(4.1.4)--7、Graymattervolumereductionrefle脑科学与影像新技术.pdf

《(4.1.4)--7、Graymattervolumereductionrefle脑科学与影像新技术.pdf》由会员分享，可在线阅读，更多相关《(4.1.4)--7、Graymattervolumereductionrefle脑科学与影像新技术.pdf（7页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、Gray matter volume reduction reflects chronic pain in trigeminal neuralgiaMark Obermanna,1,Rea Rodriguez-Raeckec,1,Steffen Naegela,Dagny Hollea,Daniel Muellera,Min-Suk Yoona,Nina Theysohnb,Sebastian Blexb,Hans-Christoph Dienera,Zaza KatsaravaaaDepartment of Neurology,University of Duisburg-Essen,Ger

2、manybInstitute for Diagnostic and Interventional Radiology and Neuroradiology,University of Duisburg-Essen,GermanycDepartment of Neurology,Hannover Medical School,Germanya b s t r a c ta r t i c l ei n f oArticle history:Accepted 14 February 2013Available online 26 February 2013Keywords:Voxel-based

3、morphometryTrigeminal neuralgiaFacial painChronic painNeuronal plasticityTrigeminal neuralgia(TN)is supposedly caused by an ectatic blood vessel affecting the trigeminal nerve at theroot entry zone of the brain stem.Recent evidence suggests an additional central component within trigeminalpain-proce

4、ssing in the pathophysiology of TN.Therefore,we aimed to identify specific brain regions possiblyassociated with the development or maintenance of TN using magnetic resonance imaging(MRI)voxel-basedmorphometry(VBM).Sixty patients with classical TN were compared to 49 healthy controls.Eighteen patien

5、ts had TN with concomitantconstant facial pain,a condition previously described as a predictor of worse treatment outcome.We found gray matter(GM)volume reduction in TN patients compared to healthy controls in the primarysomatosensory and orbitofrontal cortices,as well as the in the secondary somato

6、sensory cortex,thalamus,insula,anterior cingulate cortex(ACC),cerebellum,and dorsolateral prefrontal cortex.GM volume decreasewithin the ACC,parahippocampus,and temporal lobe correlated with increasing disease duration in TN.There were no differences comparing patients with and without concomitant c

7、onstant facial pain.No GMincrease was found comparing patient subgroups with each other and with healthy controls.The observed changes probably reflect the impact of multiple,daily attacks of trigeminal pain in these patientssimilar to what was previously described in other chronic pain conditions a

8、nd may be interpreted as adaptationmechanism to chronic pain in regard to neuronal plasticity.The ACC,parahippocampus and temporal lobevolumereduction inparallel with disease duration may pointtoa pivotalrole ofthesestructures inchronicpain.2013 Elsevier Inc.All rights reserved.IntroductionAccording

9、 to current opinion classical trigeminal neuralgia(TN)iscaused by a proximal compression of the trigeminal nerve root closeto the brainstem(root entry zone)by a tortuous or ectatic blood vessel(artery or vein)leading to mechanical compression of nerve fibers andsecondary demyelination,probably media

10、ted by microvascular ische-mic damages(Marinkovic et al.,2007).While Jannetta(1967)initiallydescribed88%ofhisinvestigatedpatientstohavea nervevesselconflict(usually the superior cerebellar artery)(Jannetta,1967),more recentinvestigations demonstrated that not all patients have a nerve vesselconflict

11、 and that between 25 to 49%of people without any clinicalsignsofTNshowanervevesselcontactonmagnetic-resonanceimaging(Adamczyk et al.,2007;Kakizawa et al.,2008).However,the quickpainrelief following microvascular decompression surgery in almost 90%ofpatients is a strong indicator for the relevance of

12、 this mechanism,butlacksexplanationwhycertainpatients(i.e.,30%)experiencerecurrenceof their complaints in the long run(Gronseth et al.,2008;Zakrzewskaand Akram,2011).It was suggested that hyperexcitability of the com-pressed nerve is necessary but alone insufficient to cause the disease.In turn a ne

13、rve-vessel conflict may represent a risk factor for the devel-opment of TN but alone does not elicit the disease(Hamlyn and King,1992).Abnormal expression of voltage-gated sodium channels wasdetected in patients with TN and a channelopathy was discussed as itspathophysiological correlate(Siqueira et

14、 al.,2009).Nav1.7,Nav1.3,and Nav1.8 were affected and are responsible for rapid activation andinactivation as well as maintenance of the action potential.They areco-expressed in nociceptive neurons of the dorsal root ganglions(Siqueira et al.,2009).Possible involvement of central factors comesmore a

15、nd more into focus of current research,but with indistinct rolefor the pathophysiology of TN.A recent voxel-based morphometry(VBM)study showed gray matter(GM)changes in the thalamus,puta-men,anterior and posterior insula,primary somatosensory cortex in amixed patient group with TN(N=8)and trigeminal

16、 neuropathy(N=13)(Gustin et al.,2011).Central facilitation and resulting centralNeuroImage 74(2013)352358 Corresponding author at:Department of Neurology,University of Duisburg-Essen,Hufelandstr.55,45122 Essen,Germany.Fax:+49 201 723 5953.E-mail addresses:mark.obermannuni-due.de(M.Obermann),r.rodrig

17、uez-raeckeuke.de(R.Rodriguez-Raecke),steffen.nageluk-essen.de(S.Naegel),dagny.holleuk-essen.de(D.Holle),daniel.muelleruk-essen.de(D.Mueller),min-suk.yoonuk-essen.de(M.-S.Yoon),nina.theysohnuk-essen.de(N.Theysohn),sebastian.blexuk-essen.de(S.Blex),hans.dieneruni-duisburg-essen.de(H.-C.Diener),zaza.ka

18、tsaravauk-essen.de(Z.Katsarava).1Authors contributed equally.1053-8119/$see front matter 2013 Elsevier Inc.All rights reserved.http:/dx.doi.org/10.1016/j.neuroimage.2013.02.029Contents lists available at SciVerse ScienceDirectNeuroImagejournal homepage: of the trigeminal system initiated or sustaine

19、d byperipheral mechanisms(e.g.,nerve-vessel conflict,channelopathy)werediscussed(Borsook et al.,2007;Obermann et al.,2007).Whether centralmechanisms are part of the underlying cause of TN or merely a conse-quenceofthediseaseremainsindiscussion.Recentinvestigationsfocusedon TN with persistent concomi

20、tant dull background pain between thetypical paroxysmal neuralgic attacks as strong indicator of the possiblerole of central mechanisms in TN(Obermann et al.,2007).This conditionmust not be confused with trigeminal neuropathy that describes similarpain characteristics but is due to a lesion or traum

21、a to the nerve oftenassociated with sensory neurological deficit(Nurmikko and Eldridge,2001).Central allodynic mechanisms that may engage the nociceptiveneurons at the trigeminal nucleus,thalamic and cortical levels weresuspected to be the correlate of this persistent pain(Borsook et al.,2007;Oberma

22、nn et al.,2007).The clinical relevance was highlighted indifferent studies clearly demonstrating that concomitant backgroundpain is associated with poor medical and surgical outcome(Aggarwal etal.,2010;Obermann et al.,2008;Sandell and Eide,2008;Szapiro et al.,1985).Our objective was(i)to detect regi

23、onal GM volume changes inpatients with TN compared to healthy controls in order to identifyspecific brain areas that may be associated with the developmentandpersistenceofthisdebilitatingfacialpaincondition,and(ii)todetectdifferences between TN patients with and without concomitant persis-tent facia

24、l pain as morphological correlate for the central component inthe pathophysiology of TN as suspected previously.MethodsSubjectsSixty patients(36 women)with classical TN were recruited prospec-tively from a tertiary headache center(West German Headache Center)between February 2007 and March 2010.The

25、diagnosis of TN wasreconfirmed in a face-to-face interview by headache experienced neu-rologists(MO,DM,MY,ZK)according to the International Classificationof Headache Disorders(ICDH-II)(Benoliel et al.,2008;InternationalHeadache Society,2004).All patients had active TN painful attacks attime of study

26、 inclusion.Additional neurological examination wasunremarkable in all patients.No patient had prior micro-vasculardecompression(MVD)surgery or other invasive treatments for TN(i.e.,gasserian ganglion procedures).Patients were compared to 49healthy controls of the same age-and gender(28 women).Exclus

27、ioncriteria were symptomatic TN(e.g.,due to MS),post-herpetic neuralgiaand other causes of facial pain,any other primary headache disorder(e.g.,migraineortensiontypeheadache),otherseveresomaticorpsychi-atricillnesses,pastsurgicaltreatmentofTNorrefractorytomedicaltreat-ment.Patients with neurological

28、 deficits such as hyper-or hypoesthesia,aswellasallodyniawerenotincludedinthestudy.Fordemographicsandpatient characteristics see Table 1.Standard protocol approvals,registrations,and patient consentsAll participants gave their written informed consent according tothe Declaration of Helsinki prior to

29、 study inclusion.The local ethicscommittee of the University of Duisburg-Essen approved the studyprotocol.Voxel based morphometry(VBM)-data acquisitionImaging was performed on a Siemens Sonata scanner operating at1.5 T using a standard 8-channel birdcage head coil.T1-weightedmagnetic resonance imagi

30、ng 3D datasets were obtained usingmagnetization prepared rapid acquisition gradient echo(MP-Rage)sequence(TR/TE/TI=2400 ms/4.38 ms/1200 ms,flip angle=8,field of view=256 mm,160 slices,voxel size 1 1 1 mm3).Thesame scanner and the same scanning protocol were used for all patientsand healthy controls.

31、All images were evaluated by an experiencedneuro-radiologist(NT)and screened for symptomatic causes of TN.Processing of structural dataVBM is a whole-brain technique capable of revealing subtle,specificchanges in GM by applying voxelwise statistics within the context ofGaussian random fields(Ashburn

32、er and Friston,2000).It has been crossvalidated with region-of-interest measurements and functional data inseveral studies(Maguire et al.,2000;May et al.,1999;Richardson et al.,2004).Data processing and analysis was performed with SPM8(http:/www.fil.ion.ucl.ac.uk/)including the VBM8 toolbox(http:/db

33、m.neuro.uni-jena.de/vbm/)and DARTEL(Diffeomorphic Anatomical RegistrationThrough Exponentiated Lie Algebra)using Matlab(Matlab 7,The Math-Works,Natick,MA,USA).Preprocessing involved spatial normalizationintotheMontrealNeurologicalInstitute(MNI)templatespace,graymat-ter segmentation,spatial smoothing

34、,and modulation in order to adjustfor volume changes during spatial normalization(Ashburner andFriston,1997;Friston,1995;Wrightetal.,1995).Thespatiallynormalizedand modulatedGMpartitions were smoothed withanisotropic Gaussiankernel of 8 mm full-width at half maximum(FWHM).Additionally,25images were

35、flipped to the symptomatic side after preprocessing,as pre-viously recommended(Good et al.,2001a).We calculated the lateralityindex for the flipped subjects and also used a symmetric template.Allimages have to be regarded as pain located on the right side of the face.Unflipped images were also analy

36、zed to avoid bias from brain asymme-try,but provided the same results with lesser effect strength.Statistical analysisStatisticalanalysistestedGMvolumedifferencesbetweenTNpatientsand healthy controls.The full factorial model included three groups:1.)healthy controls,2.)classical trigeminal neuralgia

37、 patients(TN),and 3.)trigeminal neuralgia patients with concomitant persistent facial pain(cTN).To control for age related GM changes,age was implemented inour statistical model as covariate.To avoid any unintentional bias by apriori hypotheses in the primary analysis a whole brain analysis wasperfo

38、rmed.To correct for multiple comparisons,a region of interest(ROI)analysis was performed based on anatomic structures previouslyTable 1Demographics and patient characteristics.TN totalTNcTNHealthycontrolsNumber of patients60421849Gender women/men36/2424/1812/628/21Age years62 13.2(range:3186)61.5 13

39、(range:3186)63.3 14.1(range:3585)61.8 9(range:4376)Duration of illnessyears8.3 6.7(range:129)8.5 5.8(range:122)8.2 7.5(range:129)n.a.Attack frequencyper day8.5 10.1(range:180)9.1 10.2(range:180)8.1 7.6(range:140)n.a.Pain locationV1/V2/V3%26/86/6023/81/6628/86/49n.a.Pain intensity VRS7.7 1.87.8 2.16.

40、9 1.2n.a.Nerve-vessel contactdetected on MRI493811n.a.Medication%Carbamazepine535149n.a.Gabapentin272838n.a.Pregabalin141513n.a.None660n.a.TN=trigeminal neuralgia;TN=classical trigeminal neuralgia;cTN=trigeminalneuralgia with concomitant chronic facial pain;VRS=verbal rating scale(0=nopain;10=worst

41、possible pain);n.a.=not applicable.353M.Obermann et al./NeuroImage 74(2013)352358described to be associated with central pain processing(i.e.,primary andsecondary somatosensory cortex,thalamus,striatum,anterior cingulatecortex,insula,orbitofrontal and dorsolateral prefrontal cortex,middleand inferio

42、r temporal gyrus,precuneus,hippocampus and cerebellum)(Apkarian et al.,2004;Draganski et al.,2006;Kim et al.,2008;May,2011;Obermann et al.,2009a;Rocca et al.,2006;Schmidt-Wilcke et al.,2005,2007,2010;Valfre et al.,2008;Weissman-Fogel et al.,2011).TheROI was adapted from the AAL(anatomical automatic

43、labeling)ROIlibrary(Tzourio-Mazoyer et al.,2002)where the above mentionedregions were conjoined to one ROI using the MarsBaR toolbox(Brettet al.,2002).Correction for multiple comparisons was performed withfamily wise error correction(FWE)p b 0.05 within the predefinedROI.No additional extent thresho

44、ld was applied.The same analysiswas done using unflipped data to avoid effects due to brain asymmetry(Table 4 supp.).Subgroup analysis comparing patients with nerve-vessel conflict and those without was performed,as well as analysis ofpatients with and without concomitant constant background pain.Ad

45、ditionally,attackfrequency,diseaseduration,meanpainintensityon a verbal rating scale(VRS;0=no pain,10=worst imaginablepain)and age were implemented as covariates in a linear regressionmodel.Therefore,investigating specific correlations with certain covar-iates should not be confounded by other covar

46、iates.The primary analy-sis was performed for the whole brain uncorrected at a threshold ofp b 0.001 and FWE corrected for multiple comparisons at a thresholdof p b 0.05.The same ROI described above was applied with FWE cor-rection at a threshold of p b 0.05.ResultsDemographics and clinical characte

47、ristics are summarized in Table 1.No differences in age(t-test t=0.302,df=112,p=0.763)or gender(chi=3.63,df=1,p=0.055)were found comparing TN patientsandhealthycontrolgroups.Cross-sectionalanalysisrevealedregionaldif-ferences in GM volume between healthy controls and TN patients.GMvolume decrease in

48、 the patient group was found in the primary somato-sensory and orbitofrontal cortex as well as the ACC,insula,secondarysomatosensory cortex,thalamus,putamen,caudate nucleus,dorsolateralprefrontal cortex,precuneus,and cerebellum(Fig.1;Table 2).Groupcomparison of TN patients with and without concomita

49、nt constant facialpain revealed no differences between these two patient subgroups.No increase in regional GM volume was detected at a threshold ofp b 0.001 uncorrected for the whole brain as well as in the ROI analysis.No major difference in results was detected using unflipped data in ourstatistic

50、al model(Table 4 supplement).No differences between patientswith and without nerve-vessel conflict were detected.Regression analysisRegression analysis testing for a correlation of number of attackfrequency,mean individual headache severity on a verbal ratingscale(VRS),age or gender with decrease or