慢阻肺高风险患者的管理

AndrewMcIvorMD,MSc,FRCPProfessorofMedicineFirestoneInstituteforRespiratoryHealth慢阻肺高风险患者的管理和挑战仅供医疗专业人士参考114233.022有效期2015/06/18慢阻肺的全球影响 慢阻肺位列慢性非传染性疾病的第二位:全球有600x106患者 每年3百万人死于慢阻肺(WHO健康报告1997) 慢阻肺的负担正在逐年上升(DALY):慢阻肺的负担1990年位列12，2050将跃至第5(Murray&LopezScience1996;174:740-3)慢阻肺是近年来唯一一个死亡率明显上升的主要病因1965到1998美国经年龄校正后死亡率百分比变化1.02.03.00.01965死亡率构成比2.51.50.5心脏病中风其他心血管病慢阻肺所有其他死因–59%–64%–35%+163%–7%慢阻肺.comCVD=cerebrovasculardiseaseGlobalInitiativeforChronicObstructiveLungDisease(GOLD)dataindicatethatwhilethedeathrateisdecliningfordiseasessuchascoronaryheartdiseaseandcerebrovasculardisease,thedeathratefor慢阻肺isincreasingmarkedly(163%increasefrom1965to1998intheUS).1TheGOLDstudygroupcommentedthatavailableprevalencedatagreatlyunderestimatethetotalburdenof慢阻肺becausethediseaseisusuallynotdiagnoseduntilitisclinicallyapparentandmoderatelyadvanced.Furthermore,mortalitydataunderestimate慢阻肺asacauseofdeathbecausethediseaseismorelikelytobecitedasacontributoryratherthananunderlyingcauseofdeath,ormaynotbecitedasacauseofdeathatall.2References1. GlobalInitiativeforChronicObstructiveLungDisease.Globalstrategyforthediagnosis, managementandpreventionofchronicobstructivepulmonarydisease.Availableat: http://www.gold慢阻肺.com.2. PauwelsRA,BuistAS,CalverleyPM,etal.Globalstrategyforthediagnosis,management andpreventionofchronicobstructivepulmonarydisease.NHLBI/WHOGlobalInitiativefor ChronicObstructiveLungDisease(GOLD)workshopsummary.AmJRespirCritCareMed 2001;163:1256–76.BarnesPJ.NEnglJMed2000;343:269-280.SethiS.ClinPulmMed1999;6:327-332.NHLBI2000.AmericanThoracicSociety.AmJRespirCritCareMed1995;152:S77-S121.BallP.QJMed1995;88:61-68.BritishThoracicSociety.Thorax1997;52:S1-S28.McCroryetal.Chest.2001Apr;119(4):1190-209.慢阻肺=慢性阻塞性肺疾病;AECB=慢性支气管炎急性加重DiagnosisofChronicBronchitisandAECB Chronicobstructivepulmonarydisease(慢阻肺)isavaguetermthatisusedtocharacterizeasyndromeofprogressive,irreversibleairflowlimitation. Bydefinition,anypersonwhopresentswithsputumproductiononmostdaysforatleast3consecutivemonthseachyearfor2consecutiveyearssuffersfromchronicbronchitisor慢阻肺 Chronicbronchitisisresponsiblefor85%ofthe慢阻肺cases. Patientswithchronicbronchitisexperienceintermittentairwayinflammationthatleadstofrequent,prolongedepisodesofproductivecough. Severalstagingsystemsexist.TheATSusesFEV1whichcorrelatesmostcloselywithmortalityandfrequencyofacuteexacerbation. Thediagnosisofacuteexacerbationsofchronicbronchitisisalsosubjectiveandnotfacilitatedbyasingledefinition.Itisoftenadiagnosisbyexclusion,basedonanincreaseinbaselinesymptoms(e.g.,dyspnea,sputumproduction,sputumpurulence,andcough)intheabsenceofothercauses.BarnesPJ.NEnglJMed2000;343:269-280.SethiS.ClinPulmMed1999;6:327-332.NHLBI2000.AmericanThoracicSociety.AmJRespirCritCareMed1995;152:S77-S121.BallP.QJMed1995;88:61-68.BritishThoracicSociety.Thorax1997;52:S1-S28.McCroryetal.Chest.2001Apr;119(4):1190-20915慢阻肺急性加重恢复缓慢伴有呼吸困难加重的急性加重率天数基线急性加重-14-9-4161116212631SeemungalTAR,etal.AmJRespCritCareMed2000;161:1608-13.Recoveryafteranacuteexacerbationisincompleteinasignificantproportionofpatients.Thisslideshowsthatasingleacuteexacerbationof慢阻肺hasasustainedeffectonhealthstatusasthetimecourseofrecoveryisveryprolonged.25Thereisalsoevidencethatifapatientwithchronicbronchitisexperiencesasecondexacerbationepisodewithin6months,thislimitstherecoverymarkedlyandrecoverycanoccuroverseveralmonthsinsteadofdaysorweeks.31Thefigureshows:25 Thetimecourseofsymptomsfrom14daysbeforeto35daysaftertheonsetofexacerbation. Overtheprodromalperiod(the7daysbeforeonsetofexacerbation),dyspnea,symptomsofacommoncold,sorethroat,andcoughincreasedsignificantly(p<0.05).Theslideshowsonlytheresultsforincreaseddyspnea. Onthedayofonsetofexacerbation,reportsofsymptomsincreasedsharply,with64%ofexacerbationsassociatedwithsymptomsofdyspnea. Patientsrecoveredtheirlungfunctionwithinamediantimeof7days.Only75%ofpatientswithexacerbationsrecoveredtobaselinefunctionat5weeksand7%didnotrecoverat3months.慢性阻塞性肺疾病的自然病程：严重急性加重和死亡SuissaS,etal.Thorax.2012Nov;67(11):957-63.急性加重的频率和严重度增加每10000患者天发生下次严重急性加重的机率每10000患者天发生下次严重急性加重或死亡的机率SuissaS,etal.Thorax.2012Nov;67(11):957-63.ww.ECLIPSE-慢阻肺.com‘频繁急性发作 表 关于同志近三年现实表现材料材料类招标技术评分表图表与交易pdf视力表打印pdf用图表说话 pdf 型’:ECLIPSE急性加重的频率/严重度的GOLD分级(1)两者p<0.01HurstJR,etal.NEnglJMed.2010;363:1128-38ReproducedwithpermissionMassachusettsMedicalSociety(MMS),Copyright©MMSECLIPSE1年数据THISSLIDECONTAINS3BUILDS‘频繁急性发作表型’:ECLIPSE急性加重表型的稳定性HurstJR,etal.NEnglJMed.2010;363:1128-38ECLIPSE3yeardataTHISSLIDECONTAINS7BUILDS基于BlackBerry慢阻肺患者监测 所有输入和传输均经轨迹球和触控屏 自动上传条件问句 完全由研究中心控制 所有数据完全保密 PEF等安全上传 患者数据库实时更新 最简单的可能操作，特别是针对老年有慢性病在智能手机上应用信息系统来探讨慢阻肺急性加重和患者症状日记记录的急性加重的一致性NeilW.Johnston,MSc;KimLambert,RN,MSc;PatriciaHussack,RN;MariaGerhardssondeVerdier,MD,PhD;TimHigenbottam,MD,DSc;JonathanLewis,BSc;PaulNewbold,PhD;MartinJenkins,MMath;GeoffreyR.Norman,PhD;PeterV.Coyle,MD;R.AndrewMcIvor,MDChest.2013;144(2):507-514.doi:10.1378/chest.12-2308慢阻肺诊断、处理及预防全球倡议,2011:主要章节 定义和概况 诊断和评估 治疗选择 管理稳定期慢阻肺 管理急性加重 管理合并症 缓解症状 改善活动耐量 改善健康状态 预防疾病进展 预防和治疗急性加重 降低死亡率慢阻肺诊断、处理及预防的全球倡议管理稳定期慢阻肺:治疗目标EffectivenessofBFC(ICS/LABA)vsTiotropium(LAMA)inCOPDStrangeetalATS2014 USClaimsData 慢阻肺2009年3月-2012年3月 主要终点 到首次急性加重的事件(Cox) 医院访视 门诊口服激素/抗生素 次要终点(比例风险)比较布地奈德福莫特罗与塞托溴铵在慢阻肺治疗中的有效性ATS2014EffectivenessofBFC(ICS/LABA)vsTiotropium(LAMA)inCOPDStrangeetalATS2014BFC布地奈德福莫特罗粉吸入剂信必可1381BFCvs.2670Tiotropium1198人配对Refills3.3BFC4.2Tiotropium比较布地奈德福莫特罗与塞托溴铵在慢阻肺治疗中的有效性ATS2014EffectivenessofBFC(ICS/LABA)vsTiotropium(LAMA)inCOPDStrangeetalATS2014 与塞托溴铵相比，长期布地奈德福莫特罗治疗，与更少的急性加重相关风险比0.78,95%CI=0.7,0.87,p=0.0001 至急性加重的时间：352天BFCvs.243天Tiotropium >1次的急性加重：50.7vs59.3% 急性加重率：1.2vs.1.5RR0.82 p=0.0004 口服激素/抗生素 0.93vs.1.12 p=0.004 急性治疗 0.19vs0.25 p=0.014 住院 0.11vs.0.13 p=0.16NS比较布地奈德福莫特罗与塞托溴铵在慢阻肺治疗中的有效性ATS2014 尚没有RCT比较了不同联合制剂在COPD中的疗效 一项布地奈德/福莫特罗和FLU/SAL为期7天的交叉研究显示两种ICS/LABA在起效速度方面有所不同，这种差异带来在晨间活动方面疗效有所不同1 与安慰剂比较的有关急性加重的研究显示两种ICS/LABA都减少急性加重相似2，但是联合使用LAMA时这样的观点可能会受到挑战3,4 一项加拿大倾向配对队列研究显示，不同ICS/LABAs治疗COPD的疗效具有差异5 一些对RCTs研究的独立meta分析显示不同ICS治疗COPD的肺炎风险可能存在类别内的差异6,71.Partridgeetal.TherAdvRespirDis2009;3:147–157.2.Calverleyetal.ProcAmThoracSoc2004;1:121–124.3.Aaronetal.AnnInternMed2007;146:545–555.4.Welteetal.AmJRepirCritCareMed2009;180:741–750.治疗COPD所有的ICS/LABA都相同吗? Blaisetal.ClinTher2010;32:1320–8. SinghSetalCurrOpinPulmMed.2010;16:118 Nanninietal.CochraneDatabaseSystRev2012;9:CD006829–555.选择何种ICS/LABA治疗COPD?SpeakerNotes: ThissectionwillreviewrealworldevidencesupportingthatICS/LABAcombinationsarenotthesame;influencingexacerbationratesandoutcomes. NoH2HRCTshavecomparedthefixedcombinationsonoutcomesinCOPD. OnesmallcrossoverRCTofBUD/FORMandFLU/SALMhashighlightedthatdifferencesintheonsetofeffectoftheseICS/LABAcantranslatetodifferencesinactivitylevelsinthemorning1 Exacerbationtrialsperformedvs.placebosuggestthatbothBUD/FORMandFLU/SALMwillreduceeventstoasimilarextent2,thismaynotbetruewithLAMA(tripletherapy)3,4References: PartridgeMR,etal.Effectonlungfunctionandmorningactivitiesofbudesonide/formoterolversussalmeterol/fluticasoneinpatientswithCOPD.TherAdvRespirDis2009;3:1-11. CalverleyPM.Reducingthefrequencyandseverityofexacerbationsofchronicobstructivepulmonarydisease.ProcAmThoracSoc2004;1:121–124. AaronSD,etal.TiotropiuminCombinationwithPlacebo,Salmeterol,orFluticasone–SalmeterolforTreatmentofChronicObstructivePulmonaryDisease.AnnInternMed.2007;146:545-555. WelteT,etal.Efficacyandtolerabilityofbudesonide/formoteroladdedtotiotropiuminpatientswithchronicobstructivepulmonarydisease.AmJRespirCritCareMed.2009;180:741-50.在真实世界中选择不同ICS/LABA治疗对急性加重有何不同影响? 在加拿大，应用BUD/FORM较FLU/SALM相比，前者减少25–39%的急性加重，并且减少29%额外的LAMA治疗。HealthcareutilisationinthefirstyearafterinitiatingICS/LABAtherapyBlaisL,etal.ClinTher.2010;32(7):1320-1328.该研究是在加拿大COPD患者中开展的一项回顾性、观察性、配对队列研究，对比了不同ICS/LABA治疗的有效性。AretrospectiveobservationalmatchedcohortstudyinCanadianCOPDpatientsexaminingthecomparativeeffectivenessoftheICS/LABAsSpeakerNotes: AcomparativeeffectivenesscohortstudyperformedinCanadianpatientssuggestsexacerbationoutcomesmaydifferbetweentheICS/LABAs5 InCanada,5255COPDpatientsweretreatedwithanICS/LABAbeforematching 3969(76%)patientsinthiscohortusedFLU/SALM 36%hadexacerbationspre-index Morebud/formvs.flu/salpats.hadafirstICS/LABAinitiatedbyaspecialist(8%more) Antibioticswerenotevaluatedasadefinitionofexacerbations Patientsweretrackedfor1yearReference:BlaisL,etal.RelativeEffectivenessofBudesonide/FormoterolandFluticasonePropionate/Salmeterolina1-Year,Population-Based,MatchedCohortStudyofPatientsWithChronicObstructivePulmonaryDisease(COPD):EffectonCOPD-RelatedExacerbations,EmergencyDepartmentVisitsandHospitalizations,MedicationUtilization,andTreatmentAdherence.ClinTher2010;32(7):1320-1328.加拿大观察研究*比较不同ICS/LABA联合治疗BlaisL,etal.ClinTher.2010;32(7):1320-1328.2校正风险比 长期使用BUD/FORM，与FLU/SALM相比，与较少的需入院治疗的急性加重率相关。*Retrospectiveobservationalmatched-cohortstudySpeakerNotes: AcomparativeeffectivenessretrospectivecohortstudyperformedinCanadianpatients(usingRAMQdata)suggestsexacerbationoutcomesmaydifferbetweentheICS/LABAs. InCanada,5255COPDpatientsweretreatedwithanICS/LABAbeforematching: 3969(76%)patientsinthiscohortusedFLU/SALM 36%hadexacerbationspre-index MoreBUD/FORMvs.FLU/SALMpatientshadafirstICS/LABAinitiatedbyaspecialist(8%more) Antibioticswerenotevaluatedasadefinitionofexacerbations Patientsweretrackedfor1year.Reference:BlaisL,etal.RelativeEffectivenessofBudesonide/FormoterolandFluticasonePropionate/Salmeterolina1-Year,Population-Based,MatchedCohortStudyofPatientsWithChronicObstructivePulmonaryDisease(COPD):EffectonCOPD-RelatedExacerbations,EmergencyDepartmentVisitsandHospitalizations,MedicationUtilization,andTreatmentAdherence.ClinTher.2010;32(7):1320-1328.一项比较联合制剂布地奈德/福莫特罗和氟替卡松/沙美特罗治疗COPD的回顾性配对队列研究ProvidingAnswersToHealthcarebyObservationalStudies真实世界验证Larssonetal.JInternMed2013;DOI: 10.1111/joim.12067Clinicaltrialidentifier:NCT01146392目的 PATHOS研究的目的是在通过“倾向分数”配对的患者中，分析瑞典健康数据库的数据，回顾性观察比较BUD/FOR和FLU/SAL的以下方面的有效性: COPD急性加重 除ICS/LABA外的COPD处方 包括涉及死亡在内的肺炎事件1.JansonC,etal.BMJ 2013;346:f3306.2.LarssonK,etal.JInternMed 2013;273(6):584-94.BUD/FOR，布地奈德/福莫特罗FLU/SAL，氟替卡松/沙美特罗所有人群登记乌普萨拉大学DepartmentofPublicHealthandCaringSciences初级保健中心的21361名患者随访直至死亡或者迁出1.JansonC,etal.BMJ 2013;346:f3306.2.LarssonK,etal.JInternMed 2013;273(6):584-94.Thedatacollectionwasbasedon: Electronicmedicalrecordsinprimarycare MandatorynationalhealthcareregisterssuchasHospitaldischargeregistry;Hospitaloutpatientcare;Theprescriptionregistry;Thecauseofdeathregistry,andtheSocioeconomicregister. TheLinkageofthedatawasdoneatNationalBoardofHealthandWelfareandtheanalysisatUppsalaUniversity ThestrengthofPathosstudyistheabsenceofREPORTBIASwhichisaproblemwithusualregisterstudies,inkludingGPRDinUK(GeneralPracticeResearchDatabase).Medicalrecordsdatawasextractedfromtheelectronicmedicalrecordsattheprimaryhealthcarecentresbyacomputerprogramme.Thisdatawasthen,bytheSwedishNationalBoardofHealthandWelfare,linkedwithmandatorynationalhealthcareregistersasthehospitaldischargeregister,hospitaloutpatientcareregister,socio-economicregister,causeofdeathandprescriptionregister.Thislinkagewaspossiblebyuseofthepatients’socialsecuritynumber. ThelinkeddatabasewasthenmanagedbytheDepartmentofPublicHealthandCaringSciencesattheUppsalaUniversityinUppsala.方法 主要的数据集为基线人群，包含确诊为COPD的任何年龄、性别的患者，没有预设的排除标准 进行倾向指数配对的患者为接受ICS/LABA联合制剂干粉剂(布地奈德/福莫特罗或者氟替卡松/沙美特罗治疗的患者 索引日定义为COPD诊断后首次给予ICS/LABA联合制剂处方的日期 患者随访自1999年1月1日至2009年12月31日 研究结束：任何ICS/LABA治疗结束日期,迁出或死亡 主要观察终点：急性加重及肺炎发生率1.JansonC,etal.BMJ 2013;346:f3306.2.LarssonK,etal.JInternMed 2013;273(6):584-94.Admissionanddischargedates,mainandsecondarydiagnoses,numberofcontacts,diagnoseswassourcedfromtheNationalPatientRegister.Dateandcause(s)ofdeathwasfromtheCauseofDeathregisterDrugprescriptioninformationwasfromtheSwedishPrescribedDrugRegister结果 COPD急性加重定义为:COPD相关的住院、急诊治疗、或者口服激素或抗生素使用14天以内的事件记为一次事件 COPD处方 定义为ICS,LABA,LAMA,SABA,ICS/LABA联合制剂 评估期 事件归于患者发生时接受的治疗。如果治疗改为其它ICS/LABA联合制剂，新的药物处方日期被记为新的开始日期（针对有效性和安全性，但不针对死亡率） 比较年事件率采用泊松回归分析 采用倾向指数配对以减少因不均衡的协变量引起的潜在混杂，统计分析在配对的人群中进行 COPDexacerbationsweredefinedas-COPD-relatedhospitalisations(ICD-10codeJ44asprimarydiagnosisorJ44.0/J44.1assecondarydiagnosis),-emergencyvisits(ICD-10codeJ44.0/J44.1inoutpatienthospitalcare),and-collectionoforalsteroids(ATCcodeH02AB)orantibiotics(ATCcodeJ01AA,J01CA).Exacerbationsoccurringwithin14dayswerecalculatedasonesingleevent.Dispensedprescriptionsofinhaleddrugsusedinobstructivepulmonarydiseasesweredefinedas ICS(ATCcodeR03BA), LABA(R03AC12andR03AC13), short-actingβ2-agonists(SABA;ATCcodeR03AC)and fixedICS/LABAcombinations(R03AK06andR03AK07).-Alltreatmentperiodsonthesamedrug,fromstarttoend(estimatedbasedonprescribedpacksizeandprescribednumberofdailyinhalations),weresummarisedfortreatmentlengthandevents.Anyeventswereassignedtothetreatmentthatthepatienthadatthetimeoftheevent.Eventsduringinterruptionsintreatment(nocollectionbyestimatedendoftreatmentperiod)werenotcounted.IftreatmentwasswitchedtotheotherfixedICS/LABAcombination,thedateofprescriptionofthenewdrugwasusedasstartdate.Medicationusageafterindexwascalculatedperyear.-Thedenominatorisalleligiblepatientsintheactualyear.Theenddateofdrugusagewascalculatedfromtheprescribeddose.倾向指数配对 倾向指数配对用于评估每一个COPD患者的疾病严重程度，配对前基于个体基础，使用高达31个变量* 索引日前的2年基线期用于所有变量 纳入倾向指数的变量包括:性别、年龄、COPD诊断时间治疗药物(基于率):抗生素、SABA、口服/吸入激素、抗胆碱能药物、心血管用药因急性加重、任何心血管原因、肺炎和哮喘的住院(基于率)合并症：哮喘诊断、糖尿病、癌症、心衰、高血压、卒中诊断FEV1%预计值(如果有)*Itwasnotpossibletomatchforweight,height,BMI(measurementsavailableforonlyaminority). Thisisaverybusyslidebutveryimportant.RCTsarethe“goldstandard”inoutcomeevaluation.However,inhealthresearch,RCTsarenotalwayspractical,orethical. InRCTs,therandomizationenablesunbiasedestimationoftreatmenteffects. Inobservationalstudies,propensityscorematchingisthemethodologytoprovideunbiasedestimationoftreatment-effects,tomitigateconcernsrelatedtonon-randomassignmentofpatientstotreatments.Propensityscoresarebeingusedinobservationalstudiestoreducebiasduetoconfoundingfactors.Ithasbeenshownthatmatchingonapropensityscorecanresultinsimilarmatchedpopulations. PMSisperformedtoassessindividualsdiseaseseverity Abaselineperiodof2yearsforpatientcharacterisationandconfounderidentificationbeforeindexdatewasusedformatching. Thevariablesincludedinthismatchingwas: Age,gender,timefromCOPDdiagnosis Medication:antibiotics,SABA,oral/inhaledsteroids,anti-cholinergics,CVmedications Hospitalizationforexacerbations,anycardiovascularreason,pneumoniaandasthma Co-morbidities:diabetes,asthma,cancer,heartfailure,hypertension,stroke,FEV1%pn(whereavailable) Propensityscorematchingofthetwotreatmentpopulationsresultedintwosimilarcohortsof2734patientseach(covering19,170patient-years)witheitherICS/LABAcombinationafterpropensityscorematching. Ratebasedanalysisbetterthanthetimetofirsteventsincethelaterunderestimatethediseaseburden(youthroughawaythesecond,thirdexacerbationsandsoonwhenusetimetofirstevent) Follow-uptimewas3.51+/-2.44Propensityscorematching,howmuchdifferencewasacceptedwhenmatchingpatients.Thesimpleansweristhatthereshouldbeagoodoverlapinthehistogramsforthe‘score’ofthepatients’aswehaveseenforSymbicortandSeretide,andthatfailedfortripletreatmentandspiriva.Thetechnicalanswersisthateachpatientisgettingascorebetween0and1andthatadistancebelow0.0001isusedasadifferencebetweenmatchedpatients.ThisisdescribedbyRosenbaum1982andimplementedinSASbyGriffinetal,2008.Reallytheprogrammingofthepropensityscorematchingtooklessthan1hourwhileimportingthelungfunctiondataftoo3weeks.Avisualinspectionoftheunmatcheddistributionsisagoodidea.患者在规定的时间段符合COPD诊断标准患者n=21,361FLU/SAL(2738[28%]*)配对人群n=2734瑞典连网的来自76个初级卫生保健中心~8%瑞典人群有记录使用ICS/LABA治疗的患者(索引日)n=9,893BUD/FOR(7155[72%])配对人群n=27341:1倾向指数配对*除4例FLU/SAL组的患者无法与更大的BUD/FOR组患者进行配对1.JansonC,etal.BMJ 2013;346:f3306.2.LarssonK,etal.JInternMed 2013;273(6):584-94.索引日前的患者特征数据以患者%表达，除非另有 说明 关于失联党员情况说明岗位说明总经理岗位说明书会计岗位说明书行政主管岗位说明书 1.JansonC,etal.BMJ 2013;346:f3306.2.LarssonK,etal.JInternMed 2013;273(6):584-94. Baselinecharacteristicsinthe2yearsbeforeafirstICS/LABAprescriptionfollowingCOPDdiagnosisdividedbyfixedICS/LABAcombinationtreatment.Unmatchedandpair-wise(1:1)propensitymatchedpopulationsareshown.Propensityscorematchingwasachievedbasedonmultiplebaselinefactorscollectedupto2yearspriortotheindexdata,including:age;gender;lungfunction;numberofscriptsforantibiotics,oralsteroids,LAMAs,inhaledcorticosteroids,short-actingbronchodilators,cardiovascularmedications;diagnosisfordiabetes,asthma,cancer,rheumatoidarthritis,heartfailure,hypertension,andstroke;andnumberofpreviousCOPD-relatedhospitalisations.Intotaljustfourfluticasone/salmeterolpatientswereunabletobematched1:1withasimilarbudesonide/formoterolpatientDataarepresentedasmeanplusstandarddeviation(SD)orasnumber(n)ofpatientspluspercentage.AllprescriptionsforCOPDmedicationsintheSwedishprescriptiondatabasegenerallycorrespondstothreemonthsdrugsupply(avalueof1.0couldequal>1inhalerspermonthforupto3months).Groupmeansarecalculatedincludingpatientsnotinreceiptofthelistedmedicationpre-index.SABA,short-actingβ2agonist;LABA,long-actingβ2agonist;ICS/LABA=inhaledcorticosteroid/long-actingβ2agonistfixedcombinationinhalerThistableshowssomeselectedfactorswhichwasusedinthematching,totally31variableswereusedAsyouseethereisnosignificantdifferencesbetweenthegroups.Intotaljust4fluticasone/salmeterolpatientswereunabletobematched1:1withasimilarbudesonide/formoterolpatientsincetheyhadover100hospitaldayseach.结果:药物暴露布地奈德氟替卡松总体平均随访时间3.5±2.4年1.JansonC,etal.BMJ 2013;346:f3306.2.LarssonK,etal.JInternMed 2013;273(6):584-94.信必可®都保®在中国被批准用于COPD治疗的使用剂量是160微克/4.5微克，2吸/次，一日两吸，具体请参见信必可中国 说明书 房屋状态说明书下载罗氏说明书下载焊机说明书下载罗氏说明书下载GGD说明书下载 。COPD急性加重使用泊松回归分析校正后的医疗资源使用事件（急性加重或处方）的年发生率。NNT：每患者年预防一次急性加重所需治疗的患者数**P<0.0001;*P=0.0003.NNT=3.4LarssonK,etal.JInternMed 2013;273(6):584-94.ComparedwithFLU/SALDiskus,BUD/FORTurbuhalorwasassociatedwithreducedrisk: ofallkindofexacerbationsby26%(26,6%),herepresentedasevent/100patient/year;80/100vs109/100:NNT=3.4 Budesonide/formoterolhad26.0%feweroralsteroidcourses and29.0%fewerantibioticcourses reducedriskofhospitalizationsduetoCOPDby29% and21.0%lowerriskforERvisitsinthebudesonide/formoteroltreatmentgroup AllhighlysignificantThemeancollectedbudesonidedosewas568µg/dayformatchedpatientswhowereprescribedbudesonide/formoterolandthemeanfluticasonedosewas783µg/dayforpatientswhowereprescribedfluticasone/salmeterol,correspondingto89%vs78%ofrecommendedlabeleddoseinCOPD;budesonide640µg/dayandfluticasone1000µg/d(delivereddose).时间依赖的急性加重结果RR=0.82(CI:0.75,0.88)RR=0.79(CI:0.74,0.85)RR=0.74(CI:0.69,0.79)LarssonK,etal.JInternMed 2013;273(6):584-94.*指使用联合制剂的治疗时间。基于急性加重史的急性加重*D26%RR=0.74(CI:0.70,0.80)D30%RR=0.70(CI:0.62,0.79)D27%RR=0.74(CI:0.69,0.79)*急性加重史被定义为入主前服用过口服激素和/或抗生素Larssonetal.JInternMed2013;DOI: 10.1111/joim.12067(n=2734)(n=2734)(n=1914)(n=1914)(n=820)(n=820)基于年龄的急性加重 在大于或小于60岁的患者中，BUD/FORM的疗效均优于FLU/SAL。但在大于60岁的患者中疗效优势更大。D25%RR=0.75(CI:0.66,0.85)D31%RR=0.69(CI:0.65,0.75)D27%RR=0.74(CI:0.69,0.79)(n=2734)(n=2734)(n=600)(n=600)(n=2134)(n=2134)Larssonetal.JInternMed2013;DOI: 10.1111/joim.12067基于哮喘病史的急性加重率比=0.72(Cl:0.67,0.79)率比=0.76(Cl:0.69,0.83)率比=0.74(Cl:0.69,0.79)(n=2734)(n=2734)(n=1060)(n=1060)(n=1674)(n=1674)LarssonK,etal.JInternMed 2013;273(6):584-94.PATHOS研究中的三联治疗布地奈德/福莫特罗+噻托溴铵vs.氟替卡松/沙美特罗+噻托溴铵 ~50%配对的ICS/LABA使用者同时伴随LAMA处方JansonCetalATSabstract2013F/S=fluticasone/salmeterol,T=tiotropium,B/F,budesonide/formoterol,NNT=numberneededtotreatfor1yearwithB/F+Tvs.F/S+Ttoavoidoneexacerbationorpneumoniaeventarateratiosandupperconfidencelimits<1.00favorB/F+Tvs.F/S+Tbrateratiosandlowerconfidencelimits>1.00favorB/F+Tvs.F/S+TPATHOS与加拿大研究:住院率ICS/LABA配对队列研究>1年治疗期住院率急性加重/100患者/年29%事件降低p<0.001Blaisetal.ClinTher2010;32:1320–8.CanadiancohortdidnotreportinfectiousCOPDexacerbationsrequiringantibioticsorpneumoniarelatedevents(n=2734)(n=2734)(n=1131)(n=1131)自首次处方ICS/LABA随访1年自首次处方ICS/LABA随访≤11年非急性加重相关结果额外支扩剂的使用RR=0.84(CI:0.79,0.89);p<.0001(n=2734)(n=2734)LAMA处方/患者-年SABA处方/患者-年RR=0.78(CI:0.72,0.84);p<.0001(n=2734)(n=2734)Larssonetal.JInternMed2013;DOI: 10.1111/joim.12067ICS/LABA转换 转换治疗 方案 气瓶 现场处置方案 .pdf气瓶 现场处置方案 .doc见习基地管理方案.doc关于群访事件的化解方案建筑工地扬尘治理专项方案下载 的患者的年急性加重率高于未转换的患者。 使用FLU/SAL治疗的患者出现转换治疗方案的几率是使用BUD/FORM患者的4–5倍。 (n=2734)(n=2734)nnumbersindicatethetotalnumberofpatientsintheoriginaltreatmentgroupLarssonetal.JInternMed2013;DOI: 10.1111/joim.12067一项比较联合制剂布地奈德/福莫特罗和氟替卡松/沙美特罗治疗COPD的回顾性配对队列研究一项基于日常临床医疗数据的配对研究ProvidingAnswersToHealthcarebyObservationalStudies1.JansonC,etal.BMJ 2013;346:f3306.2.LarssonK,etal.JInternMed 2013;273(6):584-94.BUD/FOR，布地奈德/福莫特罗FLU/SAL，氟替卡松/沙美特罗 尚没有RCT比较了不同联合制剂在COPD中的疗效 一项布地奈德/福莫特罗和氟替卡松/沙美特罗为期7天的交叉研究显示两种ICS/LABA在起效速度方面有所不同，这种差异带来在晨间活动方面疗效有所不同1 与安慰剂比较的有关急性加重的研究显示两种ICS/LABA都减少急性加重相似2，但是伴随使用LAMA时这样的观点可能会受到挑战3,4 一项加拿大倾向配对队列研究显示，不同ICS/LABAs治疗COPD具有疗效差异存在差异5 一些对RCTs研究的独立meta分析显示不同ICS治疗COPD的肺炎风险可能存在类别内的差异6,71.Partridgeetal.TherAdvRespirDis2009;3:147–157.2.Calverleyetal.ProcAmThoracSoc2004;1:121–124.3.Aaronetal.AnnInternMed2007;146:545–555.4.Welteetal.AmJRepirCritCareMed2009;180:741–750.治疗COPD所有的ICS/LABA都相同吗? Blaisetal.ClinTher2010;32:1320–8. SinghSetalCurrOpinPulmMed.2010;16:118 Nanninietal.CochraneDatabaseSystRev2012;9:CD006829–555.COPD患者中的肺炎风险(按照ICS类型分层)ICS使用的Meta分析(为期1-3年的研究)SinghSetalCurrOpinPulmMed.2010;16:118**氟替卡松的增加又显著差异，p<0.0001[ns]=没有统计学的显著差异Nanninietal.CochraneDatabaseSystRev2012;9:CD006829ICS/LABAvs.LABA结果:肺炎根据ICS/LABA种类分解FavourscombinationFavoursLABA随治疗时间的累计肺炎率FLU/SAL任何肺炎住院的肺炎BUD/FOR任何肺炎住院的肺炎JansonC,etal.BMJ 2013;346:f3306. Thesefigureshowscumulativeeventrates(pneumonias)/100patientsversustime. Comparedtobudesonide/formoteroltreatment,therewasa73%higherrateofpneumoniainthefluticasone/salmeteroltreatmentgroup Wecanseeauniformpatternversustimefor“allpneumonias”andfor“hospitalisedpneumonias”.Thedifferencesobservedforthebudesonide/formoterolvsthefluticasone/salmeterolgroupswithregardtodiagnosisofpneumoniawasindependentofwherethepneumoniadiagnosiswasrecorded,inprimarycareorathospital(67%ofallpneumoniaswerepneumoniasdiagnosedathospitalswhereX-rayisused). Comparedtobudesonide/formoteroltreatment,pneumonia-relatedhospitalisationswere74%higherinthefluticasone/salmeteroltreatmentgroup.NNTtoavoidonepneumoniaeventperyearof22Thefigureshowsthat100patientswillhavemorethan100pneumoniaduring9yearsinSAL/FLUtreatedpatients.Thefigureshowsnotthatsomepatientswillbewithoutpneumonia.TORCH:Usingananalysisofrates,itisestimatedtherewouldbeoneextracaseofpneumoniaforevery31patientsreceivingtreatmentwithSFCover1yr.氟替卡松/沙美特罗组患者的肺炎发生率显著增加TORCHCrimetalEurRespirJ2009,34:641FPSHR1.64(95%CL1.33to2.02)FPSHR1.94(95%CL1.19to3.17)INSPIRECalverleyetal.Chest2011;139:505PATHOSJansonCetalBMJ2013FPSHR1.73(95%CL1.57to1.90)PATHOS:肺炎发生率D64%D94%D73%肺炎发生率/100病人-年NNT肺炎和肺炎相关事件JansonC,etal.BMJ 2013;346:f3306. Theresultsarepresentedasevents/100patients/year ThehazardratioforFLU/SALversusBUD/FORwas1.73,representinga73%increaseintheriskofpneumonia Itisestimatedtherewouldbeoneextracaseofpneumoniaforevery22patientsreceivingtreatmentwithFLU/SAL,NNT=22 FLU/SALpatientshadalso74%higherriskforpneumonia-relatedhospitalisationsversusthe布地奈德/福莫特罗group,NNT=32 Inthelast2columnsyoucanseethedistributionofthepneumoniadiagnosis. Thedifferencesobservedforthe布地奈德/福莫特罗vstheFLU/SALgroupswithregardtodiagnosisofpneumoniawasinthesamerangeindependentofwherethepneumoniadiagnosiswasrecorded 2,115patients(39%)hadatleastonepneumoniaduringthestudyperiod 1491pneumoniasinSeretidein752patients 997pneumoniasinSymbicotin565patients___________________________ Therewasan82%increaseinhospitaldaysinFLU/SALvs布地奈德/福莫特罗patients(53vs29days/100patienttreatment-years).SuchbigdifferenceinhospitaldayssinceFLU/SALgeneratedtotallymuchmoreevents(=hospitalisations).Themeandurationofeachpneumoniarelatedhospitalisationwashoweversimilar,7.1daysforSAL/FLUversus6.5for布地奈德/福莫特罗. Theexacerbationrateswereafactor6higherthanthepneumoniarates. Pneumoniaswerebasedonclinicaldiagnoses,i.e.ICD-10codefrommedicalrecords(inprimarycareorathospitalisation). Pneumoniaweremoreoftendiagnosedathospitals(60%),withaccesstoX-raythaninprimarycareTORCH:ThehazardratioforSFCversusplacebowas1.64(95%CI1.33–2.02),representinga64%increaseintheriskofpneumoniaTheincreaseinreportsofpneumoniaasanAEintheICScontainingtreatmentarmsoccurreddespiteadecreaseinoverallCOPDexacerbations,ofwhichpneumoniawasasubset.Therewerenoprotocol-definedcriteriaforpneumoniatoconfirmtheclinicaldiagnosis.AsforRCTs,investigatorsrecordedallAEs,includingpneumoniareports,inthecasereportform.IntheBigTORCHtrial,afteradjustingforotherriskfactors,therewasnoevidencethatcurrentsmokerswereatgreaterriskthanformersmokers,orthatsexwasariskfactor.TherewasnointeractionbetweenageandICSontheprobabilityofpneumonia.肺炎事件发生率/100患者-年肺炎敏感性分析–4quartilesofpropensityscorediseaseburden20121684Eachquartilecontainseither683or684matchpatientpairsusingFLU/SALorBUD/FOR0PropensityscorequartilesJansonetal.BMJ2013;346:f3306DOI: 10.1136/bmj.f3306RR=1.13;P<0.08RR=1.57;P<0.001RR=1.87;P<0.001RR=2.09;P<0.001至所有原因&肺炎相关的死亡时间患者(%)102051525JansonCetalBMJ(2013) All-causemortalitydidnotdifferbetweenthegroupsalthoughthelinesseparatedmoreattheendoftheperiod(hazardratio1.08[95%CI0.93,1.14];P=0.593). PatientsintheSAL/FLUgrouphada76%higherriskforpneumonia-relatedmortalitythanthoseinthematched布地奈德/福莫特罗group(P=0.0025).Summary在这项与FLU/SAL准纳器对比的慢阻肺患者队列研究中: 长期服用BUD/FORM与更少的所有类型急性加重相关 长期服用BUD/FORM与更少的慢阻肺伴随干预(塞托溴铵和SABA)有关 长期服用BUD/FORM与更少的转换治疗方案患者相关 长期服用BUD/FORM与更少的肺炎发生率相关 敏感性分析并未发现基于索引日前慢阻肺严重程度、哮喘病史、急性加重史、肺炎病史、年龄、疗程、FEV1等的偏差。 加拿大真实世界研究结果和PATHOS研究结果一致。Larssonetal.JInternMed2013;DOI: 10.1111/joim.12067Jansonetal.BMJ2013;346:f3306DOI: 10.1136/bmj.f3306WedzichaJA,etal.AJRCCM.2008;CalverleyPM,etal.Chest.2011;139:505;PattersonC,etal.RespirRes.2012;13:40;EkA,etal.Allergy.1999;54:691;Miller-Larsson,etal.AJRCCM.2000;162:145;JohnssonM,etal.Allergy.1995;50:s11-14;DalbyC,etal.RespirRes.2009;10:104.ASL 感染时发生局部细菌增殖 约50%COPD病人随疾病严重程度细菌菌落增加一种免疫抑制剂/感染假说 容易吸收(数分钟) 推测对局部免疫反应和细菌生长有较小影响 停留在粘液(数小时) 推测抑制局部免疫反应导致细菌生长SpeakerNotes: TherearetwohypothesisforthisobserveddifferencebetweenBUD/FORMandFLU/SALMwithrespecttopneumoniarates: FLUis10-foldmoreimmunosuppressivethanBUD.1 PharmocokineticdifferencesbetweenthemoleculesmayleadtoBUDhavingafasteruptakefromtheairwaysintothemucosa.BUDislesslipophilicthanFLU,dissolvingmorerapidlyinairwaymucusandmorerapidlyabsorbedintotheairwaytissueandsystemiccirculation.FLUishighlylipophilic/lesswatersolubleandmaystayintheairwayslonger.Becauseitremainsintheairwaysitismoreapttobeexpectoratedorclearedfromthesputum.2 BecauseFLUremainsintheairwayliningfluidlongeritmorepotentlyinhibitscytokinereleasefromlungepithelialcellssuppressingtheimmuneresponsetopneumoniacausingbacteria.1,2 BUDtakesminstodissolveintheairwaysurfaceliquidandbeabsorbed,FLUislesswatersolubleandtakes>8hourstodissolveintheliquidandbeabsorbed.3References: EkA,etal.Fluticasoneandbudesonideinhibitcytokinereleaseinhumanlungepithelialcellsandalveolarmacrophages.Allergy.1999;54(7):691-9. DalbyC,etal.Thebioavailabilityandairwayclearanceofthesteroidcomponentofbudesonide/formoterolandsalmeterol/fluticasoneafterinhaledadministrationinpatientswithCOPDandhealthysubjects:arandomizedcontrolledtrial.RespirRes.2009;10:104 HöggerP,etal.Biochemicalcharacterizationofaglucocorticoid-inducedmembraneprotein(RM3/1)inhumanmonocytesanditsapplicationasmodelsystemforrankingglucocorticoidpotency.PharmRes.1998;15:296-302.此课件下载可自行编辑修改，此课件供参考！部分内容来源于网络，如有侵权请与我联系删除！GlobalInitiativeforChronicObstructiveLungDisease(GOLD)dataindicatethatwhilethedeathrateisdecliningfordiseasessuchascoronaryheartdiseaseandcerebrovasculardisease,thedeathratefor慢阻肺isincreasingmarkedly(163%increasefrom1965to1998intheUS).1TheGOLDstudygroupcommentedthatavailableprevalencedatagreatlyunderestimatethetotalburdenof慢阻肺becausethediseaseisusuallynotdiagnoseduntilitisclinicallyapparentandmoderatelyadvanced.Furthermore,mortalitydataunderestimate慢阻肺asacauseofdeathbecausethediseaseismorelikelytobecitedasacontributoryratherthananunderlyingcauseofdeath,ormaynotbecitedasacauseofdeathatall.2References1. GlobalInitiativeforChronicObstructiveLungDisease.Globalstrategyforthediagnosis, managementandpreventionofchronicobstructivepulmonarydisease.Availableat: http://www.gold慢阻肺.com.2. PauwelsRA,BuistAS,CalverleyPM,etal.Globalstrategyforthediagnosis,management andpreventionofchronicobstructivepulmonarydisease.NHLBI/WHOGlobalInitiativefor ChronicObstructiveLungDisease(GOLD)workshopsummary.AmJRespirCritCareMed 2001;163:1256–76.DiagnosisofChronicBronchitisandAECB Chronicobstructivepulmonarydisease(慢阻肺)isavaguetermthatisusedtocharacterizeasyndromeofprogressive,irreversibleairflowlimitation. Bydefinition,anypersonwhopresentswithsputumproductiononmostdaysforatleast3consecutivemonthseachyearfor2consecutiveyearssuffersfromchronicbronchitisor慢阻肺 Chronicbronchitisisresponsiblefor85%ofthe慢阻肺cases. Patientswithchronicbronchitisexperienceintermittentairwayinflammationthatleadstofrequent,prolongedepisodesofproductivecough. Severalstagingsystemsexist.TheATSusesFEV1whichcorrelatesmostcloselywithmortalityandfrequencyofacuteexacerbation. Thediagnosisofacuteexacerbationsofchronicbronchitisisalsosubjectiveandnotfacilitatedbyasingledefinition.Itisoftenadiagnosisbyexclusion,basedonanincreaseinbaselinesymptoms(e.g.,dyspnea,sputumproduction,sputumpurulence,andcough)intheabsenceofothercauses.BarnesPJ.NEnglJMed2000;343:269-280.SethiS.ClinPulmMed1999;6:327-332.NHLBI2000.AmericanThoracicSociety.AmJRespirCritCareMed1995;152:S77-S121.BallP.QJMed1995;88:61-68.BritishThoracicSociety.Thorax1997;52:S1-S28.McCroryetal.Chest.2001Apr;119(4):1190-209Recoveryafteranacuteexacerbationisincompleteinasignificantproportionofpatients.Thisslideshowsthatasingleacuteexacerbationof慢阻肺hasasustainedeffectonhealthstatusasthetimecourseofrecoveryisveryprolonged.25Thereisalsoevidencethatifapatientwithchronicbronchitisexperiencesasecondexacerbationepisodewithin6months,thislimitstherecoverymarkedlyandrecoverycanoccuroverseveralmonthsinsteadofdaysorweeks.31Thefigureshows:25 Thetimecourseofsymptomsfrom14daysbeforeto35daysaftertheonsetofexacerbation. Overtheprodromalperiod(the7daysbeforeonsetofexacerbation),dyspnea,symptomsofacommoncold,sorethroat,andcoughincreasedsignificantly(p<0.05).Theslideshowsonlytheresultsforincreaseddyspnea. Onthedayofonsetofexacerbation,reportsofsymptomsincreasedsharply,with64%ofexacerbationsassociatedwithsymptomsofdyspnea. Patientsrecoveredtheirlungfunctionwithinamediantimeof7days.Only75%ofpatientswithexacerbationsrecoveredtobaselinefunctionat5weeksand7%didnotrecoverat3months.THISSLIDECONTAINS3BUILDSTHISSLIDECONTAINS7BUILDSSpeakerNotes: ThissectionwillreviewrealworldevidencesupportingthatICS/LABAcombinationsarenotthesame;influencingexacerbationratesandoutcomes. NoH2HRCTshavecomparedthefixedcombinationsonoutcomesinCOPD. OnesmallcrossoverRCTofBUD/FORMandFLU/SALMhashighlightedthatdifferencesintheonsetofeffectoftheseICS/LABAcantranslatetodifferencesinactivitylevelsinthemorning1 Exacerbationtrialsperformedvs.placebosuggestthatbothBUD/FORMandFLU/SALMwillreduceeventstoasimilarextent2,thismaynotbetruewithLAMA(tripletherapy)3,4References: PartridgeMR,etal.Effectonlungfunctionandmorningactivitiesofbudesonide/formoterolversussalmeterol/fluticasoneinpatientswithCOPD.TherAdvRespirDis2009;3:1-11. CalverleyPM.Reducingthefrequencyandseverityofexacerbationsofchronicobstructivepulmonarydisease.ProcAmThoracSoc2004;1:121–124. AaronSD,etal.TiotropiuminCombinationwithPlacebo,Salmeterol,orFluticasone–SalmeterolforTreatmentofChronicObstructivePulmonaryDisease.AnnInternMed.2007;146:545-555. WelteT,etal.Efficacyandtolerabilityofbudesonide/formoteroladdedtotiotropiuminpatientswithchronicobstructivepulmonarydisease.AmJRespirCritCareMed.2009;180:741-50.SpeakerNotes: AcomparativeeffectivenesscohortstudyperformedinCanadianpatientssuggestsexacerbationoutcomesmaydifferbetweentheICS/LABAs5 InCanada,5255COPDpatientsweretreatedwithanICS/LABAbeforematching 3969(76%)patientsinthiscohortusedFLU/SALM 36%hadexacerbationspre-index Morebud/formvs.flu/salpats.hadafirstICS/LABAinitiatedbyaspecialist(8%more) Antibioticswerenotevaluatedasadefinitionofexacerbations Patientsweretrackedfor1yearReference:BlaisL,etal.RelativeEffectivenessofBudesonide/FormoterolandFluticasonePropionate/Salmeterolina1-Year,Population-Based,MatchedCohortStudyofPatientsWithChronicObstructivePulmonaryDisease(COPD):EffectonCOPD-RelatedExacerbations,EmergencyDepartmentVisitsandHospitalizations,MedicationUtilization,andTreatmentAdherence.ClinTher2010;32(7):1320-1328.SpeakerNotes: AcomparativeeffectivenessretrospectivecohortstudyperformedinCanadianpatients(usingRAMQdata)suggestsexacerbationoutcomesmaydifferbetweentheICS/LABAs. InCanada,5255COPDpatientsweretreatedwithanICS/LABAbeforematching: 3969(76%)patientsinthiscohortusedFLU/SALM 36%hadexacerbationspre-index MoreBUD/FORMvs.FLU/SALMpatientshadafirstICS/LABAinitiatedbyaspecialist(8%more) Antibioticswerenotevaluatedasadefinitionofexacerbations Patientsweretrackedfor1year.Reference:BlaisL,etal.RelativeEffectivenessofBudesonide/FormoterolandFluticasonePropionate/Salmeterolina1-Year,Population-Based,MatchedCohortStudyofPatientsWithChronicObstructivePulmonaryDisease(COPD):EffectonCOPD-RelatedExacerbations,EmergencyDepartmentVisitsandHospitalizations,MedicationUtilization,andTreatmentAdherence.ClinTher.2010;32(7):1320-1328.Thedatacollectionwasbasedon: Electronicmedicalrecordsinprimarycare MandatorynationalhealthcareregisterssuchasHospitaldischargeregistry;Hospitaloutpatientcare;Theprescriptionregistry;Thecauseofdeathregistry,andtheSocioeconomicregister. TheLinkageofthedatawasdoneatNationalBoardofHealthandWelfareandtheanalysisatUppsalaUniversity ThestrengthofPathosstudyistheabsenceofREPORTBIASwhichisaproblemwithusualregisterstudies,inkludingGPRDinUK(GeneralPracticeResearchDatabase).Medicalrecordsdatawasextractedfromtheelectronicmedicalrecordsattheprimaryhealthcarecentresbyacomputerprogramme.Thisdatawasthen,bytheSwedishNationalBoardofHealthandWelfare,linkedwithmandatorynationalhealthcareregistersasthehospitaldischargeregister,hospitaloutpatientcareregister,socio-economicregister,causeofdeathandprescriptionregister.Thislinkagewaspossiblebyuseofthepatients’socialsecuritynumber. ThelinkeddatabasewasthenmanagedbytheDepartmentofPublicHealthandCaringSciencesattheUppsalaUniversityinUppsala.Admissionanddischargedates,mainandsecondarydiagnoses,numberofcontacts,diagnoseswassourcedfromtheNationalPatientRegister.Dateandcause(s)ofdeathwasfromtheCauseofDeathregisterDrugprescriptioninformationwasfromtheSwedishPrescribedDrugRegisterCOPDexacerbationsweredefinedas-COPD-relatedhospitalisations(ICD-10codeJ44asprimarydiagnosisorJ44.0/J44.1assecondarydiagnosis),-emergencyvisits(ICD-10codeJ44.0/J44.1inoutpatienthospitalcare),and-collectionoforalsteroids(ATCcodeH02AB)orantibiotics(ATCcodeJ01AA,J01CA).Exacerbationsoccurringwithin14dayswerecalculatedasonesingleevent.Dispensedprescriptionsofinhaleddrugsusedinobstructivepulmonarydiseasesweredefinedas ICS(ATCcodeR03BA), LABA(R03AC12andR03AC13), short-actingβ2-agonists(SABA;ATCcodeR03AC)and fixedICS/LABAcombinations(R03AK06andR03AK07).-Alltreatmentperiodsonthesamedrug,fromstarttoend(estimatedbasedonprescribedpacksizeandprescribednumberofdailyinhalations),weresummarisedfortreatmentlengthandevents.Anyeventswereassignedtothetreatmentthatthepatienthadatthetimeoftheevent.Eventsduringinterruptionsintreatment(nocollectionbyestimatedendoftreatmentperiod)werenotcounted.IftreatmentwasswitchedtotheotherfixedICS/LABAcombination,thedateofprescriptionofthenewdrugwasusedasstartdate.Medicationusageafterindexwascalculatedperyear.-Thedenominatorisalleligiblepatientsintheactualyear.Theenddateofdrugusagewascalculatedfromtheprescribeddose. Thisisaverybusyslidebutveryimportant.RCTsarethe“goldstandard”inoutcomeevaluation.However,inhealthresearch,RCTsarenotalwayspractical,orethical. InRCTs,therandomizationenablesunbiasedestimationoftreatmenteffects. Inobservationalstudies,propensityscorematchingisthemethodologytoprovideunbiasedestimationoftreatment-effects,tomitigateconcernsrelatedtonon-randomassignmentofpatientstotreatments.Propensityscoresarebeingusedinobservationalstudiestoreducebiasduetoconfoundingfactors.Ithasbeenshownthatmatchingonapropensityscorecanresultinsimilarmatchedpopulations. PMSisperformedtoassessindividualsdiseaseseverity Abaselineperiodof2yearsforpatientcharacterisationandconfounderidentificationbeforeindexdatewasusedformatching. Thevariablesincludedinthismatchingwas: Age,gender,timefromCOPDdiagnosis Medication:antibiotics,SABA,oral/inhaledsteroids,anti-cholinergics,CVmedications Hospitalizationforexacerbations,anycardiovascularreason,pneumoniaandasthma Co-morbidities:diabetes,asthma,cancer,heartfailure,hypertension,stroke,FEV1%pn(whereavailable) Propensityscorematchingofthetwotreatmentpopulationsresultedintwosimilarcohortsof2734patientseach(covering19,170patient-years)witheitherICS/LABAcombinationafterpropensityscorematching. Ratebasedanalysisbetterthanthetimetofirsteventsincethelaterunderestimatethediseaseburden(youthroughawaythesecond,thirdexacerbationsandsoonwhenusetimetofirstevent) Follow-uptimewas3.51+/-2.44Propensityscorematching,howmuchdifferencewasacceptedwhenmatchingpatients.Thesimpleansweristhatthereshouldbeagoodoverlapinthehistogramsforthe‘score’ofthepatients’aswehaveseenforSymbicortandSeretide,andthatfailedfortripletreatmentandspiriva.Thetechnicalanswersisthateachpatientisgettingascorebetween0and1andthatadistancebelow0.0001isusedasadifferencebetweenmatchedpatients.ThisisdescribedbyRosenbaum1982andimplementedinSASbyGriffinetal,2008.Reallytheprogrammingofthepropensityscorematchingtooklessthan1hourwhileimportingthelungfunctiondataftoo3weeks.Avisualinspectionoftheunmatcheddistributionsisagoodidea. Baselinecharacteristicsinthe2yearsbeforeafirstICS/LABAprescriptionfollowingCOPDdiagnosisdividedbyfixedICS/LABAcombinationtreatment.Unmatchedandpair-wise(1:1)propensitymatchedpopulationsareshown.Propensityscorematchingwasachievedbasedonmultiplebaselinefactorscollectedupto2yearspriortotheindexdata,including:age;gender;lungfunction;numberofscriptsforantibiotics,oralsteroids,LAMAs,inhaledcorticosteroids,short-actingbronchodilators,cardiovascularmedications;diagnosisfordiabetes,asthma,cancer,rheumatoidarthritis,heartfailure,hypertension,andstroke;andnumberofpreviousCOPD-relatedhospitalisations.Intotaljustfourfluticasone/salmeterolpatientswereunabletobematched1:1withasimilarbudesonide/formoterolpatientDataarepresentedasmeanplusstandarddeviation(SD)orasnumber(n)ofpatientspluspercentage.AllprescriptionsforCOPDmedicationsintheSwedishprescriptiondatabasegenerallycorrespondstothreemonthsdrugsupply(avalueof1.0couldequal>1inhalerspermonthforupto3months).Groupmeansarecalculatedincludingpatientsnotinreceiptofthelistedmedicationpre-index.SABA,short-actingβ2agonist;LABA,long-actingβ2agonist;ICS/LABA=inhaledcorticosteroid/long-actingβ2agonistfixedcombinationinhalerThistableshowssomeselectedfactorswhichwasusedinthematching,totally31variableswereusedAsyouseethereisnosignificantdifferencesbetweenthegroups.Intotaljust4fluticasone/salmeterolpatientswereunabletobematched1:1withasimilarbudesonide/formoterolpatientsincetheyhadover100hospitaldayseach.ComparedwithFLU/SALDiskus,BUD/FORTurbuhalorwasassociatedwithreducedrisk: ofallkindofexacerbationsby26%(26,6%),herepresentedasevent/100patient/year;80/100vs109/100:NNT=3.4 Budesonide/formoterolhad26.0%feweroralsteroidcourses and29.0%fewerantibioticcourses reducedriskofhospitalizationsduetoCOPDby29% and21.0%lowerriskforERvisitsinthebudesonide/formoteroltreatmentgroup AllhighlysignificantThemeancollectedbudesonidedosewas568µg/dayformatchedpatientswhowereprescribedbudesonide/formoterolandthemeanfluticasonedosewas783µg/dayforpatientswhowereprescribedfluticasone/salmeterol,correspondingto89%vs78%ofrecommendedlabeleddoseinCOPD;budesonide640µg/dayandfluticasone1000µg/d(delivereddose).F/S=fluticasone/salmeterol,T=tiotropium,B/F,budesonide/formoterol,NNT=numberneededtotreatfor1yearwithB/F+Tvs.F/S+Ttoavoidoneexacerbationorpneumoniaeventarateratiosandupperconfidencelimits<1.00favorB/F+Tvs.F/S+Tbrateratiosandlowerconfidencelimits>1.00favorB/F+Tvs.F/S+T Thesefigureshowscumulativeeventrates(pneumonias)/100patientsversustime. Comparedtobudesonide/formoteroltreatment,therewasa73%higherrateofpneumoniainthefluticasone/salmeteroltreatmentgroup Wecanseeauniformpatternversustimefor“allpneumonias”andfor“hospitalisedpneumonias”.Thedifferencesobservedforthebudesonide/formoterolvsthefluticasone/salmeterolgroupswithregardtodiagnosisofpneumoniawasindependentofwherethepneumoniadiagnosiswasrecorded,inprimarycareorathospital(67%ofallpneumoniaswerepneumoniasdiagnosedathospitalswhereX-rayisused). Comparedtobudesonide/formoteroltreatment,pneumonia-relatedhospitalisationswere74%higherinthefluticasone/salmeteroltreatmentgroup.NNTtoavoidonepneumoniaeventperyearof22Thefigureshowsthat100patientswillhavemorethan100pneumoniaduring9yearsinSAL/FLUtreatedpatients.Thefigureshowsnotthatsomepatientswillbewithoutpneumonia.TORCH:Usingananalysisofrates,itisestimatedtherewouldbeoneextracaseofpneumoniaforevery31patientsreceivingtreatmentwithSFCover1yr. Theresultsarepresentedasevents/100patients/year ThehazardratioforFLU/SALversusBUD/FORwas1.73,representinga73%increaseintheriskofpneumonia Itisestimatedtherewouldbeoneextracaseofpneumoniaforevery22patientsreceivingtreatmentwithFLU/SAL,NNT=22 FLU/SALpatientshadalso74%higherriskforpneumonia-relatedhospitalisationsversusthe布地奈德/福莫特罗group,NNT=32 Inthelast2columnsyoucanseethedistributionofthepneumoniadiagnosis. Thedifferencesobservedforthe布地奈德/福莫特罗vstheFLU/SALgroupswithregardtodiagnosisofpneumoniawasinthesamerangeindependentofwherethepneumoniadiagnosiswasrecorded 2,115patients(39%)hadatleastonepneumoniaduringthestudyperiod 1491pneumoniasinSeretidein752patients 997pneumoniasinSymbicotin565patients___________________________ Therewasan82%increaseinhospitaldaysinFLU/SALvs布地奈德/福莫特罗patients(53vs29days/100patienttreatment-years).SuchbigdifferenceinhospitaldayssinceFLU/SALgeneratedtotallymuchmoreevents(=hospitalisations).Themeandurationofeachpneumoniarelatedhospitalisationwashoweversimilar,7.1daysforSAL/FLUversus6.5for布地奈德/福莫特罗. Theexacerbationrateswereafactor6higherthanthepneumoniarates. Pneumoniaswerebasedonclinicaldiagnoses,i.e.ICD-10codefrommedicalrecords(inprimarycareorathospitalisation). Pneumoniaweremoreoftendiagnosedathospitals(60%),withaccesstoX-raythaninprimarycareTORCH:ThehazardratioforSFCversusplacebowas1.64(95%CI1.33–2.02),representinga64%increaseintheriskofpneumoniaTheincreaseinreportsofpneumoniaasanAEintheICScontainingtreatmentarmsoccurreddespiteadecreaseinoverallCOPDexacerbations,ofwhichpneumoniawasasubset.Therewerenoprotocol-definedcriteriaforpneumoniatoconfirmtheclinicaldiagnosis.AsforRCTs,investigatorsrecordedallAEs,includingpneumoniareports,inthecasereportform.IntheBigTORCHtrial,afteradjustingforotherriskfactors,therewasnoevidencethatcurrentsmokerswereatgreaterriskthanformersmokers,orthatsexwasariskfactor.TherewasnointeractionbetweenageandICSontheprobabilityofpneumonia. All-causemortalitydidnotdifferbetweenthegroupsalthoughthelinesseparatedmoreattheendoftheperiod(hazardratio1.08[95%CI0.93,1.14];P=0.593). PatientsintheSAL/FLUgrouphada76%higherriskforpneumonia-relatedmortalitythanthoseinthematched布地奈德/福莫特罗group(P=0.0025).SpeakerNotes: TherearetwohypothesisforthisobserveddifferencebetweenBUD/FORMandFLU/SALMwithrespecttopneumoniarates: FLUis10-foldmoreimmunosuppressivethanBUD.1 PharmocokineticdifferencesbetweenthemoleculesmayleadtoBUDhavingafasteruptakefromtheairwaysintothemucosa.BUDislesslipophilicthanFLU,dissolvingmorerapidlyinairwaymucusandmorerapidlyabsorbedintotheairwaytissueandsystemiccirculation.FLUishighlylipophilic/lesswatersolubleandmaystayintheairwayslonger.Becauseitremainsintheairwaysitismoreapttobeexpectoratedorclearedfromthesputum.2 BecauseFLUremainsintheairwayliningfluidlongeritmorepotentlyinhibitscytokinereleasefromlungepithelialcellssuppressingtheimmuneresponsetopneumoniacausingbacteria.1,2 BUDtakesminstodissolveintheairwaysurfaceliquidandbeabsorbed,FLUislesswatersolubleandtakes>8hourstodissolveintheliquidandbeabsorbed.3References: EkA,etal.Fluticasoneandbudesonideinhibitcytokinereleaseinhumanlungepithelialcellsandalveolarmacrophages.Allergy.1999;54(7):691-9. DalbyC,etal.Thebioavailabilityandairwayclearanceofthesteroidcomponentofbudesonide/formoterolandsalmeterol/fluticasoneafterinhaledadministrationinpatientswithCOPDandhealthysubjects:arandomizedcontrolledtrial.RespirRes.2009;10:104 HöggerP,etal.Biochemicalcharacterizationofaglucocorticoid-inducedmembraneprotein(RM3/1)inhumanmonocytesanditsapplicationasmodelsystemforrankingglucocorticoidpotency.PharmRes.1998;15:296-302.