剖宫产分类

ContentslistsavailableatSciVerseScienceDirectBestPractice&ResearchClinicalObstetricsandGynaecologyjournalhomepage:www.elsevier.com/locate/bpobgyn11MethodsofachievingandmaintaininganappropriatecaesareansectionrateMichaelRobson,MBBS,MRCOG,FRCPI,FRCS(Eng),Dr*,LuciaHartigan,MBBchBAO,Dr,MartinaMurphy,RM,SeniorMidwifeNationalMaternityHospital,HollesStreet,Dublin2,IrelandKeywords:caesareansectionratesMultidisciplinaryQualityAssuranceProgramme10groupclassificationCaesareansectionratescontinuetoincreaseworldwide.Theappropriatecaesareansectionrateremainsatopicofdebateamongwomenandprofessionals.Evidence-basedmedicinehasnotprovidedanansweranddependsoninterpretationoftheliterature.Overallcaesareansectionratesareunhelpful,andcaesareansectionratesshouldnotbejudgedinisolationfromotheroutcomesandepidemiologicalcharacteristics.Betterunderstandingofcaesareansectionrates,theirconsequencesandtheirbenefitswillimprovecare,andenablelearningbetweendeliveryunitsnationallyandinternationally.ToachieveandmaintainanappropriatecaesareansectionraterequiresaMulti-disciplinaryQualityAssuranceProgrammeineachdeliveryunit,recognisingcaesareansectionratesasoneofmanyfactorsthatdeterminequality.Womenwillalwayschoosethetypeofdeliverythatseemssafesttothemandtheirbabies.Professionalsneedtomonitorthequalityoftheirpracticecontinuouslyinastandardisedwaytoensurethatwomencanmaketherightchoice.Ó2012ElsevierLtd.Allrightsreserved.IntroductionCaesareansectionscontinuetoincreaseworldwide.Noagreementhasbeenreachedonanappro-priatecaesareansectionrate,1–4andviewsaremixedonwhethertoomanyarebeingcarriedout.5–9Manywomenenquireaboutcaesareansectionasanoptionfordelivery,andasignificantnumberrequestacaesareansection.Mostwomendonotwantanoperation,theyrequestacaesareansectionbecausetheydonotwanttolabouranddelivervaginally.Nulliparouswomenrequestacaesarean*Correspondingauthor.Tel.:þ35316373100.E-mailaddress:MRobson@nmh.ie(M.Robson).1521-6934/$–seefrontmatterÓ2012ElsevierLtd.Allrightsreserved.http://dx.doi.org/10.1016/j.bpobgyn.2012.09.004298M.Robsonetal./BestPractice&ResearchClinicalObstetricsandGynaecology27(2013)297–308

sectionbecausetheyareworriedaboutsomethingthatmayhappen.Multiparouswomenrequestacaesareansectionbecauseofsomethingthatdidhappen.Nationalguidelineshavereinforcedtherightofwomentodecidethemodeoftheirdeliveryprovidedthattheyhavebeencounselledappropriately.Ifanobstetriciandisagreeswiththewoman’sdecisiontodeliverbycaesareansection,thensheshouldbereferredtoanobstetricianwhowouldbepreparedtocarryoutthecaesareansection.10Againstthisbackgroundtherealsoseemstobealowerthresholdofcarryingoutcaesareansectionsformedicalreasons.Thesereasonsinclude(1)differentviewsonthemanagementoflabouranddelivery,organisationalissues;and(2)societalintoleranceofpooroutcomesandexperience,andacultureofblamingindividualsorsystems(asignificantconcernamongprofessionals).3,11Caesareansectionsarethoughttobeaprocedurethatprotectsbothbabiesandmothersfromadverseevents.Althoughinsomecasesthisisundoubtedlytrue,itneedstobecontinuallyjustifiedandsafelyimple-mented.Extremeviewsonloworhighratesofcaesareansectionarenothelpful,especiallyiftheargumentsarebasedonselectedevidence.Anindifferentviewoncaesareansectionsisnothelpfulasallprocedureshavetheirimplications.Untilwomenandprofessionalsalikeappreciatethis,apolariseddebatewillcontinuethatwilldolittlemorethanconfuse.Inordertorationalisedecision-making,moreusefulinformationisneededonacontinuousandtimelybasisaboutthequalityofcarethatisbeingprovidedineachdeliveryunit.Obstetriciansandmidwivesmayknowlessabouteventsandoutcomesintheirownunitcomparedwiththeirknowledgeofpublishedresearch.Professionalshavearespon-sibilitytopracticeevidence-basedmedicinebuttheyshouldnotforgettheirresponsibilitytocollecttheevidencetoensurethattheyareprovidinggoodqualitycaretotheirpatients.Caesareansectionrateshavebeenatthecentreofthedebatefortoolong.Discussionsaboutreducingcaesareansectionrateswithouttakingotherfactorsintoaccountareatbestinappropriateandatworstdangerous.TheaimshouldbeaMultidisciplinaryQualityAssuranceProgramme(MDQAP)withwomen,babiesandtheirfamilyatthecentre.Caesareansectionratesareonlyoneofmanyfactorsthatdeterminequality.Anappropriatecaesareansectionratemaychangeovertimeandvaryindifferentdeliveryunits.Ultimately,itwilldependontheMDQAPthattakesintoaccountallthecriteriausedforassessingmaternitycare.Thepurposeofthischapteristoreinforceanddeveloptheprinciplesthatweredescribed10yearsago.12TheMultidisciplinaryQualityAssuranceProgrammeInordertoachieveanappropriatecaesareansectionrate,theconceptofanMDQAPneedstobeimplemented.ThisconceptisdescribedinFig.1inthecontextoflabouranddeliveryofthepregnantwoman,andsimilarprogrammeshavebeensuggestedelsewhere.13Qualityassuranceshouldbeappliedtothesubjectasawhole.Audit,classificationofinformation,assessingmanagementandmodifyingmanagement,whenapplicable,shouldbeappliedtotheprocessesinvolvedinachievingit.Alltheabovecomponentsarecrucialtoachievingquality,butthequalityofinformationcollectionisparamount.Atpresent,settingstandardsandbenchmarkingofinterventionsandoutcomesareusedasassessmentofqualityinahealthcareorganisation.Goodinformationcollectionitselfmustbethefirstqualitystandard.Informationhastobeeasilyavailable,qualitycontrolledandvalidated.Fourcriteriawillbeusedfortheassessmentofmaternitycare:levelofinterventionsandoutcomes(includingsafety),choice(experience),costandefficiency.Thisphilosophycanbeextrapolatedtothedebateoncaesareansectionrates:‘itisnotthatacaesareansectionrateishighorlowbutratherwhetheritisappropriateornot,afterconsideringalltherelevantinformation’.12Audit

Auditisdefinedastheformalexaminationandrecordingoftheresults,andisdividedintostructure(representingresources),process(thewaythatresourcesareapplied)andoutcome(theresultofintervention).Recently,moreemphasishasbeenplacedonauditingprocessesratherthanoutcomes,whereaspatientsaremoreinterestedinoutcome.Qualityisrelatedtooutcome,andoutcomewillguideprocesses.Amorepracticaldefinitionofauditiscontinuouslylookingatyouroutcomesinastandardisedwayatthemostseniorlevelonaregularbasis,resultinginaformalwrittenannualreportdocumentingthequantityandqualityofcare.14M.Robsonetal./BestPractice&ResearchClinicalObstetricsandGynaecology27(2013)297–308

Classification of information: the 10 groups and describing

acceptable ranges of outcomes and events

299

Pre labour outcome and events

Labour and delivery outcome and events

Audit:

When, who, how,

why

Assessment of management:

interpretation of data

Post-delivery outcome and events

Modification of management:

improving processes.

Fig.1.MultidisciplinaryQualityAssuranceProgramme:labouranddelivery.

High-qualityaudithaslongbeenundervaluedindevelopingandsupportingclinicalpractice.Thereasonisthatauditrequirestimeandresource,butmostofalldisciplineandleadership.Thechallengefromapracticalpointofviewistocombineroutinedocumentationofnoteswithauditandtheabilitytousesameforteaching,educationandresearchwithoutduplicationofeffort.Theinformationneedstoberelevant,carefullydefined,accuratelycollected,timelyandavailable.Informationcollectionneedsadequateresourcesandmeticulousorganisation.AdetaileddescriptionofLabourwardaudithasbeengivenelsewhere.14Laboureventsandoutcome

Twomaintypesofdataareavailableforlabouranddelivery.First,epidemiologicaldata,suchasage,height,bodymassindex,medicalconditions,ethnicityandothercase-mixvariables.Second,‘inter-ventions’,whichrefertoevents(oroutcomes)takenbyprofessionalsinvolvedinthemothers’care.Althoughthesearecarriedoutwiththeintentionofimprovingcare,manyinterpretthemasinterferencewithanormalphysiologicalprocess.Thedifficultywithagenerictermlike‘interventions’isthatnodistinctionismadeinhowthemother,midwifeormedicalstaffperceivetheparticulareventoroutcomeinquestion.Evenmoreconfusingisthefactthatwhatmaybeaninterventiontoonewomanmaynotbeaninterventiontoanother;indeed,itmaybeadesiredeventoroutcome.Inordertoclarifymatters,theterm‘intervention’shouldbeavoided.Insteadalleventsthattakeplaceshouldberecordedwhethertheyareprocessescarriedoutbyprofessionalsoroccurasaresultofthecareprovided.Somelaboureventsarealsolabouroutcomes,inthatthemother,midwifeormedicalstaffconsiderthemtoaffectthehealthorsatisfactionofeitherthemotherorbaby.Alleventsandoutcomesneedtobedefinedinastandardway.14Caesareansectionisacaseinpoint.Acaesareansectionisaneventthatmaytakeplaceintheprocessoflabouranddelivery.Itmayalsobeanoutcomeeithernegativeorpositiveorindeedneither,dependingonthecircumstancesofthedelivery.Inductionoflabour,artificialruptureofmembranes,useofoxytocin,andlengthoflabourareotherexamplesofeventsthatmayalsobeoutcomesormayaffecttheincidenceofotheroutcomes.ThemostusefulmaternalandfetalinformationthatneedstobecollectedispresentedinTable1.Additionalinformationwouldbehelpful,butqualityinformationshouldnotbecompromisedandprioritisationofinformationtobecollectedisessential.Thestructureofinformationcollectionis300M.Robsonetal./BestPractice&ResearchClinicalObstetricsandGynaecology27(2013)297–308

Table1Maternalandfetalinformation.MaternalinformationAgeofwomen.Ethnicity.Bookingweightandheight(bodymassindex).Totalnumberofwomen(toconstructthe10groups).Spontaneouslabour.Inductions(fetal,maternalnomedicalreasons).Pre-labourcaesareansectionindications(e.g.fetal,maternal,nomedicalreason).Numberofcaesareansections(toanalysedistributionofcaesareansections).Numberofcaesareansectionsinfirst-stageoflabour(e.g.fetal,dystocia).Numberofcaesareansectionsinsecondstageoflabour(e.g.fetal,dystocia).Artificialruptureofmembranes.Oxytocin(firststage).Oxytocin(secondstage).Epidural.Vaginaloperativedelivery(ventouseorforceps).Durationoflabour.Episiotomy.Third-orfourth-degreetears.Postpartumhaemorrhage.Bloodtransfusion.Peripartuminfectiousmorbidity.Peripartumhysterectomy.Daysatfacilityforthemother.Maternaldeaths.FetalinformationBirthweight.Gestation.Apgarscore(lessthan7at5mins).CordpH(phlessthan7.0).Erb’spalsy.Encephalopathy.Admissionstointensivecareunit.Admissionstointensivecareunitover24h.Daysatfacilityforthenewborn.Stillbirths(lessthan37and37weeksorover).Intrapartumdeaths.Neonataldeaths(7weeksorlessand28weeksorless).Cerebralpalsy.importantand,inparticular,indicationsforproceduressuchasinductionsandcaesareansections.Itwouldbeusefulfromanepidemiologicalpointofviewtoclassifyallindicationsintofetal,maternalornomedicalindication.Thiswillnotbeeasyandweneedtolookatwaysinwhichthiscouldbedone.Thethirdtypeofinformationthatiscollectedinlabouranddeliveryisinformationusedtoclassifytheepidemiologicaldataandtheeventsandoutcomes.Caesareansections:indications

Thenumberofcaesareansectionscarriedoutcaneasilyberecorded,buttheirindicationshavebeendifficulttodefineandimplementconsistently.7Afurtherproblemistheincreaseinnumbersofindicationsusedandalsothenumberusedforeachoperation.Althoughclinicallynotanissue,thisdoespresentaproblemforclassificationandobtaininganoverviewofcare,whytheproceduresarebeingcarriedoutandwhethertheycanbejustifiedintermsofotheroutcome.Ifanappropriatecaesareansectionrateistobedescribed,thenindicationsforcaesareansectionhavetobestand-ardised.Pre-labourcaesareansectionsshouldbeclassifiedintofetal,maternalornomedicalindication.Ifmorethanoneindicationexists,thenonemainindicationshouldbechosenwiththeotherindi-cationsaddedinahierarchicalmanner.Adefinitionfornomedicalindicationormaternalrequestisrequired.15Practically,itmaybebestdefinedas‘atthetimeoftherequestbythewoman,intheopinionoftheobstetricianthereisagreaterrelativeriskofasignificantadverseoutcometomotherorbabybycarryingoutacaesareansectionthanawaitingspontaneouslabouranddeliveryorinducinglabour’.Therelevanceindefiningitinthiswayisthatitplacesanonusonthedeliveryunittomakesuretherelevantinformationfromtheirownresultstogetherwithexternalevidenceisavailabletojustifytheuseoftheindication.Amedicalindicationforacaesareansectionmustbeonethatisusedconsistentlyinsimilarcircumstances.Otherwise,theindicationmustberecordedasmaternalrequestespeciallywhenthewomanhasrequestedit.Thisdoesnotmeantosayitisinappropriatecaretocarryoutacaesareansectionaftercounsellingthewoman,10butonlythatitshouldbeclassifiedasmaternalrequestandalsoM.Robsonetal./BestPractice&ResearchClinicalObstetricsandGynaecology27(2013)297–308301

includesthereasonforthatrequest.Themostcommonexampleforthisinclinicalpracticeisthedecisiontodeliverthebabyofawomanwhohashadonepreviouscaesareansectionat39weekswhohasnoothermedicalreasonforcaesareansection.Thisshouldberecordedasmaternalrequest.Ifthisisrecordedasamedicalindication,thenthelogicalextrapolationofthisisthatbabiesofallwomenwithonepreviouscaesareansectionandnoothermedicalindicationforcaesareansectionshouldbedeliveredbycaesareansectionat39weeks.Thesamewomanwithonepreviouscaesareansectionreaching41weekswithanunfavourablecervixanddeliveredbycaesareansectionshouldbeclassifiedasamedicalindication.Variancesintheapplicationofindicationscanbestudiedbyanalysingthemindifferentgroupsofwomen.16Importantlythough,itisnotinconceivablethatanindicationforcaesareansectionrecordedasmaternalrequesttodaymaywell,withchangeinpracticeandoutcomesfromlabouranddelivery,becomeamedicalindicationinthefutureandalsoviceversa.Thetermselectiveandemergencycaesareansectionaredifficulttodefineandarerarelyappliedinastandardway.Anelectivecaesareansectionmightbestbedefinedasaplannedprocedure(greaterthan24h),carriedoutduringroutineworkinghours,atgreaterthan39weeks,inawomanwhoisneitherinlabourorhashadlabourinduced.Allothercaesareansectionswouldbeauditedasemer-gencyorpossiblymoreappropriatelynon-electivecaesareansections.Thereasonswhytheywererecordedasnon-electivecouldberecordedusingthereasonsdescribedabove.Thisaddsanorgan-isationalelementaswellasclinicaltothedefinitionofelectiveandemergency,andwouldbehelpfulinassessinganappropriatecaesareansectionrate.Indicationsforcaesareansectionsinlabourneedtobesimple,replicableandallowforimprovementofcare.Managementoflabourdependsonensuringfetalwell-beingandachievingefficientuterineaction,andtheyarealsothereasonswhycaesareansectionsarecarriedoutinlabour.Itis,therefore,logicalthatindicationsforcaesareansectionsinlabourmightbeclassifiedintofetalordystociasothatmanagementcanbeassessed.Afetalindicationwouldbedefinedbyconventionwhenacaesareansectioniscarriedoutforsuspectedfetaldistress(forwhateverreason),butwithouttheuseofoxytocin.Allothercaesareansectionscarriedoutinlabourareclassifiedasaformofdystocia.Noformaldefinitionofdystociaissuggested,aseachdeliveryunitwillhavetheirowninterpretation,butthiswillnotprecludethemfromusingthefollowingclassification.Rather,thesub-classificationofdystociawilldependuponwhethertheprogressinlabourhadbeenlessthan1cm/h(inefficientuterineaction)ormorethan1cm/h(efficientuterineaction).Inefficientuterineactionisthensubdividedintopoorresponse(despitemaximumtreatmentwithoxytocin),inabilitytotreatadequately(forfetalreasons),inabilitytotreatadequately(becauseoftheuterusover-contracting),or,lastly,notreatment(oxytocinnotgivenbecauseitisthoughttobeinappropriate,forexample,inlabourwithamalpresentation,inawomanwithapreviouscaesareansection,whenawomandeclinesoxytocinorindeeddeclineslabouritself).Thisclassification(Table2)differentiatesbetweensuspectedfetaldistresswithoutoxytocincomparedwithsuspectedfetaldistressafteroxytocinwasstarted,butwhentheprimaryproblemwasdystocia.Thedistributionoftheresultsinitsuse16reflectsthewaythatdystociaisdiagnosedandhowoxytocinisusedinlabourinthedeliveryunit.Inparticular,theincidence,timing,doseandregimenofoxytocin.Applyingthisclassificationtodifferentgroupsofwomen16givesdifferentresultsthatcanbeusedtoanalysecaesareansectionratesandtheirimplicationsmorerationally.Table2Classificationforcaesareansectionsinlabour.Fetaldistress(nooxytocin)DystociaInefficientuterineactionlessthan1cm/hPoorresponse.Maximumdoseareached.InabilitytoreachmaximumdoseabecauseoffetalintoleranceInabilitytoreachmaximumdoseabecauseofover-contractingornotfollowingunitprotocol.Nooxytocingiven.Cephalopelvicdisproportion.Malposition(occipitoposteriororoccipitotransverse).Efficientuterineactionover1cm/haMaximumdosereferstoindividualunit’sprotocol.302M.Robsonetal./BestPractice&ResearchClinicalObstetricsandGynaecology27(2013)297–308

Classificationofinformation

FortheMDQAPtobesuccessful,qualityinformationisclearlyneeded,butasimportantistheneedtoclassifyandorganiseinformationsothatitcanbeeasilyusedbycliniciansonadailybasistoassessandimprovecare.Classificationsystemsareusedinmedicinetotransformcrudedataandinformationintousefulinformationsothatclinicalcarecanbeimproved.Theyarebasedontheidentificationofdifferentconceptsthatmayeachhaveseveralparameters.17Differentpermutationsoftheseparameters,andoftheirsystematicarrangement,resultinspecificgroupsorcategoriesthatsharesomedefinedpropertyfeatureorquality.Thepurposeofaclassificationsystemusuallydeterminesitsstructure,buttheidealclassificationwillsatisfydifferentpurposes.Themaingroupsoftheclassificationmustberobustenoughtobeunlikelytoneedchanges.Thegroupsorcategoriesoftheclassificationneedtobeprospectivelyidentifiablesothatoutcomescanbeimprovedinthosesamepatientsinthefuture.Thegroupsorcategoriesmustbemutuallyexclusive,totallyinclusiveandclinicallyrelevant.Theclassificationsystemmustbesimpletounderstandandeasytoimplement.Ten-groupclassification

The10-groupclassificationsystem(TGCS)17complieswiththeprinciplesofaclassificationsystemdescribedabove.Ifimplementedonacontinuousbasis,itwouldallowthecriticalassessmentofperinatalcareleadingtochangeifthoughtnecessary.18Theobstetricconcepts,withtheirparameters,usedtoclassifythewomenintheTGCS,arethecategoryofthepregnancy,thepreviousobstetricrecordofthewoman,thecourseoflabouranddelivery,andthegestationalageofthepregnancy.Theconceptsandtheirparametersareallprospective,mutuallyexclusive,totallyinclusive,simpleandeasytounderstandandorganise(Table3).Importantly,theyareclinicallyrelevanttomidwivesandobstetriciansbecausetheinformationtheydependonisrequiredwheneveranassessmentismadeofapregnantwomanwhoiseitherinlabourorabouttodeliver.Itthereforemakessensethatallmaternalandfetalinformation,asdescribedinTable1,isviewedwithintheseconceptsandparametersorcombinationsofthem,andtheTGCSwasformedasshowninTable4.17Theywerechosenonthebasisthattheyprovidethebestclinicalandorgan-isationaloverviewrelativetothenumberofgroups.Theyallowacomparisontobemadebetweendeliveryunits,allowingmorespecificanalysisofthelaboureventsandoutcomes,includingtheirindicationsandepidemiologicalvariables.Eachofthe10groupscanandshouldbefurthersubdividedwhenrequired.Groups1and2shouldbeanalysedseparatelyandalsotogether,asshouldGroups3and4.ThephilosophyoftheTGCSinassessingmaternitycareisbasedonthepremisethatallepidemi-ologicalinformation,maternalandfetaleventsandoutcomeswillbemoreclinicallyrelevantiffirstanalysedwithinthe10groups,theirobstetricconceptsorparameters.Thisisparticularimportantinassessingcaesareansectionratesbutalsootherperinataloutcomes.19TheTGCScanalsobeusedtoTable3Obstetricconceptsandtheirparameters.ObstetricconceptCategoryofpregnancyParameterSinglecephalicpregnancy.Singlebreechpregnancy.Singleobliqueortransverselie.Multiplepregnancy.Nulliparous.Multiparous(withoutauterinescar).Multiparous(withauterinescar).Spontaneouslabour.Inducedlabour.Caesareansectionbeforelabour(electiveoremergency)Gestationalageincompletedweeksattimeofdelivery.PreviousobstetricrecordCourseoflabouranddeliveryGestationM.Robsonetal./BestPractice&ResearchClinicalObstetricsandGynaecology27(2013)297–308

Table4Ten-groupclassificationsystem.Groups303

Overallcaesareansection(CS)rate(%)1977/9250(21.4%)NationalMaternityHospital2011NumberofCSRelativesizeCSrateinContributionovertotalnumberofofgroups%eachgroup%madebyeachwomenineachgroupgrouptotheoverallCSrate%1.Nulliparous,singlecephalic,󰀂37weeks,inspontaneouslabour2.Nulliparous,singlecephalic,󰀂37weeks,inducedorCSbeforelaboura3.Multiparous(excludingprev.CS),singlecephalic,󰀂37weeks,inspontaneouslabour4.Multiparous(excludingprev.CS),singlecephalic,󰀂37weeks,inducedorCSbeforealabour5.PreviousCS,singlecephalic,󰀂37weeks6.Allnulliparousbreeches7.Allmultiparousbreeches(includingprev.CS)8.Allmultiplepregnancies(includingprev.CS)9.Allabnormallies(includingprev.CS)179/2389475/136830/275125.82389/925014.81368/925029.72751/92509.4871/925010.1936/92502.4219/92501.4133/92502.3212/92500.435/92503.6336/92507.5179/238934.7475/13681.130/275112.5109/87161.0571/93693.2204/21985.0113/13363.2134/21210035/3537.8127/3361.9179/92505.1475/92500.330/92501.2109/92506.2571/92502.2204/92501.2113/92501.5134/92500.435/92501.4127/9250109/871571/936204/219113/133134/21235/3510.Allsinglecephalic,󰀃36weeks(includingprev.CS)127/336aGroups2and4arecommonlydividedintoa(inductions)andb(prelabourcaesareansections).classifyanygroupofwomendefinedbydataderivedfromTable1.Forexample,allwomenovertheageof35yearsordifferentethnicgroupscanallbeclassifiedintothe10groupsandanalysedandcomparedwithastandardpopulation.Classificationofcaesareansectionandinductionoflabour

Atpresent,noacceptedclassificationsystemexistsforcaesareansections.20Thisisquiteextraor-dinaryconsideringthecontinuingvolumeofliteratureonthesubjectandtheconcernvoicedbygovernmentsabouttheriseinthenumbersofcaesareansectionsandpossibleimplicationsonwoman’shealth.21–24Manydescriptivestudieshavebeenpublished,butnostandardclassificationsystemhasbeenusedthatfitstheprinciplesdescribedabove,andthathasbeenusedtomakechangesinspecificprospectivegroupsofwomen.Caesareansectionrateshavebeenanalysedbycomparingoverallrates,byindicationforcaesareansection,bysub-groupsofwomenandbyprimaryandrepeatcaesareansectionrates.Theyallhavetheirdisadvantages.25Twonationalguidelineshavebeenpub-lishedoncaesareansectionsintheUK,10,26butnoreferencewasmadetoaclassificationofcaesareansections.TheWorldHealthOrganizationcarriedoutasystematicreviewofclassificationsofcaesareansections25andconcludedthat:‘Women-basedclassificationsingeneral,andRobson’sclassification,inparticular,wouldbeinthebestpositiontofulfillcurrentinternationalandlocalneedsandthateffortstodevelopaninternationallyapplicablecaesareansectionclassificationwouldbemostappropriatelyplacedinbuildinguponthisclassification.Theuseofasinglecaesareansectionclassificationwillfacilitateauditing,analyzingandcomparingcaesareansectionratesacrossdifferentsettingsandhelptocreateandimplementeffectivestrategiesspecificallytargetedtooptimizecaesareansectionrateswherenecessary’.25304M.Robsonetal./BestPractice&ResearchClinicalObstetricsandGynaecology27(2013)297–308

Thebiggestsinglesteptotryandachieveandmaintainappropriatecaesareansectionrateswouldbetoagreeaclassificationforcaesareansectionsandtousethatclassificationinreportingforalldeliveryunits.Theindicationsforcaesareansectionshouldbeanalysedwithineachgroupofwomenbecausethedefinitionandmanagementwillvaryineachgroupandwillhavedifferentrisk–benefitratios.TheTGCScanbeusedtoassessanycaesareansectionrateinabsoluteterms,butalsotocomparewithotherlowerorhighercaesareansectionrateseitherwithinthesamedeliveryunitfrompreviousyears,orwithotherdeliveryunitselsewhere.Itwouldbepossibletoseehowthesizesofthedifferentgroupsvary,andalsoinwhichgroupsofwomenthereisadifferenceincaesareansectionrates.Itwillnotimmediatelyexplainthereasons,andfurtheranalysiswouldberequired,butitwillallowausefuloverviewfromwhichtostart.Fromthis,itwillbepossibletoidentifydifferentgroupsofwomenandchangethemanagementaccordingtoavailableevidence.18ThebenefitofusingtheTGCSwithintheMDQAPisthattheclassification,althoughcommonlyusedforanalysingcaesareansectionrates,wasoriginallydevisedsothatallperinataleventsandoutcome,notonlycaesareansectionrates,couldbeanalysedwithinstandardgroupsofwomen.Furthermore,itcanalsobeusedtoanalyseotherqualityindicatorsofmaternitycare,inadditiontoallowingfordifferencesincasemixandotherepidemiologicalvariables.27Inductionoflabour,andthecontributionitmakestocaesareansectionrates,remainsacontro-versialissue.TheTGCSallowsauniqueanalysisofthatcontribution.ThetwogroupsofwomenthatarerelevantinthestudyofinductionaresinglecephalicnulliparouswomenGroup(2a)andsinglecephalicmultiparous(withoutapreviousscar)womenGroup(4a)(Table4).ThedenominatorthatisusedtostudytheincidenceandindicationsfortheinductionsisthetotalnumberofwomeninGroups1and2,andGroups3and4,respectively.Classifyingtheindicationsforinductions,initiallyatleast,asshowninTable5,hasprovenusefultoobtainanoverview.16Moredetailabouteachinductioncanbeincludedinahierarchicalmannerwithinthesesixgroups.Assessmentofmanagement:interpretationofdata

Theuseofthe10-groupclassificationsystemtoanalyseacaesareansectionrate

AsimplestepwisewaytoanalysecaesareansectioninitiallyisusingtheTGCS,asshowninTables4and6inconjunction.Whencomparingdatawithotherhospitalsorwithinthesamehospitalovertime,column5willimmediatelytellyouthecontributionofeachgrouptotheoverallcaesareansectionrate.Columns3and4willtellyouwhetherthedifferenceincontributioniseitheraresultofachangeinthesizeofthegrouporthecaesareansectionratewithinthegroup,oracombinationofboth.Itisessentialtoremembertheimportanceofthesizeofthegroups.Ingeneral,groups1,2and5contributetotwo-thirdsoftheoverallcaesareansectionrate,withgroup5beingthelargestindividualcontributor.28–30Themoredetailedanalysisofandmanagementofthedifferentgroupshavebeendiscussedelsewhere.12Theanalysisisnotcomplicated,andconclusionscaneasilybedrawn.Whatisneededisothermaternalandfetalinformation(Table1)thathavealsobeenclassifiedintothesamegroupstodecideonappropriatecaesareansectionrateswithinthegroups.Groups1,2and5needtobeanalysedtocontrolcaesareansectionratessafely.Thekeyissuesareingroup1reducingtheincidenceofdystociabyachievingefficientuterineaction,ingroup2limitingtheincidenceofinductionsandpre-labourcaesareansections,andingroup5encouragingwomentowaitforspontaneouslabour.Table5Indicationsforinductionoflabour.FetalreasonsPre-eclamptictoxaemia/hypertensionPostdates(42weeksorover)SpontaneousruptureofmembranesMaternalreasonsandpainsNon-medicalreasonsordateslessthan42weeksM.Robsonetal./BestPractice&ResearchClinicalObstetricsandGynaecology27(2013)297–308

Table6Interpretationofthe10-groupclassificationsystem(Table4).305

1.Addupthenumbersincolumntwo.Thegroupsaremutuallyinclusiveandtotallyexclusivesoallnumerators(totalcaesareansectionineachgroup)anddenominators(totalwomenineachgroup)shouldadduptothetotalnumerator(caesareansection)andthetotaldenominator(totalnumberofwomen).2.LookatGroup9anditssizein3rdcolumn.Itshouldbe0.2–0.6%.Thecaesareanrateshouldbe100%.3.Column3,Groups1þ2(nulliparasinglecephalic󰀂37weeks)combinedusuallycontain35–42%oftotalwomen,sometimeshigher.Ifmorethan45%,suspectauniqueandself-selectedpopulationordatacollectionproblem.IfratioofGroup1andGroup2incolumn3islessthan2:1,thereisahighinductionandpre-labourcaesareansectionrateand,therefore,morelikelytohaveahighcaesareansectionrateinGroups1and2combined.AhighcaesareansectionrateinGroup2(morethan35%)suggestsahighpre-labourcaesareansectionrate.InducedwomeninGroup2usuallyhaveacaesareansectionrateof25–30%.4.Column3,Groups3þ4(multiparasinglecephalic>¼37withnouterinescar)combinedusuallycontainabout30–40%ofwomen.Couldbemorebutusuallylessthan40%butdependsonGroup5.RatioofGroup3andGroup4isusuallygreaterthanratioofGroup1andGroup2,butlessinfluenceoncaesareansectionrate.AhighcaesareansectionrateinGroup4(over20%)suggestsahighrequestforcaesareansection(pre-labour),astherearerelativelyfewabsolutemedicalindicationsforcaesareansectioninGroup4,andinducedwomeninGroup4usuallyhaveacaesareansectionrateof4–6%.5.Column3,lookatGroup5.Sizeunder10%reflectspreviouslowcaesareansectionrate.Ifhigher,therehasbeenahighcaesareansectionrateinthepastyears,mainlyfromgroups1and2.6.Column3,lookatGroups6and7.Thesegroupscombinedshouldcontain3–4%ofwomen,maybeashighas5%.Iftotalisover4%,suspectahighprematuredeliveryrate.Therefore,lookatsizeofgroup10,column3.Ifover4–5%thatisconfirmed.Thereisusuallya2:1ratiobetweenthesizeofgroups6and7(incidenceofbreechishigherinnulliparouswomen).7.Column3,lookatGroup8.Shouldcontain1.5–2%ofwomen.Ifhigherthaneithertertiaryreferralcentreorlargein-vitrofertilisationprogramme.8.Column3,lookatGroup10.Shouldcontainaround4–5%ofwomen.Ifhigher,thenlikelyatertiaryreferralcentreorthereishighriskofprematuredelivery.IfcaesareansectionrateinGroup10(column4)is15–20%,thereisahighpretermlabourrate,ifcaesareansectionrateisover40%,thereismorepre-labourcaesareansection(pre-eclampsiaandintrauterinegrowthretardation).9.Column4.CaesareansectionratestoaimforareimportantinGroups1and3.ForGroup1under10%issatisfactorybutisaffectedbytheratioofGroup1andGroup2.ForGroup3,thecaesareansectionrateshouldbenohigherthan3%and,ifitis,suspectwomenhavebeenwronglyincludedwithpreviousscarsorevenbreechesinthisgroup.Ingroup5,caesareansectionrateof50–60%issatisfactoryprovidedthereissatisfactoryperinataloutcome.AhighercaesareansectionrateinGroup5shouldencouragesubdividingwomenwithtwopreviouscaesareansections,andtheproportionofwomenwhoarehavingapre-labourcaesareansection,theirindicationsandgestationalage.CaesareansectionrateinGroup8isgenerallyabout60%.10.Column5:groups1,2and5normallycontributetotwo-thirdsoftheoverallcaesareansectionrate.11.Usetheserulesafterunderstandingtheclassification.17Ifthedatadistributionlooksoddalwaysfirstsuspectpoordatadefinitionorcollection.12.The10-groupclassificationisself-validatinginthatthedatainthegroupswillsuggestoutliersifyouuseitonacontinuousbasis.Remember,nodeliveryunitcontinuouslycollectscompletelyaccuratedata.Ingroup5,theuseofoxytocin(ifusedatall)andthelengthoflabourshouldbelimited.Inductionoflabourshouldbeavoidedand,ifnecessary,belimitedtoamniotomyaloneratherthanusingphar-macologicaltreatment.TheTGCSallowsimmediatecomparisonofcaesareanratesandstimulatescomparisonanddiscussion.Itcanidentifychangesinindividualgroupsanddeveloprisk–benefitratiosinindividualgroups.Evenifchangesdonotoccurimmediatelyinclinicalpractice,theuseoftheTGCSwillleadtoimprovementofinformationcollectionandorganisationofthedeliveryunit.Theanalysisandassessmentmustbecarriedoutonamultidisciplinarybasis,andallstaffshouldbeawareoftheresults.TheTGCSwasdesignedsothatitcanbeuseduniversally31withoutcomputersystems.ThechallengeistoconvinceprofessionalswhyitisimportanttoimplementanMDQAPandhowthatcanbebestachieved.AsmoreprofessionalsusetheTGCSforanalysingcaesareansectionsandotherperinataloutcomes,18,32–39ithasbecomeevidentthattheclassificationisself-validating.Thismeansthattherelativesizesofthegroupsanddifferenteventsandoutcomeswithinthegroupsgenerallyfallintoexpectedranges.Ifoutsidethoseranges,thereisusuallyanexplanation.Mostcommonly,aproblemwithdatacollectionordefinitionmayexist,buttherecouldbeuniquepopulationgroupsorindeeddifferencesinclinicalpracticethatmayormaynotbejustified.306M.Robsonetal./BestPractice&ResearchClinicalObstetricsandGynaecology27(2013)297–308

Modificationofmanagement

ThelastpartoftheMDQAPisthemodificationofmanagementifrequired.Itisthemostchallengingpartofanyquality-assuranceprogramme.Withqualityinformation,mostwomenandprofessionalswillarriveatthesameconclusion.TheTGCSallowsmorefocusedanalysisofmanagementandallowsspecificchangesinprocessesifnecessaryincertaingroupsofwomen.18ItssuccessdependsontheintegrityoftheMDQAPand,inparticular,continuousauditandclassificationofinformation.Itisnotpossibletogointodetailinthischapteraboutmodifyingmanagementandimplementingchange.Theprincipleshavebeendescribedelsewhere12andwilldependonlocalcultureandorganisationalstructure.Multidisciplinaryinvolvementwillbeessential.Decisionswilldependontheavailableevidencebothexternalandinternal.Itwillalsodependonthewishesofthemothers,butachievingqualityshouldbethepriority.ConclusionThecaesareansectiondebatehasbeenatastandstilloverthepast20years.Weneedtolookatadifferentwayofsolvingtheproblem.Indeed,weneedtoredefinetheproblemandourpriorities.CentraltothedebatemustbethedevelopmentofanMDQAPinwhichwecancomparethequalityofcare.Professionalbodiesshouldpromotedeliveryunitstoproduceastandardisedclinicalreport,withemphasisonqualityinformationbeforeinterpretingoutcomes.Thischaptersuggestsacommonwayofassessinglabouranddeliverypracticeinordertodecideappropriatecaesareansectionrates.Overallcaesareansectionratesaretoocrudetobeusefulandneedtobelookedatinstandardgroupsofwomen.Otherevents,outcomes,costandefficiencyneedtobeanalysedinthesamecontextsothatprofessionalscandeterminetheirrelationshiptocaesareansectionratesonanongoingbasis.Caesareansectionrateswillneverbeidenticaleverywhere,andtheMDQAPwillimproveourknowledgeandcare.Onlywhenwehaveadequateinformationcanwebegintodecideonanappropriatecaesareansectionrate.Itistimetoreplacenaturalandnormalasourcriteriaforpracticeinmidwiferyandobstetricswithanopenconceptofthegood.Furthermore,whatmakestheprofessionalhealthcarepractitionerprofessionalishisorherknowledgeofmeansandconsequences,nothisorheropinionaboutwhatisgoodorbad.40Womenwillalwayschoosethetypeofdeliverythattheythinkissafestforthemandtheirbaby.Professionalshavetorecognisethisandbeobligedtoprovideadequateinformationformotherstomaketherightchoice.Practicepoints󰀄󰀄󰀄󰀄󰀄AnMDQAPshouldbeinplaceatalldeliveryunits.Anannualclinicalreportshouldbeproducedbyalldeliveryunits.High-qualityaudithaslongbeenundervaluedindevelopingandsupportingclinicalpractice.Qualityisrelatedtooutcomeandoutcomewillguideprocesses.Professionalshavearesponsibilitytopracticeevidence-basedmedicinebuttheyshouldnotforgettheirresponsibilitytocollecttheevidencetoensurethattheyareprovidingqualityofcaretotheirpatients.Overall,caesarean-sectionratesareunhelpfulandshouldnotbejudgedinisolationfromotheroutcomesandepidemiologicalcharacteristics.Discussionsaboutreducingcaesareansectionrateswithouttakingotherfactorsintoaccountare,atbest,inappropriateand,atworst,dangerous.ThephilosophyoftheTGCSinassessingmaternitycareisbasedonthepremisethatallepidemiologicalinformation,maternalandfetaleventsandoutcomeswillbemoreclinicallyrelevantifanalysedwithinthe10groupsortheirsubgroups.Thebiggestsinglesteptotryandachieveandmaintainappropriatecaesareansectionrateswouldbetoagreeaclassificationforcaesareansectionsandtousethatclassificationinreportingforalldeliveryunits.󰀄󰀄󰀄󰀄M.Robsonetal./BestPractice&ResearchClinicalObstetricsandGynaecology27(2013)297–308307

Researchagenda󰀄󰀄󰀄󰀄󰀄󰀄ImplementinganMDQAP.Standardisationofevents,outcomes,indications.Developmentofnationalandinternationaldatabases.Classifyingallindicationsintofetal,maternalornomedicalindication.Apracticalandpragmaticdefinitionfornomedicalindicationormaternalrequest.Indicationsforcaesareansectionsinlabourneedtobesimple,replicableandallowforimprovementofcare.󰀄Combiningroutinedocumentationofnoteswithauditandtheabilitytousesameforteaching,educationandresearchwithoutduplicationofeffort.AcknowledgementFionnualaByrneassistedwithclassificationofdata.ConflictofinterestNonedeclared.References1.GibbonsL,BelizánJM,LauerJAetal.Inequitiesintheuseofcesareansectiondeliveriesintheworld.AmJObstetGynecol2012;206:331.e1–331.e19.2.EckerJL&FrigolettoFD.Cesareandeliveryandtherisk-benefitcalculus.NEnglJMed2007;356:885–888.3.FuglenesD,ØianP&KristiansenIS.Obstetricians’choiceofcesareandeliveryinambiguouscases:isitinfluencedbyriskattitudeorfearofcomplaintsandlitigation?AmJObstetGynecol2009;200:48.e1–48.e8.4.BetránAP,MerialdiM,LauerJAetal.Ratesofcaesareansection:analysisofglobal,regionalandnationalestimates.PaediatrPerinatEpidemiol2007;21:98–113.5.BlanchetteH.TherisingcesareandeliveryrateinAmerica:whataretheconsequences?ObstetGynecol2011;118:687–690.6.QueenanJT.Howtostoptherelentlessriseincesareandeliveries.ObstetGynecol2011;118:199–200.7.BarberEL,LundsbergLS,BelangerKetal.Indicationscontributingtotheincreasingcesareandeliveryrate.ObstetGynecol2011;118:29–38.8.NIHstate-of-the-scienceconferencestatementoncesareandeliveryonmaternalrequest.NIHConsensSciStatements2006Mar27–29;23(1):1–29.9.WagnerM.Choosingcaesareansection.Lancet2000;356:1677–1680.10.NationalInstituteofHealthandClinicalExcellence.Caesareansection.NICEGuideline,2011.p.1–282.11.MurthyK,GrobmanWA,LeeTAetal.Associationbetweenrisingprofessionalliabilityinsurancepremiumsandprimarycesareandeliveryrates.ObstetGynecol2007;110:1264–1269.*12.RobsonMS.Canwereducethecaesareansectionrate?BestPractResClinObstetGynaecol2001;15:179–194.*13.MainEK,MortonCH,HopkinsDetal.Cesareandeliveries,outcomes,andopportunitiesforchangeinCalifornia:towardapublicagendaformaternitycaresafetyandquality.PaloAlto,CA:CMQCC,2011,www.cmqcc.org,[lastaccessed12.09.12].*14.RobsonM.InCreasyR,editor.Labourwardaudit.Managementoflaboranddelivery.US:BlackwellScience,1997,pp.559–570.1–12.15.ViscoAG,ViswanathanM,LohrKNetal.Cesareandeliveryonmaternalrequest:maternalandneonataloutcomes.ObstetGynecol2006;108:1517–1529.*16.RobsonM.NationalMaternityHospitalclinicalreport2010.p.105–129.*17.RobsonM.Classificationofcaesareansections.FetalMaternMedRev2001;12:23–39.*18.RobsonMS,ScudamoreIW&WalshSM.Usingthemedicalauditcycletoreducecesareansectionrates.AmJObstetGynecol1996;174:199–205.19.HomerCSE,KurinczukJJ&SparkP.Anoveluseofaclassificationsystemtoauditseverematernalmorbidity.Midwifery2010;26:532–536.20.KnightM&SullivanEA.Variationincaesareandeliveryrates.BMJ2010;341:c5255.21.Caesareansection-thefirstcutisn’tthedeepest.Lancet2010;375:956.22.ChongY-S&KwekKYC.Saferchildbirth:avoidingmedicalinterventionsfornon-medicalreasons.Lancet2010;375:440–442.23.SteerPJ&ModiN.Electivecaesareansections:riskstotheinfant.Lancet2009;374:675–676.24.VictoraCG&BarrosFC.Beware:unnecessarycaesareansectionsmaybehazardous.Lancet2006;367:1796–1797.*25.TorloniMR,BetranAP,SouzaJP,WidmerM,AllenTetal.Classificationsforcesareansection:asystematicreview.PLoSONE2011;6(1):e14566.26.NationalInstituteofHealthandClinicalExcellence.Caesareansection.NICEGuideline,2004.p.1–160.308M.Robsonetal./BestPractice&ResearchClinicalObstetricsandGynaecology27(2013)297–308

*27.ColaisP,FantiniMP,FuscoDetal.RiskadjustmentmodelsforinterhospitalcomparisonofCSratesusingRobson’stengroupclassificationsystemandothersocio-demographicandclinicalvariables.BMCPregnancyChildbirth2012;12:54.*28.BrennanDJ,RobsonMS,MurphyMetal.Comparativeanalysisofinternationalcesareandeliveryratesusing10-groupclassificationidentifiessignificantvariationinspontaneouslabor.AmJObstetGynecol2009;201:308.e1–308.e8.*29.BrennanDJ,MurphyM,RobsonMSetal.Thesingleton,cephalic,nulliparouswomanafter36weeksofgestation:contributiontooverallcesareandeliveryrates.ObstetGynecol2011;117:273–279.30.StivanelloE,RucciP,CarrettaEetal.Riskadjustmentforinter-hospitalcomparisonofcaesareandeliveryratesinlow-riskdeliveries.PLoSOne2011;6:e28060.31.SørbyeIK,VangenS,OnekoOetal.Caesareansectionamongreferredandself-referredbirthingwomen:acohortstudyfromatertiaryhospital,northeasternTanzania.BMCPregnancyChildbirth2011;11:55.32.McCarthyFP,RiggL,CadyLetal.Anewwayoflookingatcaesareansectionbirths.AustNZJObstetGynaecol2007;47:316–320.33.BetránAP,GulmezogluAM,RobsonMetal.WHOglobalsurveyonmaternalandperinatalhealthinLatinAmerica:classifyingcaesareansections.ReprodHealth2009;6:18.34.HowellS,JohnstonT&MacleodS-L.TrendsanddeterminantsofcaesareansectionsbirthsinQueensland,1997–2006.AustNZJObstetGynaecol2009;49:606–611.35.CostaML,CecattiJG,SouzaJPetal.Usingacaesareansectionclassificationsystembasedoncharacteristicsofthepop-ulationasawayofmonitoringobstetricpractice.ReprodHealth2010;7:13.36.AllenVM,BaskettTF&O’ConnellCM.Contributionofselectmaternalgroupstotemporaltrendsinratesofcaesareansection.JObstetGynaecolCan2010;32:633–641.37.StavrouEP,FordJB,ShandAWetal.EpidemiologyandtrendsforcesareansectionbirthsinNewSouthWales,Australia:apopulation-basedstudy.BMCPregnancyChildbirth2011;11:8.38.ScarellaA,ChamyV,SepúlvedaMetal.Medicalauditusingthetengroupclassificationsystemanditsimpactonthecesareansectionrate.EurJObstetGynecolReprodBiol2011;154:136–140.39.DelbaereI,CammuH,MartensEetal.Limitingthecaesareansectionrateinlowriskpregnanciesiskeytoloweringthetrendofincreasedabdominaldeliveries:anobservationalstudy.BMCPregnancyChildbirth2012;12:3.40.WackerhausenS.Whatisnatural?Decidingwhattodoandnottodoinmedicineandhealthcare.BrJObstetGynaecol1999;106:1109–1112.