Nasalairflowandhandpreference:Istherealink?

SubmittedtoCardiffUniversityforthedegreeofMasterofPhilosophy

AnniePriceMBBCh,MRCS(ENT)

CardiffSchoolofBiosciencesCardiffUniversity

Wales,UnitedKingdomNovember2016

ii

DECLARATIONThisworkhasnotbeensubmittedinsubstanceforanyotherdegreeorawardatthisoranyotheruniversityorplaceoflearning,norisbeingsubmittedconcurrentlyincandidatureforanydegreeorotheraward.Signed:AnniePriceDate:18/11/2016STATEMENT1ThisthesisisbeingsubmittedinpartialfulfillmentoftherequirementsforthedegreeofMPhil.Signed:AnniePriceDate:18/11/2016STATEMENT2This thesis is theresultofmyown independentwork/investigation,exceptwhereotherwisestated,andthethesishasnotbeeneditedbyathirdpartybeyondwhatispermittedbyCardiffUniversity’sPolicy on the Use of Third Party Editors by Research Degree Students. Other sources areacknowledgedbyexplicitreferences.Theviewsexpressedaremyown.Signed:AnniePriceDate:18/11/2016STATEMENT3Iherebygiveconsentformythesis,ifaccepted,tobeavailableonlineintheUniversity’sOpenAccessrepositoryandfor inter-library loan,andfor thetitleandsummarytobemadeavailabletooutsideorganisations.Signed:AnniePriceDate:18/11/2016STATEMENT4:PREVIOUSLYAPPROVEDBARONACCESSIherebygiveconsentformythesis,ifaccepted,tobeavailableonlineintheUniversity’sOpenAccessrepository and for inter-library loans after expiry of a bar on access previously approved by theAcademicStandards&QualityCommittee.Signed:AnniePriceDate:18/11/2016

iii

SummaryInrecentyears,evidencehasemergedsuggestingthatnasalairflowasymmetryandbrainasymmetrymaybelinked.Thenoseisuniqueinthatitexhibitsasymmetricalairflowwith thedominantairflowalternating fromonenasalpassage to theotherover a periodof hours. The cerebral hemispheres also exhibit both functional andstructural asymmetry, and one of themost obviousmanifestations of this is handpreference.Overtenyearsago,itwassuggestedthatnasalairflowdominanceandhandpreferencewerelinked.Theaimsofthisthesisweretoexploretheliteraturerelating to nasal airflow and brain activity and to conduct a cross-sectionalobservational study looking for a correlation between nasal airflow and handpreferenceinhealthyindividuals.ThemodifiedGlatzelMirrorwasusedtorecordthedominantnasalpassageat15-minuteintervalsovera6-hourperiodin29healthysubjects,ofwhom15wereleft-handed and 14 were right-handed, as measured by the Edinburgh HandednessInventory(ShortForm).As expected based on previous studies, there was considerable variability in thenasal airflow patterns of the individuals observed, but over half demonstrated atleastonedefiniteandsustainedreversalofnasalairflowdominance.Nocorrelationbetweennasalairflowdominanceandhandpreferencewasidentified.Asymmetricalcerebralorganisationmayofferadvantages forcomplexactionssuchasspeechandhandgestures.Nasalairflowiscontrolledbytheautonomicnervoussystemandthefunctionofasymmetricalnasalairflowismostlikelyimmunedefenceand air-conditioning. Having a dominant nasal passage, comparable to having adominanthand,wouldnot thereforebephysiologically advantageous. The resultspresentedhere,alongwithfurtherevidencefromtheliterature,donotsupportthepreviouslyreportedcorrelationbetweenhandpreferenceandnasalairflow.

iv

ContentsChapter1:IntroductionA(Overview) 11.1:Nasalairflowasymmetry 11.2:Definingthenasalcycle 11.3:Functionofthenasalcycle 41.4:Durationofthenasalcycle 61.5:Controlofnasalairflow 71.6:Sensingnasalairflow 101.7:Measuringnasalairflow 111.8:Factorsthataffectnasalairflow 181.9:Nasalairflowandhandpreference 251.10:Handedness 301.11:Aimsofthesis 32Chapter2:IntroductionB(Nasalairflowandbrainactivity–literaturereview) 342.1:Introduction 342.2:Methods 352.3:Results 352.4:Conclusions 45Chapter3:Methodology 463.1:Ethicalapproval 463.2:Studydesign 463.3:Studypopulationandrecruitment 473.4:Samplesizeandpower 483.5:Measurementofhandedness 483.6:Measurementofnasalairflow 493.7:Trialenvironmentandprocedure 493.8:Blinding 523.9:Statisticalanalysis 52Chapter4:Subjects 534.1:Introduction 534.2:Methods 534.3:Results 544.4:Discussion 58Chapter5:Nasalairflowpatterns 625.1:Introduction 625.2:Methods 635.3:Results 655.4:Discussion 75

v

Chapter6:Nasalairflowandhandpreference 816.1:Introduction 81 6.2:Methods 816.3:Results 826.4:Discussion 95Chapter7:FinalDiscussionandConclusions 997.1:Nasalairflowasymmetryandcerebralasymmetry 997.2:Handpreferenceandnasalpassagedominance 1007.3:Limitationsofthisstudy 105References 106Appendix1:Inclusionandexclusioncriteria 116Appendix2:Participantinformationsheet 117Appendix3:Consentform 118Appendix4:EdinburghHandednessInventory(ShortForm) 119

1

Chapter1:IntroductionA(Overview)

1.1:Nasalairflowasymmetry

Thenoseisauniqueorganinthatitexhibitsasymmetricalairflowwiththedominant

airflowalternatingfromonenasalpassagetotheotheroveraperiodofhours (1).

Changesinnasalairflowaremediatedbyalternatingdilationandconstrictionofthe

venous erectile tissue in the nasal mucosa, by action of the sympathetic nervous

system(2).Thisalternationofnasalairflowisoftendescribedasa“nasalcycle”asit

canbe regularand reciprocal (1,3-5). The firstdescriptionof thisphenomenon is

attributed toKayser in1895 (6). Sen (1901)documentedobservationsofhisown

nasalmucosa,statingthatthealternatingcongestionanddecongestionofthenasal

erectile tissueoccurredwith“clockwork-like regularity” (7). Morerecentevidence

has suggested that regular periodicities are not always present and that there is

significant inter-individual variation in both reciprocity and rhythmicity (8, 9),

contradictory to the term “cycle”. Patterns of nasal airflow also vary within

individuals,with a lack of reproducibility seenwhenmeasurements of airflow are

repeated on different days (10, 11). Nonetheless, the presence of asymmetrical

nasalairflowthatfluctuatesspontaneouslythroughoutthedayhasbeenestablished

bymultiplestudies(4,8-13).

1.2:Definingthenasalcycle

As mentioned above, there is some confusion in the literature surrounding the

definition of the nasal cycle. Gilbert and Rosenwasser (1987, p.180) suggest that

accordingtotheclassicalideaofreciprocalandregularalternationsofnasalairflow

referredtoasthenasalcycle,“theleftandrightsideshaveidenticalperiods,are180

degreesoutofphase,andhavesimilarmeanairflowandamplitude”(8).However

this isnotnecessarilythecase,andinmanystudiesthetermnasalcycleisusedto

2

describe any asymmetries or patterns seen in nasal airflow. Without a clear

definition of what constitutes a nasal cycle, results of research studies can be

difficulttointerpret.ThisissuewashighlightedbyFlanaganandEccles(1997)who

developednumericalparameterstodefinethespontaneouschangesinnasalairway

resistancecommonlyreferredtoasthenasalcycle(13).Thenumericalparameters

suggestedwere the correlation co-efficient (r value) and airflow distribution ratio

(ADR)(13).Thecorrelationco-efficientisameasureofreciprocityrangingfrom-1,

whichwould indicateastrict reciprocal relationship, to+1,whichwould indicatea

strictly in-phase relationship (13). TheADR isameasureof the relativevolumeof

airflowthrougheachnasalpassagecalculatedasapercentageofthetotalvolumeof

airflowoveraneight-hourperiod,witharangeofzerotoone,whereoneindicates

thatthedistributionofairflowbetweenthetwonasalpassageshasbeenequalover

time(13).Theauthorsarguethatinordertobeclassedasanasalcycle,patternsof

nasalairflowshouldexhibitacertaindegreeofreciprocityaswellasequalvolume

distributionbetweenthenasalpassages,anddefinethisasanrvalueabove0.6and

an ADR above 0.7 (13). Using these numerical definitions, only 21% of healthy

subjects demonstrated a nasal cycle, although the majority exhibited a trend

towards reciprocity and equalisation of airflow suggesting some sort of pattern in

nasalairflowchanges(13).

HasegawaandKern (1978)defined thenasal cycle in rhinomanometric termsas a

changeofnasalairwayresistanceof20%ormorecomparingeachnasalpassagefor

two consecutiveobservations at least 15minutes apart (10). Using thisdefinition

theyfoundthat72%of50subjectsobservedovera7-hourperiodhadatleastone

nasalcycle(10).

Huang et al. (2003) defined the nasal cycle as significant negative correlation

between the two nasal passages, but did not specify what they deemed to be

significant(14).Intheiranalysisof10subjects,only5exhibitedanasalcycleasper

thisdefinition(14).

3

Kern(1981)suggestedtheterm“non-cyclenose”todescribeindividualswithoutthe

classical features of a nasal cycle (9). After studying nasal airflow patterns in 14

subjectswhodidnothaveanasalcycle,hesuggestedthreecategoriesforthenon-

cyclenose(9):

1. Nochangeinnasalresistance.

2. Moderatefluctuationsinnasalresistanceinonenasalpassagebutnochange

intheoppositeside.

3. Fluctuations in nasal resistance in both nasal passages but no reversal of

dominance i.e. greater airflow from one side to the other (in-phase

relationship).

Gallego et al. (2006) looked at the patterns of nasal airflow in children aged 2-11

yearsandcategorisedthenasalcycleintofourtypes(15):

• Classic–congestioninonenasalpassagewithaccompanyingdecongestionin

the other but no change in the total nasal airway volume, i.e. a reciprocal

relationship.

• Parallel – alternating congestion and decongestion with an in-phase

relationship.

• Irregular – changes in total nasal airway volume but with no discernable

pattern.

• Nopattern–nochangesintheunilateralortotalnasalairwayvolumei.e.no

nasalcycle.

Interestingly, all subjects demonstrated some sort of nasal cycle, but themajority

(50%)hadanirregularpattern(15).Itshouldbenotedhoweverthatthiswasashort

study with a measurement period of only 3 hours, and it could be argued that

patternsmayhaveemergedorchangedoveralongertimeperiod.

Whilstseveralauthorshaveattemptedtodefinewhatconstitutesanasalcycleusing

differentmethods,ageneralconsensushasnotyetbeenreached.Thepresenceof

fluctuatingpatternsofnasalairflowisnotdisputed,buttheterm“cycle”isprobably

notappropriategiventhat inmanyindividualsrhythmicandreciprocalalternations

4

in nasal airflow do not occur. For the purposes of brevity and clarity, fluctuating

patterns of nasal airflow will be referred to as the nasal cycle in subsequent

introductorychapters.

1.3:Functionofthenasalcycle

Thenasalpassagesarethenaturalpointofentryforairintotherespiratorytractin

healthyhumans,despitehavingagreaterresistancetoairflowcomparedtotheoral

cavity (16). The principal function of the nose is to prepare inspired air for rapid

gaseousexchangeinthelungs(17).Thisrequirescleaningandfiltering,heatingand

humidification.Howevertheexactpurposeofthenasalcycleremainsunclear.

Since the nasal mucosa continuously comes into contact with pathogens

encounteredininspiredair,Eccles(1996)suggestedthatthenasalcyclemayhavea

role in respiratory defence (18). The nasal mucosa has a rich blood supply, with

arteriesthatemergefromthebonywallsofthenasalcavitytoterminateincapillary

networks situated in the subepithelial layersof themucosa (17). Thesecapillaries

drainintovenoussinusoidswhichformvenousplexusesdeeperinthemucosa(17).

Thevenousplexusesareoftenreferredtoaserectiletissueastheycontainsmooth

muscularwallsandcanbeclosedbysphinctermuscles(17,19).Histologicalstudies

in rabbits revealed the presence of endothelial fenestrations in the subepithelial

venous sinusoids of the nasal mucosa in the regions of the nasal septum and

turbinates (20, 21). Interestingly, these fenestrationswere seen to open towards

theadjacentepithelium,andwerenotpresent in theveins situateddeeper in the

nasalmucosa (21). Based on these findings, Grevers (1993) suggested that these

fenestrated veinswereprobably involved in the regulationofnasal fluid secretion

(21).Theexudationofplasmafromthenasalmucosalvesselsisimportantasafirst

line defence mechanism (22). Not only does the plasma exudate contain

immunoglobulinsandpro-inflammatoryproteins,butinadditiontheflowofexudate

ontothemucosalsurfacecouldhelpwashawaypathogensandforeignmaterial(18).

5

The alternating congestion and decongestion of the venous sinusoids, known to

produce the nasal cycle, could in fact be acting as a pump mechanism for the

secretionofplasmaexudate(18).Duringthecongestedphase,hydrostaticpressure

willbe increased inthevenoussinusoidsduetothehigherbloodvolume,andthis

couldaidtheformationofplasmaexudate(18).Followingonfromthis,contraction

ofthesmoothmuscleinthewallsoftheveinsduringthedecongestivephasecould

squeeze the exudate through the fenestrations towards themucosal surface (18).

Thistheoryissupportedbyobservationsofanincreaseintheamplitudeofthenasal

cycleassociatedwithupperrespiratorytractinfection(23),whereintheexaggerated

congestion and decongestion in the venous system could lead to increased

productionofplasmaexudatethatisrichinimmunoglobulinsandpro-inflammatory

proteins(18).

It has also been suggested that thenasal cyclemaybe involved in controlling the

balanceofheatandwaterexchangerequiredtoconditioninspiredairinpreparation

for gaseous exchange in the lungs (16). The nasal airway is lined with liquid,

sometimestermedairwaysurfaceliquid(ASL),whichisderivedfromtheepithelium

andsubmucosalglandsandconsistsofapericiliarylayerwithoverlyingmucus(24).

This fluid is required for thehumidificationof inspiredairand for the transportof

trappedpathogens(24).Effectivemucusclearance,andhenceclearanceoftrapped

pathogens and particles, is inhibited if the ASL becomes dehydrated (25). Using

computermodels,White et al. (2015) demonstrated a possible role for the nasal

cycle in dealing with the difficulties of simultaneously carrying out both of these

important functions, i.e. air-conditioning andmucociliary clearance (26). Efficient

heating and humidification requires higher airflow velocities, whereas efficient

mucociliarytransportrequireslowerairflowvelocities(26).Duringthedecongested

phaseof thenasal cycle, severedehydrationof theASLoccursdue to the greater

airflowthoughthenasalpassage(26).Conversely,duringthecongestedphase,the

reduced airflow results in only minor dehydration of the ASL which leads to

improvedmucociliaryclearance(26).Alternatingthepartitioningofairflowthrough

eachnasalpassageallowsforoneside(thecongestedside)tomaintainitsASLand

carryouteffectivemucociliaryclearance,whilsttheothersidereceivesthegreatest

6

proportion of inspired air which can be efficiently heated and humidified but

consequently leadstoASLdehydration(26). Thisthenswitchesover,reducingthe

amountoftimethateachnasalpassageisexposedtodryingconditionsandperhaps

moreefficientlycarryingout theprinciple functionsof thenose– filtering,heating

andhumidificationofinspiredair(26).

1.4:Durationofthenasalcycle

Thereporteddurationofthenasalcyclevarieswidelyintheliterature.Thismayin

part be due to the problems surrounding the definition of the nasal cycle, as

discussedabove.Inadditionthereissignificantvariabilitybothwithinandbetween

individuals, and a lack of reproducible findingswhen the samemeasurements are

repeatedinthesameindividualsbutondifferentdays(10,11).Table1givessome

examples of average nasal cycle durations previously reported in the literature.

There are varying methods of measurement and sampling intervals, which again

couldbeafactorforthesignificantvariabilityseen.AsshowninTable1theshortest

duration recorded is 20minutes (27) and the longest is 7.3 hours (8). The short

duration of 20 minutes was recorded using continuous portable rhinoflowmetry

duringwakefulness,andthedevicewasgiventoparticipantssothatrecordingcould

take place during their normal daily activities (27). A longer cycle duration was

recordedduringsleep(27),whichhasbeenechoedinotherstudies(28).Tahamiler

etal.(2009)recordedanothershortdurationof30minutesusingOdiosoft-Rhino,a

software programme that detects the sounds generated during normal nasal

breathing (29). Their sampling period was 30 minutes, longer than some of the

other studies that failed todemonstrateaperiodicityof less than1hour (10,30).

Interestingly, they made their recordings by installing the programme on the

subjects’owncomputerssothatagainthetestcouldbecarriedoutinthesubjects’

ownenvironmentratherthan ina laboratory(as isthecaseformoststudies) (29).

They defined a nasal cycle as a statistically significant difference in the values

obtained from each nasal passage (29), which does not take into account the

7

reciprocityandregularityusuallyconsideredtobethehallmarksofacycle.

Table1:Examplesofnasalcycledurationsreportedintheliteratureandthedifferentmethodsusedtorecordthem.Study Number

ofsubjects

Agerange(years)

Methodofmeasurement

Durationofmeasurement(hours)

Samplingperiod(minutes)

Cyclelength(hours)Mean S.D. Range

HasegawaandKern1978(10)

50 24 Posteriorrhinomanometry

7 15 2.9 1-6

GilbertandRosenwasser1987(8)

16

18-34 Anteriorrhinomanometry

8 10 4.3 1.3 2.4-7.3

Gilbert1989(31)

9 18-30 Anteriorrhinomanometry

8 5 4.5 1.0 3.5-6

Huangetal.2003(14)

10 18-32 Posteriorrhinomanometry

6 10 3.5 2.3-4.4

Ohkietal.2005(30)

20 Meanage:45.8

Portablerhinoflowmeter

12 Continuous 1.8 - -

Tahamileretal.2009(29)

20 24-30 Odiosoft-Rhino 12 hours on 4differentdays

30 - 0.5-2.5

Rorhmeieretal.2014(27)

20 22-66 Portablerhinoflowmeter

23 Continuous 1.5 0.3-5.6

1.5:Controlofnasalairflow

Changesinnasalairflowaremediatedbyalternatingdilationandconstrictionofthe

venous erectile tissue in the nasal mucosa, by action of the sympathetic nervous

system(2).Thepresenceofavasomotorsupplytothenasalmucosawasfirstnoted

byTschalussowin1913(32).Experimentsoncatsdemonstratedthatstimulationof

thecervicalsympatheticchainledtoimmediatevasoconstrictioninthenasalveins,

indicatingtheoriginofthevasomotorfibres inthecervicalsympatheticchain(33).

Application of topical sympathomimetics causes decongestion of the nose and a

marked increase in nasal airflow (34, 35). Conversely, blockade of the stellate

ganglionor performanceof a cervical sympathectomyare known to cause venous

congestionandreducednasalairflow(36,37). Insummary, increasedsympathetic

8

tone causes vasoconstriction of the nasal veins, leading to an increase in nasal

conductance,andreductionorblockadeofsympathetictonecausesvasodilationof

thenasalveins,leadingtoanincreaseinnasalresistance.

Thereisgrowingevidenceforacentralcontrolmechanismofnasalairflow.Studies

onanaesthetisedcatsledtothenotionofcollectionsofsympatheticneuronsinthe

brainstem, so-called “oscillators”, with an asymmetry in sympathetic discharge

between the left and right oscillators resulting in asymmetry of nasal airflow

between the left and right nasal passages (38). The dominance of sympathetic

outputalternatesfromlefttorightandviceversa,causingthereciprocalchangesin

nasal airflow, both spontaneously and in response to electrical stimulation of the

brainstem (38). However ithasbeenpostulated thatoverall controloccursat the

levelofthehypothalamus,asinanimalstudies,electricalstimulationherecausedan

overall increase in sympathetic tone and greater nasal airflow bilaterally (33, 39).

Therefore, with the hypothalamus as the generator of a rhythmic nasal cycle,

increasedordecreasedhypothalamicoutputwillstimulatethebrainstemoscillators

symmetrically, but these brainstem oscillators will influence nasal airflow

asymmetrically due to their reciprocal differences in sympathetic discharge (11).

ThisissummarisedinFigure1.

9

Figure1:Modeltoexplainthecontrolofnasalairflow.Theoverallalternationinsympatheticactivityoveraperiodofhoursiscontrolledbythehypothalamusandtheasymmetryinsympatheticoutflowisdeterminedby theactivityofbrainstemoscillatorswhichactasa flip flopmechanism,witheachcentreinhibitingtheactivityoftheothercentreandonlyonecentrehavingdominantactivityatanyonetime.Thesehypothalamicandbrainstemmechanismshavebeenstudiedintheanaesthetisedcat(38,39).

Support for the existence of a central control mechanism determining the

rhythmicity and asymmetry of nasal airflow has been gained from observational

humanstudies.Inastudyofnasalairwayresistanceandaxillarysweatproductionin

7subjects,reciprocalalternatingcycleswiththesameperiodicitywereseenforboth

nasal airflow and axillary sweat production (40). Although not present in every

subject, itwasfoundthatanincreaseinrightaxillarysweatingwasassociatedwith

an increase in left nasal airflow (40). As sweating is controlled by the

thermoregulatory centre in thehypothalamus via the sympathetic nervous system

(41), the authors postulated the presence of a central cycle (40). Galioto et al.

(1991) observed an anomalous or absent nasal cycle in a small cohort of patients

10

withKallmansyndrome,adisordercausedbyhypothalamichypoplasia(42).Studies

on patients who had undergone laryngectomy revealed preservation of the

alternations in nasal airflow patterns as seen in healthy individuals but in the

absenceofnasalairflow,reinforcingthenotionofacentralratherthanlocalcontrol

mechanism (43, 44). Although the periodicities of the nasal cyclewere similar in

boththelaryngectomyandcontrolgroups,theamplitudewassignificantlylowerin

theformergroup,andtheauthorshavesuggestedthatnasalairflowreceptorscould

bemodifyingtheunderlyingcentrally-generatedpatternsofnasalairflow(44).

Corticalinfluenceonthehypothalamusandbrainstemintermsofthenasalcycleis

yet to be established, however there is some evidence to support a link between

corticalfunctionsandnasalairflow,whichwillbediscussedindetaillater.

1.6:Sensingnasalairflow

Thesensationofnasalairflowisbelievedtobemainlydeterminedbystimulationof

nasaltrigeminalnerveendingsthataresensitivetotemperaturechanges,especially

cooling of the nasal mucosa, as occurs during inspiration of air (1, 45, 46). This

phenomenonhasbeenreflectedinstudiesontheeffectofmentholonnasalairflow.

Mentholcausedacoolingsensationandasubjectiveimprovementinnasalairflowin

healthy subjects without any change in the total nasal airway resistance (47).

Followingtheapplicationoflocalanaesthetictothenasalmucosa,thesensationof

nasalairflowwasdiminished,aswastheeffectofmenthol(48).Underphysiological

conditions, the spontaneous changes innasal airway resistancebetween thenasal

passages occur without being detected since the total nasal resistance remains

relativelyconstant(18).

11

1.7:Measuringnasalairflow

Multipletechniquescanbeusedtomeasurenasalpatency,eitherinclinicalpractice

orforresearchpurposes.Subjectivemethodsincludeuseofavisualanaloguescale

(VAS)toassesstheperceptionofnasalairflow(49),andquestionnairessuchasthe

sino-nasaloutcometest20(SNOT-20)(50).Objectivemethodsincludeevaluationof

nasal cavity anatomy for example with computed tomography or acoustic

rhinometry, andmeasurements of nasal airflow such as rhinomanometry or peak

nasalflowrates(49,51,52).Anobjectivemeasurementofnasalairflowisusefulfor

theclinicianas itwillaid inthediagnosisofnasalobstructionandquantificationof

its severity, as well as in the assessment of treatment outcomes such as topical

decongestantagentsorsurgicalintervention.Itisforthesereasonsthattheability

tomeasuretheairflowthrougheachnasalpassageseparatelyisuseful(53).

Any method used to measure nasal airflow should ideally preserve normal nasal

anatomyandphysiology,beeasytouseandcomfortableforthesubjectorpatient,

measure physiological flow and pressure, and provide reproducible and relevant

results(54,55).Oneofthechallengesencounteredwhenmeasuringnasalairflowis

the physiological variability of resistance that occurs between inspiration and

expiration and between each nasal passage due to the nasal cycle (53). As air is

inspired through the nasal passages, the cartilages are drawn inwards, thereby

increasingtheresistancetonasalairflow.Theoppositeoccursduringexpiration.

The first attempts at objectively measuring nasal airflow were documented by

Zwaardemaker in1889,whousedacoldmirrorplacedhorizontallyunderthenose

tomeasurethecondensationareaproducedbyexpiredairfromeachnasalpassage

(54,56).Thiswasthefirsthygrometricmethodusedtomeasurenasalairflow(57).

Althoughmanymoretechniqueshavesincebeendeveloped,nosinglemethodhas

become commonplace in clinical practice (53). Some examples of othermethods

usedtodaytomeasurenasalpatencyaredescribedbelow.

12

Peaknasalinspiratoryflowrate(PNIF)

PNIF is a non-invasive method of measuring nasal airflow during maximal forced

nasalinspiration(51).Itthereforeprovidesanindirectmeasurementofnasalairway

resistance, as an increased resistance alters the peak flow rate achievable (58).

However it isnota reflectionofnasal resistanceduringnormalbreathing (51)and

can miss smaller obstructions to nasal airflow (52). The device used is portable,

relatively simpleand requiresminimal trainingof staff and subjects (58). Another

advantage is the availability of baseline values for adults which can be used for

comparison(52). Subjectcooperation is required,and lackofeffortor fatiguecan

affect the results (58). The subject’s respiratory function and the presence of

secretionscanalsohaveaconfoundingeffect(51,58).

Rhinomanometry

Nasalairflowoccursdue to thepressuredifferencebetween theenvironmentand

the nasopharynx (51, 59). Rhinomanometry measures the pressure and airflow

differences between the anterior and posterior nasal cavity that occur during the

inspiration and expiration of air through the nose (59). These measurements of

nasalairflowandpressurecanthenbeusedtocalculatenasalairwayresistance(58).

Rhinomanometry can provide an assessment of each nasal passage separately or

together,dependingonthetechniqueused. Duringanterior rhinomanometry, the

sensing tube is taped toeachnostril in turnprovidingavalue for thenasalairway

resistance for the left and right nasal passages separately (59). During posterior

rhinomanometry, the sensing tube is placed in the subject’s mouth, providing a

measurement of the total nasal airway resistance (59). This technique requires

subject co-operation and trainingwhich can be difficult (58). Rhinomanometry is

considered to be more sensitive and specific than acoustic rhinometry (52).

However it involves the use of more complex and bulky equipment, which is

expensive, and as such it is more useful in research laboratories than in clinical

practice(51,58).

13

Acousticrhinometry

Acoustic rhinometry involves the use of acoustic reflection to demonstrate the

geometry and cross-sectional area of the nasal cavity (60). A sound wave is

transmitted into thenasalpassages,and the reflectedwavepatternsaredetected

andanalysed(58).Variationsinthesizeorcontourofthenasalairway,forexample

septal deviation or the presence of a nasal polyp,will distort the reflected sound

wave (51). Acoustic rhinometry does not require any nasal airflow to obtain

measurementvalues,meaningthatitcanbeusedforthesubjectwithacompletely

occludednose (60). In addition, it doesnot requiremuch subjectparticipationor

cooperation (60) and therefore isuseful in childrenwhocould strugglewithother

methods.Studieshavedemonstratedgoodreproducibility,withalowcoefficientof

variationbetweenrepeatedmeasurements(60).Acousticrhinometrydoeshowever

have several disadvantages. Should an obstruction in the anterior nasal cavity

prevent transmission of sound waves, there is no way of measuring the area

posteriortothatobstruction,thuslimitingtheviewoftheoverallnasalairway(60).

It does not provide information about nasal airflow i.e. a measurement of nasal

function (58), which can be more relevant to the patient and clinician than

anatomicalfindings.Itisalsounabletomeasureeachnasalpassageseparately,but

providesacross-sectionalreflectionofthenasalcavity(52).

Nasalspirometry

The techniqueofnasal spirometrywasdevised in2001asamethodofmeasuring

the distribution of nasal airflow through each nasal passage, termed the nasal

partitioning of airflow ratio (NPR) (61). It provides ameasure of the laterality of

nasalairflow.Adeviceisusedtomeasurethevolumeofairexpiredfromeachnasal

passagefollowingmaximalinspiration(61).Itiseasytousebyboththesubjectand

the investigator, requiringminimal training (62). Otheradvantagesof thismethod

include its good correlation with rhinomanometry (61) and the availability of a

reference range for healthy adults (63). The accuracy and reproducibility of nasal

spirometrywasdemonstrated inastudyusingamodelof thenasalcavity (64). It

has also been used successfully in clinical studies, for example as an objective

14

measureofnasal septaldeviation (62) and fordetectingpostural changes innasal

airflowinpatientswithacuterhinitis(65).

ThemodifiedGlatzelMirror(GM)

Following on from the cold mirror method invented in the late 1800s, in 1904,

Glatzel described a metallic mirror marked with four lines that could be used to

quantify nasal airflow. This was modified by Cocks (1915) who added multiple

horizontalcurvedlinesandparallelvertical lines,both1cmapart(66). Toobtaina

measurement,themirrorisheldhorizontallybeneaththenostrils,andthesubjectis

instructedtobreathoutthroughthenosenormally,withthemouthclosed(54,66).

The temperature difference between the nasal cavity and the plate causes

condensation of the vapour in the expired air, producing a brief reflection of the

nasalairwayontheplate(54).

Gertneretal.(1984)describedtheirtechniqueindetail(54),andthereproducibility

ofthismethodspecificallyhasbeenanalysedmorerecently(55).Theplateusedwas

a 10cm x 12cm polished chrome-coated metal plate marked with arches at 1cm

apart(54). Theplatewaspositionedhorizontally immediatelyunderthecollumela

withtheverticalaxisat90degreestowardstheupperlip(54).Thepatientwasthen

askedtoexhalethroughthenoseslowlywiththemouthclosed(54)(seeFigure2).

Theareaofcondensationproducedduringnasalexhalationwasthenmeasured,and

intheabsenceofnasalpathologythiswastypicallyanovalshapedarea7-8cmlong

and4-5cmwide(54)(seeFigure3).

The advantages of this method are immediately obvious; it is easy to use, non-

invasive, quick, inexpensive and does not distort nasal anatomy (54, 55). It does

howeverhavelimitations,forexampletheusertechniquecancausevariabilityinthe

resultsobtained,andthereisnomethodofregulatingairflowwhichcouldalsolead

toerroneousreadings(67).

15

Thisfigurehasbeenremovedbytheauthorforcopyrightreasons.

Figure2:TakenfromGertneretal.’spaper(1984)(54).AphotographofasubjectexhalingthroughthenoseontoamodifiedGM.Thisfigurehasbeenremovedbytheauthorforcopyrightreasons.

Figure 3: Taken fromGertner et al.’s paper (1984) (54). An illustration of the condensation areasexpectedtobeproducedbyanormalindividualwithnonasalpathology.

16

IthasbeensuggestedthattheonlyreliableresultsobtainedbythemodifiedGMare

qualitative in nature, as quantitative values may be unreliable due to potential

variabilitybetweenreadingsandtheshort-livedappearanceofcondensationareas

(49).Howeveronecouldarguethatthereliabilityandusefulnessofresultsdepend

on the purpose, for example a brief view of nasal patency for the purpose of

diagnosingthepresenceandlateralityofnasalobstructioninclinicalpracticecanbe

extremelyuseful.

Brescovici and Roithmann (2008) assessed the reproducibility of themodifiedGM

method in the assessment of nasal patency, specifically utilising the technique

described byGertner et al. (1984) (55). They analysed theminute-by-minute and

hour-by-hour variability of measurements in a cohort of healthy individuals (55).

The mean variation coefficient was found to be less than 15% for unilateral

measurementsandlessthan12%fortotalmeasurementsatdifferenttimeintervals,

whichiscomparabletoothermethodsofmeasuringnasalpatency(55).Itwasalso

abletoreliablydemonstratethechanges innasalairflowcausedbythenasalcycle

(55). Another study comparing the efficacy of themodified GMwith peak nasal

inspiratoryairflowmeasurementsinchildrenidentifiedthemodifiedGMassuperior

inthediagnosisofnasalobstruction(68).

Several factors can affect the reliability and reproducibility of measurements

obtained by use of themodifiedGM. These include patient factors, such as vital

capacity and nasal obstruction, environmental factors such as temperature and

humidity,andpositionalfactorssuchasthepositionoftheplaterelativetothenasal

passages(66). Particularlyforresearchpurposes,thesecanbeminimisedbyusing

anatomical landmarks to correctlyposition theplate,using the sameexaminer for

repeated measurements, and allowing subjects to acclimatise to the room

temperature and humidity for 30minutes prior to takingmeasurements (55). In

addition,usingtheaverageof3-5readingscanimprovereliability(55).

The modified GM has been used as an outcome measure in several studies

attemptingtoanalysetheeffectsofnasalsurgery(54,67,69).Gertneretal.(1984)

17

demonstratedthatthemodifiedGMwasusefulinclinicalpractice,asitcoulddetect

differences in nasal airflow before and after septal surgery used to treat nasal

obstruction (54). However another study assessing changes to nasal airflow

followingrhinoplastyprocedureswiththemodifiedGMfailedtoshowasignificant

differenceinthevaluesobtainedpre-andpost-operatively(67).Theauthorsofthis

study commented that this could have been due to the effects of the nasal cycle

(67).Itshouldbenotedhoweverthatthisstudyonlyinvolvedasmallcohortof20

patients who were undergoing cosmetic rhinoplasty, and therefore the aim of

surgerymaynothavebeentoimprovenasalairflow.Aswithothermethods,studies

haveidentifiedapoorcorrelationbetweensubjectivemeasuresofnasalairflow(e.g.

patient questionnaires) and objectivemeasures taken using themodifiedGM (55,

67).

ThemodifiedGM isparticularlyuseful in childrenas it isnon-invasiveandeasy to

use, not requiring any training. It has been used successfully in several studies

involvingchildren (68,70). Thecoldmirrormethodhasevenbeenused inanimal

studies in order to demonstrate the presence of a nasal cycle (71). A coldmetal

mirrorwasplacedunderthenostrilsoftranquilisedratsandrabbits,andthesideof

the nose (left or right) producing the greatest area of condensationwas recorded

(71). Thiswas comparedwithmeasurements of airflow recorded inmillilitres per

minute by collecting air that escaped from each nostril under water following

insufflation into the trachea (71). The cold mirror recordings were shown to

correlatewiththemeasurementsofflow(71),supportingitsreliabilityandaccuracy.

Recently,thecoldmirrormethodhasbeenadaptedfurtherbytheadditionofvideo

technology which is termed the video-rhino-hygrometer (49). This involves the

recording of the condensation areas produced by nasal expiration followed by

computerprocessinginordertoextractmoredetailedquantitativeparameters(49).

When this method was used to evaluate the post-operative outcome following

septoplasty,itwasabletoidentifyanimprovementincomparisontopre-operative

measurements (49). However in the same group of patients, there was no

18

statisticallysignificantcorrelationbetweenmeasurementsobtainedbyvideo-rhino-

hygrometryandthoseobtainedbyanteriorrhinomanometry(49).

1.8:Factorsthataffectnasalairflow

Numerousstudieshavedemonstratedthatnasalairflowcanbe influencedbyboth

pathologic and physiologic conditions. Some significant examples are discussed

below.

Age

Cross-sectional studies of adults aged 16 to 82 years have demonstrated that age

doesnotaffectnasalairflowrate,nasalcross-sectionalareaornasalresistance(72,

73).Howevercross-sectionalstudiesmaynotbereliableduetothehighdegreeof

variabilitybetweensubjectsseeninstudiesofnasalairflow,andtheauthorssuggest

thatchangestothenasalmucosathicknessandelasticityarealikelyconsequenceof

ageing (72). It seems that ageing can affect the rhythmicity and reciprocity in

patternsofnasalairflowtypicallyassociatedwiththenasalcycle. Inastudyof60

subjectsaged18to85years,nasalairflowwasmeasuredevery15minutesover6

hours (74). In subjects under the age of 50 years, 22%exhibited a classical nasal

cycle, defined by the authors as statistically significant negative r values (i.e.

significant reciprocity), compared to 9% of subjects aged over 50 years (74). In

addition,thepresenceofacyclicpatterns,definedasnofluctuationsinnasalairflow

ineithernasalpassage,wasmore likelywithadvancingage (74). Ina longitudinal

studyof a single subject, nasal airflowmeasurementswere takenalmost 40 years

apart, and a significant decline in reciprocity was identified (75). Although the

reasonsfortheseage-associatedchangeshavenotbeenfullyestablished,itislikely

that they are due to age-related changes in the central nervous system, since

fluctuations in nasal airflow patterns are controlled by the hypothalamus and

brainstem(74,75).

19

Posture

Positional changes can affect nasal airflow. Changing from a sitting position to a

recumbentpositioncausesariseinjugularvenouspressure,increasingbloodflowto

thehead, includingthenose (76). Thiswill leadtovasodilationof thenasalveins,

increasing the total nasal airway resistance, and one study identified an 8.4%

increaseinnasalresistancewhenchangingfromasittingtoasupineposition(77).A

pressurestimulustoonesideofthebody,forexamplewheninthelateraldecubitus

position, leads to ipsilateral vasodilation (congestion) and contralateral

vasoconstriction (decongestion) in the nasal passages (78-80). This has been

demonstrated by application of axillary pressure bymeans of a small crutch, and

whenappliedtothesamesideasthedominantnasalpassage,itcausedreversalof

nasal passage dominance within 5 minutes (80). Twelve minutes of lateral

recumbency was shown to induce the same changes in nasal airway resistance,

whichinterestinglywerefoundtocontinueafterthepressurestimuluswasremoved

(79). It has been proposed that this phenomenon is mediated by a reflex arc,

sometimes called the corporo-nasal reflex, involving sensory receptors in the skin

and sympathetic innervation to the nasal mucosa (78). These are likely to be

pressure sensors situated in the thorax, pectoral and pelvic girdles with efferent

fibrestravellingviathecervicalsympatheticchaintothevenoussinusesinthenose

(79).

Sleep

Spontaneousfluctuationsinnasalresistancealsooccurduringsleep(28,81).Sleep

seemstoaffectthenasalcycle,causinganincreasedperiodicitywhencomparedto

wakefulness inthesamesubjects(27,28). Inastudyof20healthysubjects,there

was a significant reduction in thenumberof cycles, i.e. reversals of nasal passage

dominance,thatoccurredduringsleepcomparedtowakefulnesswhennasalairflow

wasmeasured continuously over a 24 hour period (28). In another similar study,

reversalsofnasalpassagedominancewereprecededbyachange inbodyposition

almost 60% of the time (27). There is also an increase in nasal cycle amplitude

during sleep, caused by increased congestion in the congested nasal passage,

howeverthisismorelikelyaresultofrecumbencyratherthansleep,consistentwith

20

theexistenceofthecorporo-nasalreflex(27).Itshouldbenotedhoweverthatboth

of these studies used portable rhinoflowmetry to continuously measure nasal

airflow,whichrequirestheinsertionofnasalprongs intotheanteriornares. Nasal

prongshavebeenshowntosignificantlyincreasenasalairwayresistanceeveninthe

absence of nasal pathology, as they reduce the cross-sectional area of the nasal

passages(82).Thereisalsoachancethattheycouldbecomedislodgedormisplaced

especially during sleep, or cause irritation to the nasalmucosa,which again could

affecttheaccuracyofmeasurements.

There is some evidence to suggest that reversal of nasal passage dominance

predominantlyoccursduringrapideyemovement(REM)sleep(28,83),evenwhen

postural effects are taken into account (28). REM sleep is associated with

sympatheticnervoussystemactivationwhichcouldexplainthisfinding(27).

Exercise

Exercise reduces the total nasal airway resistance, abolishing the asymmetry

betweenthenasalpassages(84,85).Re-breathing(resultinginanelevatedplasma

carbondioxide level)alsoreducesthetotalnasalairwayresistance(84). It is likely

that the increased metabolic demand caused by these scenarios leads to a

sympathetic response resulting in vasoconstriction and reduced nasal airway

resistance(84).

Environment

Onestudyusedacousticrhinometryasanobjectivemeasureofnasalpatencyin50

healthy subjects at two different room temperatures; 30-33°C and 18-22°C, and

found no statistically significant difference caused by the change in room

temperature (86). Ina studyofeighthealthysubjects, ingestionofhotwaterand

nebuliser treatment (i.e. heat and humidification) caused a significant decrease in

thevolumeofthenasalcavitythatwasalreadythemostcongested,althoughthere

wasnochangeintotalnasalcavityvolume,againmeasuredbyacousticrhinometry

(87).Drinkingcoldwaterandplacementofanicepackontheneckdidnothaveany

significanteffect(87).

21

Hormones

There is some evidence to suggest that female reproductive hormones may be

associated with changes in nasal airflow, however conflicting findings have been

reported.Duringpregnancy,itisthoughtthathormonalchangesareresponsiblefor

pregnancy-induced rhinitis, which is relatively common in the later stages of

gestation(88,89).Inthiscondition,nasalsymptomsdevelopduringpregnancyand

resolvespontaneouslysoonafterdelivery(90).Inalongitudinalcohortstudyofover

2000 pregnant women, self-reported “nasal stuffiness” increased throughout

pregnancy,occurringin42%ofwomenat36weeksgestation(91).

This phenomenon led to analyses of the effects of the menstrual cycle and

contraceptivedrugsonnasalairflow. Ina studyof41healthywomen,nasalpeak

expiratory flow rate was found to be significantly lower during menstruation

comparedwiththerestofthemenstrualcycle,indicatingincreasednasalcongestion

at this time (92). It hasbeen suggested that elevated levelsofoestrogenmaybe

responsibleforfluctuationsinnasalcongestionoccurringduringthemenstrualcycle,

however oestrogen levels are lowest during menstruation (92). Using

rhinostereometry, Haeggstrom et al. (2000) observed nasal hyperreactivity during

ovulation(whenoestrogenlevelsarehighest)bychallengingthenasalmucosawith

histamine (93). However when acoustic rhinometry was used to measure nasal

congestion, no significant differences were identified at different phases of the

menstrual cycle (93). The authors suggest therefore that the sensation of nasal

congestionassociatedwithpregnancycouldbedue tohyperreactivityof thenasal

mucosa caused by high oestrogen levels, rather than increased congestion of the

nasalmucosa(93).

Theoppositewas found inanother study,wherenasal resistancewas found tobe

significantlyhigheraroundthetimeofovulationcomparedtothemenstrualphase

asmeasuredbyanteriorrhinomanometry(94).Biopsiesofthenasalmucosataken

duringdifferentphasesofthemenstrualcycleshowednodifferencescomparedto

malecontrols insevenoutoftenwomen(95). Thethreewomenwithhistological

changesdemonstrated increased vascularitywithdilated and congested capillaries

22

and glandular hyperactivity, which the authors suggested may be related to

emotionaldisturbancesinfluencingtheautonomicnervoussystem(95).

A large studyofover300women found thatolfactory sensitivity reached itspeak

during the ovulatory phase of the menstrual cycle and was higher at this stage

compared tocontrols (96). The lowest sensitivitywas foundduring themenstrual

phaseof themenstrual cycle (96). It is thought that changes in oestrogen levels

throughoutthemenstrualcyclecouldbealteringolfactoryfunctionperipherallyand

centrally(96).Thesefindingsareparticularlyinterestinggiventhatsomestudieson

nasalairflowhavesuggestedthatnasalcongestionandhyperreactivityoccuratthe

timeofovulation(93,94),whichwouldusuallyreduceolfactorysensitivity. Clearly

this relationship between female sex hormones, nasal airflow and olfactory

sensitivityiscomplex,anditsdefinitiveexistencehasnotbeenfullyestablished.

Animalstudieshaverevealedthatadministeringoestrogenscausesvasodilationwith

an increase in the size of vascular spaces in the nasal mucosa, with the same

histological changes occurring in pregnancy (97). Similar effects have been

demonstrated in human histological studies. Toppozada et al. (1984) took nasal

biopsiesfrom25womenonthecombinedoralcontraceptivepill,15ofthesewomen

had developed nasal symptoms after starting the drug (98). Although minor

changessuchasglandularhyperactivitywereobservedintheasymptomaticgroup,

the symptomatic group had changes similar to those seen in hypertrophic non-

allergicrhinitis,suchasinterepithelialoedema,glandularhyperplasiaandcongested

capillaries(98).Howeveramorerecentstudyfoundthatthemoderncombinedoral

contraceptivepilldidnothaveanyeffectonnasalairflowasmeasuredbyavariety

ofmethods(99).

It would seem that the effects of hormones such as oestrogen and progesterone

have the potential to cause changes in the nasal mucosa and even symptomatic

rhinitis, however significant variability has been identified in the prevalence,

duration and severity of symptoms. Whether these changes alter nasal airflow is

even more controversial, and conflicting results have been presented in the

23

literature.Evenifthereareminorchangesinnasalairflowoccurringthroughoutthe

menstrual cycleorwithuseof contraceptives, thisdoesnotnecessarilymean that

thenasalcyclewillbeaffected.Equally,thereisevidencetosuggestthatmenhave

agreaternasalairflowratethanwomen(72),mostlikelyduetotheirincreasedvital

capacitywhichshouldnotaffectthenasalcycleduration,onlyit’samplitude.Most

importantly, a relationship between gender and the nasal cycle has not been

identified.Mirzaetal.(1997)measurednasalairflowevery15minutesfora6-hour

periodin60subjectstodeterminefactorsthataffectnasalcyclepatterns,andfound

that gender had no significant effect (74). Tahamiler et al. (2009) found no

significantdifferencesinnasalcyclingbetweenmaleandfemalesubjectswhennasal

airflow was measured every 30 minutes for 12 hours on 4 different days in 20

subjects(29).

Drugs

Many drugs have been shown to affect nasal airflow generally and also alter the

nasal cycle. For example alcohol, known to have vasodilatory effects, has been

shownto increasetotalnasalairwayresistance (100,101). Cigarettesmokingalso

hasaneffect.Inastudyofover2500adults,thosewhowerecigarettesmokershad

lowernasalcavityvolumesandpeaknasalinspiratoryflowvaluescomparedtonon-

smokers,evenafterdecongestion(102).Itislikelythatchronicdamagetothenasal

mucosa caused by cigarette smoke is responsible for these changes (102). Topical

decongestantssuchasadrenalineandxylometazolinecausevasoconstrictionofthe

dilated veins, leading to symmetrical nasal airflow (12, 14). Anti-inflammatory

medicationsuchasanti-histaminesandcorticosteroidsalsodecongest thenoseby

reducinginflammation(103,104).Nasalobstructionandrhinitishavebeenreported

inpatientstakinganti-hypertensivessuchasACE-inhibitors(105)andbeta-blockers

(106).

Disease

Severalpathologicfactorscan influencenasalairflowandthenasalcycle, including

local factors suchasanatomicalobstructionormucosal inflammation,andgeneral

factorssuchasrespiratoryfunction.Anatomicalobstructionsuchasseptaldeviation

24

reduces nasal airflow through the affected nasal passage (54). In a comparative

study, the amplitudeof fluctuation innasal airflowwas greater in thewidernasal

passage in patients with anterior septal deviation compared to those without,

althoughtherewasnochangeinthedurationofthenasalcycle(107).Itispossible

that severe nasal septal deviation can eliminate the nasal cycle i.e. prevent any

reversal in the dominance of nasal airflow (28), although the evidence for this is

weak and as discussed previously the great variation in nasal airflow patterns

generallycouldaccountforthisobservation.

Nasal resistance is increased in allergic and chronic hypertrophic rhinitis (77). In

addition,acuterhinitis,forexampleinupperrespiratorytractinfection,exaggerates

thenasalcycle,causingagreatermaximalnasalairwayresistanceandanincreasein

the amplitude of reciprocal changes between to the two nasal passages (12, 23).

Restrictedorreducedlungmovement leadstoreflexvenousdilatationinthenasal

mucosaandreducednasalairflow(37).

Overthepastdecade,evidencehasemergedtosuggestthatnasalairflowpatterns

arealteredinneurodevelopmentalandpsychiatricdisorders(70,108,109),although

this notion remains controversial with only a few studies published on the topic.

Thisconceptwillbediscussedinmoredetaillater.

Summary

There aremany factors that can influence nasal airflowbutmost of these do not

affectthecharacteristicsofthenasalcycle.Thereareafewexceptions.Asymmetry

between the nasal passages can be abolished by stimulation of the sympathetic

nervoussystem,forexampleduringexercise(84,85)orfollowingtheapplicationof

topical sympathomimetics (12). Increasing age is associated with reduced

reciprocity in the alternating fluctuations of nasal airflow (74, 75). Changes in

posture,specificallyresultinginapressurestimulustolateralaspectsofthebodyfor

exampleduring lateral recumbency, triggera reversal innasalpassagedominance,

such that the greater airflow is observed in thenasal passage contralateral to the

pressurestimulus (78,80). Pathologicalconditionssuchassevereseptaldeviation

25

have the potential to prevent fluctuations in nasal airflow (28), and acute rhinitis

exaggerates the amplitude of reciprocal changes between the two nasal passages

(12,23). These influencesneedtobeconsideredcarefullywhendesigningstudies

thataimtoassessthenasalcycle,andshouldbeadequatelycontrolledforwherever

possible.

1.9:Nasalairflowandhandpreference

The cerebral hemispheres exhibit both functional and structural asymmetry (110).

Perhapsoneof themostobvious signsof cerebral asymmetry is handpreference.

Broca proposed that the hemisphere controlling speech was contralateral to the

dominant hand, suggesting that in most people the left hemisphere is dominant,

however this relationship is actually more complex (111). The majority of the

populationisright-handdominant,hasspeechcontrolledbythelefthemisphereand

visuospatialprocessingcontrolledbytherighthemisphere(110,112).

In 2005, Searleman and colleagues hypothesised the existence of a correlation

betweennasalairflowandhandedness(113).Theydescribedtwoprinciplereasons

forthebasisoftheirhypothesis.First,thereisatendencyforapositivecorrelation

between lateral preferences, including both limb sidedness and sense organ

sidedness,sothat left-handers tendalsotobe left-footed, left-eyedand left-eared

(113). Secondly, there are two separate nasal passages which function

independently. As discussed in detail earlier, airflow through each nasal passage

fluctuatesspontaneouslyoveraperiodofhours(1,3,4),unlikeairflowthroughthe

bronchi and lungs, which under normal circumstances (i.e. in the absence of

pathology) is divided equally between the left and right sides. At any given time

point one nasal passage is decongested, i.e. has greater airflow making it the

dominantside,whilsttheotheriscongested,i.e.haslessairflowthusmakingitthe

non-dominantside.Sincetheexactnatureandpurposeofthesefluctuatingpatterns

26

ofnasalairflowremainelusive,itseemsreasonabletoconsiderstructuralinfluences

suchascerebralorganisationandlateralpreferences.

Searlemanetal. (2005)aimed toanswer twoquestions; is thereadominantnasal

passageand if so, is it consistentwithother lateralpreferences, forexamplehand

dominance(113)?Theypredictedthattheleftnasalpassagewouldbedominantfor

the majority of the time in left-handers, and vice versa in right-handers (113).

Searleman et al.’s (2005) study involved 20 healthymale participants, aged 18-22

years, 11 of whom were right-handed and 9 of whom were left-handed (113).

Women were excluded due to concerns that alterations in hormone levels could

influence nasal airflow (113). Handedness was determined by self-report and

observationofhand-writing(113). Nasalairflowwasmeasuredusingtwohotwire

anemometerspositioned just inside thenares, and themaximumairflow ratewas

used as the measure of nasal passage dominance (113). Measurements were

performedat15-minuteintervalsoveracontinuous6-hourperiod(113).Attempts

were made to minimise potential confounding factors by performing anterior

rhinoscopytoruleoutobstructivenasalpathology,excludingsmokersorthosewith

anabnormalsenseofsmell,ensuringcorrectandconsistentpositioningduringnasal

airflow measurements, and prohibiting exercise and lying down on the test day

(113).

In order to look for a correlation between hand preference and nasal passage

dominance, Searleman et al. (2005) analysed the percentage of time each nasal

passagewasdominant, basedonmaximumairflow rates recorded fromeach side

(113). In left-handers, the leftnasalpassagewasdominant for59.3%of thestudy

period,whereasinright-handers,therightnasalpassagewasdominantfor59.5%of

the studyperiod (113). Thep value forbothof these findingswas less than0.01,

howeverthemethodsofstatisticalanalysiswereunclear(113).Theyalsolookedfor

anycorrelationbetweenthenumberofnasalcyclesandhandedness.Left-handers

had an average of 5.67 nasal cycles in the 6-hour measurement period, whereas

right-handershadanaverageof3(113).Usingthisinformation,theycalculatedthat

the average duration of a nasal cycle was 63.1 minutes in left-handers and 120

27

minutes in right-handers (113). These findings were not statistically significant,

althoughagainthestatisticalmethodsusedwerenotspecified.

Thereareseveral issueswithboththemethodologyanddata interpretation inthis

article:

1. A small sample size of only 20 subjects has been used. As there is no

previoussimilarliterature,powercalculationswouldnothavebeenpossible.

Inaddition,studiesthatrequiremeasurementandanalysisofthenasalcycle

are,bydefinition, timeconsuming,and therefore themajorityofpublished

reportsonlyhaveasmallnumberofsubjects.

2. Themethod of measuring nasal airflow required the insertion of hot wire

anemometers just inside the nares, which could have interfered with the

accuracyofmeasurements.

3. Noformaltestwasusedtodeterminehandednessdespitetheavailabilityof

validatedobjectiveandsubjectivemeasures.Handednessisnotassimpleas

left or right, but is in fact a continuum, with degrees of left and right-

handedness (114). Studies that utilise hand preference as a variable can

thereforebedifficulttointerpret.

4. Dataanalysisandstatisticalmethodshavenotbeendescribedinanydetail,

making itdifficulttoreliably interprettheresults. Itwouldappearthatthe

number of times each nasal passage was dominant during the 6-hour

measurementperiodhasbeencalculatedandusedtoproduceapercentage

value for theamountof timeeach sidewasdominant. A subjecthas then

beendefinedasbeing leftnasalpassagedominant if the leftnasalpassage

hadthegreatestairflowformorethan50%ofthetimepoints,andviceversa

for the right nasal passage. This figure has then been used to show a

correlationwithhandedness.Asthereissignificantinter-andintra-individual

variabilityinnasalairflowpatterns,using50%asthecutofffornasalpassage

dominance could be unreliable. With only a few more measurements,

extending the sampling period to 7 hours for example, the percentage of

timeonenasalpassagewasdominant couldeasily change to just aboveor

justbelow50%,completelyalteringtheoutcomes.

28

5. When calculating the average number of nasal cycles in left-handers and

right-handers, there isnodescriptionordefinitionofwhat ismeantby the

term“nasalcycle”inthecontextofthisdata.

The findingofa correlationbetweennasalairflowandhandpreference isunusual

forseveralreasons. Manydifferentresearchgroupshaveperformedobservational

studiesonnasalairflowinhealthyindividuals,andalthoughtheyhavenotexplicitly

lookedforarelationshipbetweennasalairflowandhandpreference,ithasnotbeen

identified incidentally. Since 90% of the population is right-handed, surely other

studiesofnasalairflowwouldhavenotedthatthemajorityoftheirparticipantshad

rightnasalpassagedominance. Althoughtheexact functionof thenasalcyclehas

not been firmly established, there is evidence to suggest that it has functional

significancewithpotentialrolesinimmunedefenceandairconditioning(18,26).A

correlation between nasal airflow and hand preference would conflict with these

theories by suggesting that nasal airflow patterns were linked with cortical

organisation rather than having a functional role in respiratory defence or

preparationofinspiredairforgaseousexchange.

ThearticlebySearlemanetal.(2005)(113),publishedover10yearsago,istheonly

availableliteratureconcerninghandpreferenceandnasalairflowinhealthysubjects,

anditisnotcommonlycitedinthemorerecentliteraturebaseforthenasalcycle.In

2007,DaneandBalcipublisheda study that comparedhandpreferenceandnasal

airflow patterns in 37 autistic children and 20 healthy controls (70). Autism is a

neurodevelopmental disorder in which cerebral lateralisation is abnormal (70).

Children with autism have a lesser degree of handedness, i.e. are less lateralised

thancontrolsandmorelikelytodemonstratemixedorleft-handedness(70).When

nasalairflowwasmeasured24timesovera12-hourperiodusingthemodifiedGM,

themajorityofautisticchildrendemonstratedleftnasalpassagedominanceforthe

majorityof the timeperiod (70). Of the24measurements, the leftnasalpassage

wasonaveragedominant19.59timesintheautisticgroup,comparedto11.75times

inthecontrolgroup,andthisdifferencewasfoundtobestatisticallysignificant,with

apvalueof0.00(70).Theauthorsdidnotlookspecificallyforacorrelationbetween

29

handpreferenceandnasalpassagedominancebutdidfindthatintheirgroupof37

autisticchildren,therewasahigherprevalenceofleftormixedhandednessandof

leftnasalpassagedominancecomparedtocontrols(70).

Two similar studies investigatednasal airflowpatterns in right-handed adultswith

schizophrenia andmood disorders. ThemodifiedGMwas used tomeasure nasal

airflow every 30minutes over a 12-hour period in 83 patientswith schizophrenia

(108), 26 patients with unipolar depression and 44 patients with bipolar disorder

(109)andcomparedwith64healthycontrols.Psychoticdisordersmaybeassociated

withabnormalbrainasymmetry,andtherehavebeenreportsofincreasedratesof

mixed and left-handedness in patients with schizophrenia (112, 115, 116). All

subjects in thesestudieswere right-handed. In schizophrenicpatients, the rateof

leftnasalpassagedominancewas65.1%,comparedtoarateof22.9%forrightnasal

passagedominance,afindingwithapvalueof0.00.Inhealthycontrols,therateof

leftnasalpassagedominancewas25%, the rateof rightnasalpassagedominance

was21.9%,andthemajorityhadnosignificantlateralisationofnasalairflowoverthe

12-hourmeasurementperiod(108).Patientswithbipolardisorderhadahigherrate

of right nasal passage dominance, whereas those with unipolar depression had a

higher rate of left nasal passage dominance, however these findings were only

significant in female subjects (109). The authors of these studies have suggested

that the altered nasal airflow patterns apparently associatedwith these disorders

may be contributing the hypo- or dys-function in the cerebral hemispheres (108,

109). Specifically,Ozanetal.(2009)proposedthatgreaterairflowthroughtheleft

nasal passage for the majority of the time corresponds to increased sympathetic

tone and therefore reduced electrical activity in the left cerebral hemisphere,

contributingtothelefthemisphericdysfunctionseeninschizophrenia(108).

Theabovefindings inhealthycontrolsubjectsareparticularlyrelevant. Overa12-

hourperiod,53.1%hadnooverall lateralisationofnasalairflow,meaning that the

left and right nasal passages were dominant for roughly equal amounts of time

(108).Therateofleftnasalpassagedominancewas25%,andtherateofrightnasal

passage dominance was 21.9% (108). All subjects were right-hand dominant as

30

measured by the Edinburgh Handedness Inventory, a validated questionnaire for

handedness.TheseresultsdirectlycontradictthefindingsofSearlemanetal.(2005)

(113).

1.10:Handedness

Around 90% of the population is right-handed and 10% is left-handed (117).

Although these figures may vary slightly, right-handers have been predominant

throughouthistoryandacrosstheworld(117).

Handednesscanbedefinedbyeitherpreferenceorskill.Preferenceisasubjective

measurethatisassessedbyaskingsubjectswhichhandtheyuseforaparticulartask

i.e.aquestionnaire,whereasskillstestscompareasubject’sabilitytoperformtasks

witheachhand(118).TheEdinburghHandednessInventoryisthemostwidelyused

preferencequestionnaire(119),inwhichparticipantsareaskedwhichhandtheyuse

foranumberofactivitiessuchaswriting, throwingaballandusingaspoon(120).

Thereisgoodcorrelationbetweenmeasuresofpreferenceandskill,particularlyfor

complexmovementssuchaswriting,howeverwhetherpreferenceprecedesskillor

viceversaisdifficulttoestablish(118).

Somestudies fail todifferentiatemixed-handers from left-or right-handers,which

couldconfoundresults(121).PreferencetestsproduceaJ-shapeddistribution,with

themajorityof individuals showinga clear rightpreference, someshowinga clear

leftpreferenceandhardlyanybeingindifferent(122).

Defininganindividualaseitherright-handedor left-handedmayseemverysimple,

butinactualfactitcanbequitecomplicated.Whenpreferencesforseveraldifferent

actions are observed (i.e. not just hand-writing), the patterns of preference are

surprisingly variable (122). In 1970,Annett (p. 316) suggested that “to talk about

asymmetry intermsof leftandrightmightbe liketalkingaboutheight intermsof

31

“tall”or“short””(122).Handednessisthereforebestviewedasbeingacontinuum,

withmixed handers being part of the left-right continuum rather than a separate

classificationofhandpreference(122).Thatsaid,forthepurposeofresearchacut-

offpointisrequiredinordertoclassifyhandedness.Thedegreeofhandednesscan

bemeasured using laterality indices; questionnaire responses are assigned scores

that are combined, for example a pure right-hander will score +100, a pure left-

handerwillscore-100andanambidextroussubjectwillscorezero(118).

Handednessisoftenregardedasareflectionofcerebralhemispheredominance,and

Broca proposed that the hemisphere controlling speech was contralateral to the

dominanthand(111).Around5-6%ofright-handershavespeechcontrolledbythe

righthemisphere,comparedto30-35%of left-handers,sothecorrelationbetween

cerebralhemispheredominanceandhandednessisactuallymorecomplex(117).It

isimportanttorememberthereforethathandednessdoesnotnecessarilycorrelate

withcerebralhemispheredominance. Inaddition,whilst right-handersareusually

consistent intheirhandpreferencee.g.forperformingdifferenttasks, left-handers

arelessso(123).

There is a fundamental difference in the cerebral organisation of those that are

right-handedandthosethatareleft-handed,howevertheunderlyingreasonforthis

is controversial. Factors other than neural control can influence handedness, and

these include imitation, specific instruction, social prejudice and tool design (124).

Annett(1972)proposedthreepotentialmechanismsthatcouldexplainthevariation

inhandednessseeninhumans(121):

1. Genetics.Afamilialtrendforhandednesshasbeendemonstrated,however

simplegenetictransmissioncannotexplainthesefindings.

2. Translational errors. The accidental effects of minor translational errors

couldleadtovariation.

3. Culturaldifferences.Imitationandpracticecouldaffecthandedness.

Itislikelythatallofthesefactorsareinvolvedtosomeextentinthedevelopmentof

handedness(121).Twinandfamilialstudiessuggestthatageneticfactorisinvolved

32

in the development of handedness, however this cannot be explained by a

Mendelianmodel(118).

Thedifferencesbetweenhumansandanimalswithregardtolateralpreferencesare

interesting. Animalswithbilaterallysymmetrical limbsalsoexhibitpreferences for

the left or right side, however this seems to occur by chance and is not heritable

(125).Bothhumansandanimalsexhibitabell-shapeddistributionofasymmetryof

skill,howeverinhumansthebell-shapedcurveisshiftedtotheright,meaningthat

there isabias towards right-handedness, termedtherightshift (121). If this right

shift was not present, the distribution of hand preference would be the same in

humans andnon-humans; 25%would be left-handed, 25%would be right-handed

and50%wouldhavemixedhandedness(121).Therightshiftfactor,whichisspecific

tohumans,mayberelatedtospeechdevelopmentwhichusuallyoccurs inthe left

hemisphere (121). It is possible that the lateralisation of speech may confer

advantagesforspeechdevelopment(125).Whenacomplexactionoriginatesinthe

brain, theremaybeadvantages tohaving itarise inonlyonehemisphere, suchas

avoiding unnecessary conduction between the hemispheres and duplication of

specialisedareaswhenneuralspaceislimited(119).

1.11:Aimsofthesis

Theprincipleaimofthis thesiswasto lookforacorrelationbetweennasalairflow

andhandpreferenceinhealthyindividuals.Sincehandpreferenceisconsideredto

be a reflection of brain asymmetry, a secondary aim was to conduct a literature

review in order to gain a better understanding of the theories and evidence

suggestingalinkbetweenbrainactivityandnasalairflow.

Overthepastfewdecades,evidencehasemergedfromdifferentresearchgroupsin

different areas suggesting that asymmetrical brain activity may influence

asymmetrical nasal airflow, and vice versa. The research findings have been

33

obtained usingmany differentmethods with numerous variables, and have often

yieldedcontradictoryresults,makingthisfielddifficulttointerpret.Theseideasand

controversiesarepresentedandsummarisedinChapter2.

A possible relationship between nasal airflow and hand preference in healthy

individualswas identified ina single studybySearlemanetal. (2005) (113). Since

then there has been no further specific research in this area. There are several

reasons to doubt the reliability of this study. First, issues with themethodology

wereidentifiedandhavebeendescribedindetailabove(seeChapter1.9).Secondly,

numerous studies have analysed nasal airflow patterns in healthy individuals, and

although90%of thepopulation is right-handed,nonehavecommentedthat there

wasapreponderanceofrightnasalpassagedominance,whichwouldbeexpectedif

Searlemanetal.’s (2005) findings (113)wereapplicable to thegeneralpopulation.

Thirdly,alternatingnasalpassagedominanceappearstohaveafunctionrelatedto

immunedefenceandair-conditioning,bothofwhichwouldrequireequaldivisionof

labourbetweenthe twonasalpassages,and thereforehavinganoveralldominant

nasalpassagewouldnotmakesense.

Using a different method of measuring nasal airflow and analysing the data, this

study aimed to reproduce the experiment conducted by Searleman et al. (2005)

(113) in order to test the theory of a correlation between nasal airflow and hand

preference.ThemethodologyandresultsarepresentedinChapters3to6.

34

Chapter2:IntroductionB(Nasalairflowandbrain

activity–literaturereview)

2.1:Introduction

FormanycenturiestheancientartofyogainIndiahasstudiednasalbreathingand

developed techniques to switch the dominant nasal passage fromone side to the

otherbyuseofayogadandaorsmallcrutchappliedtotheaxilla (126). Theyoga

belief is that nasal airflow influences brain activity and cognition depending on

whetherairflowisdominantthroughtherightorleftnasalpassage,andthereforeby

controllingleftandrightnasalairflowwiththeyogadanda,theyogastudentcould

controlbrainactivity(127).Thistheorymayappearimplausibletoscepticsfromthe

scientificcommunity,howevertheasymmetricaleffectsoftheyogadandaonnasal

airflowhavebeenconfirmedinseveralstudiesindifferentresearchcentresandthis

ancientpracticenowhasscientificsupport.Reciprocalchangesinnasalairflowcan

bereadilycausedbypressureappliedtotheaxillabymeansofasmallcrutch,orby

adoptingthelateralrecumbentposition(78-80),butwhetherthisinturninfluences

brainactivityremainscontroversial.

Over the past few decades, some evidence has emerged suggesting that nasal

airflow asymmetry and brain asymmetry are linked. Different theories have been

proposed,with varying strengths of evidence from human studies. The following

sectionwilldiscusstheevidencethatlinksnasalairflowandbrainactivityinrelation

to two ideas; firstly the proposal that asymmetrical brain activity causes

asymmetrical nasal airflow, and secondly that asymmetrical nasal airflow causes

asymmetricalbrainactivity. Both the theoriesandevidencearecontradictoryand

mustbeinterpretedwithcaution.

35

2.2:Methods

AMedline®searchwasconductedinNovember2015usingthefollowingkeywords:

nasal airflow, nasal cycle, nasal hyperventilation, forced nostril breathing, brain

asymmetry, electroencephalogram (EEG), cerebral activity, cognition, cerebral

lateralisation,epilepsy,autismandschizophrenia.

Referencelistswerehandsearchedforotherarticlesofinterest.

2.3:Results

Doesasymmetricalbrainactivitycauseasymmetricalnasalairflow?

The peripheral control of nasal airflow via the autonomic nervous system is well

documented(2,36),andinvolvesthevasoconstrictorsympatheticnervesthatsupply

the largeveins intheturbinates. Theasymmetry inbrainactivityandsympathetic

toneextendtothebrainstemregionwhereleftandrightoscillatorscausereciprocal

changesinnasalairflow(38).Thehypothalamusmaygivetheoverallrhythmicityto

acycleofreciprocalchangesinnasalairflow,butthereisnoevidenceforasymmetry

at this level (39). Cortical involvement in nasal airflow asymmetry has been

suggested by studies on hand preference (113), lateralisation disorders such as

schizophrenia(112)andautism(70)andultradianrhythmsofcerebralactivity(128),

but the evidence for these influences is weak and this area of research is

controversial.

Fixedcerebralasymmetriesandnasalairflow

In the early 1800s, the widespread belief was that the two cerebral hemispheres

functioned as a single unit (111). However following studies on brain damaged

patients, itbecameapparent thatdifferentareasof thebrainwere specialised for

different functions (111),andnowadays it isknownthat thecerebralhemispheres

exhibit both functional and structural asymmetry (110). In the majority of the

population, the lefthemisphere isdominant for languageandspeechwhereas the

36

right hemisphere is dominant for visuospatial processing (110). Broca suggested

thatthehemispherecontrollingspeechwasoppositetothedominanthand,sothat

inleft-handersspeechiscontrolledbytherighthemisphereandinright-handersitis

controlledbythelefthemisphere(111).Inreality,therelationshipbetweenspeech

and handedness is more complex and can be influenced by brain damage and

neuronalplasticity(111).

Searlemanandcolleagues (2005)hypothesisedacorrelationbetweennasalairflow

andhandedness,basedontheobservationthatthereisoftenaconsistencyinlateral

preferencese.g.left-handerstendtobeleft-footed,left-eyedetc.(113).Giventhe

existing knowledge of the alternations in nasal airflow between the right and left

nasalpassages,theypredictedthattheleftnasalpassagewouldbedominantforthe

majority of the time in left-handers, and vice versa in right-handers. Their study

proved thehypothesis tobecorrect, in that thedominantnasalpassagepositively

correlated with the dominant hand for almost 60% of the time (113). However,

among other issues outlined in section 1.9, the study only involved a small group

monitoredovera short timeperiodandaspreviouslydemonstrated there isgreat

variability inpatternsofnasalairflow (13). It isalsounusual that this relationship

has not been noted in other observational studies of the nasal cycle in healthy

individuals,since90%ofthemwereprobablyright-handed.

Non-right handedness (left-handedness or inconsistent handedness) seems to be

more prevalent than expected in certain neurodevelopmental and psychiatric

disorders such as autism and schizophrenia and this may be related to cerebral

lateralisationabnormalities(70,112,116).Onestudyanalysedhandpreferenceand

nasalairflowinautisticchildren,andfoundthatthemajoritywereleft-handedand

had left nasal passage dominance for most of the time (12-hour measurement

period)(70).Asimilarstudyinright-handedschizophrenicsrevealedthattherewas

a significant increase in leftnasalpassagedominance in thisgroupcomparedwith

controls(108).

37

Handednessisnotassimpleasleftorright,butisinfactacontinuum,withdegrees

ofleftandright-handedness(114).Studiesthatutilisehandpreferenceasavariable

can therefore be difficult to interpret. There are also different methods of

measuring handedness, including objective performance tests and subjective

inventories(129),andthesearenotstandardisedacrossdifferentstudies.Itshould

be noted as well that the aforementioned studies are relatively small and there

couldhavebeenconfoundingfactorsthatwerenotcontrolledfor,suchastheuseof

psychoactivemedication.

Fluctuatingcerebralasymmetriesandnasalairflow

Theideaofrhythmic,spontaneousfluctuationsincerebralhemisphereactivityfirst

appearedinthe1960s.FollowingthediscoveryoftheREM/non-REM(NREM)sleep

cycle(130),Kleitman(1967)proposedthatthisphenomenonwasthenocturnalpart

ofa“BasicRestActivityCycle”(BRAC),whichinvolvesfluctuationsincentralnervous

systemactivityapproximatelyevery90minutes(131).Thesefluctuationsareoften

referred toasultradian rhythms in cerebralactivity. However theexactnatureof

thesechangesinbrainactivityremaincontentious,andconflictingresultshavebeen

presented. A few studies with small numbers of participants have suggested

rhythmic fluctuations inelectroencephalographic (EEG) activity (132) and cognitive

performance in this time frame (133,134), such that verbalperformance isbetter

than spatial performance, switching to the reverse approximately every 90-100

minutes(133).

Several studies have suggested the existence of a correlation between the

alternatingpatternofnasalairflowandcerebralhemisphereactivity(128),thisagain

is based on the regular periodicity associatedwith the BRAC (131) which has not

beenaconsistent finding (133,135). Werntzetal. (1983) reported increasedEEG

activityinthehemispherecontralateraltothedominantnasalpassageasmeasured

bynasalairflow(128).TheEEGactivitywasobservedacrossthecortexratherthan

focussed at the olfactory centres, suggesting a mechanism other than olfactory

stimulation(128).Alternationofthepredominantnasalairflowtotheoppositeside

occurredafteralternationofcerebralhemispheredominance(128). Alargerstudy

38

involving126right-handedparticipantsfoundatendencyforenhancedperformance

in verbal tasks at times of right nasal passage dominance, and enhanced

performance in spatial tasks at times of left nasal passage dominance, i.e. a link

between nasal passage dominance and activity in the contralateral cerebral

hemisphere(136).

Not only do the authors of these articles often reach different conclusions, they

utilisedifferentmethodsoftesting,analysisandreporting,usuallyinsmallgroupsof

subjects,attimesfailingtoconsiderpotentiallyconfoundingfactorssuchasgender

andhandedness.EEGstudiesinparticulararedifficulttointerpretandhavevarying

methodsofanalysis.Thereisalsoahighlevelofinter-individualvariability(132).It

mustbenotedthattheauthorswhohaveidentifiedanultradianrhythmincognitive

performance have supported Kleitman’s theory of the BRAC (131), potentially

introducingbias(128,133).

Modelofhowthebraininfluencesnasalairflow

A model of how the brain influences nasal airflow is illustrated in Figure 4. The

model summarises the evidence and ideas presented in this section and for

simplicity thecontrol isdiscussed fromtheperipheralnervesandmovingupwards

throughthehierarchyofcentralnervouscontrolcentres.Theperipheralcontrolof

nasal airflow via the autonomic nervous system involves the vasoconstrictor

sympathetic nerves that supply the large veins in the turbinates (2, 36). The

asymmetry in brain activity and sympathetic tone extend to the brainstem region

where leftand rightoscillatorscause reciprocal changes innasalairflow (38). The

hypothalamusmay give the overall rhythmicity to a cycle of reciprocal changes in

nasal airflow, but there is no evidence for asymmetry at this level (39). Cortical

involvement in nasal airflow asymmetry has been suggested by studies on

handedness (113), ultradian rhythms of cerebral activity (128) and lateralisation

disorders such as schizophrenia (108) and autism (70), but the evidence for these

influencesisweak.

39

Figure4:Modeltoexplainthe influenceofbrainactivityonnasalairflow.Theoverallalternation insympatheticactivityoveraperiodofhoursiscontrolledfromthehypothalamusandtheasymmetryinsympatheticoutflowisdeterminedbytheactivityofbrainstemoscillatorswhichactasaflipflopmechanism,witheachcentre inhibiting theactivityof theothercentreandonlyonecentrehavingdominant activity at any one time. These hypothalamic and brainstem mechanisms have beenstudied in theanaesthetised cat (38,39). Higher centres in the cerebral cortexmayalso influencenasal airflow leading to asymmetry and most of the supporting evidence for this has come fromexperimentalwork inman(70,113,128). (SNS=sympatheticnervoussystem,+ve=positive,-ve=negative.)

Doesasymmetricalnasalairflowcauseasymmetricalbrainactivity?

Thesensationofnasalairflowisbelievedtobemainlydeterminedbystimulationof

nasaltrigeminalnerveendingsthataresensitivetotemperaturechanges,especially

coolingof thenasalmucosa,whichoccursduring inspirationof air (1, 45, 46). A

40

purely trigeminal stimulus has been shown to increase arousal frequency and

durationduringsleep(137),whereasnoeffectwasseenwithanolfactorystimulus

(138).Therefore, itseemsthatanasalairflowstimuluscan influencebrainactivity,

andfurtherevidenceisdiscussedbelow.

Nasalhyperventilationandepilepticactivity

Ithasbeenknownforsometimethatdeepbreathingcanactivateepileptic foci in

the brain, triggering seizure activity (139). This phenomenon was previously

explainedbyhypocarbia leading to vasoconstriction and cerebral ischaemia, but it

appears that airflow stimulation of the nasal mucosa could be the trigger (139).

Animal studieshavedemonstrated that insufflationofair into thenasal cavity can

triggerepilepticareasinthebrain(139).Instudiesofsomeepilepticpatients,nasal

hyperventilation was more likely than oral hyperventilation to stimulate epileptic

EEG activity, and unilateral nostril breathing had a greater effect on the

abnormalities in the ipsilateral hemisphere (139, 140). This was suppressed

following theapplicationof local anaesthetic to thenasalmucosa in the regionof

thesuperiormeatus(139,141).Theexactmechanismofthisphenomenonhasnot

beenfullyestablished,howevertheauthorsofthesestudieshavesuggestedareflex

involvingstimulationoftheolfactorynervebynasalairflow(139,141).

Unilateralforcednostrilbreathing

Asymmetrical nasal airflowwith unilateral forced nostril breathing (UFNB), where

onenostrilisoccludedeithermanuallybythesubjectorwithcottonwool,hasbeen

used to analyse the influence of asymmetrical nasal airflow on the brain as

measuredbyEEGactivity(142)andcognitiveperformance(136,143).Infact,these

concepts have their basis in ancient yogic practices, and there is a relatively large

literaturediscussingthenostrilbreathingmethodsutilisedinyogaandtheireffects

on mood and cognition, for examples see (144-149). Several studies have

demonstratedaneffectontheautonomicnervoussystem, forexamplechanges in

cardiovascularparameters(150-152)andintraocularpressure(153-155).Inastudy

offivesubjects,UFNBcausedashift inthedominantcerebralhemisphere,defined

byrelativelygreaterEEGactivity,oftenwithintwominutes(142).Whetherandhow

41

thiscorrelateswithcognitiveperformanceremainscontentious.Ingeneral,studies

haveusedhemisphere-specific tasks tomeasure cognitiveperformance, i.e. verbal

taskstoreflectlefthemisphereactivityandspatialtaskstoreflectrighthemisphere

activity. Using these methods, one study identified significant improvements in

verbal test scores with right UFNB and spatial test scores with left UFNB (156).

Others have found that left UFNB significantly improved right hemisphere

performancewhereasrightUFNBhadnoeffect (149,157),but theoppositeeffect

has also been demonstrated, wherein right UFNB improved left hemisphere

performance but left UFNB had no effect (146). One study suggested that UFNB

affectedtaskperformancedifferentlyinmalesandfemales(143),althoughagainthis

findinghasnotbeenreproducible(157).OthershavefailedtoshowthatUFNBhad

anyeffectonEEGmeasurements(158)orcognitiveperformance(128).Ithasbeen

suggested that UFNB can influence mood, and it has been reported to affect

emotionalresponses(159).

Oftenthesestudiesaredifficulttointerpretaccuratelyandhaveconflictingresults.

Criticismsincludesmallnumbersofparticipants(142,143,156),differingmethodsof

UFNB (142, 143, 149), differingmethods of measuring cerebral activity (142) and

cognitive performance (143, 156), and failure to consider potential confounding

factorssuchassex(149)andhandedness(142).

Following their study of nasal airflow patterns in children with autism, Dane and

Balci(2007)suggestedthatautismisassociatedwiththeabsenceofanormalnasal

cycle,theleftnasalpassagebeingdominantforthemajorityofthetime(70).They

have likened this finding to continuous leftUFNB, and suggested that thismaybe

causingcontinuousstimulationoftherightcerebralhemispherewhilstdeprivingthe

left cerebral hemisphere of stimulation, possibly accounting for the delayed

languagedevelopmentoften seen in thisgroup (70). The issuewith this theory is

thatthere isnoevidencethatrightnasalblockage,forexampleseptaldeviationor

nasalpolyposis,leadstodevelopmentaldelay.Inthesepathologicalconditions,the

differencebetweentheamountofairflowineachnasalpassageisfargreaterthan

thedifferencecausedbythenasalcycleunderphysiologicalconditions.

42

Shannahoff-KhalsaandhiscolleagueshavesuggestedinmultiplearticlesthatUFNB

haspotential as anon-invasive treatment for psychiatric disorders (142, 160), and

haverecordedacorrelationbetweenleftnasalpassagedominanceandhallucination

occurrenceinoneschizophrenicfemale(160).ArecentarticlesuggestedthatUFNB

mayhavebeneficialeffectsforrecoveryofspeechinstrokepatients(161).

Howcouldanasalairflowstimulusaffectcerebralactivity?

A proposed mechanism for a correlation between nasal airflow and cerebral

hemisphere activity involves the sympathetic nervous system (128, 142, 157),

supported in part by the other autonomic effects demonstrated to occur during

UFNB(150-152).Asautonomicnervefibresconnectingthenoseandhypothalamus

donotdecussate,vasoconstrictioninthenasalvesselshasbeenpostulatedtoreflect

concurrent vasoconstriction in the ipsilateral cerebral hemisphere, leading to a

decrease in cerebral blood flow ipsilaterally and relative increase contralaterally

(128, 142, 157). In this way, the increased blood flow could improve cognitive

functionasmeasuredbyperformanceinhemisphere-specifictasks(157).

However thephysiologicalbasis for this theory isquestionable. Taskperformance

has been shown to increase overall blood flow to both hemispheres, and more

specifically verbal tasks cause a left lateralisation and spatial tasks a right

lateralisation in right-handed subjects (162, 163). However the effect of the

sympatheticnervoussystemoncerebralbloodflowunderphysiologicconditions is

thought tobeminimaldue to theactionof cerebralautoregulation (164). In fact,

whilst blockade of the stellate ganglion i.e. inhibition of sympathetic activity

increasesbloodflowinextracranialvessles,ithasnoeffectintra-cranially(165).

Thereforeadifferentmechanism for theeffectofnasal airflowonbrainactivity is

proposedhere,incorporatingtheactivatingeffectofanasalairflowstimulusonthe

cerebralcortexviathereticularformation.ThisisillustratedinFigure5.

43

Figure5:Model toexplain the influenceofnasalairflowonbrainactivity. Anasalairflowstimulussuch as UFNB stimulates trigeminal nerve endings on one side of the nose. Trigeminal neuronstransmittingtemperaturesignalssynapseinthespinaltrigeminalnucleusandthencrossthemidline,travelling up to the thalamus through the brainstem. Studies on cats have suggested that via thebrainstem reticular formation, a nasal airflow stimulus could lead to enhanced arousal and brainactivity in both cerebral cortices (166). EEG studies and cognitive performance testing in humansubjectshaveintimatedthatthegreateststimulatingeffectoccursinthehemispherecontralateraltothenasalairflowstimulus(128,156).

One major challenge is that the laterality of cerebral hemisphere stimulation by

nasal airflow isunclear,with somestudies suggestingan ipsilateral response (139,

143),andothersacontralateralresponse(142,156,157).Olfactorynervefibresdo

not decussate and therefore stimulate mostly the ipsilateral cortex, whereas

trigeminal fibres relaying temperature signals cross over in the spinal cord before

passing through the brainstem. The trigeminal nerve detects nasal airflow, but

experimentalinsufflationofairintothenasalpassagescouldstimulatetheolfactory

44

nerve due to the inadvertent presence of an olfactory stimulus. In addition, the

olfactorycortex isunabletosensethelateralityofastimulusunlessthetrigeminal

nerve isalso stimulated (167).Suppressionof theEEGstimulationcausedbynasal

airflow by application of local anaesthetic to the nasal mucosa (141) is more

suggestive of trigeminal nerve involvement. Although sleep studies have

demonstratedarousalsecondarytoatrigeminalnervestimulus,thisstimuluswasan

irritant (carbon dioxide) (137) and is therefore difficult to compare with nasal

inspirationofairaswithUFNB.

It is possible that nasal airflow causes bilateral cortical stimulation,with a greater

effect on one side. Experimental stimulation of the reticular formation in

anaesthetisedcatscausedEEGchangesindicatingincreasedalertness,andatlower

levelsofstimulationthiseffectwasonlyseenintheipsilateralhemisphere(166).It

isunclearwhetherthetrigeminalorolfactorynervesareinvolvedinthismechanism.

Modelofhownasalairflowinfluencesbrainactivity

Amodel of how nasal airflow influences the brain is illustrated in Figure 5. The

model summarises the evidence and ideas presented in this section. Inspired air

stimulatescoldreceptorsinthenasalmucosathatareinnervatedbythetrigeminal

nerve,providingthesensationofnasalairflow.Environmentalsensorystimulisuch

asnoiseorsmellscanenhancearousal,andthiseffect ismediatedbythereticular

formation–anarea in thebrainstem involved inarousal andconsciousness (166).

ExperimentalstimulationofthereticularformationinanaesthetisedcatscausedEEG

changesindicatingincreasedalertness(166).Insufflationofairintothenosehadthe

sameeffectonEEGpatternsi.e.increasedarousal(168).Therefore,anasalairflow

stimulus, for example insufflation of air or UFNB could activate the reticular

formation and increase arousal, leading to EEG changes and possibly improved

cognitive performance. There is evidence to suggest both an ipsilateral (139, 143)

andcontralateral (142,149,156,157)stimulatingeffect. However it ismore likely

that a unilateral nasal airflow stimulus has an activating effect on both cerebral

hemispheres, but with a greater effect on one side. As trigeminal neurons

45

transmitting temperature signals cross themidline in themedulla, it seems logical

thatthegreatesteffectwouldbeseencontralaterally.

2.4:Conclusions

Theancientyogicpracticeof“pranayama”orbreathcontrolexercisesarethoughtto

promotehealth andwell-being, improve circulationandprepare for concentration

(169). Whilst this notionmay have beenmet with scepticism from the scientific

community, it inspired clinical studies into the effects of nasal breathing on

cognition. There is a growing body of evidence to suggest that nasal airflow can

influence brain activity, however the mechanism, extent and significance are

debatable. Considering the evidence from studies in epileptic patients (139, 141)

andarousalduringsleep(137),itseemsthatanasalairflowstimuluspotentiallyhas

somesortofactivatingeffecton thebrain. Putting this intoaneverydaycontext,

stepping outside into a cold environment and inhaling cool air through the nose

often makes us feel more alert, and the cooling effects of menthol on nasal

receptorsmayalsocausearousal(170).SmellingsaltswereusedinVictoriantimes

toreviveunconsciouspatients,andevennowadayssomeathletesusesmellingsalts

as a stimulant prior to competing (171). However the role of higher centres and

cortical organisation remains uncertain, with some conflicting theories suggested.

For example, if nasal airflow dominance correlateswith hand preference (113), it

couldnotalsocorrelatewithfluctuatingultradianrhythmsofcerebralactivity(136),

ashandpreferenceisnormallyfixed.

46

Chapter3:Methodology

3.1:Ethicalapproval

ThestudywasdesignedandconductedaccordingtotheWorldMedicalAssociation’s

Declaration of Helsinki and the International Council for Harmonisation’s Good

Clinical Practice guidelines. Prior to commencement all documentation was

reviewedandapprovedbytheSchoolofBiosciencesResearchEthicsCommitteeand

Cardiff University’s Research Governance Department. The patient information

leafletandinformedconsentformareincludedinAppendices2and3.

3.2:Studydesign

Thestudywasdesignedasaprospectivepilotstudy,basedonthemethodsusedby

Searlemanetal.(2005)(113).Thehandednessofeachsubjectwasrecordedaspart

ofthescreeningprocedure.Followingthis,nasalairflowwasmeasuredat15-minute

intervals over a continuous 6-hour period using a modified GM, as described by

Gertner et al. in 1984 (54). At eachmeasurement, the investigator examined the

condensationareasformedonthemetalplateanddeterminedwhethertheleftor

right nasal passage had produced the largest area, and therefore was dominant.

Thesameinvestigator(AP)carriedoutallmeasurementsinanattempttominimise

uservariation.

AlthoughsimilarindesigntoSearlemanetal.’s2005study(113),adifferentmethod

ofmeasuringnasalairflowwaschosen.Partofthereasonforthiswasrelatedtothe

aim of the study. In order to determine the dominant nasal passage in terms of

airflow,theonlyvariablesofimportanceareleftorright.ThemodifiedGMprovides

a snapshot view of nasal airflow through each nasal passage, allowing the

investigator to instantaneously judgewhichnasal passagehas the greatest airflow

47

andthereforeisdominant.Otheradvantagesofthistechniquearediscussedinthe

introduction(seesection1.7).

3.3:Studypopulationandrecruitment

Participants were recruited from Cardiff University campus by an advertisement

email and posters seeking out normal, healthy, non-smoking adults. Participants

wereexcludediftheyhadahistoryofnasaldiseaseorweretakingmedicationthat

couldaffectnasalphysiology. Incontrast toSearlemanetal.’s (2005) study (113),

womenwereincludedprovidingtheywerenotpregnantorlactating.Thepotential

effects of female sex hormones on nasal airflow were discussed in detail in the

Introduction(seesection1.8),howeverinbrief,duringpregnancyit isthoughtthat

hormonalchangesareresponsibleforpregnancy-inducedrhinitis,whichisrelatively

commoninthelaterstagesofgestation(88,89).Thereissomeevidencetosuggest

thatnasalairwayresistancemaybehigherduringmenstruation,althoughchangesin

thenasalcyclehavenotbeenidentified(92).Inaddition,thereislittleevidenceto

suggest that the contraceptive pill has a significant effect on the nasal cycle (99).

Many other studies specifically analysing the nasal cycle have involved bothmale

andfemaleparticipants,withoutsignificantdifferencesidentified.

Forthefullexclusionandinclusioncriteria,seeAppendix1.

Participants were paid an honorarium of £70 on completion of the study. Those

excludedfromthestudyfollowingthescreeningassessmentwerepaid£5fortheir

time.

48

3.4:Samplesizeandpower

Asthiswasapilotstudy,theaimwastorecruitbetween25and30participants,with

thefinalnumberdependentonthenumberofleft-handedparticipantsincluded.A

powercalculationbasedonthepreviousstudy(113)wasnotpossibleduetolimited

data presented, however statistical significance was reportedly achieved with 20

participants(113).

3.5:Measurementofhandedness

Atestofhandednesswasperformedtoexcludethosecategorisedasmixed-handed

andtoidentifythehandpreferenceofeachparticipant.Thisinvolvedobservationof

handwriting,followedbycompletionoftheEdinburghHandednessInventory(Short

Form) toobtainahandedness categorisation (172) (seeAppendix4). If therewas

any disparity between the participant’s statement of hand preference and

observation of handwriting, the participant was excluded. Participants were also

excludediftheydemonstratedmixedhandedness.

Handedness can either be measured by preference or skill, and there is good

correlationbetweenboth typesof test especially for complexmovements such as

writing(118). Observationofhandwritingaloneasatest forhandednesscanmiss

thosewithmixedhandedness,whichcouldbeaconfoundingfactor.Aquestionnaire

was therefore used to ascertain subjects’ hand preference for several different

activities.TheEdinburghHandednessInventoryisthemostwidelyusedpreference

questionnaire(119),andtheShortFormisarevisedversionthathasbeenshownto

havegoodvalidityandreliability(172).Subjectswereaskedwhichhandtheyprefer

touse for fourdifferentactions;writing, throwing,usinga toothbrushandusinga

spoon. Theywereassignedscores thatwereusedtocalculatewhether theywere

categorisedasleft-handed,right-handedormixed-handed(seeAppendix4).

49

3.6:Measurementofnasalairflow

Nasal airflowwas assessed using themodifiedGM, as described byGertner et al.

(1984)(54).Thismethodallowstheinvestigatortoobtainamomentaryassessment

of nasal airflow by comparing the condensation area formed by expired air from

each nasal passage. The instrument usedwas a polishedmetal platemade from

aluminiummeasuring10cmx12cmandmarkedwitharches1cmapart (seeFigure

6). The plate was positioned horizontally just beneath the columella with the

verticalaxisat90degreestowardstheupperlip.Eachparticipantwasinstructedto

exhalethroughthenoseslowlywiththeeyesandmouthclosed. Thetemperature

differencebetweentheexpiredairandthemetalplateiswhatcausesthevapourto

condense on the plate. By observing which nasal passage produced the largest

condensationarea,thedominantnasalpassagewasrecorded.

Figure6:Aphotographofthemetalplate(i.e.modifiedGM)usedinthestudy.

3.7:Trialenvironmentandprocedure

All visits and testing procedures were carried out at the Common Cold Centre.

Potentialparticipantsattendedforascreeningvisit,wheretheywereprovidedwith

aparticipant information sheet (seeAppendix2) and signeda study consent form

(seeAppendix3).Theywerethenassessedbythestudyclinician(AP)toensureall

50

inclusionandexclusioncriteriaweremet,whichinvolvedelicitingamedicalhistory

and examining the nose by anterior rhinoscopy. Those enrolled onto the study

attendedatalater,convenientdatefortesting.

Everyattemptwasmadetominimisefactorsknowntoaffectnasalairflowthatcould

potentiallyaffectthestudyresults.Sinceupperrespiratorytractinfectionhasbeen

showntoincreasethenasalairwayresistanceandtheamplitudeofthenasalcycle

(23),participantswithcommoncoldsymptomsatthescreeningvisitoronthetest

dayweredelayed for twoweeks toensure resolution. Alcohol ingestion increases

nasal airway resistance (100, 101), therefore participantswere asked not to drink

more than four units of alcohol the night before testing. Exercise effectively

abolishes the reciprocal changes in nasal airway resistance, leading to an overall

increaseinnasalairflow(84),thereforeparticipantswereaskedtorefrainfromany

vigorous exercise such as running, cycling or swimming for three hours prior to

testing.

On the test day, participants arrived 30minutes prior to the start time for nasal

measurements to ensure anyeffects of exertion and theexternal environmenton

the nasal mucosa were eliminated, and to allow for acclimatisation to the

temperature and humidity of the test environment. During the test period,

participantsremainedatrestandinthesameroom,withallowancesforuseofthe

bathroom. Theywerepermitted to read, use their computersorwatch television

butwerenotpermittedtoliedown,aschangesinpostureandpressurestimulican

leadtoalterationofnasalairflow(78,79).Althoughmentholhasnotbeenshownto

affectnasalairwayresistanceperse,itcancauseasubjectivechangeinnasalairflow

(47), therefore participants were not permitted to eat any menthol containing

productssuchasmints.Theywereprovidedwithastandardcoldlunchbetween12

and12.30pm.Nohotdrinkswerepermitted.

RecordingsofthedominantnasalpassageusingthemodifiedGMwereperformedat

15-minuteintervalsovera6-hourperiod,givingatotalof24recordings.Participants

51

were positioned sitting upright with the head in a neutral position. Once sitting

comfortably,theyweregiventhefollowinginstructions:

“Iwillaskyoutotakeadeepbreathinusingthewordinhale.Youwillthenholdthis

breathuntil I informyou toexhale. Whenyouexhale,keepyoureyesandmouth

closedandgraduallybreathoutthroughyournostrils.”

Whilsttheparticipantheldtheirbreath,themodifiedGMwaspositionedunderthe

collumela as described above (see Figure 7). This procedure was repeated three

timesateach15-minuteinterval. Ajudgementwasmadebytheinvestigatorasto

which nasal passage produced the largest condensation area, and was therefore

dominant,andthiswasrecordedaseitherleftorright.Ifitwasunclearwhichnasal

passage had produced the largest condensation area, the result was recorded as

equal.Ofthethreemeasurementstaken,themajorityreadingwasusedasthefinal

result.

Figure7:AphotographofthemodifiedGMinuse.Theparticipantisexhalingthroughthenosewiththe modified GM positioned horizontally just beneath the columella. The condensation areasproduced during exhalation can be seen on the plate and aremarkedwith a dotted line. On thisoccasion the rightnasalpassageproduceda largercondensationarea than the left, indicating rightnasalpassagedominanceatthattime.

A trial runwasperformedduring the 30-minute rest period after eachparticipant

arrivedatthetestcentretoallowthemtobecomefamiliarwiththetestingmethod.

Results from this trial runwere not be recorded or included in the data analysis.

Carewastakentominimisepositionalerrorswhilstrecording. Thesubjectwassat

52

uprightwiththeheadinaneutralpositionandthesameinvestigator(AP)performed

andinterpretedeachmeasurement.

3.8:Blinding

In order to avoid potential investigator bias (given the subjective nature of the

recordings), as much as possible the investigator taking the nasal airflow

measurements (AP) was blinded to the participants’ handedness. Handedness

questionnaireswerecompletedbyparticipantsandcheckedbyanotherresearcher.

Participants were asked not to reveal their hand preference or write when the

investigatorwasintheroom.Throughoutdatacollection,participantdemographics

weremonitoredbyanotherresearcheranddiscussedwiththeinvestigatortoensure

roughly equal numbers of right-handed and left-handed participants and to guide

recruitment. Thereforethe investigatorwasnotblindedtothe lastthreesubjects’

hand preference as only left-handed volunteers were recruited (subject numbers

034, 035and036). Thehandednessof all participantswas revealedonceall data

hadbeencollectedandanalysed.

3.9:Statisticalanalysis

The data collected was compiled into a Microsoft Excel® spreadsheet for further

analysis. In order to define each subject as either left or right nasal passage

dominantbasedonthemeasurementscollected,binomialdistributionwasusedto

lookforconsistentdifferencesineachsubject.

Once all data had been analysed, the hand preference of each participant was

revealed. The Chi-square test was used to look for a statistically significant

correlationbetweennasalpassagedominanceandhandpreference.

53

Chapter4:Subjects

4.1:Introduction

Overthepastfewdecades,manystudieshavefoundthatcertainphysiologicaland

pathological conditions can influence nasal airflow. These include subject factors,

suchasage(74)andnasalpathology(54),positionalfactorssuchasposture(79)and

other influences such as sleep (28) and exercise (84). During recruitment, efforts

weremadetominimisefactorsknowntoinfluencenasalairflowusingstrictinclusion

and exclusion criteria. The test environment and methods used to take

measurementswerealsocarefullyconsidered. The followingsectiondiscusses the

recruitment procedure and subjects in detail with reference to the relevant

literaturethatinformedtheprocess.

4.2:Methods

Volunteersforthestudywererecruitedbyemailandposteradvertisementsplaced

around Cardiff University campus. There were two periods of recruitment and

testingthatcoincidedwithuniversitysemesters,oneintheautumnandanotherin

the spring. The first group of volunteers were recruited and tested between

November 5th 2015 and December 17th 2015 (subject numbers 001 – 019). The

secondgroupwererecruitedandtestedbetweenApril11th2016andMay20th2016

(subjectnumbers020–036).Bothgroupscontainedamixtureofmaleandfemale

subjectsandleft-handedandright-handedsubjects.Allvolunteerswerestudentsat

theUniversity.

Eachsubjectwasrequiredtoattendthetestcentreontwoseparatedays. Onthe

firstday,afterprovidinginformedconsent,subjectswerescreenedforsuitabilityto

takepart.Toensurethatallinclusionandexclusioncriteriaweremet,subjectswere

asked about their medical history and had a nasal examination using anterior

54

rhinoscopy (for full list seeAppendix1). Aconvenientdatewas thenarranged for

thetestday.

4.3:Results

Intotal,36subjectswererecruitedintothestudy.Ofthose,7wereexcluded.The

reasonsforexclusionarelistedinTable2.

Table 2: A summary of the demographics and reasons for exclusion. *Subject 014 was takingisotretinoinforacne,whichcancausedrynessofthenasalmucosa(173).

Subjectnumber

Age(years)

Gender Handpreference Reasonforexclusion

001 22 M Left Significantnasalseptaldeviation

014 18 F Left Concominantconfoundingmedication*

016 19 F Left Rhinitis

017 19 M N/A Mixedhandedness

021 22 M N/A Mixedhandedness

026 27 F Right Significantnasalseptaldeviation

029 20 M N/A Mixedhandedness

Efforts were made to recruit roughly equal numbers of left- and right-handed

subjects andmalesand females (seeTables5and6). Of those included,14were

maleand15werefemale.Theaverageagewas20.8years,witharangeof18–30

years. Fifteen subjects were left-hand dominant (LHD), including 6 males and 9

females,and14wereright-handdominant(RHD),including8malesand6females.

ThedemographicsofallincludedpatientsarelistedinTable3.

55

Table3:Demographicsofallincludedsubjects.

Subjectnumber

Age(years) Gender Handdominance Significantmedicalhistory

Concominantmedication

002 19 Female Right None None003 19 Male Left Seasonalrhinitis–

asymptomaticduringstudyperiod

None

004 18 Female Left None None005 19 Female Left None None006 18 Female Left None Progesterone–only

contraceptivepill007 20 Female Left Mildasthmaduring

childhood–asymptomaticfor8years

None

008 20 Female Left None Combinedoralcontraceptivepill

009 18 Female Left Cholecystectomyforgallstones

Combinedoralcontraceptivepill

010 20 Female Left Mildasthmaduringchildhood–asymptomaticfor8years

Combinedoralcontraceptivepill

011 20 Male Left Seasonalrhinitis–asymptomaticduringstudyperiod

None

012 21 Female Left Mildasthmaduringchildhood–asymptomaticfor4years

None

013 20 Male Right None None015 21 Female Left None None018 21 Female Right Appendicectomy None019 19 Female Right None Contraceptiveimplant(slow

releaseprogesterone)020 21 Female Right None Contraceptiveimplant(slow

releaseprogesterone)022 20 Female Right None Combinedoralcontraceptive

pill023 23 Male Right None None024 25 Male Right None None025 21 Male Right Depression Sertraline(selectiveserotonin

reuptakeinhibitor)027 21 Female Right None Combinedoralcontraceptive

pill028 20 Male Right None None030 22 Male Right None None031 30 Male Right None None032 21 Male Left None None033 25 Male Right None None034 21 Male Left None None035 22 Male Left None None036 19 Male Left Migraine None

56

Table4:Summaryofincludedsubjects.

Summaryofincludedsubjectdemographics

Averageage(years) 20.8

Medianage(years) 20

Agerange(years) 18-30

Numberofmales 14

Numberoffemales 15

Maletofemaleratio 1:1.07

Numberofleft-handers 15

Numberofright-handers 14

Lefttoright-handedratio 1:0.93

Table5:Comparisonofbasicdemographicsformaleandfemalesubjects.

Malesubjects Femalesubjects

Numberofsubjects 14 15

Averageage(years) 22 19.7

Medianage(years) 21 19

Agerange(years) 19–30 18–21

Numberofleft-handers 6 9

Numberofright-handers 8 6

Table6:Comparisonofbasicdemographicsforleft-andright-handedsubjects.

Left-handers Right-handers

Numberofsubjects 15 14

Averageage(years) 19.8 21.9

Medianage(years) 20 21

Agerange(years) 18–22 19–30

Numberofmales 6 8

Numberoffemales 9 6

57

During screening, all subjects were asked specifically about any history of nasal

problems in the past and concurrent nasal symptoms. No subject reported

concurrentorrecent(precedingonemonth)nasalsymptomssuchasnasalblockage,

rhinorrheaorchangeinsenseofsmell.Thiswascheckedasecondtimeonthetest

day to ensure no subject had a concurrent or recent history of rhinitis that had

developed between the screening and test days. Anterior rhinoscopy was

performed on all subjects during screening, and those with significant septal

deviation (n=2)or signsof rhinitis (n=1)wereexcluded. Noothernasalpathology

wasdetectedduringscreening.

Themajority of subjects (20/29) reported no significant previousmedical history.

Twosubjectshadahistoryofseasonalrhinitishowevertheywerebothtestedduring

theautumnperiodanddeniedanyrecentnasalsymptoms.Theydidnothavesigns

of rhinitis on anterior rhinoscopy. Three subjects reported mild asthma during

childhood however they had been asymptomatic and had not required any

treatmentforat leastfouryears. Twosubjectshadpreviouslyrequiredabdominal

surgery in the form of an appendicectomy (n=1) and a cholecystectomy (n=1)

however they had fully recovered from these procedures. One male subject

reportedahistoryofmilddepressionandanotherreportedahistoryofmigraines,

bothofwhichweredeemedunlikelytoaffecttheresultsofthestudy.

Nine subjects were taking prescribed medication at the time of the study. The

majority of these subjects were females taking contraception in the form of the

combined oral contraceptive pill (n=5), the progesterone implant (n=2) and the

progesterone-only contraceptive pill (n=1). One subject was taking sertraline, a

selective-serotoninreuptakeinhibitorusedtotreatdepression.

58

4.4:Discussion

During screening and testing, efforts were made to minimise factors that could

influence nasal airflow. Potential influencing factors are discussed below, with

referencestoevidencefromtherelevant literature. Thedemographicsof the left-

andright-handedgroupsweresimilar.

Subjectfactors

Theaverageageofthiscohortwas20.8years,witharangeof18–30years. The

averageageofright-handedsubjectswasslightlyhigherat21.9years,comparedto

19.8years in left-handedsubjects. Studieshavesuggestedthatadvancingagecan

affect patterns of nasal airflow, however changes seem to be more likely in

individualsagedover50years(74,75)andthereforetheagerangeinthisgroupis

unlikelytobeapotentialconfoundingfactor.

Although BMIwas notmeasured specifically, none of the recruited subjectswere

overweight.

Bothmaleandfemalesubjectswereincludedinthestudy.Searlemanetal.(2005)

only includedmalesduetoconcernsthatchanges infemalesexhormonelevelsas

part of the menstrual cycle could affect nasal airflow (113). However conflicting

results have been published regarding this theory,with one study reporting nasal

congestionduringmenstruation (92)and another reportingnodifference innasal

airflow throughout different phases of the menstrual cycle (93). Perhaps more

importantly,twootherstudiesthatmeasurednasalairflowoverseveralhoursusing

differentmethods found that gender did not significantly affect nasal airflow (29,

74).Enquiringaboutphaseofthemenstrualcycleinfemalesubjectswastherefore

consideredunnecessary.Femalesubjectswereaskedwhethertheywerepregnant,

howeverformaltestingwasnotcarriedoutasthestudydidnotinvolveanyrisksto

pregnantwomen (suchasmightbe thecasewithadrug trial). Whilstpregnancy-

inducedrhinitisisrelativelycommon,itpredominantlyoccursinthelatterstagesof

59

pregnancy and is associatedwith symptoms such as nasal stuffiness (88, 89). All

subjects were asked specifically about nasal symptoms and examined by anterior

rhinoscopytoexcludethosewithrhinitis.

Most subjects had no significant past medical history. Of note, three subjects

reportedmildchildhoodasthmabuthadbeenasymptomaticforatleastfouryears

andwerenottakinganyinhalers.Theirrespiratoryfunctionthereforewasassumed

to be normal. Twopatients reported seasonal rhinitis, however theywere tested

during the autumn and were asymptomatic with normal anterior nasal

examinations.

Anteriorrhinoscopywasperformedoneachsubject inordertoexcludethosewith

nasal pathology such as nasal polyps or septal deviation, which can affect nasal

airflow (28, 54, 107). A more detailed nasal examination in the form of nasal

endoscopywasdeemedunnecessaryasitisaninvasiveprocedureandonlysubjects

withnohistoryofnasalproblemsorconcurrentnasalsymptomswererecruitedinto

thestudy.

Eightoutoffifteenfemalesubjectsweretakingsomeformofcontraception,either

progesterone-only(n=3)orcombinedoestrogenandprogesterone(n=5).Although

there have been concerns that oestrogens can cause nasal congestion (92, 97), a

studyontheeffectofthemoderncombinedcontraceptivepillfailedtoshowthatit

alterednasalairflow (99). Therefore itwasdeemedunlikely thatcontraceptionof

anykindwouldinfluencetheresultsofthisstudy. Allothersubjectsdeniedtaking

anyregularprescribedorover-the-countermedication,apartfromonemalesubject

who was taking sertraline, a selective-serotonin reuptake inhibitor used to treat

depression,whichhasnonasal sideeffects listed in theBritishNationalFormulary

(173).

Positionalfactors

Changesinpostureareknowntoaffectnasalairflow.Sometimesreferredtoasthe

corporo-nasal reflex, stimulation of pressure sensors in the thorax, pectoral and

60

pelvic girdles leads to ipsilateral nasal congestion and contralateral nasal

decongestion(78-80).Changingfromasittingtoasupinepositionwillincreasenasal

airwayresistanceviaanincreasedbloodflowtothenasalvessels(76,77).Inorder

to avoid these effects, subjects were not permitted to lie down during the

measurement period. Positional factors can also affect the reliability of

measurementsobtainedusingthemodifiedGM(66)andthereforecarewastaken

to ensure that each subject sat up straight with their head in a neutral position

during measurements. To ensure continuity in the measurement technique, all

measurements were performed by one investigator, and anatomical landmarks

(columellaandupperlip)wereusedtomaintainthesamepositioningoftheplateat

eachmeasurement.

Environmentalfactors

Recruitment and testing took place in two groups, 15 subjects were tested in

NovemberandDecember2015,and14weretestedinAprilandMay2016.Testing

was performed in a laboratory, and all subjects spent the 6-hour measurement

periodinthesameroom,withallowancesfortoiletbreaksasrequired.Althoughthe

exact temperature andhumidity level in the laboratorywerenot recorded, itwas

alwayskeptatacomfortablelevel.Roomtemperaturedifferenceswerenotshown

to affect nasal airflow when acoustic rhinometry was performed at room

temperaturesof18-22°Cand30-33°C(86).Duringthefirsttestperiod,theaverage

maximumoutdoor temperaturewas12°C, theaverageminimumtemperaturewas

10°Candtheaveragehumiditywas88%(174). Duringthesecondtestperiod, the

average maximum outdoor temperature was 12°C, the average minimum

temperaturewas8°Candtheaveragehumiditywas77%(174).Whilsttheoutdoor

environmentwas slightlydifferent foreachsubject, importantly theyweregivena

30-minuterestperiod inthetestenvironmentpriortostartinganymeasurements,

allowingacclimatisationtotheindoorenvironment,assuggestedinapreviousstudy

usingthemodifiedGM(55).Asingestionofhotwaterhasbeenshowntoincrease

nasal congestion (87), subjectswere not permitted to drink hot drinks or eat hot

food during the test period. No such effect was foundwith cold drinks (87) and

61

thereforesubjectswerepermittedbottledwaterthroughoutthedayandweregiven

astandardisedcoldmealatmidday.

Temperatureandhumiditychangesalsohavethepotentialtoreducethereliability

of themodified GM (55, 66). In order to prevent heating of the plate, excessive

handling was avoided and a different plate was used for each of three

measurementsperformedateachtimepoint.Thisalsoensuredthattherewereno

remainingcondensationareas left fromthepreviousmeasurementandallowedall

threemeasurementstobeperformedinquicksuccession.Theplateswerekeptin

thesameroomas thesubjectduring the testingperiodandcleanedat theendof

eachtestday.

Otherfactors

Sleepwasnotpermittedduring the testperiodas sleephasbeen shown toaffect

nasalairflowpatterns,irrespectiveofpostureinsomeinstances(27,28).Duringthe

testperiod,subjectswereaskedtoremainatrestbutwerepermittedtostudy,read

orwatchtelevision. Subjectswereaskedtoavoidstrenuousexercise, forexample

swimming,cyclingorrunninginthethreehourspriortothestartofthestudy.This

was necessary as studies have shown that exercising abolishes the asymmetry of

airflowbetweenthenasalpassages,increasingtheoverallnasalairflowtomeetthe

increasedmetabolic demand (84, 85). Subjectswere askednot to consumemore

thanfourunitsofalcoholinthe12hourspriortothetestperiodduetotherisksof

increasingnasalcongestion(100,101). Ascigarettesmokingcandamagethenasal

mucosa(102),onlynon-smokerswererecruitedintothestudy.

62

Chapter5:Nasalairflowpatterns

5.1:Introduction

Multiple observational studies of nasal airflow patterns have been published over

thelastcenturye.g.(3,5,8-11,13).Whatisclearfromthesestudiesisthatagreat

dealofvariabilityexistsamongstnasalairflowpatterns,notonlybetweenbutalso

withinindividuals(8-11).Thishasledtocontroversyovertheuseoftheterm“nasal

cycle”,as thekey featuresof regularityand reciprocityhavenotbeenconsistently

demonstrated(13).

Nevertheless,somesimilaritiesinpatternsofnasalairflowhavebeenobserved.For

example,severalstudieshavenotedsomedegreeofreciprocitybetweenthenasal

passages, defined as congestion on one sidewith accompanying decongestion on

theotherwithoutachangeintheoverallnasalairwayvolume(13,15).Othershave

observed a parallel or in-phase relationship, with fluctuations in nasal airflow

occurringinbothnasalpassageswithnoreversalofnasalairflowdominance(9,15).

In some individuals, fluctuations in nasal airflow have been observed butwithout

any discernable pattern (8, 15). Kern (1981) used the term “non-cycle nose” to

describeindividualswithouttheclassicalfeaturesofanasalcycle(9).Hesuggested

threecategories;nochangeinnasalresistance,changesinonenasalpassagebutnot

intheother,andanin-phaseorparallelrelationship(9).

This sectiondiscusses thenasalairflowpatternsobserved in thiscohortofhealthy

individuals.

63

5.2:Methods

In order to obtain an overall description of the nasal airflow patterns for each

subject, the nasal airflow patternswere put into different categories according to

the definitions listed below. This part of the data analysis was done prior to

unblindingtohandpreference.

Category1:

Subjectswithdefinitechangesinnasalairflowdominancethatweresustained,i.e.a

changeinthedominanceofnasalairflowfromthelefttotherightnasalpassageor

vice versa. A sustained change was defined as at least four consecutive

measurementsofdominance inone side,which then switched tobecomeat least

four consecutivemeasurementsofdominance in thecontralateral side (seeFigure

8). Aperiodofuncertaintybetweenthechangewasallowed,specificallynomore

than two equal measurements (see Figure 9). Although the time period of four

measurementswaschosenarbitrarily,itwasthoughttobereasonableasitreflected

aonehourperiodoutofthesixhourtotalmeasurementperiod.

Figure8:AnexampleofasubjectwhosenasalairflowpatternfitsintoCategory1.

64

Figure9:AsecondexampleofasubjectwhosenasalairflowpatternsfitintoCategory1.

Category2:

Subjectswithlittleornochangeinnasalairflowdominance,definedasconsecutive

readings of dominance in one nasal passage throughout themeasurement period

(see Figure 10) with the exception of no more than three readings that indicate

dominance inthecontralateralnasalpassage(seeFigure11). Equalreadingswere

disregardedinthisgroup.

Figure10:AnexampleofasubjectwhosenasalairflowpatternfitsintoCategory2.

65

Figure11:AsecondexampleofasubjectwhosenasalairflowpatternfitsintoCategory2.

Category3:

Subjectswithvariationsinthepatternsofnasalairflowdominancethatdidnotfitin

witheitheroftheabovetwocategories(seeFigure12).

Figure12:AnexampleofasubjectwhosenasalairflowpatternfitsintoCategory3.

5.3:Results

Table 7 shows the categorisation of subjects into the groups described above.

Sixteensubjects(55%)exhibitedatleastonedefiniteandsustainedreversalofnasal

airflowdominanceduring the6-hourmeasurementperiodand could thereforebe

describedasinCategory1(seeFigures13-15).Eightsubjects(28%)hadlittleorno

changeinnasalpassagedominance(seeFigures16and17).Onlyfivesubjects(17%)

66

had nasal airflow patterns that were highly variable and therefore could not be

groupedintoCategory1or2(seeFigure18).

Table7:Subjectsgroupedintocategoriesaccordingtotheirpatternsofnasalairflow.

Category1 Category2 Category3

002 022 005 003

004 023 007 006

009 025 008 015

012 028 010 027

013 031 011 030

018 032 024

019 034 033

020 035 036

67

Figure13:NasalairflowpatternsofsubjectsinCategory1.

68

Figure14:NasalairflowpatternsofsubjectsinCategory1,continued.

69

Figure15:NasalairflowpatternsofsubjectsinCategory1,continued.

70

Figure16:NasalairflowpatternsofsubjectsinCategory2.

71

Figure17:NasalairflowpatternsofsubjectsinCategory2,continued.

72

Figure18:NasalairflowpatternsofsubjectsinCategory3.

Tointerpretthisdata,onemustconsiderthephysiologicalbasisforthesepatterns.

Ineachfigure,thenasalpassagewiththedominanti.e.thegreatestnasalairflowis

representedbyadiamondateachtimepoint.Thisisequivalenttowhichevernasal

passageproducedthegreatestcondensationareaonthemodifiedGM,andinmost

caseswas judged to be either the left or the right nasal passage. The degree of

differencebetweenthecondensationareaswasnotrecorded,onlywhetheronewas

larger than the other, and therefore this is a crude method of evaluating nasal

passagedominance.

73

Whenanasalpassageisdominant,itreceivesthegreatestsympatheticoutputfrom

the brainstem, leading to vasoconstriction of the nasal veins, reduced nasal

congestion and therefore greater nasal airflow (38). At the same time, the non-

dominantnasalpassagereceiveslesssympatheticoutput,resultingindilationofthe

nasal veins, increased nasal congestion and therefore reduced nasal airflow. A

change in nasal passage dominance fromone side to the other indicates that the

balance of sympathetic output has shifted from one side of the brainstem to the

other. In most subjects, it is likely that changes in sympathetic output occurred

frequentlythroughouttheday,leadingtochangesinnasalairflow.Howeverifthis

change was not large enough to cause a complete reversal of nasal passage

dominance,itwouldnothavebeendetected.

In some instances, it was not possible to decide whether the left or right nasal

passagehadproducedthegreatestcondensationareaastheyappearedtobeequal,

for example measurements 8, 12, 14, 16 and 23 in Subject 006 (see Figure 12).

There are two possible explanations for this finding. First, it could be due to the

methodused.ThemodifiedGMdidnotprovidequantitativevaluesfortheamount

of nasal airflow in each nasal passage. Instead, the dominant nasal passage was

decided subjectively by the investigator by observing the differences in the

condensation areas produced by each side. It is possible therefore that subtle

differences could have been missed, preventing the investigator from making a

judgment on which nasal passage was dominant and leading to a recording of

“Equal”forthattimepoint.Alternatively,itcouldbethatsympatheticoutputfrom

the brainstem was divided equally between the nasal passages, leading to equal

venouscongestionandthereforeequalnasalairflow.

Ateach15-minutetimeinterval,threemeasurementsofnasalairflowweretakenin

quick succession, as taking an average of three measurements has been

recommendedtoimprovereliability(55).Iftherewasuncertaintyastowhichnasal

passage had produced the largest condensation area, this measurement was

recorded as “Equal”. Therefore there were three possible outcomes at each

measurement;“Left”,“Right”or“Equal”.Themajorityresultateachtimepointwas

74

takentobethecorrectmeasurement.Forexample,iftwoormoremeasurements

were “Left”, the result at that timepointwas recorded as “Left”. If twoormore

measurementswere“Equal”,theresultatthattimepointwasrecordedas“Equal”.

If onemeasurementwas “Left”, onewas “Right”, and the otherwas “Equal”, the

resultwasrecordedas“Equal”.

Table8:Summaryofmeasurementsthroughoutthe6-hourtimeperiodforeachsubject.

Subjectnumber

Timepointswhenall3readingsthesame

Timepointswhen2/3readingsthesame

Timepointswhenall3readingsdifferent

Totalnumberof“Equal”readingsthroughoutmeasurementperiod

/24 % /24 % /24 % /72 %002 20 83 3 13 1 4 6 8

003 21 88 3 13 0 0 0 0

004 22 92 2 8 0 0 2 3

005 16 67 7 29 1 4 7 10

006 12 50 11 46 1 4 14 19

007 12 50 11 46 1 4 12 17

008 16 67 7 29 1 4 5 7

009 24 100 0 0 0 0 0 0

010 17 71 7 29 0 0 5 7

011 24 100 0 0 0 0 0 0

012 22 92 2 8 0 0 1 1

013 18 75 6 25 0 0 5 7

015 16 67 6 25 2 8 4 6

018 22 92 1 4 1 4 1 1

019 22 92 1 4 1 4 2 3

020 17 71 7 29 0 0 1 1

022 18 75 6 25 0 0 3 4

023 19 79 4 17 1 4 4 6

024 20 83 4 17 0 0 3 3

025 18 75 6 25 0 0 6 8

027 19 79 3 13 2 8 3 4

028 23 96 1 4 0 0 1 1

030 20 83 4 17 0 0 0 0

031 20 83 4 17 0 0 4 6

032 24 100 0 0 0 0 0 0

033 24 100 0 0 0 0 0 0

034 24 100 0 0 0 0 0 0

035 20 83 3 13 1 4 6 8

036 23 96 1 4 0 0 1 1

Onaverage,allthreemeasurementsgavethesameresultateachspecifictimepoint

82%ofthetime(20/24).Twooutofthreemeasurementsgavethesameresult16%

(4/24)of the timeonaverage,whereas all threemeasurementsweredifferent an

averageof2%(0.5/24)ofthetime. Acrossallmeasurementsfortheentire6-hour

75

period,theaveragenumberofequalreadingsrecordedwas3/72(4.5%).Therewas

howeveralotofvariabilityseenhere,withtwosubjectsrecording12and14equal

readings and seven subjects recording none. In general however, the investigator

was able to make a judgment as to whether the left or right nasal passage was

dominant. Considering all 29 subjects had three measurements taken 24 times,

giving a total of 72 measurements for each subject and 2,088 measurements

altogether, a dominant nasal passage was recorded 1,992 times i.e. in 95% of

measurements.ThissuggeststhatthemodifiedGMwasanappropriatemethodfor

measuringnasalpassagedominance.

5.4:Discussion

Consideringevidencefrompreviousstudies,itisclearthatthedatapresentedabove

isconsistentwith findings in the literature. There isconsiderablevariability in the

nasalairflowpatternsobserved.Overhalfofthesubjectsdemonstratedatleastone

definiteandsustainedreversalofnasalairflowdominance, indicatingsomedegree

ofreciprocity. Thosewhodidnotexhibitadefiniteandsustainedreversalofnasal

airflow dominance either had no reversal of nasal passage dominance or had

fluctuations in nasal passage dominancewithout any obvious pattern, i.e.without

clearreciprocityorregularity.

Unlike other methods of measuring nasal airflow such as rhinomanometry, the

modifiedGMmethod used in this study did not provide quantitative values. It is

possiblethereforethatsomereversalsofnasalpassagedominancecouldhavebeen

missedifthedifferencebetweennasalairflowoneachsidewassmall.Inaddition,it

is difficult to know exactly what was happening with regard to nasal airflow

fluctuations for subjects in Category 2, i.e. with no reversal of nasal passage

dominance during the measurement period. There are several possibilities here.

Themeasurementperiodof6hourscouldhavebeentooshorttocaptureareversal

ofnasalpassagedominance,asthereissomeevidencefromtheliteraturethatthe

76

durationofthenasalcyclecanbeaslongas6or7hours(8,10).Alternatively,there

mayhavebeennofluctuationsinnasalairflow,assuggestedbyKern(1981)(9).The

most likely explanation however, given that in most studies fluctuations in nasal

airflowoveraperiodofhourshavebeenobservedtosomedegree,isthatchanges

innasal airflowdidoccur,but that thesechangeswerenot largeenough to cause

reversalofnasalpassagedominance.

Thepatternsofnasalairflowobtainedinthisstudyhavebeencomparedwiththose

from a previous study of a different group of subjects inwhich nasal airflowwas

measuredhourlyoveran8-hourperiodusinganterior rhinomanometry (175). As

shown inFigures19-22,verysimilarpatternscanbeseen. Thisdemonstratesthat

whilst there is clear inter-individual variation, some consistent patterns in nasal

airflowcanbefoundwhencomparingdifferentsubjects. Inaddition,themodified

GMhassuccessfullydetectedthesepatterns.

77

Figure19:Comparisonofnasalairflowpatterns. Topgraph:hourlymeasurementsofnasalairflowobtained using anterior rhinomanometry fromone subject over 8 hours (175). Bottom graph: 15-minutemeasurementsofnasalairflowobtainedusingthemodifiedGMinadifferentsubjectover6hours. Therightnasalpassageisrepresentedbytheblueline,theleftnasalpassageisrepresentedbytheredline.Thepatternobservedisalmostidenticalinthatthereisonereversalofnasalpassagedominance,thisoccursfromlefttorightinthetopgraphandfromrighttoleftinthebottomgraph.Usingrhinomanometry(topgraph),fluctuationsinnasalairflowaredemonstratedateachtimepoint,however it is only on one occasion that the fluctuation is great enough to cause reversal of nasalpassagedominance.FluctuationsthatdidnotleadtoareversalofnasalpassagedominancearenotdemonstratedbythemodifiedGM(bottomgraph),asquantitativevalueswerenotobtained.

78

Figure20:Comparisonofnasalairflowpatterns. Topgraph:hourlymeasurementsofnasalairflowobtained using anterior rhinomanometry fromone subject over 8 hours (175). Bottom graph: 15-minutemeasurementsofnasalairflowobtainedusingthemodifiedGMinadifferentsubjectover6hours. Therightnasalpassageisrepresentedbytheblueline,theleftnasalpassageisrepresentedby the red line. The patterns observed are similar, in that fluctuations in nasal airflow occurthroughoutthemeasurementperiodleadingtoseveralreversalsofnasalpassagedominance.

79

Figure21:Comparisonofnasalairflowpatterns. Topgraph:hourlymeasurementsofnasalairflowobtained using anterior rhinomanometry fromone subject over 8 hours (175). Bottom graph: 15-minutemeasurementsofnasalairflowobtainedusingthemodifiedGMinadifferentsubjectover6hours. Therightnasalpassageisrepresentedbytheblueline,theleftnasalpassageisrepresentedby the red line. Onceagain, thenasalairflowpatternsareverysimilar. Inbothsubjects, the rightnasal passage has remained dominant throughout the measurement period. Using anteriorrhinomanometry, small fluctuations in nasal airflowwere detected, whereas this was not possiblewiththemodifiedGM.Howevernoneofthesefluctuationsweregreatenoughastoresultinreversalofnasalpassagedominance.

80

Figure22:Comparisonofnasalairflowpatterns. Topgraph:hourlymeasurementsofnasalairflowobtained using anterior rhinomanometry fromone subject over 8 hours (175). Bottom graph: 15-minutemeasurementsofnasalairflowobtainedusingthemodifiedGMinadifferentsubjectover6hours. Therightnasalpassageisrepresentedbytheblueline,theleftnasalpassageisrepresentedby the red line. Again, thenasal airflowpatterns are very similar. In both subjects, the left nasalpassage has remained dominant throughout the measurement period, with small fluctuationsdetectedbyanteriorrhinomanometrybutnotbythemodifiedGM.

81

Chapter6:Nasalairflowandhandpreference

6.1:Introduction

Theprincipleaimofthisstudywastolookforanycorrelationbetweennasalpassage

dominanceandhandpreference.Theexistingliteratureinthisfieldisverylimited,

withonlyonepublishedstudyspecificallydesignedtoanswerthisresearchquestion

inhealthyindividuals.

Theverynotionofthisrelationshipmayseemunusual,asonecouldquestionhow

andwhynasal airflowwould be related to handedness. However humanshave a

tendency for lateral preferences (113), and the nose is considered to be two

separatepassagesratherthanasingleentity.Thereforeifpeoplecanbeleft-handed

andleft-eyed,couldtheyalsobeleft-nosed?

Searlemanetal. (2005) reported thathealthy individualsdidexhibitnasalpassage

dominance, justastheyexhibithandandeyedominance,andthatthisdominance

didcorrelatewithhanddominance(113). However,thereliabilityofthisfinding is

questionable(seesection1.9formoredetail).

Thefollowingsectionspresentnewevidenceanddiscussthesefindingsinrelationto

the previously proposed link between hand preference and nasal passage

dominance.

6.2:Methods

As part of the screening process, the hand preference of each subject was

ascertained by observation of handwriting and completion of the Edinburgh

HandednessInventory(ShortForm).

82

The dominant nasal passage was recorded using the modified GM at 15-minute

intervalsovera6-hourperiodononeday(seeChapter3fordetailedmethodology).

Oncealldatahadbeencollected, itwasanalysed inorderto lookforacorrelation

between hand preference and nasal passage dominance. Several methods of

analysis were used. Similarly to Searleman et al. (2005) (113), the percentage of

timeeachnasalpassagewasdominantwascalculatedforeachsubjectandthiswas

compared in left- and right-handed subjects. Next, left-handed and right-handed

subjects were compared according to their nasal airflow patterns, which were

categorisedasdescribedinChapter5.Theresultswerethencompareddirectlywith

those presented by Searleman et al. (2005) (113). A different method of

determiningoverallnasalpassagedominancewasusedand followingclassification

ofsubjectsaseitherleft-orright-nasalpassagedominant,achi-squaretestwasused

tolookforacorrelationbetweennasalpassagedominanceandhandpreference.

6.3:Results

Nasalairflowpatternswereanalysedin14right-handedsubjectsand15left-handed

subjects. Tables 9 and 10 show the percentage of time each nasal passage was

dominant in left-handed and right-handed subjects respectively. As shown, there

was considerable variability within each group of subjects. In left-handers, the

percentage of time that the left nasal passage was dominant ranged from 0% to

100%.Likewiseinright-handers,thepercentageoftimethattherightnasalpassage

wasdominantrangedfrom4.2%to95.8%.

83

Table9:Nasalpassagedominanceinleft-handedsubjects.Thepercentageoftimeeachnasalpassagewasdominantforeachleft-handedsubject isshown. Notethatthepercentagesdonotalwaysaddup to100%, this isbecause in somesubjects (e.g. Subject005) therewere timepointswhennasalairflowwasrecordedasbeingequallydividedbetweenthenasalpassagesandthereforeadominantnasalpassagecouldnotbedeterminedatthattimepoint.Subjectnumber

%oftimerightnasalpassagedominant

%oftimeleftnasalpassagedominant

003 25.0 75.0

004 58.3 41.7005 75.0 12.5

006 58.3 20.8007 75.0 8.3

008 12.5 79.2009 37.5 62.5

010 0.0 100.0

011 100.0 0.0012 50.0 50.0

015 70.8 16.7032 50.0 50.0

034 45.8 54.2

035 37.5 54.2036 0.0 100.0

Table 10:Nasal passage dominance in right-handed subjects. The percentage of time each nasalpassagewas dominant for each right-handed subject is shown. As above, the percentages do notalwaysaddupto100%forthesamereason.Subjectnumber

%oftimerightnasalpassagedominant

%oftimeleftnasalpassagedominant

002 41.7 45.8013 66.7 29.2

018 62.5 33.3

019 45.8 50020 41.7 58.3

022 50 50023 37.5 58.3

024 95.8 4.2025 29.2 66.7

027 29.2 62.5

028 58.3 41.7030 33.3 66.7

031 33.3 62.5033 4.2 95.8

84

Figure 23:Nasal passage dominance in left- and right-handers. Graph demonstrates the averagepercentageoftimeeachnasalpassagewasdominantinleft-handedandright-handedsubjects.Thereason that the percentages displayed do not total 100% is that in some subjects, nasal passagedominancewasdividedequally(50%and50%)betweenbothsides.

In left-handers, the left nasal passage was dominant for more 50% of the

measurementperiodin7outof15subjects(47%).Inright-handers,therightnasal

passagewasdominantformorethan50%ofthemeasurementperiodin7outof14

subjects(50%). In left-handers,theleftnasalpassagewasdominantanaverageof

11.6 times out of 24, or 48.3% of the time, whereas the right nasal passagewas

dominant for 46.4% of the time (see Figure 23). In right-handers, the right nasal

passagewas dominant an average of 10.8 times out of 24, or 44.9% of the time,

whereastheleftnasalpassagewasdominantfor51.8%ofthetime(seeFigure23).

Justfromthissimpleanalysis,noclearcorrelationbetweennasalpassagedominance

andhandpreferencecanbedemonstrated.

Next,left-andright-handedsubjectswerecomparedaccordingtothecategorisation

of their nasal airflow patterns (see Table 7). Of the 15 left-handers, 6 were in

Category1,6wereinCategory2and3wereinCategory3.Ofthe14right-handers,

10weregroupedtoCategory1,2wereinCategory2and2wereinCategory3(see

Figure 24). There appears to be a trend towards right-handers having definite

changes innasalpassagedominancewithmorevariabilityofnasalairflowpatterns

48.3 46.4 51.8

44.9

0

10

20

30

40

50

60

70

80

90

100

Le2 nasal passage Right nasal passage

% o

f ?m

e na

sal p

assa

ge d

omin

ant

Le2 handers

Right handers

85

seen in left-handers. However,usingthechi-squaretest,nostatisticallysignificant

correlationbetweenhandpreferenceand categoriesofnasal airflowpatternswas

found(p=0.21).

Figure24:Graphtodemonstratethecategorisationofleft-andright-handersaccordingtotheirnasalairflowpatterns. Category 1: Subjectswith definite changes in nasal airflowdominance thatweresustained,i.e.achangeinthedominanceofnasalairflowfromthelefttotherightnasalpassageorvice versa. Category 2: Subjects with little or no change in nasal airflow dominance, defined asconsecutive readings of dominance in one nasal passage throughout the measurement period.Category3:SubjectswithvariationsinthepatternsofnasalairflowdominancethatdidnotfitinwithCategory1orCategory2.

ComparisonwithSearlemanetal.(2005)

Theaboveanalysisusingthepercentageoftimeeachnasalpassagewasdominant

(calculatedfromthenumberofreadings)issimilartothatreportedbySearlemanet

al. (2005), the only published study to describe the possible relationship between

handpreferenceandnasalairflowinhealthysubjects(113).Theystate(p.117):

“During 24 trials of the 6-hour testing period, the left-handed males were

significantlymorelikelythanchance(50%)tohavetheirleftnostril,ratherthantheir

right, be the dominant one (59.3%, Z=2.73, p<.01). By an almost identical

percentage(59.5%),right-handersshowthereversepattern:therightnostril isthe

dominantoneinairflow(Z=3.09,p<.01).”(113)

0

2

4

6

8

10

12

Category 1 Category 2 Category 3

Num

ber o

f sub

ject

s

Nasal airflow paLerns

Le2-handers

Right-handers

86

Figures 25 and 26demonstrate the differences between the current data and the

resultspresentedbySearlemanetal. (2005) (113). There isanobviousdifference

between the current study results and thosepublishedby Searlemanet al. (2005)

(113). In Searleman et al.’s (2005) study, nasal airflowwas divided roughly 60:40

with overall nasal passage dominance positively correlating to hand preference

(113). This findingwasreportedtobesignificant,withapvalueof lessthan0.01,

howeverthestatisticalmethodsusedwerenotspecified. Inthecurrentstudy,the

divisionofnasalairflowwascloserto50:50,withnoclearcorrelationbetweennasal

passagedominanceandhandpreference.

Thereisasignificantissuewiththeabovedescriptionofthedataanalysis.Fromthe

informationprovidedinthearticle,whichisquitelimited,itappearsthatSearleman

et al. (2005) have used the average percentage of time one nasal passage was

dominanttocategorisesubjectsaseitherright-or left-nasalpassagedominantand

thenattemptedtocorrelatethiswithhandpreference(113).Theyhaveused50%as

thecutoffvalue,sothatleftnasalpassagedominanceisdefinedasgreaterairflow

throughtheleftnasalpassageformorethan50%ofthe24readingstakenovera6-

hourperiod,andviceversaforrightnasalpassagedominance(113).However,given

theknownvariabilityofnasalairflowpatternsandtherelativelyshortmeasurement

period of 6 hours on one day, it seems that having left or right nasal passage

dominanceforanaverageof50-60%ofthereadingscouldeasilyhappenbychance.

If themeasurementperiodweretobeextendedbyanhour,potentiallythis figure

and even the overall nasal passage dominance in one subject could change.

Therefore, a more robust method for determining nasal passage dominance was

usedforthisstudy,andthisisexplainedbelow.

87

Thisfigurehasbeenremovedbytheauthorforcopyrightreasons.

Figure25:GraphfromSearlemanetal.2005(113).Thegraphcomparesthepercentageoftimeeachnasalpassage(ornostril)wasdominantbasedonthemaximumairflowrate.

Figure 26: Graph using the data from this study formatted to reflect the format of the graphpublishedbySearlemanetal.(2005)(113).Thepercentageoftimeeachnasalpassagewasdominantinleft-handersandright-handersisshown.

48.3 46.4 51.8

44.9

0

10

20

30

40

50

60

70

Le2 nostril Right nostril

% o

f ?m

e no

stril

dom

inan

t

Le2 handers Right handers

88

Determiningnasalpassagedominance

Thesamplingdistributionofthedatacanbedescribedasbinomial,asithasafixed

numberofobservations(n=24),eachobservationis independent,eachobservation

represents one of two outcomes (nasal passage dominance or no difference

between the nasal passages) and the probability of success is the same for each

observation. The binomial distribution could therefore be used to determine the

numberofsuccesses i.e.dominantnasalpassagereadingsoccurringwithaspecific

probability (p). This isshown inTable11. Insomesubjects,someof thereadings

were“Equal”, inotherwords,adominantnasalpassagecouldnotbedetermined.

Asthesereadingsareessentiallyunknowns,theyhadtobeexcludedinordertouse

the binomial distribution, and this was taken into account when calculating the

probabilityofsuccesses(seeTable11).

Inthiscase,thenullhypothesisisthatthereisnooverallnasalpassagedominance,

andthealternativehypothesisisthatasubjecthasadominantnasalpassage,which

iseitherthe leftortheright. Forapvalueof0.1,whenallofthe24readingsare

known values, 7 non-dominant readings were permitted to allow a subject to be

categorised as being nasal passage dominant. For example, if a subject had 18

readingsthatwereleftand6readingsthatwereright,theywerecategorisedasleft

nasal passage dominant. However if they had 16 readings that were right and 8

readingsthatwereleft,theywerecategorisedasunclear,meaningthattherewasno

clearoverallnasalpassagedominanceandthatthefindingsweremorelikelydueto

chance.Figures27and28arerepresentative(notactual)graphsofthedata,usedto

illustrate the differences between this method of determining nasal passage

dominanceand,fromtheinformationprovidedintheirarticle,whatwaspresumably

usedbySearlemanetal.(2005)(113).

89

Table11:Usingbinomialdistributiontodeterminehowtocategoriseeachsubjectaseither left-orright-nasalpassagedominant.

Numberofreadingswhereanasalpassageis

dominant

Numberofnon-dominantreadingsallowedinordertocategoriseasubjectasleft-orright-nasalpassagedominant

p=0.1 p=0.2 p=0.4

24 7 8 9

23 7 7 922 6 7 8

21 6 7 8

20 5 6 719 5 6 7

Figure27:Anormaldistributioncurveisrepresentedbytheverticallines.Thenullhypothesisisthatsubjects do not exhibit any nasal passage dominance, i.e. nasal airflow is divided roughly equallybetween the nasal passages. The alternative hypothesis is that subjects have a dominant nasalpassage,which is either left or right. At the extremes of both ends of the curve, subjects can becategorised as left (shown in red) or right (shown in blue) nasal passage dominant, i.e. the nullhypothesisisrejected.Thecut-offpointinthenumberofreadingsrequiredtocategoriseasubjectaseitherleftorrightnasalpassagedominantisdependentonthepvalueusedforthispurposeandthisisshownforpvaluesof0.05,0.1and0.2ateachextreme,whichwouldequateto0.1,0.2and0.4whenconsideringallofthereadings. The linesshowningreyrepresentthosewherenasalpassagedominancecannotbeassigned,i.e.thenullhypothesisisaccepted.

12#p#0.05# p#0.05#

p#0.1# p#0.1#

p#0.2# p#0.2#

Number#of#readings#

Le7#nasal#passage#dominant#

Right#nasal#passage#dominant#

90

Thepvaluesusedmayseemrelativelyhigh,howeverevenatapvalueof0.4, the

nullhypothesis isnotoutside theextremequartiles,asat this level20%wouldbe

left-nasalpassagedominant,20%wouldbe right-nasalpassagedominantand60%

wouldhavenooveralldominance.

Figure28: This figuredemonstrates themethodpossiblyusedby Searlemanet al. (2005) (113). Anormaldistributioncurveisrepresentedbytheverticallines.Acut-offof50%wasused,i.e.ifmorethan50%ofreadings(n=12)wereleftnasalpassagedominant,thesubjectwascategorisedasbeingleftnasalpassagedominant(showninred),andviceversafortherightnasalpassage(showninblue).

12#Number#of#readings#

Le3#nasal#passage#dominant#

Right#nasal#passage#dominant#

91

Table 12: Nasal passage dominance in each subject calculated using binomial distribution. Thenumberofknownvaluesindicatesthenumberofreadingsatwhichadominantnasalpassagecouldbeallocated,orthosewhereairflowwasnotclassifiedasbeing“Equal”betweenthenasalpassages.Thefourthcolumnunderoverallnasalpassagedominanceindicatesthenasalpassagedominanceifthecutoffused is50% i.e. ifmorethanhalfof thereadingswereeither rightor leftnasalpassagedominant.Subjectnumber

No.ofknownvalues

No.oftimesrightnasalpassagedominant

No.oftimesleftnasal

passagedominant

Overallnasalpassagedominance Handpreference

p=0.1 p=0.2 p=0.4 50%cutoff

002 21 10 11 Unclear Unclear Unclear Left Right

003 24 6 18 Left Left Left Left Left

004 24 14 10 Unclear Unclear Unclear Right Left

005 21 18 3 Right Right Right Right Left

006 19 14 5 Right Right Right Right Left

007 20 18 2 Right Right Right Right Left

008 22 3 19 Left Left Left Left Left

009 24 9 15 Unclear Unclear Left Left Left

010 24 0 24 Left Left Left Left Left

011 24 24 0 Right Right Right Right Left

012 24 12 12 Unclear Unclear Unclear Unclear Left

013 23 16 7 Right Right Right Right Right

015 21 17 4 Right Right Right Right Left

018 23 15 8 Unclear Unclear Right Right Right

019 23 11 12 Unclear Unclear Unclear Left Right

020 24 10 14 Unclear Unclear Unclear Left Right

022 24 12 12 Unclear Unclear Unclear Unclear Right

023 23 9 14 Unclear Unclear Left Left Right

024 24 23 1 Right Right Right Right Right

025 23 7 16 Left Left Left Left Right

027 22 7 15 Unclear Left Left Left Right

028 24 14 10 Unclear Unclear Unclear Right Right

030 24 8 16 Unclear Left Left Left Right

031 23 8 15 Unclear Unclear Left Left Right

032 24 12 12 Unclear Unclear Unclear Unclear Left

033 24 1 23 Left Left Left Left Right

034 24 11 13 Unclear Unclear Unclear Left Left

035 22 9 13 Unclear Unclear Unclear Left Left

036 24 0 24 Left Left Left Left Left

92

Table 13: Summary of the number of subjects categorised as unclear, right or left nasal passagedominantatdifferentprobabilitylevels.Nasalpassagedominance

Numberofsubjects

p=0.1 p=0.2 p=0.4 50%cutoff

Unclear 16 14 10 3Right 7 7 8 10

Left 6 8 11 16

Total 29 29 29 29

Table14:Nasalpassagedominanceasdeterminedusingbinomialdistributionintheleft-handedandright-handedgroups.

Nasalpassagedominanceatdifferentprobabilitylevels

Numberofsubjects

Right-handers Left-handers

p=0.1 Unclear 10 6Right 2 5

Left 2 4

p=0.2 Unclear 8 6Right 2 5

Left 4 4

p=0.4 Unclear 5 5

Right 3 5Left 6 5

50%cutoff Unclear 1 2Right 4 6

Left 9 7

JustfromlookingatthenumbersinTable14,theredoesnotseemtobeanobvious

correlationbetweenhandpreferenceandnasalairflowdominance.Evenusing50%

ofreadingsasthecut-offtodefineanasalpassageasdominant,moreright-handers

had left nasal passage dominance andmore left-handers had right nasal passage

dominance,althoughthenumberswereverysimilar.

93

Thedisadvantageofusingthebinomialdistributiontodeterminewhetherasubject

canbeclassifiedashavingadominantnasalpassageisthatof29subjectstested,the

finalnumberusedforanalysishasbeenreduced,assomesubjectsdidnotexhibita

dominantnasalpassage.Inanattempttocounteractthis,analysisofthedatawas

doneusingincreasingprobabilitylevelsinordertoincreasethenumberofincluded

subjects. However,evenatthehighestpvalueof0.4,10subjects(5right-handers

and5left-handers)couldnotbeclassifiedashavingadominantnasalpassageasthe

division of airflow between the right and left nasal passageswas not significantly

different.Searlemanetal.(2005)didnotstatethatnasalpassagedominancecould

notbeascertainedinanyoftheirsubjects(113).Converselyinthisstudy,3subjects

had 12 readings of left nasal passage dominance and 12 readings of right nasal

passagedominanceoutofatotalof24readings,meaningthatthedivisionofairflow

betweenthenasalpassageswasexactlyequalthroughoutthemeasurementperiod.

Determiningwhetheracorrelationbetweenhandpreferenceandnasalpassage

dominanceexists.

Achi-squaretestwasusedtolookforastatisticallysignificantcorrelationbetween

handpreferenceandnasalpassagedominance.Thiswasdoneforeachofthe

differentprobabilitylevelsusedtodeterminenasalpassagedominance(p=0.1,

p=0.2,p=0.4)andalsousingthe50%cutoff,andisshowninTables15–18.

Table15:Chi-squaretableusingpvalueof0.1tocategorisenasalpassagedominance.

Right-handers Left-handers Total

Rightnasalpassagedominant 2 5 7Leftnasalpassagedominant 2 4 6

Total 4 9 13

p=0.85

94

Table16:Chi-squaretableusingpvalueof0.2tocategorisenasalpassagedominance.

Right-handers Left-handers Total

Rightnasalpassagedominant 2 5 7Leftnasalpassagedominant 4 4 8

Total 6 9 15p=0.40

Table17:Chi-squaretableusingpvalueof0.4tocategorisenasalpassagedominance.

Right-handers Left-handers Total

Rightnasalpassagedominant 3 5 8Leftnasalpassagedominant 6 5 11

Total 9 10 19p=0.46

Table18:Chi-squaretableusing50%asthecutofftocategorisenasalpassagedominance.

Right-handers Left-handers Total

Rightnasalpassagedominant 4 6 10Leftnasalpassagedominant 9 7 16

Total 13 13 26p=0.42

Astatisticallysignificantcorrelationbetweenhandpreferenceandnasalairflowwas

notidentifiedinthisdataset.Thisisnotsurprisingasnotrendtowardsacorrelation

was seen earlier in the data analysis (see Figure 23). Unfortunately the subject

numbers in the data setwere reduced because in some subjects it was not clear

whether they had any overall nasal passage dominance, rather, it appeared that

nasalairflowwasdividedroughlyequallybetweenbothnasalpassages.Evenwhen

nasal passage dominancewas defined by having either right or left nasal passage

dominance for over 50% of themeasurement period, the p value remained non-

significant at 0.42 (see Table 18), however due to the variability in nasal airflow

patterns this method of defining nasal passage dominance should be considered

unreliable.

95

6.4:Discussion

ThefindingsofthisstudydirectlycontradictthosepublishedbySearlemanetal. in

2005(113). Someofthepotential issueswiththemethodsandfindingsdescribed

bySearlemanetal.(2005)havealreadybeenhighlighted,forexamplethemethods

used to measure nasal airflow, determine hand preference and analyse the data

(113).

Inordertolookfordisparitiesinthedataanalysis,thetheoreticalresultsneededto

produceasignificanceleveloflessthan0.01,asobtainedbySearlemanetal.(2005)

(113),wereascertained.Theirstudyhadatotalof20subjects,with9left-handers

and11right-handers.Thisinformationwasusedtopopulateachi-squaretablewith

thenumbersforleft-handed,left-nasalpassagedominantsubjectsandright-handed,

right-nasalpassagedominantsubjects.Thesenumberswerethenalteredtolookat

the different levels of significance that could be achieved. The total number of

subjectsinthenasalpassagedominancegroupswaschangedasthiswasunknown,

whereas the known values for the number of left-handers and right-handers was

kept the same. Unfortunately this is speculation, as although contacted several

times by email to request further detail about the data collected and its analysis,

therewasnoreplyfromtheleadauthor.Table19demonstratesthemostextreme

results that could have been found by Searleman et al. (2005) (113), that is all

(n=11/11) right-handers being right nasal passage dominant and all (n=9/9) left-

handers being left nasal passage dominant. In this case, thep valuewould have

been0.000007,inotherwordsahighlysignificantfinding.

96

Table19:Chi-squaretableusingthemostextremepossiblevaluesinSearlemanetal.’s(2005)(113)datai.e.alltheright-handerswererightnasalpassagedominantandalltheleft-handerswereleftnasalpassagedominant.

Right-handers Left-handers Total

Rightnasalpassagedominant 11 0 11

Leftnasalpassagedominant 0 9 9Total 11 9 20

pvalue=0.000007

Table20:Chi-squaretableusingtheleastextremevaluesrequiredtoproduceapvalueoflessthan0.01.

Right-handers Left-handers Total

Rightnasalpassagedominant 8 1 9Leftnasalpassagedominant 3 8 11

Total 11 9 20pvalue=0.006

Table21:Chi-squaretableusinganotherpossiblecombinationoftheleastextremevaluesrequiredtoproduceapvalueoflessthan0.01.

Right-handers Left-handers TotalRightnasalpassagedominant 9 2 11

Leftnasalpassagedominant 2 7 9Total 11 9 20

pvalue=0.008

Tables20and21showtwodifferentcombinationsofvaluesthatwouldbetheleast

extremevaluesrequiredtoproduceapvalueoflessthan0.01.Thesevaluesshould

stillbeconsideredasextremeandthecorrelationisimmediatelyobvioussince73–

82%ofright-handerswouldhavebeenright-nasalpassagedominantand78%ofleft-

handerswouldhavebeenleft-nasalpassagedominant.Tables22and23showtwo

differentcombinationsofvaluesthatwouldbethemostextremevaluesresultingin

a p value of greater than 0.01. Again, although the p value is higher than that

reportedbySearlemanetal. (2005) (113), the results required toproduce thesep

values are still extreme, with high percentages of subjects having positively

correlated hand preference and nasal passage dominance. If any of the above

97

resultsorsimilaroneswereactuallyobtained,surelytheywouldhavebeenclearly

reportedinthisfashionbySearlemanetal.(2005)(113),insteadofthepercentage

oftimevaluesthatwerepublished.

Table22:Chi-squaretabledemonstratestheresultsrequiredtoproduceapvalueofover0.01,thesignificancelevelfoundbySearlemanetal.(2005)(113).

Right-handers Left-handers Total

Rightnasalpassagedominant 8 2 10Leftnasalpassagedominant 3 7 10

Total 11 9 20

pvalue=0.02

Table23:Chi-squaretableusinganotherpossiblecombinationtheresultsrequiredtoproduceapvalueofover0.01.

Right-handers Left-handers TotalRightnasalpassagedominant 9 3 12

Leftnasalpassagedominant 2 6 8

Total 11 9 20pvalue=0.03

Considering the results that were published by Searleman et al. (2005) (113),

specificallythatleft-handershadleftnasalpassagedominanceforalmost60%ofthe

timeandvice versa for right-handers, a chi-square testwas thenperformedusing

60% to determine the number of right-handed, right-nasal passage dominant

subjectsand thenumberof left-handed, left-nasalpassagedominant subjects (see

Table24). Using thesenumbers, thep valuewas0.4. This is interestingas thep

value obtained in this study was 0.4 to 0.86 depending on the categorisation of

subjects’nasalpassagedominance.

98

Table24:Chi-squaretableusing60%tocalculatethepresumptivevalueforright-handed,right-nasalpassagedominantsubjectsandleft-handed,left-nasalpassagedominantsubjects.

Right-handers Left-handers Total

Rightnasalpassagedominant 7 4 11

Leftnasalpassagedominant 4 5 9Total 11 9 20

pvalue=0.4

99

Chapter7:FinalDiscussionandConclusions

Theaimsofthisstudywere:

1. To summarise the literature concerning nasal airflow and brain activity,

focusing on a possible correlation between cerebral asymmetries and

asymmetriesinnasalairflowpatterns.

2. To look specifically for a correlation between hand preference and nasal

airflowinhumansubjectsbymeasuringnasalairflowpatternsoveraperiod

ofhours,andcomparethesefindingstotheavailableliterature.

7.1:Nasalairflowasymmetryandcerebralasymmetry

Correlations between nasal airflow and cerebral asymmetry have been suggested,

however unfortunately the literature base concerning this topic is hugely varied

making itdifficult toconstruct firmconclusions. Theasymmetry innasalairflow is

controlled by asymmetrical sympathetic output from collections of sympathetic

neurons in thebrainstem (38). The roleofhigher cortical centres in thispathway

however is not clear, and the potential influences of handedness (113), ultradian

rhythms in cerebral activity (128) and disease (70, 108) have all been suggested.

Conversely, there isalsosomeevidencetosuggest thatasymmetricalnasalairflow

can affect the brain. In patients with certain types of epilepsy, nasal

hyperventilationhasbeenshowntoactivateepilepticfociinthebrain(140).Further

tothis,someresearchershavepostulatedthatunilateralforcednostrilbreathingcan

affectbrainactivityandcognitivefunctions(142,156),althoughtheevidenceforthis

isweakandhasbeencontestedbyotherstudies(136,158).

100

7.2:Handpreferenceandnasalpassagedominance

This study did not identify a relationship between hand preference and nasal

passagedominance,andinfactinarelativelyhighproportionofsubjectsadominant

nasalpassage couldnotbeascertained (although theexact figuredependson the

methodofclassifyingnasalpassagedominance–seesection6.3).

The following are the possible reasons for not identifying an association between

handpreferenceandnasalairflow:

1. Lack of statistical power. It remains possible that there is a relationship

betweennasalairflowandhandpreferencehowevertheeffectwastoosmall

tobedetectedinthissample.Conductingthesameanalysisinamuchlarger

sample couldpossiblydetectaneffect. It shouldbenotedhowever thata

smallersamplesizewasusedinthestudybySearlemanetal.(2005),inwhich

statisticalsignificancewasreported(113).

2. The measurement period was too short. Alternation of nasal passage

dominance,sometimesreferredtoas thenasalcycle,hasbeenreportedto

occurupto8-hourly(8).Therefore,withameasurementperiodof6hours,it

ispossiblethatinsomesubjectsthealternationofnasalpassagedominance

wasmissed.Inthreesubjects,onenasalpassageremaineddominantforthe

entire measurement period, and it is possible that they will have had

alternation innasalpassagedominancethateitheroccurred justbeforethe

first measurement or after the last one. Also, in subjects who did

demonstratealternationsinnasalpassagedominance,somecouldhavebeen

missed by the 6-hourmeasurement period,which could have affected the

results. For example, if the next few readings after the 6-hour period had

demonstratedanotherswitchinnasalpassagedominance,thepercentageof

time each nasal passage was dominant would have changed, which could

have altered the classification of the subject as left- or right-nasal passage

dominant.

101

3. Thereisnorelationshipbetweenhandpreferenceandnasalairflowandthe

resultsreportedbySearlemanetal.(2005)(113)occurredbychance.

Itistheopinionofthisauthorthatthe3rdoptionisthemostlikely.Therearethree

reasons for reaching this conclusion. First, the physiological reasoning underlying

thisrelationshipisimplausible.Handednesshasbeenshowntocorrelatewithother

behavioural lateralpreferences, forexampleeyepreference,handclaspingand leg

crossing,howeverthecorrelationsaresmallandthesemeasurescannotbereliably

usedtodeterminehandedness(118). Nasalairflowiscontrolledbytheautonomic

nervous system and is dissimilar to behavioural lateral preferences such as hand

clasping and leg crossing. Since there isn’t a consistent relationship between

handedness and these behaviours, it doesn’t seem logical to have a consistent

relationshipbetweenhandednessandnasalairflow.

Althoughnotfullyunderstood, it is likelythatthepurposeofhavingfluctuationsin

nasalpassagedominanceisrelatedtotheair-conditioning(26)andimmunedefence

functions of the nose (18). Therefore, equal or roughly equal division of labour

betweenthetwonasalpassageswouldberequired.Thedivisionoflabourmaynot

occurhourlyorevendaily,whichiswhyitcouldbemissedinsomestudies.Infact,

studies have shown that nasal airflow patterns vary whenmeasured in the same

individualondifferentdays(11). Bearingthis inmind, itwouldbepossibleforthe

findingsbySearlemanetal. (2005) (113) tohaveoccurredbychance,andhadthe

experimentbeen repeatedon a different day, theopposite relationshipmayhave

beendiscovered.

In addition, there is no obvious reason why the air-conditioning and immune

functionsof thenosewouldbe related tocerebralorganisation. Handpreference

mayberelatedtospeechdevelopmentwhichusuallyoccursinthelefthemisphere

(121), and the lateralisation of speech may confer advantages for speech

development (125). Whena complexactionoriginates in thebrain, theremaybe

102

advantagestohavingitariseinonlyonehemisphere(119).Theconductanceofair

through the nasal passages is not a complex action such as speech or hand

movements and therefore having a dominant nasal passage would not be

advantageousinthisrespect.

Secondly,arelationshipbetweennasalairflowandhandpreferenceisnotsupported

by the other literature concerning nasal airflow. In the wealth of observational

studiesperformedoverthelastcenturylookingatnasalairflowpatternsinhealthy

individuals,nonehave reportedan incidental findingofoverall rightnasalpassage

dominance. IfSearlemanetal.’s (2005) (113) findingswerecorrect, thiswouldbe

extremely surprising given the overwhelming majority of right-handers in the

generalpopulation.Inastudycomparingnasalairflowpatternsofschizophrenicvs.

healthyindividuals,53.1%ofthehealthycohorthadnooveralllateralisationofnasal

airflow,meaning that the left and rightnasalpassagesweredominant for roughly

equal amountsof time (108). The rateof left nasal passagedominancewas25%,

andtherateof rightnasalpassagedominancewas21.9%(108). Thiswasa larger

studywith64healthycontrolsubjectswhowereallright-handed,conductedovera

longerperiodof12hours(108). Thisfinding isconducivewiththetheoriesonthe

functionofthealternationsinnasalpassagedominance,asexplainedabove.

Thirdly,reproductionofSearlemanetal.’s2005studyusingsimilarmethodsfailedto

identify a relationshipbetweenhandpreferenceandnasal airflow (113). In recent

years, there have been concerns amongst the scientific community that many

publishedstudyfindingsarenotreproducible.Inhis2005paper,Ioannidisstates:

“Itcanbeproventhatmostclaimedresearchfindingsarefalse”(176).

Colquhoun(2014)describedthemisuseandmisinterpretationofpvaluesasamajor

contributortothis,asthereisalackofunderstandingwhenitcomestosignificance

testing (177). It is generally accepted that a p value of less than 0.05 confers

statisticalsignificance,howeverthisiscommonlymisinterpretedastheerrorrate.In

fact, a p value of less than 0.05 suggests that the test sample provides enough

evidencetorejectthenullhypothesis,howeveritdoesnotprovethatthealternative

103

hypothesisistrue.Withapvalueof0.05,thereisa5%chancethatthealternative

hypothesis has occurred by chance alone. Colquhoun (2014) argues that a 5%

chance is actuallyquitehigh in itself, butdue to themisinterpretationand lackof

powerinmanypublishedstudies,thisfigureisactuallyalothigher(177).Theerror

rate,or falsediscovery rate, isequivalent to the falsepositive rate, i.e. incorrectly

accepting the alternativehypothesis (177). Sellke et al. (2001) estimated that the

errorrateforapvalueof0.05wasover29%,andforapvalueof0.01wastypically

closeto15%(178).

Searlemanetal.(2005)statedthattheyobtainedapvalueoflessthan0.01fortheir

results, although the statistical methods used to calculate this have not been

described (113). This equates to a less than 1% chance that the alternative

hypothesis (in this case a positive correlation between nasal airflow and hand

preference)hasoccurredbychancealone,howeverhowmuch lessthan1%this is

cannotbeknown.Theerrorrateorfalsediscoveryrate,ontheotherhand,islikely

tobemuchhigher (seeSellkeet al. 2001 (178)andColquhoun2014 (177)). This

maybecompoundedbythefactthatthesamplesizewassmallmeaningthestudy

was probably underpowered, and there was a risk of bias, for example as

participantswerenotrandomisedandtherewasnoblinding.Ithasbeensuggested

thatevenawell-poweredepidemiologicalstudymayonlyhaveaoneinfivechance

of being true (176). Colquhoun (2014) suggests that ap value of less than 0.001

shouldberequiredtodemonstrateapositivefinding.

Anothermajorissueisthegeneralreluctancetoreportnegativefindings(177).Itis

thereforeimpossibletoknowwhetherotherresearchgroupshaveconductedsimilar

studiesonhandpreferenceandnasalairflow,butnotreportedtheirfindingsifthey

were not positive or significant. As discussed previously, Searleman et al.’s 2005

study is theonlypublishedreportofacorrelationbetweennasalairflowandhand

preference (113), and so presumably the scientific community has accepted this

finding. It has in fact been cited on 12 occasions. Accepting and re-iterating the

104

resultsobtainedbyasingleresearchteam,eveniftheywerestatisticallysignificant,

canbemisleading(176). This isparticularlytrueforsmallstudieswithsmalleffect

sizeswheretheremaybedifferencesindefinitions,studydesign,methodologyand

outcomes(176).Intherelativelysmallfieldofnasalairflowresearch,thereisahigh

degreeofvariability in themethodsandoutcomemeasuresused (seesections1.2

and 1.7 on the nasal cycle and methods of measuring nasal airflow), and many

studies have small numbers of participants. One could argue that accepting the

findingsofSearlemanetal.(2005)(113)astrueisnotadisaster,asit isunlikelyto

haveaffectedclinicalpracticeorpatientsafety.However,understandingphysiology

isthebasisforunderstandingpathology,andcurrentlythephysiologicalfactorsthat

influencenasalairflowandthenasalcyclearenotfullyunderstood.

AsdiscussedintheChapter2,conflictingtheorieshavebeensuggestedforcortical

influenceonnasalairflow. Nasalblockage isacommonpresentation toEar,Nose

andThroatsurgeons,andat timesthere isnopathologydetected. In thesecases,

theremay be a physiological explanation for the patient’s symptoms. The results

presentedhere,alongwithevidence fromother studies,donot suggest thathand

preferenceisafactor.

105

7.3:Limitationsofthisstudy

The sample size of 29 subjectsmay have been too small to detect a relationship

betweennasalairflowandhandpreference.UnfortunatelythestudybySearleman

et al. (2005) did not provide enough information for power calculations to be

performed (113), and therefore this was a pilot study. Measuring nasal airflow

patterns and how they alter over time is, by definition, time consuming, and

therefore many studies that have looked at nasal airflow have involved small

numbersofparticipants.Itisalsopossiblethatthemeasurementperiodof6hours

wastooshort,howeverthiswaschosentoreflectthemethodsusedbySearlemanet

al.(2005)(113).

ThemethodchosentomeasurenasalairflowwasthemodifiedGM,whichdoesnot

providequantitativevalues.Instead,theinvestigatormadeajudgementastowhich

nasalpassagewasdominantateachtimepointbylookingatthecondensationareas

producedbyeachnasalpassage.Therefore,itispossiblethathumanerrorandbias

mayhaveoccurred.Theinvestigatorwasblindedtohandpreferencewherepossible

inanattempttominimisetheriskofbias.

106

References1. EcclesR.Nasalairflowinhealthanddisease.ActaOtolaryngol.2000;120(5):580-95.2. HanifJ,JawadSS,EcclesR.Thenasalcycleinhealthanddisease.ClinOtolaryngolAlliedSci.2000;25(6):461-7.3. StokstedP.Rhinometricmeasurementsfordeterminationofthenasalcycle.ActaOtolaryngolSuppl.1953;109:159-75.4. HasegawaM,KernEB.Thehumannasalcycle.MayoClinicProceedings.1977;52:28-34.5. EcclesR.Thecentralrhythmofthenasalcycle.ActaOtolaryngol.1978;86(5-6):464-8.6. KayserR.DieexackteMessungderLuftdurchgangigkeitderNase.ArchivLaryngologieRhinologie.1895;3:101-20.7. SenH.Observationsonthealternateerectilityofnasalmucousmembrane.Lancet.1901;2(4069):564.8. GilbertAN,RosenwasserAM.BiologicalRhythmicityofNasalAirwayPatency-aReexaminationoftheNasalCycle.ActaOto-Laryngol.1987;104(1-2):180-6.9. KernEB.Thenoncyclenose.Rhinology.1981;19:59-74.10. HasegawaM,KernEB.Variationsinnasalresistanceinman:arhinomanometricstudyofthenasalcyclein50humansubjects.Rhinology.1978;16(1):19-29.11. WilliamsM,EcclesR.Amodelforthecentralcontrolofairflowpatternswithinthehumannasalcycle.JLaryngolOtol.2016;130(1):82-8.12. WilliamsRG,EcclesR.Nasalairflowasymmetryandtheeffectsofatopicalnasaldecongestant.Rhinology.1992;30(4):277-82.13. FlanaganP,EcclesR.Spontaneouschangesofunilateralnasalairflowinman.Are-examinationofthe'nasalcycle'.ActaOtolaryngol.1997;117(4):590-5.14. HuangZL,OngKL,GohSY,LiewHL,YeohKH,WangDY.Assessmentofnasalcyclebyacousticrhinometryandrhinomanometry.Otolaryngology--headandnecksurgery:officialjournalofAmericanAcademyofOtolaryngology-HeadandNeckSurgery.2003;128(4):510-6.15. GallegoAJ,CavallariFE,ValeraFC,DemarcoRC,Anselmo-LimaWT.Studyofnasalcyclesinchildrenbyacousticrhinometry.AmJRhinol.2006;20(6):560-2.16. EladD,WolfM,KeckT.Air-conditioninginthehumannasalcavity.RespirPhysiolNeurobiol.2008;163(1-3):121-7.17. LucasHA.TheHistopathologyofSinusitis.TheJournalofLaryngologyandOtology.1952;66(10):480-9.18. EcclesR.Aroleforthenasalcycleinrespiratorydefence.EurRespirJ.1996;9(2):371-6.19. GreversG,KamargakisWN.Intervascularsmoothmusclefibersandmuscularbolstersinnasalswellbodiesofhumans.AnnOtolRhinolLaryngol.1995;104(2):144-8.20. GreversG,HerrmannU.Fenestratedendotheliainvesselsofthenasalmucosa.Anelectron-microscopicstudyintherabbit.Archivesofoto-rhino-laryngology.1987;244(1):55-60.

107

21. GreversG.Theroleoffenestratedvesselsforthesecretoryprocessinthenasalmucosa:ahistologicalandtransmissionelectronmicroscopicstudyintherabbit.Laryngoscope.1993;103(11Pt1):1255-8.22. PerssonCG,ErjefaltI,AlknerU,BaumgartenC,GreiffL,GustafssonB,etal.Plasmaexudationasafirstlinerespiratorymucosaldefence.ClinExpAllergy.1991;21(1):17-24.23. EcclesR,ReillyM,EcclesKS.Changesintheamplitudeofthenasalcycleassociatedwithsymptomsofacuteupperrespiratorytractinfection.ActaOtolaryngol.1996;116(1):77-81.24. WarrenNJ,CrampinEJ,TawhaiMH.Theroleofairwayepitheliuminreplenishmentofevaporatedairwaysurfaceliquidfromthehumanconductingairways.Annalsofbiomedicalengineering.2010;38(12):3535-49.25. ButtonB,CaiLH,EhreC,KesimerM,HillDB,SheehanJK,etal.Apericiliarybrushpromotesthelunghealthbyseparatingthemucuslayerfromairwayepithelia.Science.2012;337(6097):937-41.26. WhiteDE,BartleyJ,NatesRJ.Modeldemonstratesfunctionalpurposeofthenasalcycle.BiomedEngOnline.2015;14:38.27. RohrmeierC,SchittekS,EttlT,HerzogM,KuehnelTS.Thenasalcycleduringwakefulnessandsleepanditsrelationtobodyposition.Laryngoscope.2014;124(6):1492-7.28. KimuraA,ChibaS,CapassoR,YagiT,AndoY,WatanabeS,etal.Phaseofnasalcycleduringsleeptendstobeassociatedwithsleepstage.Laryngoscope.2013;123(8):2050-5.29. TahamilerR,YenerM,CanakciogluS.Detectionofthenasalcycleindailyactivitybyremoteevaluationofnasalsound.ArchOtolaryngolHeadNeckSurg.2009;135(2):137-42.30. OhkiM,OgoshiT,YuasaT,KawanoK,KawanoM.Extendedobservationofthenasalcycleusingaportablerhinoflowmeter.TheJournalofotolaryngology.2005;34(5):346-9.31. GilbertAN.Reciprocityversusrhythmicityinspontaneousalternationsofnasalairflow.Chronobiologyinternational.1989;6(3):251-7.32. TschalussowMA.DieInnervationderGefassederNasenschleimhaut.PflugersArch.1913;151(11):523-42.33. MalcolmsonKG.Thevasomotoractivitiesofthenasalmucousmembrane.JournalofLaryngologyandOtology.1959;73(2):73-98.34. FlanaganP,EcclesR.Physiologicalversuspharmacologicaldecongestionofthenoseinhealthyhumansubjects.ActaOtolaryngol.1998;118(1):110-3.35. DavisSS,EcclesR.Nasalcongestion:mechanisms,measurementandmedications.Coreinformationfortheclinician.ClinOtolaryngolAlliedSci.2004;29(6):659-66.36. StokstedP,ThomsenKA.Changesinthenasalcycleunderstellateganglionblock.ActaOtolaryngolSuppl.1953;109:176-81.37. GirgisIH,YassinA,HamdyH,MorisM.Amethodforassessmentofthenasalcirculation.JLaryngolOtol.1974;88(12):1149-58.38. BamfordOS,EcclesR.Thecentralreciprocalcontrolofnasalvasomotoroscillations.PflugersArch.1982;394(2):139-43.

108

39. EcclesR,LeeRL.Theinfluenceofthehypothalamusonthesympatheticinnervationofthenasalvasculatureofthecat.ActaOto-Laryngol.1981;91:127-34.40. LeclercJ,DoyleWJ,KarnavasWJ.Therelationshipbetweenthenasalcycleandaxillarysweatproduction.Rhinology.1987;25(4):249-57.41. ShibasakiM,CrandallCG.Mechanismsandcontrollersofeccrinesweatinginhumans.FrontBiosci(ScholEd).2010;2:685-96.42. GaliotoG,MevioE,GaliotoP,FornasariG,CisterninoM,FraiettaL.ModificationsoftheNasalCycleinPatientswithHypothalamicDisorders-KallmannsSyndrome.AnnOtoRhinolLaryn.1991;100(7):559-62.43. FisherEW,LiuM,LundVJ.TheNasalCycleafterDeprivationofAir-Flow-aStudyofLaryngectomyPatientsUsingAcousticRhinometry.ActaOto-Laryngol.1994;114(4):443-6.44. FisherEW,LiuM,LundVJ.Air-FlowandtheNasalCycle-NasalPatencyFluctuationsafterLaryngectomy.AmericanJournalofRhinology.1995;9(3):175-8.45. TsuboneH.Nasal'flow'receptorsoftherat.RespirPhysiol.1989;75(1):51-64.46. SozanskyJ,HouserSM.Thephysiologicalmechanismforsensingnasalairflow:aliteraturereview.IntForumAllergyRhinol.2014;4(10):834-8.47. EcclesR,JonesAS.Theeffectofmentholonnasalresistancetoairflow.JLaryngolOtol.1983;97(8):705-9.48. EcclesR,MorrisS,TolleyNS.Theeffectsofnasalanaesthesiauponnasalsensationofairflow.ActaOtolaryngol.1988;106(1-2):152-5.49. CasaleM,PappacenaM,SetolaR,SodaP,CusimanoV,VitaliM,etal.Video-rhino-hygrometer:anewmethodforevaluationofnasalbreathingafternasalsurgery.AmJRhinolAllergy.2010;24(6):467-71.50. PynnonenMA,KimHM,TerrellJE.ValidationoftheSino-NasalOutcomeTest20(SNOT-20)domainsinnonsurgicalpatients.AmJRhinolAllergy.2009;23(1):40-5.51. ChaabanM,CoreyJP.Assessingnasalairflow:optionsandutility.ProcAmThoracSoc.2011;8(1):70-8.52. ChavesC,deAndradeCR,IbiapinaC.Objectivemeasuresforfunctionaldiagnosticoftheupperairways:practicalaspects.Rhinology.2014;52(2):99-103.53. MackayIS.Measurementofnasalairflowandresistance.JRSocMed.1979;72(11):852-5.54. GertnerR,PodoshinL,FradisM.Asimplemethodofmeasuringthenasalairwayinclinicalwork.JLaryngolOtol.1984;98(4):351-5.55. BrescoviciS,RoithmannR.ModifiedGlatzelmirrortestreproducibilityintheevaluationofnasalpatency.BrazJOtorhinolaryngol.2008;74(2):215-22.56. EmpeyDW.Assessmentofthenasalpassages.BrJClinPharmacol.1980;9(4):317-9.57. FoxenEH,PrestonTD,LackJA.Theassessmentofnasalair-flow:areviewofpastandpresentmethods.JLaryngolOtol.1971;85(8):811-25.58. NathanRA,EcclesR,HowarthPH,SteinsvagSK,TogiasA.Objectivemonitoringofnasalpatencyandnasalphysiologyinrhinitis.JAllergyClinImmunol.2005;115(3Suppl1):S442-59.59. EcclesR.Aguidetopracticalaspectsofmeasurementofhumannasalairflowbyrhinomanometry.Rhinology.2011;49(1):2-10.

109

60. HilbergO,JacksonAC,SwiftDL,PedersenOF.Acousticrhinometry:evaluationofnasalcavitygeometrybyacousticreflection.JApplPhysiol(1985).1989;66(1):295-303.61. HanifJ,EcclesR,JawadSS.Useofaportablespirometerforstudiesonthenasalcycle.AmJRhinol.2001;15(5):303-6.62. CuddihyPJ,EcclesR.Theuseofnasalspirometryasanobjectivemeasureofnasalseptaldeviationandtheeffectivenessofseptalsurgery.ClinOtolaryngolAlliedSci.2003;28(4):325-30.63. RoblinDG,EcclesR.Normalrangefornasalpartitioningofairflowdeterminedbynasalspirometryin100healthysubjects.AmJRhinol.2003;17(4):179-83.64. OwensD,MooreM,CravenC,MagureanC,BackhouseS,WhittetH.Theaccuracyandreproducibilityofrhinospirometryindetectingflowasymmetryinanasalcavitymodel.EurArchOtorhinolaryngol.2011;268(10):1469-74.65. CuddihyPJ,EcclesR.Theuseofnasalspirometryfortheassessmentofunilateralnasalobstructionassociatedwithchangesinpostureinhealthysubjectsandsubjectswithupperrespiratorytractinfection.ClinOtolaryngolAlliedSci.2003;28(2):108-11.66. CocksGH.ImprovedGlatzelMirror.Laryngoscope.1915;25:135-41.67. dePochatVD,AlonsoN,MendesRR,GravinaPR,CronenbergEV,MenesesJV.AssessmentofnasalpatencyafterrhinoplastythroughtheGlatzelmirror.IntArchOtorhinolaryngol.2012;16(3):341-5.68. MeloDdeL,SantosRV,PeriloTV,BeckerHM,MottaAR.Mouthbreathingevaluation:useofGlatzelmirrorandpeaknasalinspiratoryflow.Codas.2013;25(3):236-41.69. CatundaIS,VasconcelosBC,CaubiAF,doAmaralMF,MorenoEF,MeloAR.Evaluationofchangesinnasalairwayinpatientshavingundergonesurgicallyassistedmaxillaryexpansion.JCraniofacSurg.2013;24(4):1336-40.70. DaneS,BalciN.Handedness,eyednessandnasalcycleinchildrenwithautism.IntJDevNeurosci.2007;25(4):223-6.71. Bojsen-MollerF,FahrenkrugJ.Nasalswell-bodiesandcyclicchangesintheairpassageoftheratandrabbitnose.JAnat.1971;110(Pt1):25-37.72. CrouseU,Laine-AlavaMT.Effectsofage,bodymassindex,andgenderonnasalairflowrateandpressures.Laryngoscope.1999;109(9):1503-8.73. Laine-AlavaMT,MinkkinenUK.Shouldahistoryofnasalsymptomsbeconsideredwhenestimatingnasalpatency?TheAngleorthodontist.1999;69(2):126-32.74. MirzaN,KrogerH,DotyRL.Influenceofageonthe'nasalcycle'.Laryngoscope.1997;107(1):62-6.75. WilliamsMR,EcclesR.Thenasalcycleandage.ActaOtolaryngol.2015;135(8):831-4.76. JonsonB,RundcrantzH.Postureandpressurewithintheinternaljugularvein.ActaOtolaryngol.1969;68(3):271-5.77. WangHW.Effectsofpostureonnasalresistance.JournalofMedicalSciences.2002;22:161-4.78. DaviesAM,EcclesR.Reciprocalchangesinnasalresistancetoairflowcausedbypressureappliedtotheaxilla.ActaOtolaryngol.1985;99(1-2):154-9.

110

79. HaightJS,ColeP.Unilateralnasalresistanceandasymmetricalbodypressure.JOtolaryngolSuppl.1986;16:1-31.80. MohanSM.Reflexreversalofnostrildominancebyapplicationofpressuretotheaxillabyacrutch.IndianJPhysiolPharmacol.1993;37(2):147-50.81. HudgelDW,RobertsonDW.Nasalresistanceduringwakefulnessandsleepinnormalman.ActaOtolaryngol.1984;98(1-2):130-5.82. LorinoAM,LorinoH,DahanE,d'OrthoMP,CosteA,HarfA,etal.Effectsofnasalprongsonnasalairflowresistance.Chest.2000;118(2):366-71.83. AtanasovAT,DimovPD.Nasalandsleepcycle--possiblesynchronizationduringnightsleep.MedHypotheses.2003;61(2):275-7.84. DallimoreNS,EcclesR.Changesinhumannasalresistanceassociatedwithexercise,hyperventilationandrebreathing.ActaOtolaryngol.1977;84(5-6):416-21.85. HasegawaM,KernEB.Theeffectofbreathholding,hyperventilation,andexerciseonnasalresistance.Rhinology.1978;16(4):243-9.86. YogeethaR,RamanR,QuekKF.Effectsoftemperaturechangesonnasalpatency.SingaporeMedJ.2007;48(4):304-6.87. LalD,GorgesML,UngkharaG,ReidyPM,CoreyJP.Physiologicalchangeinnasalpatencyinresponsetochangesinposture,temperature,andhumiditymeasuredbyacousticrhinometry.AmJRhinol.2006;20(5):456-62.88. EllegardEK.Pregnancyrhinitis.ImmunolAllergyClinNorthAm.2006;26(1):119-35,vii.89. OrbanN,MaughanE,BleachN.Pregnancy-inducedrhinitis.Rhinology.2013;51(2):111-9.90. ToppozadaH,MichaelsL,ToppozadaM,El-GhazzawiI,TalaatM,ElwanyS.Thehumanrespiratorynasalmucosainpregnancy.Anelectronmicroscopicandhistochemicalstudy.JLaryngolOtol.1982;96(7):613-26.91. BendeM,GredmarkT.Nasalstuffinessduringpregnancy.Laryngoscope.1999;109(7Pt1):1108-10.92. EllegardE,KarlssonG.Nasalcongestionduringthemenstrualcycle.ClinOtolaryngolAlliedSci.1994;19(5):400-3.93. HaeggstromA,OstbergB,StjernaP,GrafP,HallenH.Nasalmucosalswellingandreactivityduringamenstrualcycle.ORL;journalforoto-rhino-laryngologyanditsrelatedspecialties.2000;62(1):39-42.94. PhilpottCM,El-AlamiM,MurtyGE.Theeffectofthesteroidsexhormonesonthenasalairwayduringthenormalmenstrualcycle.ClinOtolaryngolAlliedSci.2004;29(2):138-42.95. ToppozadaH,MichaelsL,ToppozadaM,El-GhazzawiE,TalaatA,ElwanyS.Thehumannasalmucosainthemenstrualcycle.Ahistochemicalandelectronmicroscopicstudy.JLaryngolOtol.1981;95(12):1237-47.96. Navarrete-PalaciosE,HudsonR,Reyes-GuerreroG,Guevara-GuzmanR.Lowerolfactorythresholdduringtheovulatoryphaseofthemenstrualcycle.BiolPsychol.2003;63(3):269-79.97. TaylorM.Anexperimentalstudyoftheinfluenceoftheendocrinesystemonthenasalrespiratorymucosa.JLaryngolOtol.1961;75:972-7.98. ToppozadaH,ToppozadaM,El-GhazzawiI,ElwanyS.Thehumanrespiratorynasalmucosainfemalesusingcontraceptivepills.Anultramicroscopicandhistochemicalstudy.JLaryngolOtol.1984;98(1):43-51.

111

99. WolstenholmeCR,PhilpottCM,OlotoEJ,MurtyGE.Doestheuseofthecombinedoralcontraceptivepillcausechangesinthenasalphysiologyinyoungwomen?AmJRhinol.2006;20(2):238-40.100. RobinsonRW,WhiteDP,ZwillichCW.ModerateAlcoholIngestionIncreasesUpperAirway-ResistanceinNormalSubjects.AmRevRespirDis.1985;132(6):1238-41.101. EcclesR,TolleyNS.Theeffectofalcoholingestionuponnasalairwayresistance.Rhinology.1987;25(4):245-8.102. KjaergaardT,CvancarovaM,SteinsvaagSK.Smoker'snose:structuralandfunctionalcharacteristics.Laryngoscope.2010;120(7):1475-80.103. SchmidtBM,TimmerW,GeorgensAC,HiltM,MattingerC,WurstW,etal.Thenewtopicalsteroidciclesonideiseffectiveinthetreatmentofallergicrhinitis.Journalofclinicalpharmacology.1999;39(10):1062-9.104. MeltzerEO,CaballeroF,FromerLM,KrouseJH,ScaddingG.Treatmentofcongestioninupperrespiratorydiseases.Internationaljournalofgeneralmedicine.2010;3:69-91.105. PinargoteP,GuillenD,GuarderasJC.ACEinhibitors:upperrespiratorysymptoms.BMJCaseRep.2014;2014.106. ChengJW.Nebivolol:athird-generationbeta-blockerforhypertension.Clinicaltherapeutics.2009;31(3):447-62.107. SungYW,LeeMH,KimIJ,LimDW,RhaKS,ParkCI.Nasalcycleinpatientswithseptaldeviation:evaluationbyacousticrhinometry.AmJRhinol.2000;14(3):171-4.108. OzanE,DaneS,YildirimS,TatarA,TanismanS,YaziciAB,etal.Nasalcycleinschizophrenia:Leftnostrildominancemaybeassociatedwithcerebrallateralisationabnormalityandlefthemispheredysfunction.Neurology,PsychiatryandBrainResearch.2009;16(3):135-8.109. OzanE,YildirimS,TatarA,CanpolatS,YaziciAB,YukselS,etal.Sex-anddiagnosis-relateddifferenceinnostrildominancemaybeassociatedwithhemispheredysfunctioninaffectivedisorders.TurkJMedSci.2012;42(1):25-30.110. SunT,WalshCA.Molecularapproachestobrainasymmetryandhandedness.NatRevNeurosci.2006;7(8):655-62.111. SpringerSP,DeutschG.AHistoricalOverviewofClinicalEvidenceforBrainAsymmetries.LeftBrain,RightBrain.SanFrancisco:W.H.FreemanandCompany;1981.p.1-22.112. DaneS,YildirimS,OzanE,AydinN,OralE,UstaogluN,etal.Handedness,eyedness,andhand--eyecrosseddominanceinpatientswithschizophrenia:sex-relatedlateralisationabnormalities.Laterality.2009;14(1):55-65.113. SearlemanA,HornungDE,SteinE,BrzuszkiewiczL.Nostrildominance:differencesinnasalairflowandpreferredhandedness.Laterality.2005;10(2):111-20.114. BeatonAA.TheNatureandDeterminantsofHandedness.In:HugdahlK,DavidsonRJ,editors.TheAsymmetricalBrain.Cambridge,Massachusetts:TheMITPress;2003.p.105-58.115. BruderGE.FrontalandParietotemporalAsymmetriesinDepressiveDisorders:Behavioural,Electrophysiologic,andNeuroimagingFindings.In:HugdahlK,DavidsonRJ,editors.TheAsymmetricalBrain.Cambridge,Massachusetts:TheMITPress;2003.p.719-42.

112

116. SatzP,GreenMF.Atypicalhandednessinschizophrenia:somemethodologicalandtheoreticalissues.SchizophrBull.1999;25(1):63-78.117. McManusIC.Thehistoryandgeographyofhumanhandedness.In:SommerIECaK,R.S.,editor.LanguageLateralisationandPsychosis.Cambridge:CambridgeUniversityPress;2009.118. McManusIC,M.P.Bryden.Thegeneticsofhandedness,cerebraldominanceandlateralization.In:RapinI,S.J.Segalowitzeditor.HandbookofNeuropsychology.6:ElsevierSciencePublishers;1992.119. CorballisMC,Badzakova-TrajkovG,HaberlingIS.Righthand,leftbrain:geneticandevolutionarybasesofcerebralasymmetriesforlanguageandmanualaction.WileyinterdisciplinaryreviewsCognitivescience.2012;3(1):1-17.120. OldfieldRC.Theassessmentandanalysisofhandedness:theEdinburghinventory.Neuropsychologia.1971;9(1):97-113.121. AnnettM.Thedistributionofmanualasymmetry.BrJPsychol.1972;63(3):343-58.122. AnnettM.Aclassificationofhandpreferencebyassociationanalysis.BrJPsychol.1970;61(3):303-21.123. AnnettM.AModeloftheInheritanceofHandednessandCerebralDominance.Nature.1964;204:59-60.124. AnnettM.Thegeneticsofhandedness.TrendsinNeurosciences.1981;4:256-8.125. AnnettM.Therightshifttheoryofhandednessandbrainasymmetryinevolution,developmentandpsychopathology.Cognitie,Creier,Comportament(Cognition,Brain,Behaviour).2006;10:235-50.126. BholeMV,KarambelkarPV.Significanceofnostrilsinbreathing.Yoga-Mimamsa.1968;10(4):1-12.127. Shannahoff-KhalsaD.Lateralizedrhythmsofthecentralandautonomicnervoussystems.IntJPsychophysiol.1991;11(3):225-51.128. WerntzDA,BickfordRG,BloomFE,Shannahoff-KhalsaDS.Alternatingcerebralhemisphericactivityandthelateralizationofautonomicnervousfunction.HumNeurobiol.1983;2(1):39-43.129. BrownSG,RoyEA,RohrLE,SniderBR,BrydenPJ.Preferenceandperformancemeasuresofhandedness.BrainCogn.2004;55(2):283-5.130. AserinskyE,KleitmanN.RegularlyOccurringPeriodsofEyeMotility,andConcomitantPhenomena,duringSleep.Science.1953;118(3062):273-4.131. KleitmanN.Thebasicrest--activitycycleandphysiologicalcorrelatesofdreaming.ExpNeurol.1967:Suppl4:2-4.132. ManseauC,BroughtonRJ.BilaterallysynchronousultradianEEGrhythmsinawakeadulthumans.Psychophysiology.1984;21(3):265-73.133. KleinR,ArmitageR.Rhythmsinhumanperformance:11/2-houroscillationsincognitivestyle.Science.1979;204(4399):1326-8.134. BossomJ,NatelsonBH,LevinBE,StokesPE.Ultradianrhythmsincognitivefunctionsandtheirrelationshiptovisceralprocesses.PhysiolBehav.1983;31(1):119-23.135. NeubauerAC,FreudenthalerHH.UltradianRhythmsinCognitivePerformance-NoEvidencefora1.5-HRhythm.BiolPsychol.1995;40(3):281-98.

113

136. KleinR,PilonD,ProsserS,Shannahoff-KhalsaD.Nasalairflowasymmetriesandhumanperformance.BiolPsychol.1986;23(2):127-37.137. HeiserC,BajaJ,LenzF,SommerJU,HormannK,HerrRM,etal.Trigeminalinducedarousalsduringhumansleep.SleepBreath.2015;19(2):553-60.138. HeiserC,BajaJ,LenzF,SommerJU,HormannK,HerrRM,etal.EffectsofanArtificialSmokeonArousalsDuringHumanSleep.ChemosensoryPerception.2012;5(3):274-9.139. ServitZ,KristofM,StrejckovaA.ActivatingEffectofNasalandOralHyperventilationonEpilepticElectrographicPhenomena-ReflexMechanismsofNasalOrigin.Epilepsia.1981;22(3):321-9.140. ServitZ,KristofM,KolinovaM.ActivationofepilepticelectrographicphenomenainthehumanEEGbynasalairflow.PhysiolBohemoslov.1977;26(6):499-506.141. KristofM,ServitZ,ManasK.ActivatingEffectofNasalAir-FlowonEpilepticElectrographicAbnormalitiesintheHumanEeg-EvidencefortheReflexOriginofthePhenomenon.PhysiolBohemoslov.1981;30(1):73-7.142. WerntzDA,BickfordRG,Shannahoff-KhalsaD.Selectivehemisphericstimulationbyunilateralforcednostrilbreathing.HumNeurobiol.1987;6(3):165-71.143. BlockRA,ArnottDP,QuigleyB,LynchWC.Unilateralnostrilbreathinginfluenceslateralizedcognitiveperformance.BrainCogn.1989;9(2):181-90.144. DesaiR,TailorA,BhattT.Effectsofyogaonbrainwavesandstructuralactivation:Areview.ComplementTherClinPract.2015;21(2):112-8.145. TellesS,RaghurajP,MaharanaS,NagendraHR.Immediateeffectofthreeyogabreathingtechniquesonperformanceonaletter-cancellationtask.PerceptMotSkills.2007;104(3Pt2):1289-96.146. TellesS,JoshiM,SomvanshiP.Yogabreathingthroughaparticularnostrilisassociatedwithcontralateralevent-relatedpotentialchanges.IntJYoga.2012;5(2):102-7.147. NaveenKV,NagarathnaR,NagendraHR,TellesS.Yogabreathingthroughaparticularnostrilincreasesspatialmemoryscoreswithoutlateralizedeffects.PsycholRep.1997;81(2):555-61.148. TellesS,NaveenKV.Yogaforrehabilitation:anoverview.IndianJMedSci.1997;51(4):123-7.149. JoshiM,TellesS.Immediateeffectsofrightandleftnostrilbreathingonverbalandspatialscores.IndianJPhysiolPharmacol.2008;52(2):197-200.150. DaneS,CaliskanE,KarasenM,OztasanN.Effectsofunilateralnostrilbreathingonbloodpressureandheartrateinright-handedhealthysubjects.IntJNeurosci.2002;112(1):97-102.151. BhavananiAB,Madanmohan,SanjayZ.Immediateeffectofchandranadipranayama(leftunilateralforcednostrilbreathing)oncardiovascularparametersinhypertensivepatients.IntJYoga.2012;5(2):108-11.152. Shannahoff-KhalsaDS,KennedyB.Theeffectsofunilateralforcednostrilbreathingontheheart.IntJNeurosci.1993;73(1-2):47-60.153. ChenJC,BrownB,SchmidKL.Effectofunilateralforcednostrilbreathingontonicaccommodationandintraocularpressure.ClinAutonRes.2004;14(6):396-400.154. BackonJ,MatamorosN,TichoU.Changesinintraocularpressureinducedbydifferentialforcedunilateralnostrilbreathing,atechniquethataffectsbothbrain

114

hemisphericityandautonomicactivity.Apilotstudy.GraefesArchClinExpOphthalmol.1989;227(6):575-7.155. BackonJ,MatamorosN,RamirezM,SanchezRM,FerrerJ,BrownA,etal.Afunctionalvagotomyinducedbyunilateralforcedrightnostrilbreathingdecreasesintraocularpressureinopenandclosedangleglaucoma.BrJOphthalmol.1990;74(10):607-9.156. Shannahoff-KhalsaDS,BoyleMR,BuebelME.Theeffectsofunilateralforcednostrilbreathingoncognition.IntJNeurosci.1991;57(3-4):239-49.157. JellaSA,Shannahoff-KhalsaDS.Theeffectsofunilateralforcednostrilbreathingoncognitiveperformance.IntJNeurosci.1993;73(1-2):61-8.158. VelikonjaD,WeissDS,CorningWC.Therelationshipofcorticalactivationtoalternatingautonomicactivity.ElectroencephalogrClinNeurophysiol.1993;87(1):38-45.159. SchiffBB,RumpSA.Asymmetricalhemisphericactivationandemotion:theeffectsofunilateralforcednostrilbreathing.BrainCogn.1995;29(3):217-31.160. Shannahoff-KhalsaD,GolshanS.Nasalcycledominanceandhallucinationsinanadultschizophrenicfemale.PsychiatryRes.2015;226(1):289-94.161. MarshallRS,BasilakosA,WilliamsT,Love-MyersK.Exploringthebenefitsofunilateralnostrilbreathingpracticepost-stroke:attention,language,spatialabilities,depression,andanxiety.JAlternComplementMed.2014;20(3):185-94.162. VingerhoetsG,StroobantN.Lateralizationofcerebralbloodflowvelocitychangesduringcognitivetasks-AsimultaneousbilateraltranscranialDopplerstudy.Stroke.1999;30(10):2152-8.163. SchmidtP,KringsT,WillmesK,RoesslerF,ReulJ,ThronA.Determinationofcognitivehemisphericlateralizationby"functional"transcranialDopplercross-validatedbyfunctionalMRI.Stroke.1999;30(5):939-45.164. terLaanM,vanDijkJM,EltingJW,StaalMJ,AbsalomAR.Sympatheticregulationofcerebralbloodflowinhumans:areview.BrJAnaesth.2013;111(3):361-7.165. KangCK,OhST,ChungRK,LeeH,ParkCA,KimYB,etal.Effectofstellateganglionblockonthecerebrovascularsystem:magneticresonanceangiographystudy.Anesthesiology.2010;113(4):936-44.166. MoruzziG,MagounHW.BrainStemReticularFormationandActivationoftheEeg.ElectroenClinNeuro.1949;1(4):455-73.167. KobalG,VanTollerS,HummelT.Istheredirectionalsmelling?Experientia.1989;45(2):130-2.168. ArduiniA,MoruzziG.Olfactoryarousalreactionsinthecerveauisolecat.ElectroencephalogrClinNeurophysiol.1953;5(2):243-50.169. IyengarBKS.LightonPranayama.London:UnwinPaperbacks;1981.170. EcclesR.Roleofcoldreceptorsandmentholinthirst,thedrivetobreatheandarousal.Appetite.2000;34(1):29-35.171. McCroryP.Smellingsalts.BrJSportsMed.2006;40(8):659-60.172. VealeJF.EdinburghHandednessInventory-ShortForm:arevisedversionbasedonconfirmatoryfactoranalysis.Laterality.2014;19(2):164-77.173. BritishNationalFormlary(online)London:BMJGroupandPharmaceuticalPress;.Availablefrom:http://www.medicinescomplete.com.

115

174. PastWeatherinCardiff:TimeandDate.Availablefrom:http://www.timeanddate.com/weather/uk/cardiff/historic?month=5&year=2016.175. WilliamsM.Apilotstudyonthestabilityofthehumannasalcycle:CardiffUniversity;2015.176. IoannidisJP.Whymostpublishedresearchfindingsarefalse.PLoSmedicine.2005;2(8):e124.177. ColquhounD.Aninvestigationofthefalsediscoveryrateandthemisinterpretationofp-values.RoyalSocietyopenscience.2014;1(3):140216.178. SellkeT,BayarriMJ,BergerJO.Calibrationofpvaluesfortestingprecisenullhypotheses.TheAmericanStatistician.2001;55(1):62-71.

116

Appendix1:InclusionandExclusionCriteriaInclusioncriteria:

1. Aged18orover

2. Havegivenwritteninformedconsent

Exclusioncriteria:

1. Anyhistoryofchronicnasalconditions

2. Activenasaldiseasee.g.currentupperrespiratorytractinfection

3. Anyhistoryoftrauma(includingsurgery)tonose,sinusesorcentralnervoussystem

4. Anysignificantseptalabnormality

5. Anydiseaseormedical or surgical history that the investigator deemsmayaffect nasal physiology and influence the results of the study e.g. chronicrespiratorydiseaseor intakeofmedicinesknowntoaffectthenosesuchastopicalcorticosteroidsordecongestants

6. Knownallergytoaluminium

7. Memberofstudystafforpartnerorrelativeofstudystaff

8. Intakeofmore than 4 units of alcoholwithin 12hours ofmeasurement ofnasalairflow

9. Performance of vigorous exercise within 3 hours ofmeasurement of nasalairflow

10. Currentsmoker,definedbyadailyuseofanytobaccoproduct

11. Pregnantorlactatingwomen

12. Mixedorunclearhandedness

117

Appendix2:ParticipantInformationSheetThissectionhasbeenremovedbytheauthorforcopyrightreasons.

118

Appendix3:ConsentFormThissectionhasbeenremovedbytheauthorforcopyrightreasons.

119

Appendix 4: Edinburgh Handedness Inventory – Short Form. Adapted

fromVeale(2014)(172).

Please indicate your preferences in the use of hands in the following activities orobjects: Always

right

Usuallyright

Bothequally

Usuallyleft

Alwaysleft

Writing

Throwing

Toothbrush

Spoon

ScoringForeachitem:Alwaysright=100;Usuallyright=50;Bothequally=0;Usuallyleft=-50;Alwaysleft=-100TocalculatetheLateralityQuotientaddthescoresthendividebyfour:

Writingscore

Throwingscore

Toothbrushscore

Spoonscore

Total

Lateralquotientscore(total/4)

Classification LateralityQuotientScoreLefthanders -100to-61Mixedhanders -60to60Righthanders 61-100