The Journal of Pain, Vol 16, No 1 (January), 2015: pp 53-59Available online at www.jpain.org and www.sciencedirect.com

Transcranial Magnetic Stimulation Reveals Cortical

Hyperexcitability in Episodic Cluster Headache

Guiseppe Cosentino, Filippo Brighina, Sara Brancato, Francesca Valentino,Serena Indovino, and Brigida FierroDepartment of Experimental Biomedicine and Clinical Neurosciences (BioNeC), University of Palermo, Palermo, Italy.

Received2014.The authAddress1, 90129

1526-590

ª 2015 b

http://dx

Abstract: Evidence shows involvement of the cerebral cortex in the pathophysiology of cluster

headache (CH). Here we investigated cortical excitability in episodic CH patients by using transcranial

magnetic stimulation. In 25 patients with episodic CH and 13 healthy subjects we evaluated the mo-

tor cortical response to single-pulse (ie, motor threshold, input-output curves, cortical silent period)

and paired-pulse (ie, intracortical facilitation, short intracortical inhibition) transcranial magnetic

stimulation in both hemispheres. Thirteen patients were evaluated outside bout and the remaining

12 patients inside bout. Our results showed increased slope of the input-output curves after stimu-

lation of both hemispheres in patients outside bout and in the hemisphere contralateral to the head-

ache side in patients inside bout. Increased intracortical facilitation was observed in the hemisphere

ipsilateral to the headache side in patients evaluated both outside and inside bout; reduced short in-

tracortical inhibition was observed in patients inside bout ipsilateral to the side of pain. In conclusion,

we provide evidence of increased cortical excitability in episodic CH both outside and inside bout,

especially in the hemisphere ipsilateral to the side of headache attacks. Our results suggest that an

abnormal regulation of cortical excitability could be involved in the pathophysiology of CH.

Perspectives: We investigated cortical excitability in episodic cluster headache by using transcra-

nial magnetic stimulation, providing evidence of cortical hyperexcitability in patients both inside

and outside bout. We suggest that an abnormal state of cortical excitability could be involved in

the pathophysiology of the disease.

ª 2015 by the American Pain Society

Key words: Episodic cluster headache, pathophysiology, transcranial magnetic stimulation, cortical

excitability, motor cortex.

Cluster headache (CH) is an extremely painful head-ache disease, clinically impressing with strictly uni-lateral pain accompanied by ipsilateral cranial

autonomic symptoms. The headache attacks typicallyoccur in cluster periods lasting several weeks or months(inside bout), separated by remission periods of monthsor years (outside bout). The pathophysiology of CH isnot comprehensively understood, being still indistinctas to how different peripheral and central neural mech-anisms reciprocally interact to determine the clinical fea-tures of the disease. Modern techniques of functionalneuroimaging have supported the hypothesis thatsome peculiar features of CHmight rely on the hypothal-amus involvement.23,24,36 A hypothalamic dysfunction

August 14, 2014; Revised October 6, 2014; Accepted October 21,

ors declare that they have no competing interests.reprint requests to Filippo Brighina, MD, Via Gaetano La Loggia,Palermo, Italy. E-mail: filippo.brighina@unipa.it

0/$36.00

y the American Pain Society

.doi.org/10.1016/j.jpain.2014.10.006

could also account for the abnormal centraldisinhibition of the trigeminal nociceptive system asevidenced by most electrophysiological studies inCH. 27,31 Indeed, as the hypothalamus plays a centralrole in modulating the trigeminal-autonomic reflex,5,25

it is conceivable that its dysfunction could result in adestabilization of the balance between pronociceptiveand antinociceptive inputs acting on thetrigeminovascular system. However, it is still debatedwhether the hypothalamus actively contributes to thegeneration of the attacks, or whether its activationoccurs just after the attack onset, in the context of alarger ‘‘pain matrix,’’ for modulating CH. Consistentwith this latter hypothesis, there is experimentalevidence that brain areas involved in processing of painstimuli are extensively activated during cluster attacks,but not in pain-free states.24,36,38

Therefore, if on the one hand the hypothalamicimpairment could affect activity of several brainareas,37 on the other it has been supposed that in CHa hypothalamic dysfunction could be orchestrated by

53

54 The Journal of Pain Cortical Excitability in Cluster Headache

many other brain regions, including the brainstemnuclei, the thalamus, and the cerebral cortex.1,2,38

Neurophysiological studies of visually, somatosensory,and auditory evoked potentials have providedevidence of impaired cortical information processingin CH.9,16,39 The cerebral cortex involvement is alsosupported by the occurrence of aura symptoms in CHpatients as reported by different authors; indeed, thecortex is most likely the source of the aurasymptoms.4,30,33 In spite of the above, a very limitedamount of research has been performed toinvestigate the contribution of the cerebral cortex inthe pathogenesis of CH.Transcranial magnetic stimulation (TMS) has been

widely employed to noninvasively evaluate the corticalexcitability state in migraine, and numerous TMS studieshave given support to the concept that abnormal regula-tion of cortical excitability plays a pivotal role in the sus-ceptibility and recurrence of the migraine attacks.8,12

Instead, to our knowledge, no TMS studies have beenconducted in CH until now. Thus, the aim of the presentstudy was to investigate the excitability of the cerebralcortex in CH by using TMS. We assessed corticomotorexcitability because different TMS paradigms canprovide objective measures of activity and function ofspecific intracortical circuits in the human motor cortex.

Methods

SubjectsTwenty-five episodic CH patients (21 males, mean age

37.7 6 10.5 years) and 13 sex- and age-matched healthyvolunteers (HVs) without pastmedical history or familiar-ity for recurrent headache (11 males, mean age35.2 6 11.2 years) were enrolled in the study (Table 1).Diagnosis of episodic CH was made according to theInternational Classification of Headache Disorders (3rdedition, beta version).19 All patients experienced strictlyunilateral pain (16 right side, 9 left side). Exclusioncriteria included coexistence of other primary or second-ary headaches, family history of migraine, serious sys-temic or neurologic disease, or psychiatric disorder.Neurologic examination and brain magnetic resonanceimaging were normal in all patients. Thirteen patientswere studied outside bout (30–150 days after the endof the bout), whereas the remaining 12 patients under-went the experimental procedures inside bout, outsideattacks (2–36 hours after the end of an attack). None of

Table 1. Demographic and Clinical Characteristics

SUBJECT AGE, Y (MEAN 6 SD) GENDER, F/M (N)

HVs (n = 13) 35.2 6 11.2 3/10

CH patients

Outside bout (n = 13) 37.4 6 11.2 2/11

Inside bout (n = 12) 38 6 10.8 2/10

Abbreviations: SD, standard deviation; F/M, female/male.

the patients was taking prophylactic drugs or corticoste-roids at least 3 months prior to the study. Patients insidebout underwent the electrophysiological assessmentonly when they had not taken symptomatic medicationsin the 12 hours preceding the evaluation. To minimizeany hormonal effect, female patients and controls werenot examined during the menstrual phase.35 Beforeenrollment, all the subjects were checked for contraindi-cations to TMS20 and gave their written informed con-sent to participate. The study conformed to theDeclaration of Helsinki, and the experimental proce-dures were approved by the local ethics committee.The demographic and clinical data of subjects are sum-marized in Table 1.

Stimulation ProceduresAll subjects were comfortably seated on a chair and

told to be as relaxed as possible. They wore a tight-fitting plastic swimming cap to mark the optimum stim-ulation site. Electromyography (EMG) signals wererecorded from the abductor pollicis brevis muscle using.9-cm-diameter Ag-AgCl surface electrodes placed 3 cmapart over the belly and tendon of the muscle. TheEMG activity was recorded with a bandpass of 10 to1,000 Hz and a display gain ranging from 50 to1,000 mV/cm. EMG signals were collected, averaged,and analyzed off-line. Focal TMS was applied over thehandmotor cortex by using a figure-of-8 coil (double-cir-cular 70-mm coil) connected to 2 Magstim 200 stimula-tors through a Bistim module (Magstim Co, Dyfed, UK).The stimulating coil was placed over the optimal sitefor eliciting responses in the contralateral target muscle.Excitability of the 2 hemispheres was investigated sepa-rately. The resting motor threshold (RMT) for elicitingresponses in the relaxed abductor pollicis brevis musclewas defined as the minimum intensity of stimulationneeded to produce responses of 50 mV in at least 50%of 10 trials. Stimulation was performed following safetyguidelines.29

Experimental Paradigm andMeasurementsAll subjects underwent an experimental evaluation in

which we assessed, in the hand motor cortex of bothhemispheres, 3 measurements of cortical excitability.The first measurement was intracortical facilitation(ICF) and short intracortical inhibition (SICI). These mea-sures were assessed bymeans of a paired-pulse paradigm

and RMT Values of the Enrolled Subjects

DISEASE DURATION, Y(MEAN 6 SD)

RMT LEFT SIDE/HEADACHE

SIDE (MEAN 6 SD)

RMT RIGHT SIDE/NONHEADACHE SIDE(MEAN 6 SD)

— 49.1 6 4.1 48.1 6 4.6

11.4 6 6.6 50.1 6 9.3 49.6 6 8.1

14.2 6 11.7 48.2 6 9.2 48.5 6 6.8

Cosentino et al The Journal of Pain 55

with a subthreshold conditioning stimulus set to 80% ofthe RMT followed by a testing stimulus delivered at120% RMT intensity. The interstimuli intervals usedwere 2 and 10 milliseconds, representing, respectively,the inhibitory and facilitatory limbs of the paired-pulseparadigm.43 Three conditions (test stimulus alone andpaired stimulation at interstimuli intervals of 2 and10 milliseconds) were applied 8 times each, intermixedin a random order based on single trials, and separatedby 10-second intertrial intervals. The effect of condition-ing stimuli on motor evoked potential (MEP) amplitudewas determined as the ratio of the average amplitudeof conditioned MEP to the average amplitude of uncon-ditioned test MEP. The second measurement of corticalexcitability was cortical silent period (CSP) induced bysingle-pulse TMS delivered at 120% RMT intensity. CSPduration was measured in 8 trials in the moderatelyactive contralateral abductor pollicis brevis muscle asthe interval between the stimulation artifact and the re-turn of continuous voluntary electromyographic activity.The average value of 8 rectified traces was calculated ineach subject. The third measurement of cortical excit-ability was input-output curves (I-O curves) measuredby stimulation at 6 intensity levels, ranging from 100 to150% of RMT in steps of 10%. Eight stimuli were re-corded at each intensity level in a random order, with a10-second interstimuli interval, needed to avoid anyinterference between 2 successive pulses.10 The ampli-tude of MEPs elicited by single magnetic stimuli wascalculated peak-to-peak and then averaged for eachstimulation intensity.

Statistical AnalysisThe mean values of all electrophysiological measures

obtained in single subjects were submitted to statisticalanalyses. Two-way repeated-measures analyses of vari-ance (ANOVAs) were performed for RMT, ICF, SICI, andCSP. The within-subject factor was Side (headache sidevs nonheadache side in CH patients and left vs rightside in the HVs); the between-subject factor was Disease

Figure 1. Electrophysiological measures by paired-pulse TMS and Cmary motor cortex ipsilateral and contralateral to the headache sideamplitude of conditioned MEP to the average amplitude of unconvalues recorded from the contralateral hemisphere and with respecaverage amplitude of conditionedMEP to the average amplitude of uto values recorded from the contralateral hemisphere. (C) Mean val

(episodic CH inside bout vs episodic CH outside bout vsHVs). A 3-way ANOVA including Side and Stimulus inten-sity (6 levels from 100 to 150% RMT in steps of 10%) aswithin-subject factors, and Disease as a between-subjects factor was conducted to evaluate changes inthe I-O curves.If ANOVA showed significant differences, Duncan post

hoc testwas used formultiple comparisons ofmeans. Thesphericity assumption was checked by using Mauchly’stest, and Huynh-Feldt’s correction was adopted, if neces-sary, for the degrees of freedom.Pearson’s test was used to check for correlation of the

electrophysiological measures with the clinical charac-teristics (ie, disease duration, duration of the bout andtime from the end of the last attack in CH patients eval-uated during the bout, time from the end of the last boutin CH patients outside the cluster period). For all ana-lyses, the statistical significance was set at P values lowerthan .05. All statistics were calculated with Statistica 7.0software (StatSoft, Tulsa, OK).

ResultsThe experimental procedures werewell tolerated in all

subjects and no adverse effects were documented. Asregards RMT, there were neither significant differencesbetween the 2 patient groups and controls nor differ-ences between the 2 hemispheres in each group(Table 1). No significant differences were found for theclinical and demographic characteristics between CH pa-tients evaluated outside and inside bout (Table 1).ANOVA used to evaluate ICF (Fig 1A) showed a signif-

icant effect of both factors Side (F(1, 35) = 4.73, P < .05) andDisease (F(2, 35) = 3.55, P < .05). Post hoc analysis showedin patients both inside and outside bout that ICF wasmore pronounced in the hemisphere ipsilateral to theheadache side with respect to the contralateral one, aswell as to values recorded from both hemispheres inHVs (P < .05). ANOVA for SICI (Fig 1B) revealed a signifi-cant effect of factor Side (F(1, 35) = 4.69, P < .05). At thepost hoc analysis, significant higher SICI values (P < .05),

SP recorded from patients and HVs after stimulation of the pri-. (A) Mean values of ICF determined as the ratio of the averageditioned test MEP. *Significant variations (P < .05) compared tot to HVs. (B) Mean values of SICI determined as the ratio of thenconditioned testMEP. *Significant variation (P < .05) comparedues of CSP. Error bars indicate standard errors of means (SEs).

56 The Journal of Pain Cortical Excitability in Cluster Headache

indicating reduced intracortical inhibition, were foundin the motor cortex ipsilateral to the headache sidecompared to the contralateral one in patients evaluatedinside bout. There were no significant main effects in theANOVA results for CSP (Fig 1C).Three-way ANOVA for I-O curves (Fig 2) showed a sig-

nificant effect of the factors Side (F(1, 35) = 6.76, P < .02),Disease (F(2, 35) = 3.27, P < .05), and Stimulus intensity(F(5, 175) = 92.96, P < .0001), as well as a significant inter-action between Side and Stimulus intensity(F(5, 175) = 2.35, P < .05) and betweenDisease and Stimulusintensity (F(10, 175) = 2.97, P < .002). Post hoc analysisshowed that peak-to-peak MEP amplitude increasedwith increasing stimulus intensity in both patient groupsand controls. However, a steeper slope was recordedfrom both hemispheres of CH patients evaluated outsidebout (Fig 2B), and from themotor cortex contralateral tothe headache side in patients inside bout (Fig 2C) ascompared to HVs (Fig 2A). The correlation analysis testsshowed that in patients outside bout, the RMT recordedfrom both hemispheres was positively correlated withdisease duration (r = .66, P < .02, and r = .58, P < .05, inthe hemisphere ipsilateral and contralateral to theheadache side, respectively) (Fig 3).

DiscussionTo our knowledge, this is the first TMS study per-

formed in patients having CH. Our findings provide evi-

Figure 2. I-O curve plots of mean MEP peak-to-peak amplitude inrecorded from the primary motor cortex ipsilateral and contralateraand zP < .00001 from MEP amplitude at an intensity of 100% RMT.

dence of increased cortical excitability in episodic CHpatients. Interestingly, abnormal electrophysiologicalvalues were recorded from both hemispheres, in spiteof the strict one-sidedness of the attacks.

Cortical Excitability in CH Outside BoutIn CH patients evaluated outside bout, evidence of

increased cortical excitability was shown bilaterally butwas greater in the motor cortex ipsilateral to the head-ache side. Indeed, although the I-O curves were steeperin both hemispheres, increased ICF was observed onlyipsilateral to the attack side. This pattern of cortical hy-perexcitability resembles that seen in migraine, wheremore pronounced ICF and greater slope of the I-O curveshave been documented in the interictal period.14,21,33,34

Our findings of abnormal cortical excitability outsidebout suggest that a cortical dysfunction could be pri-marily involved in the pathogenesis of CH. Indeed, itappears unlikely that the cortical excitability changeswere in relation to previous cluster periods becauseneither ICF nor slope of the I-O curves was correlatedwith disease duration or time from the end of thelast bout. Moreover, the cortical abnormalities cannotbe directly related to the metabolic changes observedin the hypothalamus, or in the other structures of thepain neuromatrix, because these have been docu-mented during the cluster attacks but not in remissionperiods.23,24,36,38

HVs (A), and in CH patients outside bout (B) and inside bout (C),l to the headache side. Significant variations *P < .05, yP < .001,Error bars indicate standard errors of means (SEs).

Figure 3. Correlations between individual RMT values recorded after stimulation of the primary motor cortex ipsilateral (A) andcontralateral (B) to the headache side and disease duration (years) in CH patients evaluated outside bout.

Cosentino et al The Journal of Pain 57

The finding of interhemispheric asymmetry for ICF inCH patients, different from that observed in HVs, isworthy of being discussed. One of the key features ofCH is the strict one-sidedness of the attacks, both withinthe bout and among different cluster periods. Theincreased ICF seen in the motor cortex ipsilateral to theCH points to a more increased excitability of this hemi-sphere and could reflect the pathophysiological pro-cesses on which the attack’s side depends. A firsthypothesis is that a functional interhemispheric imbal-ance between cortico-subcortical pathways involved inthe regulation of transmission of nociceptive informa-tion could lead to a preferential side for the attack occur-rence. Another intriguing possibility is that cortical areasof one hemisphere, because of a more pronounced hy-perexcitability, could undergo a hyperactivation stateresponsible for changes in the concentration of variousvasoactive and nociceptive substances; these substances,spreading through the extracellular compartment, couldreach and activate the perivascular trigeminal fibers.7

Experimental evidence in favor of such mechanism hasbeen provided for pathogenesis of headache inmigrainewith aura.6

Another relevant datum to be discussed relates to thepositive correlation between RMT and disease durationseen in patients outside bout. The physiological mecha-nisms modulating RMT are not completely understood,though there is evidence that both glutamatergic neuro-transmission and factors modulating neuronal mem-brane excitability may be involved.17,42 The increase inthe RMT values with disease duration could beinterpreted either as an attempt to compensate for theunderlying cortical hyperexcitability state or as anexpression of a cumulative neuronal damage. Wecannot infer what specific mechanism(s) might beresponsible for the increase in RMT values, it beingpossible to hypothesize both changes in intracorticalcircuits and/or downregulation of subcortical structuresresponsible for the cortical preactivation level.2

Cortical Excitability in CH Inside BoutIn patients evaluated inside bout, a significantly

reduced SICI was observed along with an increased ICFin the hemisphere ipsilateral to the headache side. Thisindicates a further increase in the cortical excitability

level due to reduced activity of inhibitory intracorticalcircuits, as attenuation of MEP amplitudes by pairedstimuli at interstimuli intervals of 2 milliseconds isbelieved to be mediated by GABAergic intracorticalinterneurons.15

Theoretically, the reduction of SICI could representeither a pain-relatedmodification or the result of changesin cortical excitability that lead to the beginning of thecluster period. In a recent study, Schabrun and Hodgesshowed increase of SICI after an experimental pain stim-ulus, possibly because of cortico-cortical projections be-tween the primary sensory and motor cortices, but notSICI reduction as seen in our patients.32 Moreover, it isnoteworthy that we recorded changes in SICI not fromthe hemisphere directly receiving contralateral nocicep-tive inputs, but from the primary motor cortex ipsilateralto the pain side. These considerations argue in favor ofasymmetricdysregulationof corticalexcitability that couldbe primarily involved in CH pathogenesis, rather than be-ing expression of pain-related modifications.It should be noted that although we found lower CSP

values in CH patients compared to HVs, the differencedid not reach statistical significance. The CSP representsanother measure of central inhibition of motor path-ways, and some considerations should be made toexplain the different behavior of CSP compared to SICIseen between patients and controls: 1) CSP and SICI arethought to reflect the activities of different inhibitoryneural circuits26; and 2) duration of CSP strictly dependson stimulation intensity, so that we cannot excludethat significant differences could be observed by usingintensities of magnetic stimulation different from120% RMT. In this regard, it is interesting that shorterCSP values have been recorded in migraine with aura pa-tients only at the intensity of the RMT and not at supra-threshold intensities.3

Another notable finding of the study is that in theaffected side, the slope of the I-O curves normalized dur-ing the bout. This could appear in contrast with theincreased ICF and reduced SICI values seen in the samehemisphere. It should be considered, however, that thepaired-pulse TMS paradigm and the I-O curves investi-gate different aspects of cortical excitability. In partic-ular, ICF and SICI are thought to reflect the activity ofspecific intracortical circuits, that is, the glutamatergicand GABAergic ones, respectively.15,43 Differently, the

58 The Journal of Pain Cortical Excitability in Cluster Headache

delivery of single TMS stimuli at high stimulationintensity determines the massive, reverberatingactivation of both facilitatory and inhibitoryintracortical interneurons together with a directactivation of the pyramidal tract neurons, thusallowing investigation of the overall excitability oflarger neuronal populations.17 There is evidence thatthe amplitude ofMEPs elicited by singlemagnetic stimuliapplied to the primary motor cortex decreases followinga pain stimulus.32 However, in the present results, thereduced slope of the I-O curves cannot be easily relatedto the headache pain, as it was not observed in the hemi-sphere that primarily received the trigeminal nociceptiveinput, but rather in the contralateral one (headacheside). As an alternative explanation, we suggest that ina condition of increased cortical excitability, homeostaticmechanisms down-regulating cortical responsivity couldbe activated by stressful stimuli to avoid a further, poten-tially dangerous increase of the cortical activation level.This hypothesis is in line with findings in migraine withand without aura in which, along with cortical hyperex-citability, a reduced threshold for inducing inhibitory ho-meostatic cortical responses to TMS has been observed inthe motor cortex.11,13

Conclusions and Final RemarksTaken together, the present results suggest that a cere-

bral cortex dysfunction may be involved in the patho-genesis of CH. Our findings also support the view thatmigraine and CH might share, at least in part, commonpathophysiological pathways.22,41 This suggestion alsocomes from several lines of evidence such as the higherprevalence of migraine in patients with CH compared

to the general population,30 and the presence ofcommon clinical features and environmental triggersamong migraine and CH patients.30,40,41

There are some methodological considerations andlimitations of the current study to mention. First, re-cordings outside and inside bout were not obtainedin the same subjects. Theoretically, longitudinal studieswould be more sensitive in determining cortical excit-ability changes related to the recurrence of the clusterperiods. Then, here we investigated the excitability ofthe primary motor cortex, and further studies will beneeded to assess whether our results are generalizableto other cortical areas. Future focused studies shouldbe carried out also to clarify how different corticaland subcortical structures interact with each other inthe pathogenesis of CH. Indeed, although the hypo-thalamus likely plays a major role in determiningsome distinctive features of the disease, it is wellknown that its activity is under control of modulatoryinfluences from cortical areas and subcortical struc-tures.2,28 Conversely, it is remarkable that theorexinergic system projects widely throughout thebrain.18 Thus, we could suppose that subclinical hypo-thalamic dysfunctions, not detectable by means ofneuroradiologic evaluations in the remission periods,could affect function of several brain areas. Finally,we cannot exclude that different cortical and subcor-tical abnormalities could also coexist in CH, ratherthan being dependent on one another. If so, it couldbe supposed that in a condition of abnormal corticalhyperexcitability, a concurrent hypothalamic dysfunc-tion could be the factor that switches the phenotypefrom migraine to CH.

References

1. Absinta M, Rocca MA, Colombo B, Falini A, Comi G,Filippi M: Selective decreased grey matter volume of thepain-matrix network in cluster headache. Cephalalgia 32:109-115, 2012

2. Afra J, Ertsey C, Bozsik G, Jelencsik I: Cluster headache pa-tients show marked intensity dependence of cortical audi-tory evoked potentials during and outside the bout.Cephalalgia 25:36-40, 2005

3. Aurora SK, al-Sayeed F, Welch KM: The cortical silentperiod is shortened in migraine with aura. Cephalalgia 19:708-712, 1999

4. BahraA,MayA, Goadsby PJ: Cluster headache: A prospec-tive clinical study in 230 patients with diagnostic implica-tions. Neurology 58:354-361, 2002

5. Bartsch T, Levy MJ, Knight YE, Goadsby PJ: Differentialmodulation of nociceptive dural input to [hypocretin]orexin A and B receptor activation in the posterior hypotha-lamic area. Pain 109:367-378, 2004

6. Bolay H: The first phase of amigraine attack resides in thecortex. J Neural Transm 119:569-574, 2012

7. Bolay H, Reuter U, Dunn AK, Huang Z, Boas DA,Moskowitz MA: Intrinsic brain activity triggers trigeminal

meningeal afferents in a migraine model. Nat Med 8:136-142, 2002

8. Brighina F, Cosentino G, Fierro B: Brain stimulation inmigraine. Handb Clin Neurol 116:585-598, 2013

9. Casale MS, Baratto M, Cervera C, Gallamini M,Lynch G, Gjini K, Boutros NN: Auditory evoked potentialabnormalities in cluster headache. Neuroreport 19:1633-1636, 2008

10. Chen R, Classen J, Gerloff C, Celnik P, Wassermann EM,Hallett M, Cohen LG: Depression of motor cortex excitabilityby low-frequency transcranial magnetic stimulation.Neurology 48:1398-1403, 1997

11. Cosentino G, Brighina F, Talamanca S, Paladino P,Vigneri S, Baschi R, Indovino S, Maccora S, Alfonsi E,Fierro B: Reduced threshold for inhibitory homeostatic re-sponses in migraine motor cortex? A tDCS/TMS study. Head-ache 54:663-674, 2014

12. Cosentino G, Fierro B, Brighina F: From different neuro-physiological methods to conflicting pathophysiologicalviews in migraine: A critical review of literature. Clin Neuro-physiol 125:1721-1730, 2014

13. Cosentino G, Fierro B, Vigneri S, Talamanca S, Paladino P,Baschi R, Indovino S, Maccora S, Valentino F, Fileccia E,Giglia G, Brighina F: Cyclical changes of cortical excitabilityand metaplasticity in migraine: Evidence from a repetitive

Cosentino et al The Journal of Pain 59

transcranial magnetic stimulation study. Pain 155:1070-1078, 2014

14. Cosentino G, Fierro B, Vigneri S, Talamanca S, PalermoA,Puma A, Brighina F: Impaired glutamatergic neurotransmis-sion in migraine with aura? Evidence by an input-outputcurves transcranial magnetic stimulation study. Headache51:726-733, 2011

15. Di Lazzaro V, Oliviero A, Meglio M, Cioni B,Tamburrini G, Tonali P, Rothwell JC: Direct demonstrationof the effect of lorazepam on the excitability of the humanmotor cortex. Clin Neurophysiol 111:794-799, 2000

16. Evers S, Bauer B, Suhr B, Voss H, Frese A, Husstedt IW:Cognitive processing is involved in cluster headache butnot in chronic paroxysmal hemicrania. Neurology 53:357-363, 1999

17. Hallett M: Transcranial magnetic stimulation: A primer.Neuron 55:187-199, 2007

18. Holland PR, Goadsby PJ: Cluster headache, hypothala-mus, and orexin. Curr Pain Headache Rep 13:147-154, 2009

19. International Headache Society: International classifica-tion of headache disorders, 3rd ed, beta version. Cephalagia33:629-808, 2013

20. Keel JC, Smith MJ, Wassermann EM: A safety screeningquestionnaire for transcranial magnetic stimulation. ClinNeurophysiol 112:720, 2001

21. Khedr EM, Ahmed MA, Mohamed KA: Motor and visualcortical excitability in migraineurs patients with or withoutaura: Transcranial magnetic stimulation. Neurophysiol Clin36:13-18, 2006

22. Larner AJ: Concurrence of primary headaches: Lane andDavies (2006) revisited. J Headache Pain 10:133, 2009

23. May A: New insights into headache: An update on func-tional and structural imaging findings. Nat Rev Neurol 5:199-209, 2009

24. May A, Bahra A, B€uchel C, Frackowiak RS, Goadsby PJ:Hypothalamic activation in cluster headache attacks. Lancet352:275-278, 1998

25. Miranda-Cardenas Y, Rojas-Piloni G, Mart�ınez-Lorenzana G, Rodr�ıguez-Jim�enez J, L�opez-Hidalgo M,Freund-Mercier MJ, Cond�es-Lara M: Oxytocin and electricalstimulation of the paraventricular hypothalamic nucleusproduce antinociceptive effects that are reversed by anoxytocin antagonist. Pain 122:182-189, 2006

26. Paulus W, Classen J, Cohen LG, Large CH, Di Lazzaro V,Nitsche M, Pascual-Leone A, Rosenow F, Rothwell JC,Ziemann U: State of the art: Pharmacologic effects oncortical excitability measures tested by transcranial mag-netic stimulation. Brain Stimul 1:151-163, 2008

27. Perrotta A, Serrao M, Sandrini G, Bogdanova D,Tassorelli C, Bartolo M, Coppola G, Pierelli F, Nappi G:Reduced habituation of trigeminal reflexes in patientswith episodic cluster headache during cluster period. Ceph-alalgia 28:950-959, 2008

28. Risold PY, Thompson RH, Swanson LW: The structural or-ganization of connections between hypothalamus and cere-bral cortex. Brain Res Brain Res Rev 24:197-254, 1997

29. Rossi S, Hallett M, Rossini PM, Pascual-Leone A, Safety ofTMS Consensus Group: Safety, ethical considerations, andapplication guidelines for the use of transcranial magneticstimulation in clinical practice and research. Clin Neurophy-siol 120:2008-2039, 2009

30. Rozen TD: Cluster headache with aura. Curr Pain Head-ache Rep 15:98-100, 2011

31. Sandrini G, Alfonsi E, Ruiz L, Pavesi G, Micieli G,Manzoni GC, Mancia D, Nappi G: Impairment of cornealpain perception in cluster headache. Pain 47:299-304, 1991

32. Schabrun SM, Hodges PW: Muscle pain differentiallymodulates short interval intracortical inhibition and intra-cortical facilitation in primary motor cortex. J Pain 13:187-194, 2012

33. Silberstein SD, Niknam R, Rozen TD, Young WB: Clusterheadache with aura. Neurology 54:219-221, 2000

34. Siniatchkin M, Kr€oner-Herwig B, Kocabiyik E,Rothenberger A: Intracortical inhibition and facilitation inmigraine—A transcranial magnetic stimulation study. Head-ache 47:364-370, 2007

35. Smith MJ, Keel JC, Greenberg BD, Adams LF, Schmidt PJ,Rubinow DA, Wassermann EM: Menstrual cycle effects oncortical excitability. Neurology 53:2069-2072, 1999

36. Sprenger T, Boecker H, Tolle TR, Bussone G, May A,Leone M: Specific hypothalamic activation during a sponta-neous cluster headache attack. Neurology 62:516-517, 2004

37. Sprenger T, Goadsby PJ: What has functional neuroi-maging done for primary headache. and for the clinicalneurologist? J Clin Neurosci 17:547-553, 2010

38. Teepker M, Menzler K, Belke M, Heverhagen JT,Voelker M, Mylius V, Oertel WH, Rosenow F, Knake S: Diffu-sion tensor imaging in episodic cluster headache. Headache52:274-282, 2012

39. van Vliet JA, Vein A, Le Cessie S, Ferrari MD, van Dijk JG:Dutch RUSSH Research Group. Impairment of trigeminalsensory pathways in cluster headache. Cephalalgia 23:414-419, 2003

40. Vingen JV, Pareja JA, Stovner LJ: Quantitative evaluationof photophobia and phonophobia in cluster headache.Cephalalgia 18:250-256, 1998

41. Young WB, Peres MF, Rozen TD: Modular headache the-ory. Cephalalgia 21:842-849, 2001

42. Ziemann U, L€onnecker S, Steinhoff BJ, Paulus W: Effectsof antiepileptic drugs on motor cortex excitability inhumans: A transcranial magnetic stimulation study. AnnNeurol 40:367-378, 1996

43. Ziemann U, Rothwell JC, Ridding MC: Interactionbetween intracortical inhibition and facilitation in humanmotor cortex. J Physiol 496:873-881, 1996