Cervicalspinemanipulationalterssensorimotorintegration:AsomatosensoryevokedpotentialstudyHeidiHaavik-Taylor*,BernadetteMurphyHumanNeurophysiologyandRehabilitationLaboratory,DepartmentofSportandExerciseScience,TamakiCampus,UniversityofAuckland,PrivateBag92019,261MorrinRoad,GlenInnes,Auckland,NewZealandAccepted11September2006Availableonline29November2006AbstractObjective:To studythe immediatesensorimotor neurophysiologicaleectsof cervical spinemanipulationusingsomatosensory evokedpotentials(SEPs).Methods:Twelve subjects with a history of reoccurring neck stiness and/or neck pain, but no acute symptoms at the time of the studywereinvitedtoparticipateinthestudy. Anadditional twelvesubjectsparticipatedinapassiveheadmovement control experiment.Spinal (N11, N13) brainstem (P14) and cortical (N20, N30) SEPs to median nerve stimulation were recorded before and for 30 min afterasinglesessionofcervicalspinemanipulation,orpassiveheadmovement.Results:There was a signicant decrease in the amplitude of parietal N20 and frontal N30 SEP components following the single sessionofcervical spinemanipulationcomparedtopre-manipulationbaselinevalues. Thesechangeslastedonaverage20 minfollowingthemanipulationintervention.Nochangeswereobservedinthepassiveheadmovementcontrolcondition.Conclusions:Spinal manipulation of dysfunctional cervical joints can lead to transient cortical plastic changes, as demonstrated by atten-uationofcorticalsomatosensoryevokedresponses.Signicance:This study suggests that cervical spine manipulation may alter cortical somatosensory processing and sensorimotor integra-tion. These ndings may help to elucidate the mechanisms responsible for the eective relief of pain and restoration of functional abilitydocumentedfollowingspinalmanipulationtreatment.2006InternationalFederationofClinicalNeurophysiology.PublishedbyElsevierIrelandLtd.Allrightsreserved.Keywords: Cervicalspinemanipulation;Human;Somatosensoryevokedpotentials; Brainplasticity;Somatosensorysystem;Sensorimotorintegration1.IntroductionSpinal manipulationis acommonlyusedconservativetreatment for neck, back, and pelvic pain. The eectivenessof spinal manipulation in the treatment of acute and chron-ic low back and neck pain has been well established by out-come-basedresearch(forreview, seeHurwitzetal., 1996and;Vernon,1996).However,themechanism(s)responsi-blefortheeectivereliefofpainandrestorationoffunc-tional ability after spinal manipulation are not wellunderstood, asthereislimitedevidencetodateregardingtheneurophysiologicaleectsofspinalmanipulation.Theevidence todate indicates that spinal manipulationcanlead to alterations in reex excitability (Herzog et al.,1999; Murphyet al., 1995; Symons et al., 2000), alteredsensory processing (Zhuet al., 2000, 1993) andalteredmotor excitability (Herzog et al., 1999; Dishmanet al.,2002;Suteretal.,2000).Spinal manipulation is used therapeutically by a numberof healthprofessionals, includingphysical medicine spe-cialists, physiotherapists, osteopaths and chiropractors.The dierent professions have dierent terminology forthe entity or manipulable lesion that they manipulate.Thismanipulablelesionmaybecalledvertebral (spinal)lesionbyphysicalmedicalspecialistsorphysiotherapists,somatic dysfunction or spinal lesion byosteopaths,and vertebral subluxation or spinal xation by1388-2457/$32.002006InternationalFederationofClinicalNeurophysiology.PublishedbyElsevierIrelandLtd.Allrightsreserved.doi:10.1016/j.clinph.2006.09.014*Corresponding author. Tel.: +649 3737599x82544; fax: +649 3737043.E-mailaddress:h.taylor@auckland.ac.nz(H.Haavik-Taylor).www.elsevier.com/locate/clinphClinicalNeurophysiology118(2007)391402chiropractors (Leach, 1986). Joint dysfunction as discussedintheliteraturerangesfromexperimentallyinducedjointeusion(Shakespeare et al., 1985), topathological jointdiseasesuchasosteoarthritis(OConnoret al., 1993), aswell as the more subtle functional alterations that are com-monly treated by manipulative therapists (Suter et al.,1999,2000).Forthepurposesofthispaper,themanipu-lable lesion will be referred to as an area of spinaldysfunction.There is a growing body of evidence suggesting that thepresence of spinal dysfunction of various kinds has aneect oncentral neural processing. For example, severalauthors have suggested spinal dysfunction may lead toalteredaerent input tothe CNS(BoltonandHolland,1996, 1998; Murphyet al., 1995; Zhuet al., 1993, 2000).AlteringaerentinputtotheCNSiswell knowntoleadto plastic changes in the way that it responds to any subse-quent input (Brasil-Neto et al., 1993; Byl et al., 1997; Hal-lettetal., 1999; Pascual-LeoneandTorres, 1993). Neuralplastic changes take place bothfollowingincreased(Bylet al., 1997; Pascual-Leone and Torres, 1993) anddecreased (Brasil-Neto et al., 1993; Hallett et al., 1999; Zie-mannetal.,1998)aerentinput.Altered aerent input from joints can lead to both inhi-bitionandfacilitationof neural inputtorelatedmuscles.Numerousstudiesofpainful jointshaveshownarthroge-nous muscle inhibition (Hides et al., 1994; McPartlandet al., 1997; Stokes and Young, 1984). However, even pain-less experimentallyinducedjoint dysfunction(joint eu-sion) has been shown to inhibit surrounding muscles(Shakespeareetal.,1985).Thisalteredmotorcontrolwasalso shown to persist even after aspiration of the joint eu-sion(Shakespeareet al., 1985). Intheearly1980s, Stein-metzetal. demonstratedthatrelativelyshort(1530 min)episodesofmoderatelyintenseaerentinputtothespinalreex pathways of rats causes increases in neural excitabil-itythat persistsforseveral hours(Steinmetzet al., 1982,1985). Once these facilitatedareas are established, theremaybe noneedfor ongoingaerent input tomaintainthealteredoutputpatterns.Sincetheseearlyexperiments,numerous studies have shown rapid central plastic changesafterinjuriesandalteredsensoryinputfromthebody(forreview, see Wall et al., 2002). This can explain the ndingsofShakespeareetal. (1985)ofalteredmotorcontrol per-sisting even after aspiration of the joint eusion. This pro-cess provides a potential explanation for alteredneuralprocessingasaresultofjointdysfunction,andarationalefortheeectsofspinalmanipulationonneuralprocessingthathavebeendescribedintheliterature.Given that spinal dysfunction would alter the balance ofaerent input to the CNS we propose that this altered aer-ent input may over time lead to potential maladaptive neu-ral plasticchanges intheCNS. Wefurther proposethatspinal manipulation can eect this. By recording SEPsandmonitoringtheperipheral nerveaerent volley, it ispossible to determine where in the somatosensory pathwaychanges induced by spinal manipulation may be occurring.2.Methods2.1.SubjectsTwelve subjects (ve women and seven men), aged2053(meanage29.9), participatedinthespinal manip-ulation study. An additional twelve age-matched sub-jects (4 males and 8 females), aged 2135 (mean age27.1 years), participated in the passive head movementcontrol study. The subjects were allocated into eithergroup in a pseudo-randomized order. It was decidedinadvancethat therst 12volunteers that t theinclu-sion/exclusion criteria for the study would become themanipulation group as this would enable the next 12subjects to be age matched, if needed. However, bothgroupswereof similarages, sonoadditional age-match-ing was necessary. All 24 subjects agreedtohave theircervical spines manipulated and/or their head movedby the researcher, and no subject knew which grouptheyweretakingpart inprior totheir experimental ses-sion taking place. To be included subjects could nothaveahistoryof neurological disease. Thesubjectswererequiredtohave a history of reoccurring neckpainorstiness (e.g., repeatedlypresent duringthe performanceof certaintaskssuchasworkorstudy). However, at thetime of the experiment all subjects were requiredtobepainfree. Thiswasdoneinordertoassessthepotentialeects of joint manipulation delivered to dysfunctionaljoints alone without the presence of acute pain, as thepresenceof painaloneisknowntoinducedasignicantreduction of the post-central N20P25 complex and asignicant increase of the N18 wave (Rossi et al.,2003). Table 1 contains the experimental subjectsdetails, including their neck complaint history andknownpast neck(and/or head) trauma. Informedcon-sent was obtained and the local ethical committeeapprovedthe study.2.2.SomatosensoryevokedpotentialsAll SEPrecordingelectrodes (7 mmAg/AgCl Hydro-spotdisposableadhesiveelectrodes fromPhysiometrix)wereplacedaccordingtotheInternational FederationofClinical Neurophysiologists (IFCN) recommendations(Nuwer et al., 1994). Recording electrodes were placedonthe ipsilateral Erbs point, over the C6spinous pro-cess (Cv6), and 2 cmposterior to contralateral centraland frontal scalp cites C3/4 and F3/4, which will bereferred to as Cc0, and Fc0, respectively. All recordingelectrodes were referenced to the ipsilateral earlobe.The C6 spinous electrode was also referenced to theanterior neck(tracheal cartilage). The Erbs point elec-trode andthe central Cc0electrode were alsoreferencedtothe contralateral shoulder, as SEPcomponents origi-nating fromsubcortical regions are best recorded witha non-cephalic reference (Ulas et al., 1999). AgroundelectrodewasattachedtoFz. Stimuli consistedofelectri-392 H.Haavik-Taylor,B.Murphy/ClinicalNeurophysiology118(2007)391402cal square wave pulses of 1 ms duration deliveredthrough Grass gold cup 7 mm electrodes (imped-ance < 5 kX). The stimulatingelectrodes (cathode distal)were placed over the median nerve at the wrist of thedominant arm. Stimuli were delivered at 1.25 timesmotor thresholdusingaconstant current stimulationatarateof1.98 Hz, aratethatdoesnotleadtoSEPatten-uation (Fujii et al., 1994; GarciaLarrea et al., 1992).Sweep length was 55 ms (5 ms pre-stimulus and 50 mspost-stimulus) and ltering bandwidth was 31000 Hz(6 dBoctaveroll-o). Atotal of 500sweepswereaver-agedusingapurposewrittenLabView5.1programandtheaveragedwaveformwasdisplayedinananalysispan-el fromwhichthe waveforms of interest were measuredforamplitudeandlatency. Theamplitudeof theindivid-ual SEPcomponents was measuredfromtheir peaktothe preceding or succeeding trough according to theIFCNguidelines (Nuwer et al., 1994). Accordingtotheguidelines the N11 peak amplitude was measured tothe preceding positive trough and the N13 to the suc-ceedingpositive trough(Nuwer et al., 1994). The laten-cies were recorded at their maximal peak of eachcomponent. We identied and analyzed the followingSEP components: the peripheral N9, the spinal N11andN13, thefar-eldP14N18complex(whichwasana-lyzedfromboththefrontal andparietal recordingsites),the parietal N20 (P14N20 and N20P27 complexes),andthefrontal N30(P22N30complex).2.3.Interventions2.3.1.SpinalmanipulationinterventionThis interventionconsistedof spinal manipulationofthe subjects dysfunctional cervical joints, which wasdeterminedbyaregisteredchiropractor. Theclinical evi-dence of joint dysfunctionincludes tenderness topalpa-tion of the relevant joints, restricted intersegmentalrangeofmotion,palpableasymmetricintervertebralmus-cletension, abnormal or blockedjoint playandend-feelof a joint, and sensorimotor changes in the upperextremity (Fryer et al., 2004; Hestboek and Leboeuf-Yde, 2000). The most reliable spinal-dysfunction-indica-toristendernesswithpalpationofthedysfunctionaljoint(Hubka and Phelan, 1994; Jull et al., 1988). Cervicalrange of motion (Rheault et al., 1992; Youdas et al.,1992) has also been shown to have good inter- andintra-examiner reliability. For the purpose of this studyspinal dysfunctionwas therefore denedas the presenceof both restricted intersegmental range of motion andtendernesstopalpationof thejoint at at least onecervi-cal spine segment. This was detected in the followingmanner. The examiner, a registered chiropractor withatleastsevenyearsofclinicalexperience,wouldpassivelymove the subjects head, while palpatingandstabilizingover the zygapophyseal joints. For eachspinal segmenttheheadwouldbegentlyandpassivelymovedfromneu-tral positiontothe maximal range of lateral exioninTable1ExperimentalsubjectdetailsNo. Gender Age Handedness Neckcomplaindetails,andknownneck(orhead)injury1 F 28 R 13yearsofongoing,intermittentneckpainandstiness.Motorvehicleaccident14yearsago,whiplashinjury2 M 33 L 5yearsreoccurringmildneckstinesswithcertainactivities.Numeroussoccersportsinjuries3 M 33 R 2yearsofmildreoccurringneckstinesswhenworkingatacomputer.Felloutofswingontoheadandneckatage12receiving3stitchestohead4 F 33 R 12yearsofalmostconstantneckpain,acheandstiness,worsewithworkingatcomputer andwithstress.Falloaswingatage14landingonface5 F 26 R 3yearsreoccurringneckstinessandtensionwhenstudying.Noknownheadorneckinjuries6 F 24 R Ongoingnecktensionpast10years,occasionalneckache.Oneconcussion7 M 53 R Reoccurringmildneckacheandstinesspast3years.Threesignicantheadandneckinjuriesfromsoccer8 M 25 R Necktensionwithstudylast2years.Noknownheadorneckinjuries9 M 20 R Neckachewithcomputerworklast2years.Noknownheadorneckinjuries10 M 20 R 14yearsofreoccurringneckpainandstiness,particularlywithstress.Noknownheadorneckinjuries11 F 38 R 6yearsreoccurringnecktensionandache,worsewithstressandstudy.Hitheadontableaschildreceivingseveralstitches12 M 20 R 13yearsofongoing,intermittentneckpainandstiness.Motorvehicleaccident14yearsago,whiplashinjuryTable showing the spinal manipulation subjects details, including age, gender, handedness, history of cervical pain or tension, and any known neck and/orheadtrauma.H.Haavik-Taylor,B.Murphy/ClinicalNeurophysiology118(2007)391402 393thecoronal plane, toboththeleft andtheright. If thismovementappearedrestricted, theexaminerwouldapplygentle pressure tothe joint, while watchingfor signs ofdiscomfort fromthe subject. The examiner would alsoaskthe subject if the pressure tothe joint elicitedpain.Spinal segments that were deemed both restricted inlateral exion range of motion and elicited pain onpalpationwere denedfor the purpose of this studytobedysfunctional.Thespinalmanipulationscarriedoutinthisstudywerehigh velocity, low amplitude thrusts to the spine held in lat-eral exion, with slight rotation and slight extension. This isastandardmanipulativetechniqueusedbymanipulativephysicians, physiotherapists, and chiropractors. Themechanical properties of this type of CNS perturbation havebeen investigated, and although the actual force applied tothesubjects spinedepends onthetherapist, thepatient,andthespinal locationoftreatment,thegeneralshapeofthe force-time history of spinal manipulation is very consis-tent (Hessell et al., 1990), andthe durationof the thrustisalways less than200 ms (for review, see Herzog, 1996).The high velocity type of manipulation was chosenspecicallysince previous research(Herzoget al., 1995)hasshownthatreexEMGactivationobservedfollowingmanipulation only occurred following high-velocitylow amplitude manipulations (as compared to lowervelocity mobilizations), and would therefore be more likelyto alter aerent input to the CNS and lead to measurable SEPchanges.2.3.2.PassiveheadmovementinterventionThe passive headmovement interventionwas carriedout by the same chiropractor whohadpre-checkedthesubjects for spinal dysfunction and who performed thespinal manipulations for thespinal manipulationexperi-ment. The passiveheadmovement interventioninvolvedthe subjects head being passively laterally exed, andslightlyextendedandrotatedtoapositionthat thechi-ropractorwouldnormallymanipulatethatpersonscervi-cal spine, and then return the subjects head back toneutral position. This was repeatedtoboththeleft andthe right. However, the experimenter was particularlycareful not to put pressure on any individual cervicalsegment. Loading a joint, as is done prior to spinalmanipulation, hasbeenshowntoalterparaspinal propri-oceptiveringinanesthetizedcats (Pickar andWheeler,2001), and was therefore carefully avoided by endingthe movement prior to end-range-of-motion whenpassivelymovingthesubjects heads. Nospinal manipu-lationwasperformedduringanypassiveheadmovementexperiment. Thepassiveheadmovement experiment wasnotintendedtoactasashammanipulationbuttoactasa physiological control for possible changes occurringdue tothe cutaneous, muscular or vestibular input thatwould occur with the type of passive head movementinvolved in preparing a subject/patient for a cervicalmanipulation.2.4.ExperimentalprotocolAll thesubjects cervical spineswererstcheckedbyaregistered chiropractor to determine if and where theirspineswouldbemanipulated. Ifthesubjectswerejudgedto have cervical spine dysfunction the relevant information(including detailed medical history) was then obtained. Allsubjects were also screened for evidence of vertebral arteryischemia, with their head in a position of extension, lateralexionandrotation, whichareneckpositions showntohave the greatest mechanical stress tothe contra-lateralvertebral artery(Arnoldet al., 2004). Subjects werealsoscreened for other contraindications for cervical manipula-tion, suchasrecent historyoftrauma, knownconditionssuchasinammatoryorinfectiousarthropathies,orbonemalignancies.Threepre-intervention(baseline) SEPtrials werethenrecorded. Followingthis, the subjects wouldeither havetheir heads passivelymoved(as describedinthe passivehead movement intervention section) or the subjects cervi-cal spinewasmanipulatedatthepredeterminedlevels(asdescribedinthespinalmanipulationinterventionsection).Thepassiveheadmovements andmanipulationswerecar-riedout with thesubjectstillseatedinthe slightly reclinedLa-Z-boychair to minimize any disturbance to therecording electrodes. The spinal segments manipulatedfor eachsubject inthemanipulationgroupinthis studyareshowninTable2. Immediatelyfollowingeitherinter-vention the three post-intervention SEP trials were record-ed. The rst post-interventionSEPtrial (trial 010 min)was recorded within the rst 10 min post-intervention.The secondpost-interventionSEPtrial (trial 1020 min)wasrecordedwithin1020 minpost-interventionandthethird post-intervention SEP trial (trial 2030 min) wasrecordedwithin2030 minpost-intervention.2.5.StatisticalanalysisPrior to any SEP peak analysis the data les were codedbyanindependentpersontoreduceanybiasduringSEPTable2SpinalsegmentmanipulatedSubjectNo. Spinalsegmentsmanipulatedinthisstudy1 C0/1,C3/4andC6/72 C2/3andC6/73 C2/3andC6/74 C1/2andC6/75 C0/1,C3/4andC6/76 C2/3andC6/77 C1/2andC5/68 C1/2,C3/4andC5/69 C2/3andC4/510 C2/3andC6/711 C2/3andC6/712 C2/3,C4/5Tableshowingthespinal segmentsmanipulatedforeachsubjectinthisstudy.394 H.Haavik-Taylor,B.Murphy/ClinicalNeurophysiology118(2007)391402peak amplitude and latency analysis. The rst baseline SEPtrialwasusedtofamiliarisethesubjectswiththeelectricalstimulationneededtoelicit SEPs andwas thereforedis-carded for all subjects. Thefollowing two pre-interventionSEP trials (trials 2 and 3) of both latencies and amplitudeswereexaminedwiththenon-parametricWilcoxonSignedRankstest(astheSEPamplitudevaluesdidnotfollowaGaussiandistribution) priortocombiningthemintooneset to ensurethat there were no statistically signicant dif-ferences. Onceit was determined that therewere no statis-tically signicant dierences, the two pre-intervention trialswere collapseddownintopre-interventionmeanvalues.The pre-intervention means and post-intervention data (tri-als 010 min, 1020 min, and 2030 min) were then decod-ed, grouped according to intervention and transferred intotheSPSSstatistical packagefor statistical analysis. Notethatthefrontal P14N18andtheparietal P14N20wereanalysedpost hocat thesuggestionof reviewers. ResultsofthoseSEPcomponentsshouldthereforebeinterpretedaccordingly.As the SEP amplitude values did not follow a Gaussiandistributionnon-parametric analysis was carriedout forourstatisticalevaluation.Non-parametricanalysisorvar-iance for repeated measures (Friedman test) was applied tolatency and amplitude values for each SEP peak. Wheneverasignicanteectwasfound(p < 0.05),plannedcompari-sons (Wilcoxon signed rank test) were executed. Theplannedcomparisons that were made were betweenthepre-intervention averaged results and the subsequentpost-interventionresults. Thelevel ofsignicancewassetat p < 0.05. Plannedcomparisonswerechoseninsteadofpost hoc analysis to minimize Type 1 error, without reduc-ingthesensitivitytorelevanteects(Perneger,1998).Thisis alsoinaccordancewithprevious SEPresearch(Rossietal.,2003;Rossietal.,2002).3.Results3.1.SubjectstatisticsTable1 showsthespinalmanipulationsubjects details,includingage, gender, historyofcervical painortension,andanyknownneckand/orheadtrauma. Table2showsthespinal segments manipulatedfor eachsubject inthisstudy.3.2.PassiveheadmovementinterventionThere were no statistically signicant dierencesbetween the pre-intervention trials prior to the passive headmovementintervention.TherewerenosignicantchangesobservedineitherlatencyoramplitudeofanyoftheSEPpeaks followingpassive headmovement (see Fig. 1andTable3). Fig. 1depicts thepercentageof changeintheaveraged normalized SEP peak amplitudes SEbefore(pre-intervention mean), and after (trials 010 min, 1020 minand2030 min) thesinglesessionof passiveheadmovements (the upper bar graphs A) andcervical spinemanipulation(thelowerbargraphsB).3.3.SpinalmanipulationinterventionThere were no statistically signicant dierences betweenthe pre-intervention trials priorto the spinal manipulationintervention. Pre-intervention representational traces ofthe variousSEP componentsof asinglesubjectare showninFig.2, withbothpre-interventiontrialstoshowrepro-ducibility of SEP recordings. There were no statistically sig-nicantchangesforthelatencyofanypeakinthespinalmanipulationcondition. The peripheral N9, spinal N11andN13, andbrainstemP14N18(measuredfrombothfrontal and parietal recording sites) SEP peaks did not showanysignicantamplitudechangesfollowingcervicalspinemanipulation (see Fig. 1). However, the two cortical peaks,parietal N20andfrontal N30, signicantlydecreasedinamplitude following the cervical spine manipulations (Pari-etal P14N20 Friedman non-parametric ANOVAp = 0.025, parietal N20P27 Friedman non-parametricANOVAp = 0.006andfrontal P22N30Friedmannon-parametric ANOVA p = 0.009), particularly during the rst20 min. ThesignicantchangesaremarkedwithastarinFig. 1 and Table 3. There was variabilitybetween the sub-jects, however the trend of the changes were similar for allsubjects.TheparietalP14N20,theparietalN20P27andthefrontal P22N30peakamplitudesdecreasedforeverysinglesubject duringtherst post-manipulationtrial. Bythe second post-manipulation trial only one of the subjectsparietal P14N20 peak amplitudes returned to baseline lev-els and only two were at baseline values for the third post-manipulationtrial. Forparietal N20P27onlytwoofthesubjects peakamplitudes returnedtobaseline levels bythesecondpost-manipulationstrialandbothremainedatbaselinevalues for thethirdpost-manipulationtrial. Onthe other hand, four of the subjects frontal P22N30 peakamplitudeshadreturnedtobaselinevaluesbythesecondpost-manipulation trial, and by the third post-manipulationtrial half the subjects frontal P22N30 peak amplitudes hadreturned to baseline values. Table 3 contains the raw aver-agedSEPdata for boththe spinal manipulationgroupand the passive head movement group.The greatest amplitude attenuationoccurredwiththefrontal P22N30 SEPcomplex. It was onaverage over30% attenuated during the rst 10 min following the cervi-cal spinal manipulation(trial 010 min), thenonaverage15%attenuatedduringthenext 10 min(trial 1020 min)(see Fig. 1). These changes were signicant with theplannedcomparisons (Wilcoxonsignedranks test) (trial010 min; p = 0.002, and trial 1020 min; p = 0.019).Fig. 3 depicts the attenuation of the frontal P22N30SEPcomponent for an individual subject. The parietalP14N20complexremainedsignicantlyattenuatedwiththe planned comparisons (Wilcoxon signed ranks test) dur-ing all three post-manipulation trials, 010 min (p = 0.012),1020 min (p = 0.025), and 2030 min (p = 0.036) (seeH.Haavik-Taylor,B.Murphy/ClinicalNeurophysiology118(2007)391402 395Fig.1).TheparietalN20P27complexalsoremainedsig-nicantly attenuated with the planned comparisons (Wilco-xonsignedrankstest)duringall threepost-manipulationtrials, 010 min (p = 0.021), 1020 min (p = 0.010), and2030 min(p = 0.010) (seeFig. 1). Theparietal P14N20amplitudewasattenuatedby1218%ofthepre-interven-tionamplitude.TheparietalN20P27amplitudewasonlyattenuatedby 811%of the pre-interventionamplitude.Figs. 4 and 5 depict spinal, parietal and frontal representa-tional traces of thevarious SEPcomponents of asinglesubject prior toandafter boththe spinal manipulationand passive head movement interventions, respectively.Thehorizontal dashedlines represent theonset, peakorsubsequent trough of the various SEP components forthepre-interventionconditiontodemonstrateclearlytheattenuation of both cortical SEP components (parietalN20 and frontal N30) following spinal manipulation. Noteno changes occurred to any of the SEP components for thissubject following the passive head movement controlcondition.4.DiscussionThe major nding in this study was that a single sessionof spinal manipulationof dysfunctional jointsresultedinattenuatedcortical(parietalN20andfrontalN30)evokedresponses. Onaveragethefrontal N30changes persistedfor20 minpost-manipulationbeforereturningtobaselinelevels. The parietal N20 changes persisted for at least30 min,i.e. duringallthreepost-manipulationrecordings.Toour knowledge, noprior studyhas shownpersistentchangesinneithersomatosensoryprocessingnorsensori-motorintegrationfollowingspinalmanipulation.4.1.EvidenceforcorticalneuralplasticchangesfollowingspinalmanipulationAs the peripheral N9 peak, representing the aerent vol-leyinthebrachialplexus(CraccoandCracco,1976;Des-medt andCheron, 1981; Jones, 1977), was unalteredthechanges observedinour studymost likelyreect centralFig.1. BargraphsshowingthepercentageofchangeintheaveragednormalizedSEPpeakamplitudes SEbefore(pre-interventionmean),andafter(trials 010 min, 1020 min and 2030 min) a single session passive head movements (the upper bar graphs A) and cervical spine manipulation (the lowerbar graphs B). Note no signicant changes were observed following the passive head movement control intervention. The signicant changes that occurredfollowingthespinalmanipulationsessionareindicatedwithastar(*).396 H.Haavik-Taylor,B.Murphy/ClinicalNeurophysiology118(2007)391402changes. However, theexact mechanismsofthesecentralchanges canonlybehypothesized. Theseplasticchangeslastedfortheentire30 minof recordingsfortheparietalN20SEPcomponent.ThefrontalN30SEPpeakchangeslasted on average 20 min post-manipulation and thenreturned to baseline values. Some individual subjects dem-onstratedchangesthatpersistedduringallthreepost-ma-nipulationtrials(forexampleseeFig.3forlastingfrontalN30peakchanges). Othersubjectsdemonstratedchangesinonlytherst(010 min)post-manipulationtrial.AlthoughnobrainstemSEPpeakamplitude changeswereobservedinthisstudy, thedesignofthestudydoeslimittheabilitytoexcludethepossibilitythatsub-corticalchanges didoccur. It is generally agreed that although500sweepsmaybesucienttorecordreliableperipheralErbsandcorticalSEPpotentials,far-eldpotentialssuchas subcortical P14N18 do generally require a higher num-ber of averaged sweeps (F. Mauguiere, 1999; Nuwer et al.,1994). The possibility for sub-cortical SEP changes follow-ing spinal manipulation does therefore need furtherinvestigation.4.2.TheparietalN20SEPpeakchangesTheparietalN20SEP component,generated in thepri-mary somatosensory cortex (S1) (Desmedt andCheron,1980;F.Mauguiere,1999;Nuweretal.,1994)wassignif-icantly decreased during all post-manipulation blocks, bothwhen measured from P14 to N20 and when measured fromTable3Mean raw amplitude (lV) and latency (ms) data with standard deviations of individual SEP components for the pre-intervention mean, and the three post-interventiondata,trials010 min,1020 min,and2030 minN11 N13 P14N18(Fc0) P14N18(Cc0) P14N20 N20P27 P22N30SM Control SM Control SM Control SM Control SM Control SM Control SM ControlPre-interventionAmplitude 0.98 1.08 2.67 2.72 1.22 1.46 1.31 1.65 2.65 3.02 2.85 3.97 2.61 2.29SD 0.31 0.29 0.53 0.54 0.49 0.58 0.54 0.54 1.05 1.07 1.03 1.78 1.41 1.35Latency 12.1 12.4 13.7 13.8 15.1 15.2 17.6 17.6 19.7 19.7 19.7 19.7 30.5 31.9SD 0.88 1.04 0.77 0.98 0.98 0.98 1.23 1.06 1.09 1.42 1.09 1.42 0.72 1.41Trail010 minAmplitude 0.95 1.12 2.74 2.75 1.16 1.38 1.29 1.65 2.28*2.99 2.70*4.04 2.02*2.37SD 0.30 0.35 0.53 0.38 0.30 0.64 0.59 0.54 0.87 1.06 1.08 1.83 1.30 1.43Latency 12.0 12.3 13.7 13.8 15.1 15.2 17.8 17.7 19.7 19.7 19.7 19.7 30.6 31.4SD 0.92 1.07 0.83 0.96 0.98 0.93 1.37 1.19 1.09 1.42 1.09 1.42 1.03 1.33Trail1020 minAmplitude 0.93 1.16 2.69 2.70 1.14 1.42 1.31 1.60 2.23*3.00 2.58*3.97 2.29*2.26SD 0.38 0.32 0.62 0.63 0.43 0.63 0.49 0.44 1.02 0.97 0.95 1.74 1.3 1.4Latency 12.0 12.3 13.7 13.7 15.0 15.1 17.8 17.5 19.6 19.5 19.6 19.5 30.53 31.52SD 0.95 0.98 0.92 0.87 1.17 1.02 1.21 1.27 1.27 1.31 1.27 1.31 0.96 1.37Trial2030 minAmplitude 0.96 1.11 2.86 2.69 1.27 1.40 1.24 1.46 2.38*3.11 2.56*3.59 2.43 1.83SD 0.26 0.33 0.49 0.35 0.39 0.59 0.54 0.31 0.96 0.94 0.97 1.80 1.10 1.36Latency 11.9 12.0 13.5 13.5 15.0 15.0 17.5 17.2 19.5 19.2 19.5 19.2 30.7 31.9SD 0.89 0.78 0.92 0.82 0.95 0.91 1.37 1.15 1.20 1.14 1.20 1.14 0.91 1.26Note that the P14 SEP peak latency data is included under the P14N18 (Fc0) column and the N18 SEP peak latency data is included under the P14N18(Cc0)column.*p < 0.05comparedtobaselinevalues.Fig. 2. Erbs point (Erb), cervical (Cv5), post-central (Cc0), andpre-central (Fc0) SEPtracesfromonerepresentativesubject priortospinalmanipulation. Traces are superimposedtoshowreproducibilityof themeasured peaks. The Cc0was recorded with a non-cephalic reference. TheFc0wasrecordedwithacephalicreference(ipsilateralearlobe).TheCv5spinous electrode wasreferencedto theanterior neck(tracheal cartilage).The Erbs point electrode and the central Cc0electrode were alsoreferencedtothecontralateral shoulderforanalysisof subcortical SEPcomponents.H.Haavik-Taylor,B.Murphy/ClinicalNeurophysiology118(2007)391402 397N20 to N27. This suggests that S1 processing wasreducedpost manipulation. One possible contribution to thereducedparietal N20 couldcome fromenhancedactiveinhibition. For example, thalamocortical aerents mono-synapticallyactivateanintracortical feed-forwardinhibi-tion to area 3b pyramidal cells supposed to generateparietal N20(Desmedt andCheron, 1980; F. Mauguiere,1999;Nuweretal.,1994)viathesummationofexcitatorypostsynaptic potentials. The cervical spine manipulation(s)could have altered the aerent information originatingfromthecervicalspine(fromjoints,muscles,etc),inturnalteringthe waythat the 3bpyramidal cells respondtoanysubsequent aerent input suchas the mediannervestimulationinourstudy.Fig. 3. ChangestotheN30SEPcomponentofasinglesubjectrecordedimmediatelybeforeandafterasessionofcervical spinemanipulations. Theaveragedpre-interventionN30peakisdepictedinthickblack, thepost-interventiontrials(trials010 min, 1020 minand2030 min)010 min, 1020 min, and 2030 min post-manipulation, respectively, are depicted in blue, green and purple, respectively. This particular subject displayed attenuatedpeakamplitudeduringallthreepost-manipulationtrials.Fig. 4. Cervical (Cv5), post-central (Cc0), and pre-central (Fc0) SEP traces from one representative subject prior to and after cervical spine manipulation.Horizontal dashed lines represent the onset, peak or subsequent trough of the various SEP components for the pre-intervention condition to demonstrateclearly the attenuation of the cortical SEP components observed following spinal manipulation (parietal N20 and frontal N30). The Cc0 was recorded witha non-cephalic reference, the Fc0was recorded with a cephalic reference (ipsilateral earlobe), and the Cv5 spinous electrode was referenced to the anteriorneck(trachealcartilage).398 H.Haavik-Taylor,B.Murphy/ClinicalNeurophysiology118(2007)391402The passive headmovement SEPexperiment demon-stratedthat nosignicant changes occurredfollowingasimple movement of the subjects head. Our results arethereforenotsimplyduetoalteredinputfromvestibular,muscleorcutaneousaerentsasaresultofthechiroprac-torstouchorduetotheactualmovementofthesubjectshead.Thisthereforesuggeststhattheresultsinthisstudyarespecictothedeliveryofthehigh-velocity,low-ampli-tudethrust tothedysfunctional joints. Thepassiveheadmovement experiment was to control for the potential neu-ralchangesduetotheaerenttracresultingfromtouchandheadmovement alone. It was not intendedtobe ashammanipulation.It is however alsopossible that subject arousal levelsdieredbetweenthetwogroups andthis maybeacon-foundingvariableinthisstudy. It isforexampleknownthatsleepcanprolongthelatencyandalterthemorphol-ogy of the parietal N20 component of median nerveSEPsinhealthysubjects(Emersonetal., 1988). Thesub-jects inthe manipulationgroupmayhave beenanxiousabout the manipulations andtherefore more alert thanthe subjects inthe passive headmovement group. As itis not possible to performa shamadjustment, this isaninherent problemwiththis type of interventionthatis dicult toovercome. It is the authors opinion thatit isunlikelythat thesubjectsinthepassiveheadmove-ment control condition were merely more relaxedthroughout the study for several reasons. First of alltherewerenochangesobservedinanySEPpeaklatencyinthis study, andlatency changes are knowntooccurwithalterations inarousal levels (Emersonet al., 1988).Furthermore, the subjects frontal N30 SEPcomponentreturnedtobaselinebythethirdpost-manipulationtrial.It wouldnot make sense that the subjects were anxiouspriorthemanipulations, thenall relaxedinstantlyfollow-ing the manipulations, and then became anxious againtowards theendof thestudy.4.3.ThefrontalP22N30SEPpeakchangesThegreatest amplitudechangewas observedwiththefrontalN30componentoftheSEPpeaks.Althoughsomeauthors suggest this peak is also generated in the post-cen-tral cortical regions (i.e. S1) (Allisonet al., 1989b, 1991,1989a),mostevidencesuggeststhatthispeakisrelatedtoacomplexcortical andsubcortical looplinkingthebasalganglia, thalamus, pre-motor areas, andprimary motorcortex (Kanovskyet al., 2003; Mauguiere et al., 1983; Ros-sini etal., 1987, 1989; Waberski etal., 1999). ThefrontalN30 peak is therefore thought to reect sensorimotor inte-gration(Rossi et al., 2003). The attenuatedfrontal N30SEPpeakobservedinour studytherefore suggests thattheremaybeadecreaseof activityinthesecortical andsubcorticalloopslinkingthebasalganglia,thalamus,pre-Fig. 5. Cervical (Cv5), post-central (Cc0), andpre-central (Fc0)SEPtracesfromonerepresentativesubjectpriortoandafterpassiveheadmovement(control condition). Horizontal dashedlinesrepresenttheonset, peakorsubsequenttroughofthevariousSEPcomponentsforthepre-interventioncondition to demonstrate clearly that no changes occur following the passive head movement control condition. The Cc0 was recorded with a non-cephalicreference, the Fc0was recorded with a cephalic reference (ipsilateral earlobe), and the Cv5 spinous electrode was referenced to the anterior neck (trachealcartilage).H.Haavik-Taylor,B.Murphy/ClinicalNeurophysiology118(2007)391402 399motor areas and primary motor cortex resulting from medi-an nerve stimulation, immediately following spinal manipu-lation, lasting on average 20 min. Again this may be due toaltered aerent input following spinal manipulation.PreviousresearchhasshownthatthefrontalN30com-ponent has independent cortical generators with a separatethalamocortical input (Balzamoet al., 2004; Mauguiereet al., 1983). Intracortical human recordings have alsoshownaerentinformationfollowing median nerve stimu-lationprojectdirectlytotheprimarymotorcortex(Balz-amoet al., 2004), ashasbeendocumentedpreviouslyinprimates (Strick and Preston, 1982; Tanji and Wise,1981). Thefrontal N30componentisthereforesubjecttomorecomplex inputsthan thoseowingon fromthe pari-etalN20generatoralone.ThatthefrontalN30amplitudeattenuationwasonaverage20%greaterthantheparietalN20 amplitude attenuationcouldbe a reectionof theadditional inputs from the separate thalamocorticalpathways.4.4.TheaerentmediatorsofthecentralneuraleectsofspinalmanipulationThecurrentstudysupportspreviousworkthatsuggestmuscle aerents (probably Ia) are the most likely mediatorsof the central neural eects of spinal manipulation (BoltonandHolland,1996,1998;Murphyetal.,1995;PickarandWheeler,2001;Zhuetal.,2000,1993).Asmentioned,theN30 peakwasthe mostdramatically attenuated SEPpeakinthis study. Previous workhas demonstratedthat thefrontal N30peakisthemost vulnerableSEPcomponenttovibrationinterference(HoshiyamaandKakigi, 2000).Vibrationof relaxedmuscles, asusedinHoshiyamaandKakigisstudy, 2000, predominantlystimulatesthegroupIa bres (from muscles spindles) (Roll et al., 1989). GroupIb and II bres are not sensitive to this type of stimulation(Roll et al., 1989). This study demonstrates that the frontalN30 componentis highlysensitiveto changing levels of Iaaerentinput(HoshiyamaandKakigi, 2000). Thesigni-cant attenuation of the frontal N30 SEP componentobserved in the current study thus suggests that spinalmanipulation may alter Ia aerent processing. This is con-sistentwith previousanimal studiesthathavedemonstrat-ed that displacement of vertebrae is signaled to the centralnervoussystembyaerentnervesarisingfromdeepinter-vertebral muscles (Bolton and Holland, 1996, 1998; Pickarand Wheeler, 2001). In particular, both the velocity and rel-ative position of the vertebral displacement appeared to beencoded by aerent nerve activity from intervertebral mus-cles(BoltonandHolland,1996,1998).If Iaaerents arethemediators of thecentral neuraleectsdocumentedfollowingspinal manipulationof dys-functional joints, the persistent attenuation of the N20component, seeninourstudy, mayreectareductionofactivityatS1followingspinalmanipulation.Itispossiblethat joint dysfunction leads to a bombardment of theCNS with Ia aerent signaling from the surrounding inter-vertebral muscles. Bolton and Hollands (1996, 1998)observations of increased Ia activity from the paravertebralmusculature surrounding surgically xated vertebrae incats support this theory. Furthermore, Pickar and Wheeler(2001) have demonstratedinanesthetizedcats that bothspindle cells and golgi tendon organs can respond tohigh-velocity, short-duration loads, i.e. loads with aforce-time prole similar to that of a load delivered duringspinal manipulation. Therefore, if spinal manipulationreduces excessive signaling from the involved intervertebralmuscles this altered aerent input to the CNS maychangethewayitrespondstoanysubsequentinput,suchastheelectricalstimulationfromthemediannerveinthisstudy.4.5.ImplicationsforinvestigationsofneuralplasticityandspinalmanipulationEpisodes of acute pain, such as following an injury, mayinitiallyinduceplasticchangesinthesensorimotorsystem(for review, see Wall et al., 2002). Pain alone, without deaf-ferentation, has beenshownto induce increased SEPpeakamplitudes (Tinazzi et al., 2000, 2004) and increasedsomatosensory evoked magnetic elds (Soros et al.,2001). As the sensorimotor disturbances are known to per-sist beyond the acute episode of pain (Jull et al., 2002; Ster-ling et al., 2003), and thought to play a dening role in theclinical picture andchronicity of dierent chronic neckpain conditions (Michaelson et al., 2003), then the reducedcortical SEP peak amplitudes observed in the current studyfollowingspinalmanipulationmayreectanormalizationofsuchinjury/pain-inducedcentralplasticchanges,whichmayreectonemechanismfortheimprovementoffunc-tionalabilityreportedfollowingspinalmanipulation.The observations in the present study suggest that spinalmanipulation of dysfunctional joints may modify transmis-sion in neuronal circuitries not only at a spinal level as indi-catedbypreviousresearch(Herzoget al., 1999; Murphyet al., 1995; Symonset al., 2000), but at acortical level,and possibly alsodeeper brainstructures such as the basalganglia. Further studies areneededtoelucidatetheroleandmechanismsofthesecortical changes, andtheirrela-tionshiptoapatientsclinicalpresentation.AcknowledgementsTheheadauthorwasinitiallysupportedwithafellow-ship fromthe Foundation for Chiropractic Educationand Research, and later by a Bright Futures Top AchieversDoctoralScholarship.ReferencesAllison T, McCarthy G, Wood CC, Darcey TM, Spencer DD, WilliamsonPD. Humancortical potentialsevokedbystimulationofthemediannerve. II. Cytoarchitectonicareasgeneratingshort-latencyactivity. JNeurophysiol1989b;62(3):694710.400 H.Haavik-Taylor,B.Murphy/ClinicalNeurophysiology118(2007)391402Allison T, McCarthy G, Wood CC, Jones SJ. Potentials evoked inhumanandmonkey cerebral cortex by stimulationof the mediannerve. A review of scalp and intracranial recordings. Brain1991;114(Pt 6):2465503.Allison T, McCarthy G, Wood CC, Williamson PD, Spencer DD.Human cortical potentials evoked by stimulation of the mediannerve. II. Cytoarchitectonicareas generatinglong-latencyactivity. JNeurophysiol 1989a;62(3):11722.Arnold C, Bourassa R, Langer T, Stoneham G. Doppler studiesevaluating the eect of a physical therapy screening protocol onvertebralarterybloodow.ManTher2004;9(1):1321.Balzamo E, Marquis P, Chauvel P, Regis J. Short-latency components ofevokedpotentialstomediannervestimulationrecordedbyintracere-bral electrodes in the human pre- and postcentral areas. ClinNeurophysiol2004;115(7):161623.Bolton PS, Holland CT. Aerent signaling of vertebral displacement in theneckofthecat.SocNeurosciAbstr1996;22:1802.BoltonPS, HollandCT. Aninvivomethodfor studyingaerent breactivityfromcervical paravertebraltissueduringvertebral motioninanaesthetisedcats.JNeurosciMethods1998;85(2):2118.Brasil-NetoJP,Valls-SoleJ,Pascual-LeoneA,CammarotaA,AmassianVE, CraccoR, et al. Rapidmodulationof humancortical motoroutputs following ischaemic nerve block. Brain 1993;116(Pt 3):51125.Byl NN, Merzenich MM, Cheung S, Bedenbaugh P, Nagarajan SS,Jenkins WM. Aprimate model for studying focal dystonia andrepetitivestraininjury: eectsontheprimarysomatosensorycortex.PhysTher1997;77(3):26984.Cracco RQ, Cracco JB. Somatosensory evoked potential in man: far eldpotentials.ElectroencephalogrClinNeurophysiol1976;41(5):4606.Desmedt JE, Cheron G. Central somatosensory conduction in man: neuralgenerators and interpeak latencies of the far-eld components recordedfrom neck and right or left scalp and earlobes. Electroencephalogr ClinNeurophysiol1980;50(56):382403.Desmedt JE, Cheron G. Non-cephalic reference recording of earlysomatosensory potentials to nger stimulation in adult or agingnormal man: dierentiation of widespread N18 and contralateral N20fromtheprerolandicP22andN30components. ElectroencephalogrClinNeurophysiol1981;52(6):55370.DishmanJD,BallKA,BurkeJ.Centralmotorexcitabilitychangesafterspinal manipulation: a transcranial magnetic stimulation study. JManipulativePhysiolTher2002;25(1):19.Emerson RG, Sgro JA, Pedley TA, Hauser WA. State-dependent changesinthe N20component of the mediannerve somatosensoryevokedpotential.Neurology1988;38(1):648.Fryer G, Morris T, Gibbons P. Paraspinal Muscles andIntervertebralDysfunction: Part One. J Manipulative Physiol Ther2004;27(4):26774.FujiiM,YamadaT,AiharaM,Kokubun Y,NoguchiY,Matsubara M,et al. The eects of stimulus rates upon median, ulnar and radial nervesomatosensory evoked potentials. Electroencephalogr Clin Neuro-physiol1994;92(6):51826.GarciaLarreaL, Bastuji H, Mauguiere F. Unmaskingof cortical SEPcomponents by changes in stimulus rate: a topographic study.ElectroencephalogrClinNeurophysiol1992;84(1):7183.Hallett M, ChenR, ZiemannU, CohenLG. Reorganizationinmotorcortex in amputees and in normal volunteers after ischemic limbdeaerentation. Electroencephalogr Clin Neurophysiol Suppl1999;51:1837.HerzogW.Mechanical,Physiologic,andNeuromuscularConsiderationsofChiropractictreatment. In: LawrenceDJ, CassidyJD, McGregorM,MeekerWC,Vernon.HT, editors.AdvancesinChiropractic, vol3. Newyork: Mosby-YearBook;1996.p.26985.Herzog W, Conway PJ, Zhang YT, Gail J, Guimaraes ACS. Reexresponses associated with manipulative treatments on the thoracicspine:apilotstudy.JManipulativePhysiolTher1995;18:2334.HerzogW,ScheeleD,ConwayPJ.Electromyographicresponses ofbackandlimbmusclesassociatedwithspinal manipulativetherapy. Spine1999;24(2):14653.Hessell BW, Herzog W, Conway PJ, McEwen MC. Experimental measure-ment of the force exerted during spinal manipulation using theThompson technique. J Manipulative Physiol Ther 1990;13(8):44853.HestboekL,Leboeuf-YdeC.Arechiropractictestsforthelumbo-pelvicspine reliable and valid? Asystematic critical literature review. JManipulativePhysiolTher2000;23(4):25875.Hides JA, Stokes MJ, Saide M, Jull GA, Cooper DH. Evidence of lumbarmultidus muscle wasting ipsilateral tosymptoms inpatients withacute/subacutelowbackpain.Spine1994;19(2):16572.Hoshiyama M, Kakigi R. Vibratory stimulation of proximal muscles doesnot aect cortical components of somatosensory evoked potentialfollowing distal nerve stimulation. Clin Neurophysiol2000;111(9):160710.HubkaMJ,PhelanSP.Interexaminerreliabilityofpalpationforcervicalspinetenderness.JManipulativePhysiolTher1994;17(9):5915.HurwitzEL, AkerPD, AdamsAH, MeekerWC, ShekellePG. Manip-ulationandmobilizationofthecervicalspine.Asystematicreviewoftheliterature.Spine1996;21(15):174660.Jones SJ. Short latency potentials recorded fromthe neck and scalpfollowingmediannervestimulationinman. ElectroencephalogrClinNeurophysiol1977;43(6):85363.Jull G, BogdukN, MarslandA. Theaccuracyof manual diagnosisforcervical zygapophysial joint pain syndromes. Med J Aust1988;148(5):2336.Jull G, Trott P, Potter H, Zito G, Niere K, Shirley D, et al. ARandomizedControlledTrial ofExerciseandManipulativeTherapyforCervicogenicHeadache.Spine2002;27(17):183543.Kanovsky P,BareM, RektorI. TheselectivegatingoftheN30corticalcomponent of the somatosensory evoked potentials of median nerve isdierentinthemesial anddorsolateral frontal cortex: evidencefromintracerebralrecordings.ClinNeurophysiol2003;114(6):98191.LeachRA. Manipulationterminologyinthe chiropractic, osteopathic,andmedical literature. In: TheChiropracticTheories: ASynopsisofScienticResearch. Baltimore: WilliamsandWilkins;1986,15-2.Mauguiere F. Somatosensory evoked potentials: normal responses,abnormal waveforms and clinical applications in neurological diseases.In: NiedermeyerE, editor. Electroencephalography: basicprinciples,clinical applications, and related elds. Baltimore: Williams andWilkins;1999.Mauguiere F, Desmedt JE, CourjonJ. Astereognosis anddissociatedloss of frontal or parietal components of somatosensory evokedpotentials in hemispheric lesions. Detailed correlations withclinical signs and computerized tomographic scanning. Brain1983;106(Pt 2):271311.McPartland JM, Brodeur RR, Hallgren RC. Chronic neck pain, standingbalance, andsuboccipital muscleatrophyapilotstudy. JManipu-lativePhysiolTher1997;20(1):249.Michaelson P, Michaelson M, Jaric S, Latash ML, Sjolander P,Djupsjobacka M. Vertical Posture and Head Stability in Patients withChronicNeckPain.JRehabMed2003;35(5):22935.Murphy BA, Dawson NJ, Slack JR. Sacroiliac joint manipulationdecreases the H-reex. Electromyogr Clin Neurophysiol1995;35(2):8794.NuwerMR,AminoM,DesmedtJ,EisenAA,GoodinD,MatsuokaS,et al. IFCN recommended standards for short latency somatosensoryevoked potentials. Report of an IFCNcommittee. InternationalFederation of Clinical Neurophysiology. Electroencephalogr ClinNeurophysiol1994;91(1):611.OConnor BL, Visco DM, Brandt KD, Albrecht M, OConnor AB.Sensory nerves only temporarily protect the unstable canine knee jointfromosteoarthritis. Evidence that sensory nerves reprogramthecentral nervous systemafter cruciateligament transection. ArthritisRheum1993;36(8):115463.Pascual-Leone A, Torres F. Plasticity of the sensorimotor cortexrepresentation of the reading nger in Braille readers. Brain1993;116(Pt1):3952.Perneger TV. Whats wrong with Bonferroni adjustments. BMJ1998;316(7139):12368.H.Haavik-Taylor,B.Murphy/ClinicalNeurophysiology118(2007)391402 401Pickar JG, Wheeler JD. Response of muscle proprioceptors tospinalmanipulative-like loads in the anesthetized cat. J Manipulative PhysiolTher2001;24(1):211.RheaultW,AlbrightB,ByersC,FrantaM,JohnsonA,SkowronekBS,et al. Intertesterreliabilityof thecervical rangeof motiondevice. JOrthopSportsPhysTher1992;15(3):14750.Roll JP, Vedel JP, Ribot E. Alteration of proprioceptive messages inducedbytendonvibrationinman: amicroneurographicstudy. ExpBrainRes1989;76(1):21322.Rossi S, della Volpe R, Ginanneschi F, UlivelliM, Bartalini S, SpidalieriR,et al.Earlysomatosensoryprocessingduringtonicmusclepaininhumans: relation to loss of proprioception and motor defensivestrategies.ClinNeurophysiol2003;114(7):13518.Rossi S, TecchioF, Pasqualetti P, Ulivelli M, PizzellaV, Romani GL,et al. Somatosensory processing during movment observation inhumans.ClinNeurophysiol2002;113:1624.Rossini PM,Babiloni F,Bernardi G,CecchiL,Johnson PB, MalentaccaA, et al. Abnormalities of short-latency somatosensory evokedpotentialsinparkinsonianpatients.ElectroencephalogrClinicalNeu-rophysiol1989;74(4):27789.Rossini PM, Gigli GL, Marciani MG, Zarola F, Caramia M. Non-invasive evaluation of input-output characteristics of sensorimotorcerebral areas inhealthy humans. Electroencephalogr ClinNeuro-physiol1987;68(2):88100.ShakespeareDT,Stokes M,Sherman KP,Young A.Reex inhibition ofthe quadriceps after meniscectomy: lack of association with pain. ClinPhysiol1985;5(2):13744.Soros P, Knecht S, Bantel C, Imai T, Wusten R, Pantev C, et al.Functional reorganization of the human primary somatosensorycortex after acute paindemonstratedby magnetoencephalography.NeurosciLett2001;298(3):1958.Steinmetz JE, Beggs AL, Molea D, Patterson MM. Long-term retention ofa peripherally induced exor reex alteration in rats. Brain Res1985;327:3125.Steinmetz JE, Lupica CR, Secard CM, Patterson MM. Eects ofmanipulating stimulation intensity and duration on xation of aperipherally inducedspinal reex alterationinrats. Physiol Behav1982;29:103944.SterlingM,JullG,VicenzinoB,KenardyJ,DarnellR.Developmentofmotor system dysfunction following whiplash injury. Pain2003;103(1-2):6573.Stokes M, Young A. The contribution of reex inhibition to arthrogenousmuscleweakness.ClinSci1984;67(1):714.StrickPL, PrestonJB. Tworepresentationsof thehandinarea4of aprimate.II.Somatosensory input organization. J Neurophysiol1982;48:1509.Suter E, McMorlandG, Herzog W, Bray R. Decrease inquadricepsinhibitionaftersacroiliacjointmanipulationinpatientswithanteriorkneepain.JManipulativePhysiolTher1999;22(3):14953.Suter E, McMorlandG, HerzogW, BrayR. Conservativelower backtreatmentreducesinhibitioninknee-extensormuscles: arandomizedcontrolledtrial.JManipulativePhysiolTher2000;23(2):7680.Symons BP, Herzog W, Leonard T, Nguyen H. Reex responsesassociated with activator treatment. J Manipulative Physiol Ther2000;23(3):1559.TanjiJ,WiseSP.Submodalitydistributioninsensorimotorcortexoftheunanesthetizedmonkey.JNeurophysiol1981;45(3):46781.Tinazzi M, Fiaschi A, RossoT, Faccioli F, GrosslercherJ, Aglioti SM.Neuroplastic Changes Related to Pain Occur at Multiple Levels of theHumanSomatosensorySystem: ASomatosensory-EvokedPotentialsStudy in Patients with Cervical Radicular Pain. J Neurosci2000;20(24):927783.Tinazzi M, Valeriani M, MorettoG, RossoT, NicolatoA, Fiaschi A,et al. Plasticinteractionsbetweenhandandfacecortical representa-tions inpatients withtrigeminal neuralgia: asomatosensory-evokedpotentialsstudy.Neuroscience2004;127(3):76976.Ulas UH, Odabasi Z, Ozdag F, Eroglu E, Vural O. Median nervesomatosensoryevokedpotentials:recordingwithcephalicandnonce-phalicreferences.ElectromyogrClinNeurophysiol1999;39(8):4737.VernonLF.Spinalmanipulationasavalidtreatmentforlowbackpain.DelMedJ1996;68(3):1758.WaberskiTD, BuchnerH, PerkuhnM, GobbeleR, WagnerM, KuckerW, et al. N30 and the eect of explorative nger movements: a modelof the contribution of the motor cortex to early somatosensorypotentials.ClinNeurophysiol1999;110(9):1589600.Wall JT, Xu J, Wang X. Human brain plasticity: an emerging view of themultiplesubstratesandmechanismsthat causecortical changesandrelatedsensorydysfunctionsafterinjuriesofsensoryinputsfromthebody.BrainResRev2002;39(2-3):181215.Youdas JW, Garrett TR, Suman VJ, Bogard CL, Hallman HO, Carey JR.Normal rangeofmotionofthecervical spine: aninitial goniometricstudy.PhysTher1992;72(11):77080.Zhu Y, Haldeman S, Hsieh CY, Wu P, Starr A. Do cerebral potentials tomagnetic stimulation of paraspinal muscles reect changes in palpablemuscle spasm, low back pain, and activity scores?[comment]. JManipulativePhysiolTher2000;23(7):45864.ZhuY, HaldemanS, StarrA, SengerMA, SuSH. Paraspinal muscleevokedcerebral potentialsinpatientswithunilateral lowbackpain.Spine1993;18(8):1096102.Ziemann U, Hallett M, Cohen LG. Mechanisms of deaerentation-induced plasticity in human motor cortex. J Neurosci1998;18(17):70007.402 H.Haavik-Taylor,B.Murphy/ClinicalNeurophysiology118(2007)391402