Swallowing in individuals with disorders of consciousness: A cohort study

(1)

Original

article

Swallowing

in

individuals

with

disorders

of

consciousness:

A

cohort

study

Evelyne

Me´lotte

a,b,c,

*

,

Audrey

Maudoux

d,e

,

Sabrina

Delhalle

e

,

Aude

Lagier

e

,

Aurore

Thibaut

b,c

_,

_Charle`ne

_Aubinet

b,c

_,

_{Jean-Franc¸ois}

_Kaux

a

_,

_Audrey

_{Vanhaudenhuyse}

d,f

_,

Didier

Ledoux

g,1

,

Steven

Laureys

b,c,1

,

Olivia

Gosseries

b,c,1

a

PhysicalandRehabilitationMedicineDepartment,UniversityHospitalofLiege,Liege,Belgium

b

ComaScienceGroup,GIGA-Consciousness,UniversityofLiege,Liege,Belgium

c

CentreduCerveau2

,UniversityHospitalofLie`ge,Liege,Belgium

d

SensationandPerceptionResearchGroupGIGA,UniversityofLiege,Liege,Belgium

e

OtorhinolaryngologyHeadandNeckSurgeryDepartment,UniversityHospitalofLiege,Liege,Belgium

f

AlgologyDepartment,UniversityHospitalofLiege,Liege,Belgium

g_Intensive_Care_Unit_Department,_University_Hospital_of_Liege,_Liege,_Belgium

1. Introduction

Afteraperiodofcoma,individualswithseverebraininjurymay remaininanalteredstateofconsciousnessbeforeregainingpartial

orcompleterecovery[1].Disordersofconsciousness(DOC)consist of 3 states ranging from no awareness and no arousal to the preservation of arousal with ﬂuctuating awareness [2]: coma, vegetative state/unresponsivewakefulnesssyndrome (UWS)[3] andminimallyconsciousstate(MCS)[4].UWSischaracterizedby thepresence ofeye-opening and reﬂexivemovements, without consciousbehaviours[3].IndividualswithMCSshowreproducible but inconsistent signs of consciousness, such as command following,visualpursuit,andlocalizationtonoxiousstimulation

ARTICLE INFO Articlehistory: Received4March2020 Accepted18April2020 Keywords: Braininjury Dysphagia Swallowing Disordersofconsciousness Oralfeeding ABSTRACT

Background:Afteraperiodofcoma,aproportionofindividualswithseverebraininjuryremaininan alteredstateofconsciousnessbeforeregainingpartialorcompleterecovery.Individualswithdisorders ofconsciousness(DOC)classicallyreceivehydrationandnutritionthroughanenteral-feedingtube. However,therealimpactofthelevelofconsciousnessonanindividual’sswallowingabilityremains poorlyinvestigated.

Objective:WeaimedtodocumenttheincidenceandcharacteristicsofdysphagiainDOCindividualsand toevaluatethelinkbetweendifferentcomponentsofswallowingandthelevelofconsciousness. Methods: We analyzed clinical data on the respiratory status, oral feeding and otolaryngologic examination of swallowing in DOC individuals. We analyzed the association of components of swallowingandparticipantgroups(i.e.,unresponsivewakefulnesssyndrome[UWS]andminimally consciousstate[MCS]).

Results:Weincluded92individualswithDOC(26UWSand66MCS).Overall,99%oftheparticipants showeddeﬁcitsintheoraland/orpharyngealphaseofswallowing.AscomparedwiththeMCSgroup,the UWSgroupmorefrequentlyhadatracheostomy(69%vs24%),withdiminishedcoughreﬂex(27%vs54%) andnoeffectiveoralphase(0%vs21%).

Conclusion:AlmostallDOCparticipantshadseveredysphagia.Somecomponentsofswallowing(i.e., tracheostomy,coughreflexandefficacyoftheoralphaseofswallowing)wererelatedtoconsciousness. Inparticular,noUWSparticipanthadanefficientoralphase,whichsuggeststhatitspresencemaybea signofconsciousness.Inaddition,noUWSparticipantcouldbefedentirelyorally,whereasnoMCS participantorallyreceivedordinaryfood.Ourstudyalsoconfirmsthatobjectiveswallowingassessment canbe successfullycompletedinDOCindividuals and thatspecificcareisneededto treatsevere dysphagiainDOC.

C₂₀₂₀_The_Authors._Published_by_Elsevier_Masson_SAS._This_is_an_open_access_article_under_the_CC BY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/4.0/).

* Correspondingauthor.ComaScienceGroup,GIGAConsciousness,Avenuede l’Hopital1,4000Liege,Belgium.

E-mailaddress:evelyne.melotte@chuliege.be(E.Me´lotte).

1

Contributedequallytothisworkaslastauthors.

Available

online

at

ScienceDirect

www.sciencedirect.com

https://doi.org/10.1016/j.rehab.2020.04.008

1877-0657/ C₂₀₂₀_The_Authors._Published_by_Elsevier_Masson_SAS._This_is_an_open_access_article_under_the_CC_BY-NC-ND_license₍

(2)

[4].Whentheyrecovertheabilitytofunctionallycommunicateor touseobjectsadequately,theyemergefromtheMCS[4].

Swallowingdisordersarerelativelyfrequentafteranacquired braininjuryfromtraumaticoranoxiccauses,rangingfrom25%to 61% depending on the studies [5,6]. Oral feeding has been suggestedtoberelatedtothelevelofconsciousnessinprevious studiesevaluatingswallowinginseverebrain-injuredindividuals [5–17]. Individuals with DOC classically receive hydration and nutritionthroughanenteral-feedingtube[18].Therealimpactof the level of consciousness on individuals’ swallowing ability remains poorly investigated. Indeed, previous studies that assessedswallowinginindividualswithseverebraininjuryhad smallsamplesizes[15],focusedononlyonecomponentsuchas thetypeoffeeding[17]orthepossibilityofextubation[16]and/or theymostlyusedclinicalassessmentofconsciousnessthatwasnot basedon establisheddiagnosticcriteria [4](see[5–14]). In line with this, previous literature showed that the Coma Recovery Scale-Revised(CRS-R)[19]isthecurrentgoldstandardtoassess the level of consciousness in DOC individuals, and multiple evaluations should be performed to decrease misdiagnosis [20]. Accurate diagnosis is challenging because of confounding factorssuchasaphasiaand motordeﬁcits,butit hasimportant implicationsforprognosis[21],treatmentmanagement[22]and relatedethicalconsiderations[23].

Swallowing hasnot yet beenstudied systematically in DOC individuals,andtoourknowledge,thelinkbetweenthelevelof consciousnessand swallowingcomponents(e.g.,lipprehension, lingualpropulsion,pharyngo-laryngealsensitivity,efﬁcacyofthe pharyngeal phase and ability to clear saliva) has never been

investigated.Werecentlysuggestedthataneffectiveoralphaseof swallowingcouldbea determinanttoconsiderswallowingasa conscious behavior [17]. However, in this previous study, we includedonlyUWSparticipantsandbasedourconclusionononly 2components(i.e.,presenceorabsenceoforalfeedingandtypeof oralfeeding).

Inthepresentwork,wecollectedrespiratoryandnutritional statusaswellastheotolaryngologicalresultsofswallowingina largecohortofindividualswithprolongedDOC.Wepredictedthat mostindividualshaveseverealterationsofthedifferent compo-nentsofswallowingandthatsomecomponentsoftheswallowing process, suchas the oral phase, may belinked to thelevel of consciousness.

2. Participantsandmethods 2.1. Participants

We retrospectively collected data for individuals admitted consecutivelyfromJanuary2010toAugust2018totheUniversity HospitalofLiege(Belgium)fora1-weekmultimodalassessmentof consciousness for diagnostic and prognostic purposes. Fig. 1 illustratestheflowofparticipantsinthestudy.Inclusioncriteria were1)recoveredfromcomacausedbyasevereacquiredbrain injury, 2) with a prolonged condition (>1 month post-insult) [25,26],3)medicallystable,4)underwentfiberopticendoscopic examinationofswallowing(FEES),5)adiagnosisofUWSorMCS basedonaminimumof5CRS-Rtests(toavoiddiagnosticerrors duetofluctuationsinresponsivenessandtoobtainastableclinical

Fig.1.Flowchartofthestudy.FEES:ﬁberopticendoscopicevaluationofswallowing;CRS-R:ComaRecoveryScale-Revised;FDG-PET:ﬂuorodeoxyglucose-positronemission tomography;EMCS:emergencefromminimallyconsciousstate;LIS:locked-insyndrome;DOC:disordersofconsciousness;UWS:unresponsivewakefulnesssyndrome; MCS:minimallyconsciousstate.

(3)

diagnosis) [20], which was conﬁrmed on ﬂuorodeoxyglucose-positron emission tomography (FDG-PET) (see next sections concerningdata acquisitionandanalyses) and6)no morethan 1missingdataitem(i.e.,atleast9of10selectedcriteriaregarding respiratorystatus,typeoffeedingandcomponentsoftheoraland pharyngealphasesofswallowing).

Weextracteddemographic data(i.e.,sex,age,etiology,time sinceinsult)andDOCdiagnosisfromparticipants’medicalrecords. Etiology was classiﬁed in terms of focal or global injury to distinguish between on one hand, ischemic, hemorrhagic and traumaticbrain injury, andon theother,anoxicand metabolic encephalopathy.

ThestudywasapprovedbytheEthicsCommitteeoftheFaculty of Medicine of the University Hospital of Liege, and written informed consent was obtained from all legal surrogates. We followed the principles of the STrengthening the Reporting of OBservationalstudiesinEpidemiology(STROBE)(see Supplemen-tarymaterials)[24].

2.2. Diagnosisofconsciousness

The diagnosis was established after repeated behavioral assessments performed by trained and experienced clinician researchersusingtheCRS-R[19]andFDG-PET.TheCRS-Rconsists of6subscales(auditory,visual,motor,andoromotorfunctionsas wellascommunicationandarousal)of23itemsorderedbydegree of complexity, ranging from reflexive to cognitively mediated behaviors[27].Allparticipantsunderwentatleast5CRS-Rtests overamaximumof10daysandthebestresultwaskeptforthe behavioral diagnosis [20]. To confirm thebehavioral diagnosis, FDG-PETimageswerevisuallyinspectedbyexpertsandclassified ascompatible withUWS(when thestatisticaltool detected no voxels with preserved metabolism in the associative fronto-parietal network bilaterally) or compatible with MCS (with incompletehypometabolismorpartialpreservationofmetabolic activitydetectedinthefronto-parietalnetwork).Theacquisition procedureandanalysesofFDG-PETdataweredescribedpreviously [28,29].Thebrainactivitymapwasobtainedwithathresholdof uncorrectedp<0.05inallcontrastsforsingle-subjectanalyses,as in previous studies [28,29]. We excluded individuals with a diagnosisofUWSbasedon theCRS-Rbut withFDG-PETresults compatible with MCS because these individuals may present covertconsciousness(referredtoasMCS*)[2].

2.3. Respiratorystatus,typeoffeedingandswallowingassessments Wecollecteddataon10speciﬁccriteriabasedontheresultsof theear,noseandthroat(ENT)examination(performedbySD,AM orAL)andfromthequestionnairescompletedbythefamily(for thetypeoffeeding).

Forrespiratorystatus,wereportedthepresenceorabsenceof tracheostomy (criterion 1). The type of feeding (criterion 2) referredtothepresenceorabsenceofexclusiveenteral-feeding. Forparticipantswhoreceivedoralfeeding,wedistinguishedtype offeedingbasedonthecriteriaoftheFoodIntakeLevelScale(FILS) [30].TheFILSisanobserver-ratingscaleforassessingtheseverity of dysphagia, examining to what degree individuals take food orallyon a daily basis, rangingfrom 0 (no oral intake,and no swallowingtraining)to10(normaloralfoodintake).Scores1to 3correspondtonooralintake,4to6oralintakewithalternative nutrition,and7to10oralintakeexclusively.

The other 8 criteria were related to the otolaryngological examinationperformed by ENTexperts. A FEESwasperformed withaﬂexiblevideorhinolaryngoscope(OlympusViseraOTP-S7, Tokyo)andcolormonitor.Weexcludedallcriteriathatrequireda responsetoacommand(e.g.,assessmentofthenasopharyngealor

vocalfoldclosurewiththeproductionofsounds,apnea,volunteer salivaswallow,cough).Thefirst3criteriaoftheotolaryngological examinationwererelatedtotheoralphaseofswallowingwiththe presenceorabsenceofhypertoniaofjawmuscles(criterion3),the presenceorabsenceofanoralphaseofswallowing(lipprehension orlingualpropulsion;criterion4),andtheobservationornotofan effectiveoralphase(criterion5).Practically,wemovedaspoonto infrontoftheindividual’smouthandobservedthereaction.With absenceoflipprehension,weplaceda2-mlbolusinthemiddleof the tongue and observed if lingual propulsion occurred. We consideredtheoralphaseeffectiveifwedetectedconsecutivelylip prehension,lingualpropulsionandnopost-swallowingoralstasis. The last 4 criteria were related to the pharyngeal phase of swallowing. The presence or absence of secretions in the pharyngo-laryngealarea(criterion6)andthesalivaryaspiration (criterion 7)informedon participants’ abilityto manage secre-tions.Thecoughreflex(criterion8)wasevaluatedbystimulating thelaryngealarea,andifnocoughwasobserved,the pharyngo-laryngealsensitivitywasconsideredabsent.Finally,wenotedthe presenceorabsenceofbolusaspirationduringtheswallowingof 2mlofthickandliquidtextures(criteria9and10,respectively). Someparticipantsdidnotundergothefunctionalswallowingtest becauseofaseverebitereflex,aninefficacityoftheoralphase,or because it was considered too dangerous regarding other parameters(e.g.,toomanysalivaaspirations,absenceof sponta-neoussalivaswallowing).

2.4. Statisticalanalysis

Weperformedadescriptiveanalysisforeachdiagnosisgroup (UWSandMCS)intermsofsex,age,etiologyandtimesinceinsult. Normality was assessed with histograms, quantile plots and Shapiro-Wilk tests. Univariate comparisons between UWS and MCSgroupsinvolvedchi-squareorFisherexacttestforcategorical variables and Kruskal-Wallis test for non-normally distributed continuous variables. The association between each of the 10 criteriaandthediagnosisgroupswasassessedbyunivariate logisticregression.Theseassociationswerefurtherinvestigatedby multivariable logistic regression adjusted foretiology and time since insult.Theresultsof logisticregressionsare presentedas odds ratios (ORs) and adjusted ORs together with their 95% conﬁdence intervals (CIs). P<0.05 was considered statistically signiﬁcant.Individualswithmissingvalues wereexcludedfrom theanalysisfortheconsideredcriteria.Statisticalanalyseswere performedwithStatav14.2(StataCorp.2015,CollegeStation,TX). 3. Results

Amongthe167individualsidentified,92matchedourinclusion criteria (Fig. 1); 26 showed only reflexive behavior and had a diagnosisofUWSand66satisfiedtheCRS-RcriteriaforMCSinat leastoneevaluation(Table1).Elevenofthe26UWSparticipants werealreadyincludedinourpreviousUWS studiesfocusingon typeoffeedingexclusively[17].Diagnosiswasconfirmedby

FDG-Table1

Descriptivestatisticsofthewholesample. UWS (n=26) MCS (n=66) P-value Sex(F/M) 9/17 30/36 0.343

Age,years,mean(SD) 41(12) 38(12) 0.405

Etiology Focal:11

Global:15

Focal:49 Global:17

0.004 Timesinceinjury,months,mean(SD) 30(22) 4(34) 0.014 UWS:unresponsivewakefulnesssyndrome;MCS:minimallyconsciousstate.

(4)

PETinallincludedparticipants.Thetwodiagnosisgroupsdidnot differinageorsexbutdiddifferbyetiologyandtimesinceinsult (Table1),MCSparticipantshavinglonger timesince insultand morefocal damagethanUWSparticipants.We ﬁrstpresentthe descriptive analysis (Table 1) and then the percentage of participantsfor each of the10 selectedcriteria, the univariate analysisfor eachcriterion by diagnosis(puni), and theadjusted

multivariateanalysis(padjusted)foretiologyandtimesinceinsult

(Table 2). We also illustrate the percentage of UWS and MCS participantsforeachcriterioninFig.2.

Intotal,91of92participantspresenteddisordersinatleastone criterionlinkedtotheoralorpharyngealphaseofswallowing(i.e., criteria3to10).Regardingrespiratorystatus,34(37%)participants stillhadatracheostomyatthetimeofassessment.Ascompared withMCS individuals, UWS individualsmore frequently had a tracheostomy (padjusted=0.002). In total, 71 (77%) participants

receivedenteralfeedingexclusively,withnosigniﬁcantdifference betweenUWSandMCSgroups(padjusted=0.254).NoneoftheMCS

participants received exclusive ordinary solid food (FILS 10). Regarding the ENT examination, for the oral phase, 52 (56%) participantspresented hypertonia of thejaw muscles, with no difference between UWS and MCS groups (padjusted=0.881). In

total,43(47%)participantsshowedatleastonecomponentofthe oralphaseofswallowing(lipprehension orlingualpropulsion), with no difference between groups (padjusted=0.94). However,

UWS and MCS groups differed in efﬁcacy of theoral phase of swallowing(padjusted=0.011),characterizedbythepresenceoflip

prehensionandlingualpropulsionwithoutpost-swallowingoral stasis. For the pharyngeal phase, 34 (37%) participants had pharyngo-laryngeal secretions and 26 (28%) saliva aspiration. UWS and MCS groups differed on univariate analysis for the pharyngo-laryngealsecretions(puni=0.012)andsalivaaspiration

(puni=0.019) but not signiﬁcantly on multivariable analysis

(pharyngo-laryngealsecretions:padjusted=0.067;salivaaspiration:

padjusted=0.062). For the test of the cough reﬂex, among the

63participantsassessed,43 (52%)showeddecreased pharyngo-laryngeal sensitivity, with signiﬁcantly more MCS participants presentingacoughreﬂexthanUWSparticipants(padjusted=0.027).

Regardingthe functional test withthick and liquid texture,16 (61%) and 53 (80%) UWS and MCS participants, respectively, performedtheswallowingtestwithacreamtextureand12(46%) and42(63%)withaliquidtexture.Nine(13%)participantsshowed aspirationwithathicktextureand17(31%)withaliquidtexture, withnodifferencebetweengroups(thicktexture:padjusted=0.798;

liquidtexture:padjusted=0.226).

4. Discussion

Themainaimsofthisstudyweretodocumenttheproportion andcharacteristicsofdysphagiainindividualswithDOCandto evaluatethelinkbetweendifferentcriteriaofswallowingandlevel of consciousness. To our knowledge, this is the first study of respiratory status, oral feeding and FEES in a large cohort of individualswithDOC. In ourstudy,all butone DOC individual (MCS)presentedatleastoneswallowingdysfunctionoftheoral and/orpharyngealphase.Also,tracheostomy,coughreflexandthe efficacy of the oral phase were the 3 criteria related to consciousness.Finally,noneoftheUWSindividualscouldbefed entirely orally, whereas none of the MCS individuals received ordinaryoralfood.

Regardingtypeoffeeding,noneoftheUWSparticipantscould achieveafulloralfeeding,mostprobablylinkedtotheabsenceof aneffectiveswallowingoralphaseandlesseffectivepharyngeal phase,asshownbytheproportionofparticipantswith tracheos-tomy.Onlyasmallproportion(7%)ofMCSparticipantscouldsafely resumefulloralfeedingwitheasy-to-swallowfood(i.e.,FILS7). Despite the ability of some MCS participants to resume oral feeding, ahigher levelof consciousness(i.e.,EMCS)is probably necessarytoenableafullordinaryoralfood(FILS10).

Someswallowingcriteriawerenotablyrelatedtothelevelof consciousness. First, UWS and MCS participants differed in spontaneoussalivamanagementbecausemoreUWSparticipants stillhadatracheostomyatthetimeoftheevaluationthanMCS participants.Noneoftheparticipantswererespirator-dependent. The need for the tracheostomy in about one third of the participants can probably be explained by insufﬁcient saliva

Table2

Descriptiveandstatisticalanalysisofthe10criteriaintheUWSandMCSgroups.

UWSvsMCS Criteria UWS n(%) MCS n(%) puni padjusted OR 95%CI

1.Tracheostomystillinplace Tracheostomyremoved Neverhadatracheostomy

18/26(69.2) 5 3 16/66(24.2) 38 12 <0.001 0.002 5.67 1.86–17.27

2.Fullenteralfeeding Partialoralfeeding(FILS7) Fulloralfeeding(FILS7) Fulloralfeeding(FILS7) withgastrostomyforhydration

23/26(88.5) 3 0 0 48/66(72.7) 13 3 2 0.117 0.254 2.45 0.53–11.39 Oralphase

3.Hypertoniaofthejawmuscles 15/26(57.7) 37/66(56.1) 0.887 0.881 1.08 0.41–2.87

4.Oralphase 12/26(46.2) 31/66(47.0) 0.844 0.94 0.96 0.37–2.53

5.Efﬁcacyoftheoralphase 0/26(0) 14/66(21.2) 0.007 0.011 0.09 0–0.63

Pharyngealphase 6.Pharyngo-laryngealsecretions 15/26(57.7) 19/66(28.8) 0.012 0.067 2.53 0.94–6.82 7.Salivaaspiration 12/26(46.2) 14/66(21.2) 0.019 0.062 2.65 0.95–7.38 8.Coughreﬂex 7/23(30.4) 36/60(60) 0.019 0.027 0.30 0.10–0.87 9.Creamaspiration Notperformeda 2/16(12.5) 10 7/53(13.2) 13 0.941 0.798 1.26 0.21–7.44 10.Liquidaspiration Notperformeda 2/12(16.7) 14 15/42(35.7) 24 0.223 0.226 0.35 0.07–1.90

puni:univariateanalysisbetweenUWSandMCS;padjusted:multivariableanalysisbetweenUWSandMCSadjustedforetiologyandtimesinceinsult;PL:pharyngo-laryngeal;

OR:oddsratio;CI:conﬁdenceinterval;FILS:FoodIntakeLevelScale.

(5)

swallowingreflexesandineffectivepharyngealphaseof swallow-ing.Also, weobservedmore pharyngo-laryngeal secretionsand saliva aspiration in UWS than MCS participants although not significantlyaftercontrollingfortimesinceinjury andetiology. Thesefindingssuggestthatthelevelofconsciousnessmayaffect theabilitytocorrectlymanagesaliva.

Second,bothUWSandMCSparticipantsshowedapartialoral phase of swallowing (e.g., tongue and masticatory-like move-ments).However,noUWSparticipantsandonlyasmallnumberof MCSparticipantsshowedaneffectiveoralphaseofswallowing(lip prehension, lingual propulsion and no post-swallowing oral stasis).Fromthisresultandasalreadysuggestedinourprevious study[17],aneffectiveoralphaseshouldbeconsideredasignof consciousness and thus should be taken into account in DOC diagnosis.Theoralphaseofswallowingisusuallydescribedasthe voluntary(conscious)partofswallowing,controlledbymultiple cortical regions such as the primary sensori-motor cortex, supplementarymotorarea,premotorcortex(alsocalled‘‘cortical masticatory area’’), thalamus, cingulate, putamen and insulo-opercularcortexthatinteractwithregionsinthebrainstem[31– 34].Someauthorsshowedthatmasticatory-likemovementsand rhythmictongueactivitycanbeproducedbytherecruitmentof thebrainstem alone [35,36]. However, although thebrainstem controlsbasicactivitypatternsofthecranialmotoneurongroups involvedintheoralphaseofswallowing(hypoglossal,trigeminal, facial,vagal),descendinginputsfromthecentralnervoussystem alsoplay animportant role. Indeed,for someauthors[36], the activitypatternofeachmotoneuronismodulatedbyhigherbrain and peripheral inputs. This corticaland peripheral recruitment allowsthattheﬁnalmotoroutputsﬁttheenvironmentaldemand. If the components of the oral phase of swallowing (e.g., mastication)werebasedon only a centralpatterngenerator in thebrainstem,itwouldbestereotyped[37].Thus,thepresenceof aneffectiveoralphaseofswallowingseemshighlydependenton corticalrecruitment,whichwouldexplainwhyUWSindividuals whoshowseverealterationsofsupratentorialcerebralmetabolism donotpresentaneffectiveoralphaseofswallowing.

Finally,the presenceof a coughreflex wasa criterionmore presentinMCSthanUWSparticipants.Asshowninneuroimaging studies[38],thecoughreflexisprobablynotasimple pontome-dullaryreflex arc. Indeed, in theseprevious studies,the cough reflexwasfacilitatedbycorticalactivations(mainlytheprimary

motorcortex,posteriorinsula,paracentrallobule,posterior mid-cingulatecortex,premotorcortex).Thus,theimpactofthelevelof consciousnesshighlightedinthepresentstudymightprobablybe linked to the importance of the underlying cortical damage. Moreover,itisnowgenerallyacceptedinstrokeliteraturethatthe cortexplaysanimportantroleinthecontrolofswallowingand thatdamagestoswallowingmotorareasand/ortheirconnection tothebrainstemusuallyresultindysphagia[39,40].

Besidesthesemainﬁndings,hypertoniaofthejawmusclesand thenumberofaspirationsoncream/liquidtexturedidnotdiffer betweenUWSandMCSgroups.However,moreMCSparticipants performedthefunctionaltestwiththickandliquidtexturethan did UWS participants, which highlights that UWS participants seemtobegenerallyconsideredmoreatriskofaspirationbythe examiners or had more troubles in the oral phase (e.g., hypertonicity, absenceof lip prehension or lingual propulsion) thanMCSparticipants.

Our findings also agree with previous studies showing that objective swallowing assessment such as the FEES can be successfully completed in DOC individuals [7,9,15]. The ENT examination gives preciousinformation on 2 main points, first regarding tracheostomyandsecond regardingthepossibility of oralfeeding.AftertheFEES,theENTspecialistsuggestedremoving thetracheostomyforseveralDOCparticipants(8MCSparticipants and6UWSparticipants)becauseofgoodsalivamanagementand theabsenceofstenosisorlaryngealparalysis.Inthesecases,the tracheostomywasprobablymaintainedtopreventanyrespiratory complications because of the lack of adequate information regarding the management of thetracheostomy or becauseno one tested the possibility to begin a tracheostomy weaning. A previousstudyreportedthatindividualswithatracheostomyare more likely to develop pneumonia than patients without a tracheostomy [41]. This should bekept in mind when making decisionsontheneedtomaintainatracheostomyinthisfragile population.Inthisregard,theFEEScanhelpindecidingonpossible decannulation,butgiventhecomplexitiesofsalivamanagementin theDOCpopulation,thedecisionfortracheostomyweaningshould be discussed in a multidisciplinary team. Besides tracheotomy management,theENTexaminationiscrucialforpatientsinwhom wewouldliketostartorcontinueoralfeeding.Indeed,withthe highproportionofpatients withabsenceofcoughreflex(about 54%inthewholesample),thereisahighriskofsilentaspiration.In

Fig.2.PercentageofUWSandMCSpatientsforthe10criteria.UWS:unresponsivewakefulnesssyndrome;MCS:minimallyconsciousstate.Asterisks(*)indicatesigniﬁcant differencebetweenUWSandMCS,padjusted<0.05.

(6)

thisstudy,theENTexaminationalsoadvisedstartingpartialoral feedingin3moreUWSparticipantsand13MCSparticipantsandto stopfullorpartialoralfeedingin3MCSparticipants.Afunctional swallowingtestcanbedifﬁculttoimplementinindividualswith severetrismusortotalabsenceofanoralphaseofswallowing,but aslongaspartialoralphaseispresent(e.g.,partiallipprehension andlingualpropulsion),thickorliquidswallowingcanbetestedby aqualiﬁedclinician.

RegardingtheseverityofdysphagiainDOCindividuals,speciﬁc caresuchasnursing,chestphysiotherapy,andspeechtherapyare recommended [42]. Management of swallowing should be integratedintoaglobalapproachtakingintoaccountrespiration, mobility and tonicity of the face and considering emotional reactions,spasticity,andpotentialhypersensitivity.Clinically,we noticedthattherapeuticfeeding(i.e.,swallowingsmallamountsof tastyeasy-to-swallowfood)cansometimeshelpclearexcesssaliva secretions in the pharyngo-laryngeal area. In addition, taste stimulation(involvingonlyaverysmallamountoffoodorliquid thatisdeliveredviaacottonswabinparticularzonesoftheoral cavity)is alsoa goodoption for individualswho areat risk of aspiration.Nevertheless, thedecision to introduce oral food or liquidinDOCindividualsastherapeuticfeedingorasarealpartof the feeding should only be made after the completion of an objectiveswallowingevaluation[9].

There are several limitations of the study related to its retrospective, observational and single-centre design, which suggests that our data are moderately representative of the general DOC population. The presenceof missing data for one criterion(cough reflex)shouldbeacknowledged.We werealso limitedinthenumberofavailablecriteriathatcouldbestudied.A prospective analysis, given a prescribed test protocol for DOC individuals with more detailed criteria, may yield additional informationthatwerenotcoveredinthecurrentstudy,suchasthe duration of the oral phase before the swallowing reflex, the frequency of spontaneous saliva swallowing reflex [13] or the importanceofsecretionspost-swallowing witha validprotocol suchastheNewZealandsecretionscale[43].Futurelongitudinal studiesshouldalsoinvestigatetherecoveryofdysphagia along withconsciousnessrecoverywithinthesameindividuals.

In conclusion, this study provides promising resultslinking swallowingand consciousness, notably regarding the abilityof (minimally)‘‘conscious’’individualstobettermanagespontaneous swallowingofsaliva(reflectedherebythepresenceorabsenceof tracheostomy),thecoughreflexandtheefficiencyoftheoralphase ofswallowing.Weshouldcontinueoureffortsintheassessmentof the oro-facial area in DOC individuals to be able to propose appropriateandsensiblecaremanagement.

Funding

ThestudywassupportedbytheUniversityandUniversityHospitalof Lie`ge; the Belgian National Funds for Scientific Research (FRS-FNRS); EuropeanUnion’sHorizon2020FrameworkProgrammeforResearchand InnovationundertheSpecificGrantAgreementNo.785907&No.945539 (Human BrainProject SGA2 and SGA3); Luminous project (EU-H2020-fetopen-ga686764); the BIAL Foundation; DOCMA project [EU-H2020-MSCA–RISE–778234]; the fund Generet; AstraZeneca Foundation; King Baudouin Foundation; the Wallonie-Bruxelles International; the James McDonnellFoundation;MindScienceFoundation;IAPresearchnetworkP7/ 06 of the Belgian Government (Belgian Science Policy); the European Commission; the Public Utility Foundation ‘Universite´ Europeénne du Travail’;‘‘FondazioneEuropeadiRicercaBiomedica’’;FondationBenoit;and BelgianNationalPlanCancer(139).CAisresearchfollow,ATispostdoctoral researcher,OGisresearchassociateandSLisresearchdirectoratFRS-FNRS. Authorcontributions

E.M.,O.G.,andS.L.contributedtoconceptionanddesignofthestudy. E.M.,S.D.,A.M.,andA.L.,acquiredthe ENTdata.OG, A.T.,C.A.andA.V.

acquiredthediagnosisdata.E.M.,OGandD.L.analyzedthedata.E.M.and O.G.draftedthemanuscript.A.M.,A.L.,A.T.,C.A.,JF.K.,A.V.,D.L.,andS.L. providedmajorrevisionsinsigniﬁcantproportionsofthemanuscript.D.L., O.G.andS.L.contributedequally.

Disclosureofinterest

Theauthorsdeclarethattheyhavenocompetinginterest.

Acknowledgements

WethankPrHustinxfromtheNuclearMedicineDepartment, UniversityHospitalofLie`ge,allthebehavioralandPETacquirers fromtheComaScienceGroup,aswellasthewholeNeurologystaff, patients,andtheirfamilies.

AppendixA.Supplementarydata

Supplementarydataassociatedwiththisarticlecanbefound,in theonlineversion,athttps://doi.org/10.1016/j.rehab.2020.04.008. References

[1]SchnakersC.Comaanddisordersofconsciousness, 2nded.,NewYork,NY: SpringerBerlinHeidelberg;2017.

[2]GosseriesO,ZaslerND,LaureysS.Recentadvancesindisordersof conscious-ness:focusonthediagnosis.BrainInj2014;28:1141–50.http://dx.doi.org/ 10.3109/02699052.2014.920522.

[3]LaureysS,CelesiaGG,LavrijsenJ,Leo´n-Carrio´nJ,SannitaWG,SazbonL,etal. Unresponsivewakefulnesssyndrome:anewnameforthevegetativestateor apallicsyndrome.BMCMed2010;1:68. http://dx.doi.org/10.1186/1741-7015-8-68.

[4]GiacinoJT,AshwalS, ChildsN,CranfordR,JennettB,Katz DI,etal.The minimally conscious state: deﬁnition and diagnostic criteria. Neurology 2002;58:349–53.

[5]MackayLE,MorganAS,BernsteinBA.Swallowingdisordersinseverebrain injury:riskfactorsaffectingreturntooralintake.ArchPhysMedRehabil 1999;80:365–71.http://dx.doi.org/10.1016/S0003-9993(99)90271-X. [6]WinsteinCJ.Frequency,progression,andoutcomeinadultsfollowinghead

injury.PhysTher1983;63:8.

[7]O’Neil-PirozziTM,JackMomoseK,MelloJ,LepakP,McCabeM,ConnorsJJ,etal. Feasibilityofswallowinginterventionsfortracheostomizedindividualswith severelydisorderedconsciousnessfollowingtraumaticbraininjury.BrainInj 2003;17:389–99.http://dx.doi.org/10.1080/0269905031000070251. [8]FormisanoR,VoogtRD,BuzziMG,VinicolaV,PentaF,PeppeA,etal.Time

intervaloforalfeedingrecoveryasaprognosticfactorinseveretraumatic brain injury. Brain Inj 2004;18:103–9. http://dx.doi.org/10.1080/ 0269905031000149470.

[9]BradySL,DarraghM,EscobarNG,O’neilC,PapeTLB,RaoN,etal.Personswith disorders of consciousness: are oral feedings safe/effective? Brain Inj 2006;20:1329–34.http://dx.doi.org/10.1080/02699050601111435. [10]Terre´ R,MearinF.Prospectiveevaluationoforo-pharyngealdysphagiaafter

severetraumaticbraininjury.BrainInj2007;21:1411–7.http://dx.doi.org/ 10.1080/02699050701785096.

[11]HansenTS,EngbergAW,LarsenK.Functionaloralintakeandtimetoreach unrestricteddietingforpatientswithtraumaticbraininjury.ArchPhysMed Rehabil2008;89:1556–62.http://dx.doi.org/10.1016/j.apmr.2007.11.063. [12]Terre´ R,MearinF.Evolutionoftrachealaspirationinseveretraumaticbrain

injury-relatedoropharyngealdysphagia:1-yearlongitudinalfollow-upstudy. NeurogastroenterolMotil2009;21:361–9. http://dx.doi.org/10.1111/j.1365-2982.2008.01208.x.

[13]CraryMA,SuraL,CarnabyG.Validationanddemonstrationofanisolated acousticrecordingtechniquetoestimatespontaneousswallowfrequency. Dysphagia2013;28:86–94.http://dx.doi.org/10.1007/s00455-012-9416-y. [14]MandavilleA,RayA,RobertsonH,FosterC,JesserC.Aretrospectivereviewof

swallowdysfunctioninpatientswithseveretraumaticbraininjury.Dysphagia 2014;29:310–8.http://dx.doi.org/10.1007/s00455-013-9509-2.

[15]BremareA,RapinA,VeberB,Beuret-BlanquartF,VerinE.Swallowing dis-ordersinseverebraininjuryinthearousalphase.Dysphagia2016;31:511–20. http://dx.doi.org/10.1007/s00455-016-9707-9.

[16]GodetT,ChabanneR,MarinJ,KauffmannS,FutierE,PereiraB,etal.Extubation failureinbrain-injuredpatients:riskfactorsanddevelopmentofaprediction score in a preliminary prospective cohort study. Anesthesiology 2017;126:104–14.http://dx.doi.org/10.1097/ALN.0000000000001379. [17]Me´lotteE,MaudouxA,DelhalleS,MartialC,AntonopoulosG,LarroqueSK,

etal.Isoralfeedingcompatiblewithanunresponsivewakefulnesssyndrome? JNeurol2018;265:954–61.http://dx.doi.org/10.1007/s00415-018-8794-y. [18]WhyteJ,LabordeA,DipasqualeM.Assessmentandtreatmentofthevegetative

(7)

al.,editors.Rehabilitationoftheadultandchildwithtraumaticbraininjury. PA:Philadelphia;1999.p.435–52.

[19]GiacinoJT,KalmarK,WhyteJ.TheJFKComaRecoveryScale-Revised: mea-surement characteristics and diagnostic utility. Arch Phys Med Rehabil 2004;85:2020–9.http://dx.doi.org/10.1016/j.apmr.2004.02.033.

[20]WannezS,HeineL,ThonnardM,GosseriesO,LaureysS,ComaScienceGroup collaborators.Therepetitionofbehavioralassessmentsindiagnosisof dis-ordersofconsciousness:repeatedCRS-RassessmentsfordiagnosisinDOC. AnnNeurol2017;81:883–9.http://dx.doi.org/10.1002/ana.24962.

[21]LuauteJ,Maucort-BoulchD,TellL,QuelardF,SarrafT,IwazJ,etal.Long-term outcomesofchronicminimallyconsciousandvegetativestates.Neurology 2010;75:246–52.http://dx.doi.org/10.1212/WNL.0b013e3181e8e8df. [22]ThibautA,BrunoM-A,LedouxD,DemertziA,LaureysS.tDCSinpatientswith

disordersofconsciousness:sham-controlled randomizeddouble-blindstudy. Neurology2014;82:1112–8.http://dx.doi.org/10.1212/WNL.0000000000000260. [23]DemertziA,LedouxD,BrunoM-A,VanhaudenhuyseA,GosseriesO,SodduA, etal.Attitudestowardsend-of-lifeissuesindisordersofconsciousness:a European survey. J Neurol 2011;258:1058–65. http://dx.doi.org/10.1007/ s00415-010-5882-z.

[24]vonElmE,AltmanDG,EggerM,PocockSJ,GøtzschePC,VandenbrouckeJP.The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE)statement:guidelinesforreportingobservationalstudies. Epidemi-ology2007;18:800–4.http://dx.doi.org/10.1097/EDE.0b013e3181577654. [25]GiacinoJT,KatzDI,SchiffND,WhyteJ,AshmanEJ,AshwalS,etal.Practice

guidelineupdaterecommendationssummary:disordersofconsciousness. Neurology 2018;91:450–60. http://dx.doi.org/10.1212/ WNL.0000000000005926.

[26]GiacinoJT,KatzDI,SchiffND,WhyteJ,AshmanEJ,AshwalS,etal.Practice guidelineupdaterecommendationssummary:disordersofconsciousness. Arch Phys Med Rehabil 2018;99:1699–709. http://dx.doi.org/10.1016/ j.apmr.2018.07.001.

[27]NaccacheL.Minimallyconsciousstateorcorticallymediatedstate?Brain 2018;141:949–60.http://dx.doi.org/10.1093/brain/awx324.

[28]StenderJ,GosseriesO,BrunoM-A,Charland-VervilleV,VanhaudenhuyseA, DemertziA,etal.DiagnosticprecisionofPETimagingandfunctionalMRIin disorders of consciousness: a clinical validation study. The Lancet 2014;384:514–22.http://dx.doi.org/10.1016/S0140-6736(14)60042-8. [29]BodartO,GosseriesO,WannezS,ThibautA,AnnenJ,BolyM,etal.Measuresof

metabolismandcomplexityinthebrainofpatientswithdisordersof con-sciousness. NeuroImage Clin 2017;14:354–62. http://dx.doi.org/10.1016/ j.nicl.2017.02.002.

[30]KuniedaK,OhnoT,FujishimaI,HojoK,MoritaT.Reliabilityandvalidityofa tooltomeasuretheseverityofdysphagia:thefoodintakeLEVELScale.JPain

Symptom Manage 2013;46:201–6. http://dx.doi.org/10.1016/j.jpainsym-man.2012.07.020.

[31]ShakerR,editor.Principlesofdeglutition:amultidisciplinarytextfor swal-lowinganditsdisorders..NewYork:Springer;2013.

[32]SessleBJ,AdachiK,Avivi-ArberL,LeeJ,NishiuraH,YaoD,etal.Neuroplasticity offaceprimarymotorcortexcontroloforofacialmovements.ArchOralBiol 2007;52:334–7.http://dx.doi.org/10.1016/j.archoralbio.2006.11.002. [33]SessleBJ,YaoD,NishiuraH,YoshinoK,LeeJ-C,MartinRE,etal.Propertiesand

plasticityoftheprimatesomatosensoryandmotorcortexrelatedtoorofacial sensorimotorfunction.ClinExpPharmacolPhysiol2005;32:109–14.http:// dx.doi.org/10.1111/j.1440-1681.2005.04137.x.

[34]Avivi-ArberL,MartinR,LeeJ-C,SessleBJ.Facesensorimotorcortexandits neuroplasticityrelatedtoorofacialsensorimotorfunctions.ArchOralBiol 2011;56:1440–65.http://dx.doi.org/10.1016/j.archoralbio.2011.04.005. [35]DellowPG,LundJP.Evidenceforcentraltimingofrhythmicalmastication.J

Physiol1971;215:1–13.

[36]YamadaY,YamamuraK,InoueM.Coordinationofcranialmotoneuronsduring mastication.RespirPhysiolNeurobiol2005;147:177–89. http://dx.doi.org/ 10.1016/j.resp.2005.02.017.

[37]MooreJD,KleinfeldD,WangF.Howthebrainstemcontrolsorofacialbehaviors comprised ofrhythmic actions.Trends Neurosci 2014;37:370–80. http:// dx.doi.org/10.1016/j.tins.2014.05.001.

[38]MazzoneSB,ColeLJ,AndoA,EganGF,FarrellMJ.Investigationoftheneural controlofcoughandcoughsuppressioninhumansusingfunctionalbrain imaging.JNeurosci2011;31:2948–58.

[39]HamdyS,AzizQ,ThompsonDG,RothwellJC.Physiologyandpathophysiology oftheswallowingareaofhumanmotorcortex.NeuralPlast2001;8:91–7. http://dx.doi.org/10.1155/NP.2001.91.

[40]MartinRE,SessleBJ.Theroleofthecerebralcortexinswallowing.Dysphagia 1993;8:195–202.http://dx.doi.org/10.1007/BF01354538.

[41]MatthewsCT,CoyleJL. Reducingpneumoniariskfactorsinpatientswith dysphagiawhohaveatracheotomy:whatrolecanSLPsplay?ASHALead 2010;15, https://leader.pubs.asha.org/doi/full/10.1044/leader.FTR4. 15062010.np.

[42]RobertsH,GreenwoodN.Speechandlanguage therapybestpracticefor patientsinprolongeddisordersofconsciousness:amodiﬁedDelphistudy. IntJLangCommunDisord2019;54:841–54. http://dx.doi.org/10.1111/1460-6984.12489.

[43]MilesA,HuntingA,McFarlaneM,CaddyD,ScottS.PredictivevalueoftheNew ZealandSecretionScale(NZSS)forpneumonia.Dysphagia2018;33:115–22. http://dx.doi.org/10.1007/s00455-017-9841-z.