Thorax196924527Chronicbronchitisincoalminersantemortempostmor - PDF document

Thorax(1969),24,527.Chronicbronchitisincoalminers:ante-mortem/post-mortemcomparisonsH.I.McKENZIE,M.GLICK,ANDK.G.OUTHREDFromtheJointCoalBoard,Sydney,AustraliaFrombronchialmeasurementsin136deceasedcoalminersacomparisonwithwell-documentedante-mortemfindingsshowsthatwithincreasingclinicalseverityofchronicbronchitisthereisincreasingnarrowingofintrapulmonaryairwaysduetowallthickeningattheexpenseofthelumen.WiththeaidoftheReidindex(gland/wallratio)andanewproposedindexwhichmoredirectlymeasuresairwayobstruction-thewallinternaltocartilage/lumenradiusratio-bothofwhichtendtobeindependentofbronchussizebutchangewithincreasingabnormality-itispossibletoquantitatechronicbronchitispathologicallyandtoobtainsatisfactorycorrelationswithante-mortemdata.Chronicbronchitismaytherebybeevaluatedindependentlyfrompneumoconiosisandemphysema,thusfacilitatingthestudyoftheirseparaterelationshipswithcigarettesmokingandwithotherpossibleaetiologicalfactors.Aspartofalong-termstudyofrespiratorydiseasesinpost-mortemlungspecimensfromcoalminerswerecentlybeganmeasuringbronchiandcalculatingratiosincludingtheReidindex(1960).Thepurposeofthispaperistopresentsomeofourfindings,coveringsofarsome136individualminers.Ourobjectiveshavebeen,first,totrytodeviseasystemofseveritygradingofchronicbronchitispostmortemwhichcouldbesatisfac-torilyrelatedtotheante-mortemstatus;secondly,usingsuchasystem,tostudytherelationshipsofthepost-mortempathologytosuchante-mortemfactorsascigarettesmokingandoccupation,andpossiblyotherconstitutionalorenvironmentalin-fluences.Thispapercoversonlythefirstobjective.Severalmethodshavebeendescribedforquanti-tatingchronicbronchitis,mostlyinrelationtothedegreeofhypersecretionormucousglandorgobletcellhyperplasianoted.Thepathologicalsignsofinfectionarenotreadilyquantitated.Ourowninterestismainlyconcernedwiththeobstruc-tivephaseofthisdisease,andwiththosemeasur-ableaspectsofthepathologywhichappeartoberelevanttothisphase.AfterexamininganumberofGoughsectionsfromthelungsofminersaffectedbyseverechronicbronchitisourinterestinwall/lumenmeasurementswasarousedbythegrosswallthickeningandlumennarrowingseenmacro-scopically,extendingthroughmostofthevisibleintrapulmonarytree.Anattempttoquantitatethischangewasundertaken.CLINICALASPECTSFortheindividuallung/heartspecimensobtainedatnecropsy,thereareusuallyavailabletherecordsoffullydocumentedmedicalexaminations(con-ductedattwo-yearlyintervalsduringtheminers'workinglives)oralternativelyoraswelltheequallyfullydocumentedfindingsofworkers'compensationmedicalboards.Theseexaminationsincludearespiratorysymptomquestionnaire;fullroutineclinicalexaminationwithemphasisontheheartandlungs;inmorerecentyearssometestsoflungairwayobstructionsuchasmeasurementofthevitalcapacity,peakexpiratoryflowrate(PEFR),forcedexpiratoryvolumeinonesecond(FEV1),maximummidexpiratoryflowrate(MMEFR)and/ormaximumbreathingcapacity(MBC);postero-anteriorandfrequentlyadditionalchestradiographs;frequentlyalsofluoroscopyandelectrocardiographicexamination;andtherecord-ingindetailoftheoccupationalhistoryfromschoolleavingonwards.Fromlongexperienceinperiodicallyclinicallyexaminingminers,manyof

whomhadchronicairwaysobstruction,wehavecometoviewthisconditionincoalminersasacomplexofpneumo-coniosis(evaluatedradiologically),emphysemaorover-inflation(evaluatedradiologically,fluoro-scopically,andclinically)andchronicbronchitiswithobstructedairways.Forspecificdiagnosisandgradingofchronicbronchitisantemortemwehaverequired(after527 H.1.McKenzie,M.Glick,andK.G.Oithredexcludingthetypicalasthmatics)thefollowingcriteria:I.Simplebronchitis:ahistoryofcoughwithsputumonmostdaysormorningsforatleastthreemonthsoftheyear.Thisisclassedasgrade1butifinfectionwithoutobstructiondominatesthepictureitisassignedtograde2.However,infectionwithoutsomedegreeofobstructionisuncommoninourminers.2.Chronicbronchitiswithairwaysobstruction:theabovesputumhistoryatleast,plusthepre-senceofwheezing.Thelattermaybepresentonlyonforcedbreathingoreffortoraftercoughing.Suchapatientwouldconstituteagrade3chronicbronchiticonourscale.Whenasubjecthasalltheaboveplusareduc-tionintheFEV1valuepercentpredictedforageandheightbelow7000(atwhichstagewheezingisusuallypresentatrest)hewouldbegraded4or5,dependingonthedegreeofreduction(e.g.,agrade5bronchiticwouldhaveFEV1valueslessthan5000predicted).Eachsubjectwasalsogradedforanumberofindividualsymptomsandphysicalandradiologicalsigns,manyreferabletoemphysemaratherthantobronchitis,toaidusintheclinicaldifferentia-tionofthesetwoconditions.1Thesesystemsofclassificationhavebeeninusesince1955,having,however,undergonesomerefinementsduringthisperiod.ItwasnecessaryforustousethesesimpleclinicalmethodsofgradingchronicbronchitisanddifferentiatingitfromemphysemabecauseitwasnotpossibletorefermostofoursubjectstoSydneyforthemoreintensivediscriminatoryphysiologicalinvestiga-tionsavailableattheteachingcentres.Comparisonstobepresentedinthematerialwhichfollowsincludeonlythosesubjectsinwhomtheseante-mortemassessmentscouldbebasedondetailedinformationfromexaminationsconduc-tedwithinfiveyearsofdeath,despitethefactthatthereweremanyadditionalcasesinwhichafirmpositivesymptomorsignpresentoveralongerperiodwouldhavemateriallyimprovedtheinten-sityofcorrelationifincluded.Onlythosesubjectsinwhomtheseverityofasymptomorsign,oritspresenceorabsence,couldbefirmlyascertainedfromtherecords,wereincluded.Amongthe136subjectstherewereonly10whohadaclearhistoryanddiagnosisofepisodicbronchialasthmarespondingeffectively,atleastforatime,tobronchodilatordrugs.Thepost-mortemfindingsinthesesubjectsdidnotfitthe'Codingdetailsavailablefromtheauthors.Whilstappreciatingtheproblemofobservervariationinrecordingthesedetails,mostoftheseobservationswererecordedbyoneoftheauthors.generalpatternshownhereunderandtheyhavebeenexcludedfromtheante/post-mortemcom-parisons.Theinclusionofafairlysubstantialproportionwhodiedfromsystemichypertensiveorcoronaryheartdiseaseorothernon-pulmonarycausedidnotappeartohaveaffectedthebronchialpatho-logyfoundinanyway,thoughitmayinsomeinstanceshaveimpairedthecorrelationsfoundwithindividualsymptoms,forexample,dyspnoea.Theagerangeatdeathofthe136subjectswas43to83years,withaskewdistribution;123(900%)wereover55years.Themeanwas654+S.D.826years.AllhadbeencoalminersinNewSouthWalesorinNewSouth

Walespluselse-wherethemeandurationofminingeimploymentbeing333+S.D.11-37years;80%hadhad25yearsormoreexperience.Ofthe113whosesmokinghabitswererecorded,101(890%)hadbeencigarettesmokers.Inmanyofthespecimenspneumoconiosiswaspresentinvaryingdegree.Thelungsweremostlyperfusedbyacetateformalinmixtureforprepara-tionofwhole-lungsectionsbytheGough-Went-worthtechnique,andemphysemawaspresentinmostspecimens.Theseconditionshavebeenstudiedseparately.Theirpresencedoesnotaffectthefindingstobepresentedinthispaper.POST-MORTEMMEASUREMENTSThefollowingmeasurementsweremadewheneverpossibleonthelungspecimens,withtheaidofaneyepiecegraticuleinabinoculardissectingmicro-scope,histologicalslidesbeingexaminedusuallyatfrom5to40magnifications(examplesofthesemeasurementsareshowndiagrammaticallyinFigsIAandB):1.Thicknessofmucousglandswithintheinternalsurfaceofthecartilages,asrecommendedbyLynneReid(1960)2.Wall,thicknessfromtheepithelialbasementmem-branetotheinternalsurfaceofthecartilagesandperpendiculartheretoatthesamepointastheglandmeasurement(WIC-1).2TheReidindex(gland-wallratio)wascalculatedfromthesetwomeasurements3.Wherepossible,totalbronchialwallthickness(TW).Therewassomelossofaccuracyinthemeasurementowingtolackofcleardefinitionoftheouterbronchialwall,butitwasmadeeasierinmanyofourspecimensbythepresenceofdustaccumula-tionalongtheoutermargin.2Intheremainderofthetextandgraphs,whereaWICmeasure-mentisquotedinrelationtoglandthicknessitreferstotheWIC-1asdefinedabove;ifusedinrelationtoalumenorwailmeasurementitreferstoWIC-2definedabove.Thisexplanationisnecessarybecause,unlikeThurlbeckandAngus(1964),wemadesomeofourgland/wallmeasurementsneartheendsofthecarti-lagesaspermittedbyLynneReid(FigsLAandB).HadwetakenthemonlyatthecentresofthecartilagesthetwocategoriesofWICwouldhavebeenpracticallyidentical.528 Chronicbronchitisincoalminers:ante-mortem/post-mortemcomparisonsABFIG.1.Diagrammaticrepresentationof(A)centralandlateralmeasurementsofmucousglandthicknessandbronchialwallthicknessinternaltocartilage(WIC-1),utilizedJbrcalculationoftheReidindex;and(B)measurementsofbron-chialwallinternaltocartilageinlinewiththecentreofthebronchiallumen(WIC-2),asutilizedforcalculationoftheWICILratio.Thesemeasurementsweretakenatornearthecentresofthecartilaginousplates.4.Theradiusofthebronchiallumenmeasuredtotheinneredgeofthebasementmembranethroughanassumedgeometriccentre,andtheradiusinthesamelinetotheinnerwallofthecartilage.Thedifferencebetweenthesewasameasureofthethicknessofthewallintternaltothecartilage(WIC-2).2Variationsintheamountofepithelialdamagenecessitatedourstandardizingonthislumenmeasurementtothebase-mentmembrane.Thesemeasurementsweremadeinanumberofdirectionsacrossthosebronchiwherethefullcircum-ferencewassectioned,andtheresultswereaveraged.Thenumberofbronchiexaminedpersubjectwasdeterminedbythematerialavailable;itvariedfromonetofourandaveragedjustundertwo.Inmorerecenttimeswehaveprogrammedoursectioningtoenableatleastthreetofourbronchitobeexaminedperspecimen,takenfrombothlungs.Theresultswerecalculatedtoactualmeasurementinmillimetres.Wherenecessarytheywereaveragedfortheparticularindividualforcomparison

withante-mortemfindings.Thefollowingratioswerecalculatedtotwodecimalpoints:Gland/WallInternaltoCartilage(G/W)(Reidindex)DiameterofLumen/TotalBronchusDiameterInternaltoCartilage(LD/DIC)WallInternaltoCartilage/LumenRadius(WIC/L)TotalWall/LumenRadius(TW/L)WallInternaltoCartilage/TotalWall(WIC/TW)Afterclosestudythetworatiosrequiringmeasure-mentofthetotalwallthickness(WIC/TWandTW/L)werediscardedforthereasonsthat(a)thetotalwallthicknessasaruleisnotasaccuratelymeasure-ableasisthethicknesstotheinnerwallofthe2Seefootnoteonpreviouspage.cartilage,and(b)inourhandstheseratiosdidnotproveasdiscriminatorynorcorrelateaswellwithindividualsymptomsandsignsorwithbronchitisgradeasdidtheG/WandWIC/Lratios.TheLD/DICwaslikewisetooinsensitiveadequatelytoseparategradesofseverityofdiseaseandwasalsodiscarded.Thefollowingmaterialpresentsourfindingsinrelationtoabsolutevaluesofcertainmeasurementsandtothetworemainingratios,theReidindexandtheWIC/Lratio.Some241bronchifrom136individualsweremeasured.Theaveragedfindingsinabsolutevaluesandineitherratiowerecomparedwiththesizeofbronchiandtheratiosalsowiththeante-mortemfindings.Almostallthebronchimeasuredwereintra-pulmonary;themeanexternaldiameterwas5-42mm.(±S.D.1.94)andthemeanlumendiameter2-44mm.(±S.D.078).ABSOLUTEVALUESInFig.2thetotalwallthick-nessofeachbronchusinmillimetreshasbeenplottedagainstitsapproximateoverallsize(cal-culatedbysumminglumenradiusandtotalwallthicknessanddoubling).Itwillbeseenthatthereisalinearrelationshipbetweenthem,withahighorderofcorrelation;inotherwords,thetotalwallthicknesshasaconstantratiotobronchussizewithinthelimitsofthelatterstudied,namely2mm.to10mm.overalldiameter.Thiscoveredarangefromsmallthroughsubsegmentalandsegmentaltolobarbronchi.Figure2includesbothnormalandabnormalbronchi;nevertheless,ahighdegreeofcorrelationisstillpresent,despitethefactthatthewallthick-529 H.1.McKenzie,M.Glick,andK.G.OuthredxSeverebronchiticsoNil+simplebronchitics*Intermediatecases's*e,*000x*0X'S0Xexx**wxX0i000000'1Oi23456789101lOverallbronchussize(mm.)FIG.2.Theratioofbronchialwallthicknesstobronchussize(andthustolumen)remainsconstantwithvaryingsizeofintrapulmonarybronchusbutincreasesinobstructivebronchialdisease.Thisincreaselikewiseappearstobelargelyindependentofsizeofbronchus.(Measurementsfromsubjectswhohadseverebronchialairwayobstructionantemortemarecomparedwiththosewithnilobstructionandwiththosewithintermediatedegreesofobstruction.)7EE4IE4)EJ64-32-nI234578910Totalbronchusdiameterinternalcartilaqe(mm.)FIG.3.Changeinratiooflumendiametertodiameterofbronchusinternaltocartilaginousplatesinsevereobstruc-tive'bronchial'diseaseascomparedwith'nilobstructives'(subjectswhohadnochronicbronchitisorwerechronicexpectoratorswithoutclinicalairwaysobstruction).nessincreasesinchronicbronchitis.Tworegres-sionlineshavebeensuperimposedonthisfigure.Oneshowsthesituationwithrespecttothosesub-jectswhohadnosymptomsnorsignsofobstruc-tivebronchitiswhenlastexamined('nil'plus'simple'chronicbronchitis)andtheother,thosewhohadthemostseveregradeofobstructivebronchitis(grade5).Theshiftinthesevereob-structivestowardsathickerbronchialwallforagivensizeofbronc

husisevident.Therewereinsufficient'nil'casestocomparetheseseparately;nodoubthadwebeenabletodosotheshiftwouldhavebeensomewhatgreater.Fromvisualinspectionthethickeningtakesplacelargely,ifnotwholly,attheexpenseofthelumen.Toobtainabsoluteproofofthiswouldrequiresharpstandardizationofbronchuslevelasbetweennormalsandabnormals-somethingverydifficulttoachieve.Inasmuchaswefoundnoshiftintherangeofbronchussizesbetweenthenormalandabnormalextremes,ourfindingspro-videconfirmationofthisconceptofinternalwallthickeningandresultantrelativelumennarrow-ing.Figure3showsthatforagivensizeofbronchuswefoundlumennarrowinginabnormalsascomparedwithnon-obstructives-againwithnocorrespondingsystematicshiftinthelimitsofoverallbronchussize.Inourexperience,thesechanges,thoughwide-spread,arenotentirelyuniformthroughoutthebronchialtreeinaparticularindividual.Inad-vancedcasestheytendtobemoregeneralized.Figure4suggeststhatthecartilaginousframe-workneitherexpandsnorcontractsinbronchitiswithairwaysobstruction.TableIpresentssomeofthemeanabsolutevaluesandtheirstandarddeviationsobtainedinthederivationoftheregressionlinesshowninFigures2to4.3.5,3*0-EE-0_c0-o-nF=3:2520I5100*5-530Ili Chronicbronchitisincoalminers:ante-mortem/post-mortemcomparisonsb~~~~~~~~~~b5E~~.-4-'03.u0oI.023456789I0Bronchussize(mm.)FIG.4.Lackofchangewithante-mortemseverityofbronchialobstructioninthediameterofintrapulmonarybronchiasmeasuredfromtheinnersurfacesofthecarti-laginousplates.Thisdiameterbearsaconstantrelationshiptototalbronchusdiameterandismoreaccuratelymeasur-ableintheabsenceofpneumoconiosis.TABLEIABSOLUTEVALUESOFBRONCHIALMEASUREMENTSINNORMALORNEAR-NORMALSUBJECTSANDINCHRONICBRONCHITICSWITHSEVEREOBSTRUCTIONNilChronicplusNo.ofBronchitisNo.ofSimpleBronchiwithBronchiChronicMeasuredSevereMeasuredBronchitisObstructionGlandthickness(mm.)..0-24±0-12450-54±0-2725Bronchusoveralldiameter(mm.)5-68i2-1745570±1*7325Lumenradius(mm.)..1-41±0-60501-10±0-5232Wallinternaltocartilage(mm.)(WIC-2)..0-52±0-26500-95±03932Wallinternaltocartilage(mm.)(WIC-2)..052±026500-95±03932Totalwallthick-ness(mm.)..1-33±064501-68±0-6832Thefollowingpointsareofinterest:1.Themeanglandthicknessintheseverebronchitisgroup;isslightlymorethandoublethethicknessinthenilplussimplebronchitisgroup,thisincreaseoccurringdespitethefactthatthemeanoverallbronchusdiameterisalmostthesameinbothgroups.Theglandthicknessincreasesbyanaverageof030mm.2.Thelumenradiusdiminishesby0-31mm.inthechronicbronchiticswithsevereobstruction.Alsothetotalwallenlargesby035mm.inthesamegroup.ItwillbenotedthatthemeanWIC-2increasesbyalargeramount-043mm.Thisisexplainedbythefactthatnotallsuchmeasure-2Qmentsweremadeatthecentresofthecartilag-inousplates(seeFig.1B).3.Theamountofshrinkageofthelumenofabronchiticatourseverestgradeofobstruction(grade5)issufficienttoreduceitscross-sectionalareato50%ofitsnormalvalue.Thismaywellbeconsideredrelevanttothedegreeofobstructionpresentclinicallyinsevereairwaysobstruction,asmeasuredbysuchtestsastheMBC,FEV1,MMEFR,andPEFR.RATIOSTheabsolutevaluesvarywiththesizeofthebronchusanditwasbecauseofthisthat,likeReid,Thurlbeck,andothers,wedecidedtouseratioswhicharerela

tivelyindependentofbron-chussizeoverawiderangeyetchangesufficientlywiththedegreeofchronicbronchitistopermittheiruseinseveritygradingssuitaibleforepidemi-ologicalwork.Thisobviatestheneedforprecisespecificationofalocationinthebronchialtreeatwhichsectionsmustbecut.Wewerelookingforaratiowhichmightbeused,togetherwiththeReidindex,forassessmentofchronicbronchitisinpathologicalspecimensbutwhichmightpermitbetterdefinitionoftheobstructivestagesthantheReidindex.TheWIC/Lratioappearedprimafacietobelikelytomeetourrequirements.04.20cE30-90-8070-6050-4030-201***00..'7L=**;WIC,0L-0424-000542Br*-**.0.000*00O2345678910Overallbronchussize(mm.)FIG.5.Constancyofwallthicknessinternaltocartilagellumenradiusratiowithvaryingbronchussizein'non-obstructive'subjects.Thelumenradiusisnormallyabouttwoandahalftimestheintracartilaginouswallthickness.Instudyingratiostobeginwithweplottedeachratioinnormalornearnormalbronchi(non-obstructedsubjects)againsttheoverallsizeofthebronchus.TheresultsfortheWIC/LratioaregiveninFigure5.Onlybronchifromsubjectswhohadnomorethansimplechronicbronchitiswhenlastexaminedwereincludedinthesediagrams.Figure5confirmsthatinthenormalornearnormalbronchusthisratio,liketheReidindex,531. H.1.McKenzie,M.Glick,andK.G.Outhred(II64/05210*I-85/O-63)i615-1-41-31-2**1-0-~~~0-8*07-0-7---0-6-*S0-4-*-.03-**0-3.02~~~~.0-200-10-20-30-4050-607Gland/wallratioFIG.6.CorrelationbetweenWIC/LratioandReidindex(r=065)r=0-6568091l0TABLEIICORRELATIONWITHANTE-MORTEMDISEASEGRADINGSChronicChronicNilChronicSimpleChronicChronicBronchitisBronchitisBronchitisBronchitis'Bronchitis'MildlyObstructive'ModeratelySeverelyObstructive'Obstructive'No.ofsubjects527171513MeanFEVI%ofpredicted'..104%84%i66'55,39'/(3)(19)(14)(13)(13)No.ofbronchimeasured...644362928Meangland/wallratio...0-28±0-063034±0100-41±0-090-51±0-110-55±009(0-024)(0-016)(0-015)(002043)(0-0170)Meanratiowallinternaltocartilage!0-36±0-16040±0-180-63±0-120-88±0-211-04±0-26lumenradius(0-0653)(0-027)(0-020)(0-0390)(0-0491)Withorwithoutemphysemaand/orpneumoconiosis(asthmaticsexcluded).2No.ofcasesinparenthesesindicatethoseinwhichrespiratoryfunctiontestresultsareonrecord.N.B.(1)Mean±standarddeviationineachinstance.(2)S.E.meanshowninparentheses.tendstoremainconstantornearlysowiththesizeofthebronchus,thoughthereisconsiderablevariabilityin-individualbronchi.InFig.6theWIC/LratioisplottedagainsttheG/Wratio.Eachpointrepresentsthemeanforanindividual,andallpersonsnormalandabnormalintermsofbronchitisareincluded.Inthisdia-gramarelationshipisdemonstratedwhichappearstobelinear,thoughwithfairlywideandratherirregularspreadofindividualpoints.Afairdegreeofcorrelationisdemonstrated(r=0-65).Ingeneral,ourfindingswiththeReidindexforvariousgradesofdiseasewereidenticalwiththosepublishedonvariousoccasionsbyLynneReid,andthedistributionofthisindexwassimilartothatgivenbyThurlbeckandAngus(1964).ConsideringtheoverallpictureshowninFigs2to6,theindicationsare,first,thatbronchiwith-intherangeofapproximately2mm.to10mm.overallsizecanbesatisfactorilyusedforassessingchronicbronchitis,and,secondly,thatthesizeisnotimportantwithinthisrangeandusingratiosratherthanabsolutem

easurements.Wedonotrecommendusingbronchismallerthanabout25mm.fortworeasons;firstthatwefindthediscr-iminationbetweenobstructivesandnon-obstructivesusinganyoftheratiosisratherlesssatisfactoryintheseverysmallbronchi,andsecondlythatinsuchbronchiitisoftenimpossibletofindasatisfactorymucousglandformeasure-ment.Aboveabout7-5mm.size,thevariabilityandoverlapbetweennormalsandabnormals5320acE0cr0cr0u0-4-0vcx3:.*..0 Chronicbronchitisincoalminers:ante-mortem/post-mortemcomparisons25NilSimpleMildobst.Noof64436casesGradeofchronicbronchitisFIG.7.Meanvaluesandthe95%confiaWICILratioandtheReidindexforeachante-mortemseverityofchronicbronbronchitis'to'severeobstructivebronchiaTABLEIIISYMPTOMSANDSIGNS(WITHINFIVEYJv.BRONCHITISRATIOS.CORRELATIO]RatioGWIGRatio}wLFEV,%predicted....07430745Sputumhistory"04500-549Dyspnoea..0-2980-483Historyofwheezing'0-4910-573Wheezes..05920-638Observedbreathlessness..04120434Prolongedexpiration..0-4630569Percussionhyperresonance..03770-355Distantbreathsounds..01110395Distantheartsounds..0-2490-316RadiographicfindingsDiminishedlungmarkings0-2440285Lowflatdiaphragm03350-471Smallcentralheartshadow0-1460-268Largeretro-stemalwindow0-3950-324EnlargedA.P.thoracicdiameter....05190-241Chestexpansion0....-0183-0-132Diaphragmaticexcursion..-0353-0075lSymptomsrecordedbystandardquesseemstoincreasesothatforbestandaccuracyingradingwereconpresentrestrictingmeasurementstosizerange2-5to7-5mm.RELATIONSHIPWITHANTE-MORTEMSYMPTOMS,ANDPHYSICALF]TableIIgivesthemeanvalueinteachoffivegradesofchronicbronchitisadjudgedpresentantemortem,togetherwithotherrelevantWIC/LdataincludingthemeanFEV1%foreachgrade.Figure7showsthesamemeanratiosandtheirconfidencelimits.TableIIandFig.7discloseachangingmagni-tudeofeachratiowithincreasingseverityofchronicbronchitis.Thereis,however,considerable..-.....overlapofindividualsubjectvaluesbetween"'G/Wgradesasindicatedbythemagnitudeofthevariancesshowninthetable.Someofthisvari-ancemayhavearisenfromthenatureofthebronchitisclassification,frominaccuracyintheante-mortemassessments,orfromacceptingmedicalexaminationsconducteduptofiveyearsMod.Severebeforedeathasthebasisforsuchassessment.obst.obstSomeisinherentinthescientificmethodused-2928forexample,ifwehadrestrictedallwall/lumenmeasurementstothosefromthecentresofthelencelimitsofthecartilaginousplates(ThurlbeckandAngus,1964),offivegradesoftheoverlapmightwellhavebeenreduced.chitis-from'nilLookingatFig.7itwillbenotedthattheIldisease'.WIC/LratiocombineswideseparationofgradeswithsmallerrelativevarianceswhencomparedEARSOFDEATH)withtheG/Wratio.ThisdiscriminatingpowerNCOEFFICIENTSmakesitattractiveasabasisforthegradingofNo.ofSubjectsseverityoftheobstructedbronchitic.Inboth-ratiosthemeanvaluesinthemoderatelyandGw-cseverelyobstructedpatientsaresignificantlydiffer-5452entbyStudent'st-testfromthoseinthemenwho665452hadnilorsimplechronicbronchitis(P)7366TableIIIpresentsdetailsofthelinearcorrela-6760tioncoefficientsfoundbetweenthesetworatios7466andanumberofsymptomsandphysicaland6358rdooiah6459radiologicalsigns,includingtheFEV1.6862Thistableshows,asexpected,thattheG/W6963andtheWIC/Lratioscorrelatewellwitht

he7174FEV1andthesetofsymptomsandsigns(upper6660halfoftable)whichtendtobeassociatedwith6861bronchitiswithobstructionratherthanemphy-65694842sema.Ofthetwo,theWIC/Lappearstohavea4539slightedgeontheG/Wratiointhedetermination_6357ofobstructivedisease.Inpassing,thesamesetof6053symptomsandphysicalandradiologicalsignsti_nnaire.-havebeentabulatedagainsttheamountofemphy-,tionnaire.semaoverall(disregardingtype)asgradedinthediscriminationpost-mortemspecimens,andinthisinstancethenmendforthebestcorrelationswereobtained,aswouldbeex-bronchiinthepected,withthesignsinthelowerhalfofthetable(TableIV)andwithdyspnoeaandobservedbreathlessnessratherthanwithwheezing.IDIAGNOSIS,InderivingthecorrelationcoefficientslistedinINDINGSTablesIIIandIVwehaveassumedthattherealassociationswouldbelinearinform.If,asisprob-)othratiosforable,forexample,inthatbetweenquantityof533 H.I.McKenzie,M.Glick,andK.G.OuthredTABLEIVSYMPTOMSANDPHYSICALANDRADIOLOGICALSIGNSWITHINFIVEYEARSOFDEATHv.EMPHYSEMAGRADE(POST-MORTEMPATHOLOGY)Chronicsputumproduction'DyspnoealRegularwheezing'ObservedbreathlessnessExpirationprolongedWheezesCyanosisChestexpansionInspiratorypositionofthoraxVisibleepigastricpulsationApicalshiftPercussionhyperresonanceDistantbreathsoundsDistantheartsoundsRadiographicfindingsDiminishedlungmarkingsLowflatdiaphragmSmallcentralheartIncreasedA.P.diameterofthoraxEnlargedretrosternalwindowDiaphragmaticmovementFEV,(%predicted).No.ofSubjects13513412111912513813511912790120127131134123124119838710191CorrelationCoefficient0-240570300700-420330-32-0-250520580-270-600540440580590470*550650-520*50ISymptomsrecordedbystandardquestionnaire.emphysemaandtheFEV1theassociationisnon-linear,ourresultswouldunderestimatethetruedegreeofcorrelation.Forthosepersonsinterestedthedetailsofthesystemofgradingofindividualsymptomsandphysicalsignsareavailablefromtheauthors.InFig.8wehaveattemptedtodemonstrate,diagrammaticallytheapparentrelationshipbetweentheWIC/LandG/Wratiosandtheante-mortemstagesofdiseaseasassessedbyus,includ-ingsomeofthesymptomsandsignswhicharelikelytobepresentatthesestages.Thisfiguredemonstratesbroadlythetheoreticalprogressionofchronicbronchitisantemortem(thiscommonlyoccursverygraduallyovertheageperiod25to55years)andexpectedpathologyateachstageintermsoftheaboveratios.Thefigurealsosuggestsabasisforapathologicalaswellasaclinicalclassificationofseverityofchronicbronchitis,bothofwhichcouldbeusefulinepidemiologicalstudies.DISCUSSIONTheuseofratiosthatnormallyareindependentofbronchialsizeoveraconsiderablerange(butarealteredinchronicbronchitis)wasadoptedbyLynneReid(1960,1968)andThurlbeckandAngus(1964)toassessthepathologyofchronicbron-chitics.Comparedwithusingactualvaluesofthemeasurementsthishastheadvantagethat,withincertainwidelimits,precisecontrolofbronchussizefortheselectionofblocksforsectioningisAWIC/L-03?9Nil/SimplebronchitisMorningsputumr.2mlornilff'USputummorninganddurinqdayRFT-1norma,ranqeUkfBWC/!.'bMildobstrdctonSputumuptolOin.naiyWheezemainlyinmor-ninqWheezesheardorioced!0breethlngqR.FTtowards;owerraitnormcaaOD.aot50°/fb-%/predrc'ec.LAC/u-LVMod.obstructionSputumlOmrl.ormor-eEarlydyspnoec-notg

oodonhills,normcaorilc'0atownpaceReqularwheezeandwheezesheardatres;RFT45%0/.0%predicted'DSevereobstruction/SputumCOp!Ous.5'CouqhdisturbssieepXDyspnoeaadvancedspell50yardsonflatWheezesheardbypatientanddoctorwithoutastethoscopeDyspnoeaatrestand/ordisrobirgqV:ides-readwheezeshecrdwithastethoscoDeRFT40%/opredictedorlessFIGS8A,B,C,D.Diagrammaticrepresentationoffourstagesofseverityofintrapulmonaryobstructivebronchialdisease.Shownareselectedrepresentativeante-mortemfindingstogetherwiththeexpectedmeanWICILandGIWratiosasmeasuredinpost-mortemlungspecimens.notanecessaryprerequisite.Ofcourse,referencetocellularandotheradditionalcomponentsisnecessaryforfullpathologicaldiagnosis,butcer-tainparameters,suchaswallthicknesslumenandglandsizethatcanbequantitated,lendthemselvesparticularlywelltothestudyofepidemiologicalaspects.Ourfindingswiththegland/wallratioagreecloselywiththoseofReidandThurlbeckand534 Chronicbronchitisincoalminers:ante-mortem/post-mortemcomparisonsAngus.Inthispaperwehavepresentedtheresultsofsomeworkwithanotherratio,thewallinternaltocartilage/lumenradiusratio,aswellastheReidindex.Itispossiblebythecombineduseoftheseratiostomakeafairlygoodassessmentoftheseverityofchronicbronchitisinthepathologicalspecimen,andtoadoptasystemofgradingofseverity.Itispossibletomakeanassessmentwhichrelatesnotonlytotherecordedamountofsputumproductionduringlifebutalsotothedegreeofairwaysobstruction.Wearecontinuingtousetheseratiosforstudyingtherelationshipsbetweenobstructivebronchialdiseaseandcigarettesmokingandoccupationalgroupingsinourminers.Thesestudieshaveconfirmedourearlierim-pressionthatchronicobstructionisnotsolelyaclinicalmanifestationofemphysemaand/orbronchiolarpathology.Wefindthatthedegreeofnarrowingofintrapulmonarybronchialairwaysgenerallyisquitecloselyrelatedtotheamountofclinicalobstructionantemortem.Chronicbron-chitispathologicallyisnotjusthypersecretionandhypertrophyofglands,withorwithoutcellularevidenceofinfectionorallergy.Itisalsoapro-gressivethickeningofthebronchialwallinwardswithconsequentnarrowingofairwaysand,afteracertainthresholdvalueofthisnarrowinghascomealbout,thedevelopmentofclinicalairwaysobstruction.Thethickeningappearstobegeneral-mucosalandsubmucosal-andnotconfinedtotheregionsofthemucousglands.Astoitsnature,ithasbeensuggested(Reid,1968)thatchronichyperaemiaandoedemaduetohypoxiaorinfec-tionmaybethemechanisminvolved.Similaritiesbetweennasalandbronchialmucosaemaybeofrelevancehere.ThispaperispublishedwiththepermissionoftheJointCoalBoard.WearegratefultoProfessorJohnRead,DepartmentofMedicine,UniversityofSydney;ProfessorBryanGandevia,DepartmentofMedicine,UniversityofNewSouthWales;ProfessorJethroGough,WelshNationalSchoolofMedicine,Cardiff;Dr.CharlesFletcher,C.B.E.,RoyalPostgraduateMedicalSchool,London;andProfessorLynneReid,InstituteofDiseasesoftheChest,BromptonHospital,London,fortheirhelpfulcriticismandadvice.REFERENCESReid,Lynne(1960).Measurementofthebronchialmucousglandlayer:Adiagnosticyardstickinchronicbronchitis.Thorax,15,132.(1968).Personalcommunication.Thurlbeck,W.M.,andAngus,G.E.(1964).Adistributioncurveforchronicbronchitis.Thorax,19,436.535