o A, Po a-E essam J e al (2022) Pos -COVID-19 pe sis en headache: a mul icen ic 9-mon hs ollow-up s udy o 905 pa ien s. Cephalalgia. h ps:// doi. o g/ 10. 1177/ 03331 02421 10680 74 10. Winkle AM, G e e DN, Bjuland KJ e al (2018) Join analysis o co ical a ea and hickness as a eplacemen o he analysis o he olume o he ce eb al co ex. Ce eb Co ex 28:738–749. h ps:// doi. o g/ 10. 1093/ ce co / bhx308 Jou nal o Neu ology 1 3 11. Jia Z, Yu S (2017) G ey ma e al e a ions in mig aine: a sys- ema ic e iew and me a-analysis. Neu oimage Clin 14:130–140. h ps:// doi. o g/ 10. 1016/j. nicl. 2017. 01. 019 12. Planchuelo-Gómez Á, Ga cía-Azo ín D, Gue e o ÁL e al (2020) G ey ma e s uc u al al e a ions in ch onic and episodic mig aine: a mo phome ic magne ic esonance imaging s udy. Pain Med 21:2997–3011. h ps:// doi. o g/ 10. 1093/ pm/ pnaa2 71 13. Schwed TJ, Chong CD, Peplinski J e al (2017) Pe sis en pos - auma ic headache s. mig aine: an MRI s udy demons a ing di e ences in b ain s uc u e. J Headache Pain 18:87. h ps:// doi. o g/ 10. 1186/ s10194- 017- 0796-0 14. Messina R, Rocca MA, Colombo B e al (2015) Whi e ma e mic os uc u e abno mali ies in pedia ic mig aine pa ien s. Cephalalgia 35:1278–1286. h ps:// doi. o g/ 10. 1177/ 03331 02415 578428 15. Szabó N, Fa agó P, Ki ály A e al (2017) E idence o plas ic p ocesses in mig aine wi h au a: a di usion weigh ed MRI s udy. F on Neu oana 11:138. h ps:// doi. o g/ 10. 3389/ nana. 2017. 00138 16. Ka em Husøy A, Eikenes L, Håbe g AK e al (2019) Di usion enso imaging in middle-aged headache su e e s in he gene al popula ion: a c oss-sec ional popula ion-based imaging s udy in he No d-T øndelag heal h s udy (HUNT-MRI). J Headache Pain 20:78. h ps:// doi. o g/ 10. 1186/ s10194- 019- 1028-6 17. Yu D, Yuan K, Zhao L e al (2013) Whi e ma e in eg i y a ec ed by dep essi e symp oms in mig aine wi hou au a: a ac -based spa ial s a is ics s udy. NMR Biomed 26:1103–1112. h ps:// doi. o g/ 10. 1002/ nbm. 2924 18. Planchuelo-Gómez Á, Ga cía-Azo ín D, Gue e o ÁL e al (2020) Whi e ma e changes in ch onic and episodic mig aine: a di u- sion enso imaging s udy. J Headache Pain 21:1. h ps:// doi. o g/ 10. 1186/ s10194- 019- 1071-3 19. Lu Y, Li X, Geng D e al (2020) Ce eb al mic o-s uc u al changes in COVID-19 pa ien s—an MRI-based 3-mon h ollow-up s udy. EClinicalMedicine 25:100484. h ps:// doi. o g/ 10. 1016/j. eclinm. 2020. 100484 20. Douaud G, Lee S, Al a o-Almag o F e al (2022) SARS-CoV-2 is associa ed wi h changes in b ain s uc u e in UK Biobank. Na u e 604:697–707. h ps:// doi. o g/ 10. 1038/ s41586- 022- 04569-5 21. Huang S, Zhou Z, Yang D e al (2021) Pe sis en whi e ma e changes in eco e ed COVID-19 pa ien s a he 1-yea ollow-up. B ain. h ps:// doi. o g/ 10. 1093/ b ain/ awab4 35 22. Pelizza i L, Cazzoli M, Lipa i S e al (2022) Mid- e m MRI e alu- a ion e eals mic os uc u al whi e ma e al e a ions in COVID- 19 ully eco e ed subjec s wi h anosmia p esen a ion. The Ad Neu ol Diso d 15:17562864221111996. h ps:// doi. o g/ 10. 1177/ 17562 86422 11119 95 23. Wo ld Heal h O ganiza ion Co ona i us disease (COVID-19) echnical guidance: labo a o y es ing o 2019-nCoV in humans. h ps:// www. who. in / eme g encies/ diseases/no el-co ona i- us-2019/ echnical -guidance/labo a o y-guidance. Accessed 31 May 2020 24. Minis y o Heal h (Spain) Technical documen s. Hospi al man- agemen o COVID-19. Ve sion Ma ch 19, 2020. h ps:// www. mscbs. gob. es/ p o e siona les/ salud Publi ca/ ccayes/ ale asAc ual/ nCo - China/ docum en os/ P o o colo_ manejo_ clini co_ ah_ COVID- 19. pd . Accessed 31 May 2020 25. Headache Classi ica ion Commi ee o he In e na ional Headache Socie y (2018) The in e na ional classi ica ion o headache diso - de s, 3 d edi ion. Cephalalgia 38:1–211. h ps:// doi. o g/ 10. 1177/ 03331 02413 485658 26. Dale AM, Fischl B, Se eno MI (1999) Co ical su ace-based analysis. I Segmen Su Recons uc Neu oimage 9:179–194 27. Fischl B, Liu A, Dale AM (2001) Au oma ed mani old su ge y: cons uc ing geome ically accu a e and opologically co ec models o he human ce eb al co ex. IEEE T ans Med Imaging 20:70–80 28. Fischl B, Sala DH, Busa E e al (2002) Whole b ain segmen- a ion: au oma ed labeling o neu oana omical s uc u es in he human b ain. Neu on 33:341–355 29. Ségonne F, Dale AM, Busa E e al (2004) A hyb id app oach o he skull s ipping p oblem in MRI. Neu oimage 22:1060–1075 30. Ve aa J, No iko DS, Ch is iaens D e al (2016) Denoising o di usion MRI using andom ma ix heo y. Neu oimage 142:394– 406. h ps:// doi. o g/ 10. 1016/j. neu o image. 2016. 08. 016 31. Co de o-G ande L, Ch is iaens D, Hu e J e al (2019) Complex di usion-weigh ed image es ima ion ia ma ix eco e y unde gene al noise models. Neu oimage 200:391–404. h ps:// doi. o g/ 10. 1016/j. neu o image. 2019. 06. 039 32. Kellne E, Dhi al B, Kisele VG, Reise M (2016) Gibbs- inging a i ac emo al based on local sub oxel-shi s. Magn Reson Med 76:1574–1581. h ps:// doi. o g/ 10. 1002/ m m. 26054 33. Ande sson JL, So i opoulos SN (2015) An in eg a ed app oach o co ec ion o o - esonance e ec s and subjec mo emen in di usion MR imaging. Neu oimage 125:1063–1078. h ps:// doi. o g/ 10. 1016/j. neu o image. 2015. 10. 019 34. Zhang Y, B ady M, Smi h S (2001) Segmen a ion o b ain MR images h ough a hidden Ma ko andom ield model and he expec a ion-maximiza ion algo i hm. IEEE T ans Med Imaging 20:45–57. h ps:// doi. o g/ 10. 1109/ 42. 906424 35. Smi h SM, Jenkinson M, Wool ich MW e al (2004) Ad ances in unc ional and s uc u al MR image analysis and implemen a ion as FSL. Neu oimage 23:S208–S219. h ps:// doi. o g/ 10. 1016/j. neu o image. 2004. 07. 051 36. Tou nie J-D, Smi h R, Ra el D e al (2019) MR ix3: a as , lexible and open so wa e amewo k o medical image p ocess- ing and isualisa ion. Neu oimage 202:116137. h ps:// doi. o g/ 10. 1016/j. neu o image. 2019. 116137 37. Dhollande T, Ra el D, Connelly A (2016) Unsupe ised 3- is- sue esponse unc ion es ima ion om single-shell o mul i-shell di usion MR da a wi hou a co- egis e ed T1 image. ISMRM Wo kshop on B eaking he Ba ie s o Di usion MRI 5 38. Jenkinson M, Beckmann CF, Beh ens TE e al (2012) FSL. Neu- oimage 62:782–790. h ps:// doi. o g/ 10. 1016/j. neu o image. 2011. 09. 015 39. Smi h SM, Jenkinson M, Johansen-Be g H e al (2006) T ac - based Spa ial S a is ics: oxelwise analysis o mul i-subjec di - usion da a. Neu oimage 31:1487–1505. h ps:// doi. o g/ 10. 1016/j. neu o image. 2006. 02. 024 40. Mo i S, Wakana S, Nagae-Poe sche LM, an Zijl PC (2005) MRI a las o human whi e ma e . Else ie , Ams e dam 41. Oishi K, Zilles K, Amun s K e al (2008) Human b ain whi e ma e a las: Iden i ica ion and assignmen o common ana omical s uc u es in supe icial whi e ma e . Neu oimage 43:447–457. h ps:// doi. o g/ 10. 1016/j. neu o image. 2008. 07. 009 42. Planchuelo-Gómez Á, Ga cía-Azo ín D, Gue e o ÁL e al (2020) Al e na i e mic os uc u al measu es o complemen di usion enso imaging in mig aine s udies wi h s anda d MRI acquisi- ion. B ain Sci 10:711. h ps:// doi. o g/ 10. 3390/ b ain sci10 100711 43. Benjamini Y, Hochbe g Y (1995) Con olling he alse disco e y a e: a p ac ical and powe ul app oach o mul iple es ing. J R S a Soc 57:289–300. h ps:// doi. o g/ 10. 2307/ 23461 01 44. Nichols TE, Holmes AP (2002) Nonpa ame ic pe mu a ion es s o unc ional neu oimaging: a p ime wi h examples. Hum B ain Mapp 15:1–25. h ps:// doi. o g/ 10. 1002/ hbm. 1058 45. Smi h SM, Nichols TE (2009) Th eshold- ee clus e enhance- men : add essing p oblems o smoo hing, h eshold dependence and localisa ion in clus e in e ence. Neu oimage 44:83–98. h ps:// doi. o g/ 10. 1016/j. neu o image. 2008. 03. 061 Jou nal o Neu ology 1 3 46. Ma ín-Ma ín C, Planchuelo-Gómez Á, Gue e o ÁL e al (2022) Viabili y o AMURA bioma ke s om single-shell di usion MRI in Mig aine Clinical S udies. bioRxi . h ps:// doi. o g/ 10. 1101/ 2022. 04. 01. 486661 47. Beaulieu C (2002) The basis o aniso opic wa e di usion in he ne ous sys em—a echnical e iew. NMR Biomed 15:435–455. h ps:// doi. o g/ 10. 1002/ nbm. 782 48. Alexande AL, Lee JE, Laza M, Field AS (2007) Di usion enso imaging o he b ain. Neu o he apeu ics 4:316–329. h ps:// doi. o g/ 10. 1016/j. nu . 2007. 05. 011 49. Kochuno P, Thompson PM, Lancas e JL e al (2007) Rela ion- ship be ween whi e ma e ac ional aniso opy and o he indices o ce eb al heal h in no mal aging: ac -based spa ial s a is ics s udy o aging. Neu oimage 35:478–487. h ps:// doi. o g/ 10. 1016/j. neu o image. 2006. 12. 021 50. Vi a A, de Pe i L, Des e G, Sacche i E (2012) P og essi e loss o co ical g ay ma e in schizoph enia: a me a-analysis and me a- eg ession o longi udinal MRI s udies. T ansl Psychia y 2:e190– e190. h ps:// doi. o g/ 10. 1038/ p. 2012. 116 51. Kelly S, Jahanshad N, Zalesky A e al (2018) Widesp ead whi e ma e mic os uc u al di e ences in schizoph enia ac oss 4322 indi iduals: esul s om he ENIGMA Schizoph enia DTI Wo k- ing G oup. Mol Psychia y 23:1261–1269. h ps:// doi. o g/ 10. 1038/ mp. 2017. 170 52. an Ha en NEM, Schnack HG, Cahn W e al (2011) Changes in co ical hickness du ing he cou se o illness in schizoph enia. A ch Gen Psychia y 68:871–880. h ps:// doi. o g/ 10. 1001/ a chg enpsy chia y. 2011. 88 53. Wang W-Y, Yu J-T, Liu Y e al (2015) Voxel-based me a-analysis o g ey ma e changes in Alzheime ’s disease. T ansl Neu ode- gene 4:6. h ps:// doi. o g/ 10. 1186/ s40035- 015- 0027-z 54. Oishi K, Mielke MM, Albe M e al (2011) DTI analyses and clinical applica ions in Alzheime ’s disease. J Alzheime s Dis 26(Suppl 3):287–296. h ps:// doi. o g/ 10. 3233/ JAD- 2011- 0007 55. Wu Z, Peng Y, Hong M, Zhang Y (2021) G ay ma e de e io a- ion pa e n du ing Alzheime ’s disease p og ession: a egions-o - in e es based su ace mo phome y s udy. F on Aging Neu osci. h ps:// doi. o g/ 10. 3389/ nagi. 2021. 593898 56. A kinson-Clemen C, Pin o S, Eusebio A, Coulon O (2017) Di - usion enso imaging in Pa kinson’s disease: e iew and me a- analysis. Neu oimage Clin 16:98–110. h ps:// doi. o g/ 10. 1016/j. nicl. 2017. 07. 011 57. Laansma MA, B igh JK, Al-Bacha i S e al (2021) In e na ional mul icen e analysis o b ain s uc u e ac oss clinical s ages o Pa kinson’s disease. Mo Diso d 36:2583–2594. h ps:// doi. o g/ 10. 1002/ mds. 28706 58. Colloby SJ, Wa son R, Blami e AM e al (2019) Co ical hinning in demen ia wi h Lewy bodies and Pa kinson disease demen ia. Aus N Z J Psychia y 54:633–643. h ps:// doi. o g/ 10. 1177/ 00048 67419 885165 59. Qin Z, He X-W, Zhang J e al (2019) S uc u al changes o ce - ebellum and b ains em in mig aine wi hou au a. J Headache Pain 20:93. h ps:// doi. o g/ 10. 1186/ s10194- 019- 1045-5 60. Ma ciszewski KK, Meylakh N, di Pie o F, e al (2019) Fluc ua ion egional b ains em di usion imaging measu es o mic os uc u e ac oss he mig aine cycle. eNeu o 6:ENEURO.0005–19. h ps:// doi. o g/ 10. 1523/ ENEURO. 0005- 19. 2019 61. Yu D, Yuan K, Qin W e al (2013) Axonal loss o whi e ma e in mig aine wi hou au a: a ac -based spa ial s a is ics s udy. Ceph- alalgia 33:34–42. h ps:// doi. o g/ 10. 1177/ 03331 02412 466964 62. Sun S-W, Liang H-F, C oss AH, Song S-K (2008) E ol ing Wal- le ian degene a ion a e ansien e inal ischemia in mice cha ac- e ized by di usion enso imaging. Neu oimage 40:1–10. h ps:// doi. o g/ 10. 1016/j. neu o image. 2007. 11. 049 63. Winklewski PJ, Sabisz A, Naumczyk P e al (2018) Unde s anding he physiopa hology behind axial and adial di usi i y changes— Wha do we know? F on Neu ol 9:92. h ps:// doi. o g/ 10. 3389/ neu . 2018. 00092 64. Benne IJ, Madden DJ, Vaidya CJ e al (2010) Age- ela ed di - e ences in mul iple measu es o whi e ma e in eg i y: a di u- sion enso imaging s udy o heal hy aging. Hum B ain Mapp 31:378–390. h ps:// doi. o g/ 10. 1002/ hbm. 20872 65. Kuma R, Cha ez AS, Macey PM e al (2013) B ain axial and adial di usi i y changes wi h age and gende in heal hy adul s. B ain Res 1512:22–36. h ps:// doi. o g/ 10. 1016/j. b ain es. 2013. 03. 028 66. Bashi A, Lip on RB, Ashina S, Ashina M (2013) Mig aine and s uc u al changes in he b ain: a sys ema ic e iew and me a- analysis. Neu ology 81:1260–1268. h ps:// doi. o g/ 10. 1212/ WNL. 0b013 e3182 a6cb32 67. Po e A, Glads one JP, Dodick DW (2005) Mig aine and whi e ma e hype in ensi ies. Cu Pain Headache Rep 9:289–293. h ps:// doi. o g/ 10. 1007/ s11916- 005- 0039-y 68. Xie H, Zhang Q, Huo K e al (2018) Associa ion o whi e ma - e hype in ensi ies wi h mig aine ea u es and p ognosis. BMC Neu ol 18:93. h ps:// doi. o g/ 10. 1186/ s12883- 018- 1096-2