Endovascular repair or open repair for ruptured abdominal aortic aneurysm: a Cochrane systematic review Badger, S. A., Harkin, D. W., Blair, P. H., Ellis, P. K., Kee, F., & Forster, R. (2016). Endovascular repair or open repair for ruptured abdominal aortic aneurysm: a Cochrane systematic review. BMJ Open, 6(2), e008391. https://doi.org/10.1136/bmjopen-2015-008391 Published in: BMJ Open Document Version: Publisher's PDF, also known as Version of record Queen's University Belfast - Research Portal: Link to publication record in Queen's University Belfast Research Portal Publisher rights Copyright 2016 the authors. This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial license (http://creativecommons.org/licenses/by-nc/4.0/) , which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. General rights Copyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or other copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated with these rights. Take down policy The Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made to ensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in the Research Portal that you believe breaches copyright or violates any law, please contact [email protected]. Download date:17. Jan. 2023 Open Access Research Endovascular repair or open repair for ruptured abdominal aortic aneurysm: a Cochrane systematic review S A Badger,1 D W Harkin,2 P H Blair,2 P K Ellis,2 F Kee,3 R Forster4 Tocite:BadgerSA, ABSTRACT Strengths and limitations of this study HarkinDW,BlairPH,etal. Objectives:Emergencyendovascularaneurysmrepair Endovascularrepairoropen (eEVAR)mayimproveoutcomesforpatientswith ▪ This is the first study to do a meta-analysis of repairforrupturedabdominal aorticaneurysm:aCochrane rupturedabdominalaorticaneurysm(RAAA).The the three major randomised controlled trials systematicreview.BMJOpen studyaimwastocomparetheoutcomesforeEVAR comparing endovascular and open repair of 2016;6:e008391. withconventionalopensurgicalrepairforthetreatment abdominalaorticaneurysms. doi:10.1136/bmjopen-2015- ofRAAA. ▪ The major finding is that there is no difference 008391 Setting:Asystematicreviewofrelevantpublications seen between operative modalities for ruptured wasperformed.Randomisedcontrolledtrials(RCTs) aneurysm repair in terms of morbidity or ▸ Prepublicationhistoryfor comparingeEVARwithopensurgicalrepairforRAAA mortality. thispaperisavailableonline. wereincluded. ▪ Thequalityofthestudyisreducedbythelackof Toviewthesefilesplease Participants:3RCTswereincluded,withatotalof completereportinginthetrials. visitthejournalonline 761patientswithRAAA. ▪ Long-term outcome data would enhance the (http://dx.doi.org/10.1136/ Interventions:Meta-analysiswasperformedwith resultsofthemeta-analysisandisstillawaited. bmjopen-2015-008391). fixed-effectsmodelswithORsand 95%CIsfor ▪ Robust conclusions are difficult to draw due to dichotomousdataandmeandifferenceswith95%CIs paucity of data, but one modality should not be Received10April2015 forcontinuousdata. favouredoveranotherbasedoncurrentresearch. Revised19August2015 Accepted21August2015 Primaryand secondaryoutcome measures: Primaryoutcomewasshort-termmortality.Secondary been recent decline in prevalence to about outcomemeasuresincludedaneurysm-specificand generalcomplicationrates,qualityoflifeandeconomic 2%. This is independent of participant selec- analysis. tion criteria and reflects better cardiovascu- Results:Overallriskofbiaswaslow.Therewasno lar risk profiling and management in the differencebetweenthe2interventionson30-day(or overall population.1 2 Unfortunately, many in-hospital)mortality,OR0.91(95%CI0.67to1.22; aneurysms first present as a rupture resulting p=0.52).30-daycomplicationsincludedmyocardial indeath inthe majorityof those affected. infarction,stroke,compositecardiaccomplications, ThehighmortalityratefromrupturedAAA renalcomplications,severebowelischaemia,spinal (RAAA) is well recognised. Clinicians have cordischaemia,reoperation,amputationand been reticent to apply scoring systems rigidly respiratoryfailure.Reportingwasincomplete,andno as to do so would condemn many to certain robustconclusionwasdrawn.Forcomplication death.3–5 Patients who undergo successful outcomesthatdidincludeatleast2studiesinthe open repair of RAAA enjoy a postoperative meta-analysis,therewasnoclearevidencetosupport adifferencebetweeneEVARandopenrepair.Longer quality of life similar to the ‘normal popula- 1MaterMisericordiae termoutcomesandcostperpatientwereevaluatedin tion’.6 7 The risk of aneurysm rupture is pro- UniversityHospital,Dublin, onlyasinglestudy,thusprecludingdefinite portional to aneurysm size, with aneurysms Ireland 2BelfastVascularCentre, conclusions. measuring less than 5.4cm having an annual RoyalVictoriaHospital, Conclusions:OutcomesbetweeneEVARandopen rupture rate of approximately 1% whereas Belfast,UK repair,specifically30-daymortality,aresimilar. those greater than 7.0cm in diameter have 3CentreforPublicHealth, However,furtherhigh-qualitytrialsarerequired,asthe anannualrupturerateof32.5%.89Currently, QueensUniversityBelfast, paucityofdatacurrentlylimitstheconclusions. rupture leads to death in over 80% of those Belfast,UK 4CentreforPopulationHealth affected, including 30–65% of those who Sciences,Universityof receiveconventionalopensurgicalrepairand Edinburgh,Edinburgh,UK is responsible for 2.1% of all deaths in men BACKGROUND over 65years.8 10–12 These findings contrast Correspondenceto Abdominal aortic aneurysm (AAA) affects with the significantly better outcome if con- DWHarkin; denis.harkin@belfasttrust. between 1.2% and 7.6% of the population ventional open surgical repair of the AAA is hscni.net over 50years of age in the UK. There has plannedbeforerupturecanoccur. BadgerSA,etal.BMJOpen2016;6:e008391.doi:10.1136/bmjopen-2015-008391 1 Open Access Historically, open surgical repair was the only effective that open repair will remain the treatment for a large treatment for AAA. This major operation carries signifi- proportion of patients whose AAA is unsuitable on ana- cant morbidity and mortality, due to the combined tomical grounds for EVAR.16 19 20 In addition, long-term effects of surgical exposure, haemorrhage and aortic results from the EVAR 1 trial should add caution, with clamping with related lower torso ischaemia-reperfusion later ruptures in the EVAR group, and no long-term dif- injury. However, with improved patient selection and ference in the all-cause mortality.21 perioperative care, some specialist centres are reporting Rupture of an AAA (RAAA) is a catastrophic event, mortality rates of less than 2% and surgeons in non- withupto80%mortality.2223RAAAexposesthepatients specialist units achieving mortality rates of 5–8%.8–10 to the combined injury of a period of haemorrhagic However, in the past two decades, this treatment has shock and lower torso ischaemia followed by a reperfu- been challenged by endovascular aneurysm repair sion injury on successful revascularisation. This ‘two hit’ (EVAR). Significant advancements in commercial stent mechanism of injury initiates a systemic inflammatory design, delivery and implantation technique have made response syndrome, leading to a multiple organ dysfunc- this avaluable alternative to open repair in selected suit- tion syndrome (MODS).24 MODS is the primary cause able cases.13–15 Its use in elective cases has reduced early of 70% of such deaths and a contributory cause of the postoperative morbidity and mortality.16 17 Its minimally remainder.25 26 The multiorgan failure, despite intensive invasive nature allows it to be performed under regional care support, once established leads to death in over or even local anaesthesia, rather than general anaesthe- 70% of cases.25 27 EVAR, especially under local anaesthe- sia. This allows repair in patients with significant con- sia, reduces the physiological insult to the body as com- comitant medical disease who may otherwise have been pared with conventional open surgical repair.28 29 considered unfit for surgery.10 18 Two recent large pro- Emergency EVAR (eEVAR) appears to offer a feasible spective randomised controlled trials (RCTs) haveshown alternative to conventional open repair in selected reduced early complications and mortality with patients.2930Emergencycaseswererepairedwithanaor- EVAR.16 17 However, while elective endovascular repair touniiliac stent graft, contralateral occluder and fem-fem clearly has a role, these trials reinforce the knowledge crossover in the initial years, due to increased speed of Figure1 PRISMAflowchartof literatureselection.TSC,Trials SearchCo-ordinator. 2 BadgerSA,etal.BMJOpen2016;6:e008391.doi:10.1136/bmjopen-2015-008391 Open Access achieving a seal and haemodynamic stability. However, electronic searches of MEDLINE, EMBASE, CINAHL, the use of bifurcated stent graft is routine for ruptured AMED and through hand searching relevant journals. AAA. Early reports have suggested a trend towards The full list of the databases, journals and conference reduction in perioperative morbidity and mortality in proceedings which have been searched, as well as the selected patients. Indeed evidence is emerging that the search strategies used are described in the Specialised inflammatory response to RAAA repair and the asso- Register section of the Cochrane Peripheral Vascular ciated organ dysfunction is attenuated by eEVAR com- Diseases Group module in The Cochrane Library.35 36 pared with open repair.31 32 References of relevant studies were reviewed for other The purpose of this review was to assess the available pertinent publications. evidence to support the use of eEVAR to treat RAAA. The advantages and disadvantages of eEVAR for patients Datacollection and extraction with RAAA was determined by the effect on short-term Two authors (SAB and RB) independently reviewed the mortality, major complication rates, aneurysm exclusion studies identified by the search for their relevance using and late complications when compared with patients the selection criteria. Disagreements were resolved who have had conventional open repair of RAAA. The through discussion. Two review authors (SAB and RB) results of this review have been published as a Cochrane independently extracted data for each included study. review,which was an update ofthe initial study.33 34 METHODS Table1 Thestringsearchperformedofthetermsinthe Only prospective RCTs comparing eEVAR with conven- literature tional open surgical repair were considered eligible for #1 MeSHdescriptor:[Ednovascular 6017 inclusion. Within the studies, all patients in whom a Procedures]explodealltrees RAAA has been clinically diagnosed by CT, angiography, #2 MeSHdescriptor:[Stents]explodealltrees 3314 #3 MeSHdescriptor:[VascularSurgical 652 MR angiography or objective acute symptoms suggestive Procedures]thistermonly of impending rupture of the aneurysm to warrant inclu- #4 Endovasc*:ti,ab,kw 941 sion. All types of endovascular devices were considered #5 Endostent*:ti,ab,kw 1 in comparison with conventional open surgical treat- #6 Endograft* 81 ment for patients consideredfit for surgery. #7 EVRAR:ti,ab,kw 1 The primary outcome for the study was the short-term #8 (EVARorREVAR):ti,ab,kw 74 morality (30-day, or in-hospital mortality). Secondary #9 (Palmaz):ti,ab,kw 91 outcomes included: #10 *stent*orgraft* 54489 ▸ Aneurysm exclusion, or further extravasation beyond #11 Palmaz:ti,ab,kw 91 the sac on follow-up imaging 30days after the #12 ViabahnorNitinolorHemobahnorIntracoil 207 procedure; orTantalumorpowerlinkorexcluderor ▸ Majorcomplications,suchasopenconversion,haemor- talentoraorfixorendologixoranaconda:ti, ab,kw rhage,myocardialinfarction,stroke,renalfailure(20% #13 ZenithorDynalinkorHemobahnor 128 rise in creatinine levels), respiratory failure (require- LuminexorMemothermorWallstent:ti,ab, ment for postoperative mechanical ventilation), pneu- kw monia,bowelischaemia,lowerlimbischaemia; #14 MeSHdescriptor:[BloodVesselProsthesis] 452 ▸ Minor complications, such as catheter site haema- explodeallterms toma, wound infection; #15 MeSHdescriptor:{BloodVesselProsthesis 508 ▸ Long-term complications and mortality; reinterven- Implantation]thistermonly tion rates for problems related to the RAAA or its #16 Endoprosthesis:ti,ab,kw 142 treatment will be sought where possible, as will cause #17 #1or#2or#3or#4or#5or#6or#7or#8 59170 of death, with or without reintervention, that is, or#9or#10or#11or#12or#13or#14or #15or#16 device related; ▸ Qualityoflife (standardised questionnaires); #18 MeSHdescriptor:[Aneurysm,Ruptured] 201 ▸ Economic analysis(cost per patient). explodeallterms #19 MeSHdescriptor:[Aneurysm,Dissecting] 73 thistermonly SEARCHMETHODS #20 (aneurysm*orabdom*orthoracoabdom*or 1195 thorac-abdom*oraort*)near(ruptur*or For this update, the Cochrane Peripheral Vascular tearorbleed*ortrauma) Diseases Group Trials Search Co-ordinator (TSC) #21 MeSHdescriptor:[Aorta]explodealltrees 317 searched the Specialised Register (last searched andwithqualifier(s):[Surgery—SU] February 2014) and the Cochrane Central Register of #22 RAAA 12 Controlled Trials (CENTRAL) 2014, Issue 2, part of The #23 #18or#19or#20or#21or#22 1545 Cochrane Library.35 The Specialised Register is main- #24 #17and#23inTrials 309 tained by the TSC and is constructed from weekly BadgerSA,etal.BMJOpen2016;6:e008391.doi:10.1136/bmjopen-2015-008391 3 Open Access Table2 ThedetailsoftheHinchliffestudy Methods Studytype:single-centre,randomisedcontrolledtrial,openlabel,intention-to-treat Studyaim:totestthehypothesisthatEVARcanreducetheperioperativemortalityof rupturedAAA,comparedwithopenrepair Country:England Setting:hospital Participants Numberrandomised:totaln=32(eEVARn=15;openrepairn=17) Age(medianyears(IQR)):eEVAR=74years(68.8–79.5);openrepair=80years(73.8–83.8) Gender(M/F):eEVAR=11/4;openrepair=13/4 Inclusioncriteria:allpatientsadmittedwithclinicallysuspectedorradiologicallyconfirmed ruptureofaninfrarenalabdominalaorticaneurysmthat,intheopinionoftheduty consultantvascularsurgeonwouldnormallybetreatedwithopenrepair Exclusioncriteria:noendovascularteamavailable;fullselectionofemergencystentgrafts notavailable;age<50years;inabilitytogiveverbalorwrittenconsent;unconsciouspatient; allergytoradiologicalcontrast,stainlesssteelorpolyester;severecomorbiditythatwould precludeintensivecaretreatmentfollowingopenrepair;previousendovascularAAArepair; womenofchildbearingpotentialnottakingcontraception;pregnantandlactatingwomen eEVARanatomicalsuitability(exclusioncriteria):absolutecontraindications:noevidenceon aneurysmrupture,juxtarenalaneurysm,neckdiameter>32mm,externaliliacartery diameter>6mm;relativecontraindications:proximalnecklength<10mm,excessive thrombusintheproximalneck,commoniliacarterylength<25mm,heavilycalcifiediliac arteries Interventions eEVARdescription:thosewithadiagnosticCTweretransferreddirectlytooperating theatre,andthosewithoutfirsthadaCTscantodetermineaorticmeasurement;performed indedicatedvascularoperatingtheatreusingaSiremobil2000imageintensifier,withdigital subtractionangiographyfacilities;mostpatientsheparinised;two-pieceaortouniiliacstent graftmadewithGianturcostentswithuncoveredsuprarenalcomponent;occludingdevice usedincontralateralcommoniliacartery;afterdeploymentofstentgraft,afemoro-femoral crossovergraftwasperformed Openrepairdescription:afterrandomisationtoopenrepair,patientsweretransferred directlytotheoperatingtheatre,accordingtolocalpractice;performedtransperitoneally eitherbymidlineortransverseincisions;aortaclampedbelowrenalarteries;patientsno heparinised;inlaytechniquewasusedandgraftsweregelatin-coatedpolyester Outcomes Perioperativemortality,definedas30-dayorin-hospital Notes ‘PatientsweredeemedsuitableforEVARif,intheopinionoftheoperatingsurgeon,they couldperformtherepair’;participantsrecruitedSeptember2002toDecember2004;five surgeonsonunit,requiredthatsurgeonandteamavailablehadsufficientexpertisetooffer EVAR,ifnot,conventionalopenrepairwasoffered;unstablepatientsthatmightbe disadvantagedbydelayincurredbyCTscancould,atthesurgeon’sdiscretion,notbe randomisedandtakendirectlyforopenrepair Riskofbias Bias Authors’ Supportforjudgement judgement Randomsequencegeneration Unclear ‘Randomisationwasthenperformedfromsealedopaqueenvelopeskeptinthe (selectionbias risk AccidentandEmergencyDepartment’.Unclearhowrandomisationsequence wasgenerated Allocationconcealment Lowrisk ‘Randomisationwasthenperformedfromsealedopaqueenvelopeskeptinthe (selectionbias) AccidentandEmergencyDepartment’ Blindingofparticipantsand Lowrisk Studywasunblinded,duetonatureofinterventionbutunlikelytoinfluence personnel(performancebias); outcomes.‘Thesurgeonswereblindedtothedimensionsofpatient’saorta alloutcomes untilrandomisationhadtakenplace’toavoidbias Blindingofoutcome Lowrisk Notpossibletoblindteamregardingallocationgroup,butunlikelytoinfluence assessment (detectionbias);all outcomemeasures outcomes Incompleteoutcomedata Lowrisk Allpatientsaccountedfor;crossoverpatientsaccountedfor;similardropout (attritionbias);alloutcomes ratesandreasonsbetweentreatmentgroups Selectivereporting(reporting Lowrisk Mostoftheprotocoloutlinedinthetext;allrelevantoutcomesreported;with bias) theexceptionofmortality,outcomesarenotwelldescribedinthemethods Otherbias Unclear Underpoweredstudy:32oftherequired100patientsrecruited risk AAA,abdominalaorticaneurysm;eEVAR,emergencyendovascularaneurysmrepair;F,female;M,male. 4 BadgerSA,etal.BMJOpen2016;6:e008391.doi:10.1136/bmjopen-2015-008391 Open Access Table3 ThedetailsoftheAmsterdamAcuteAneurysmTrial(AJAX)study Methods Studytype:multicentre,randomisedcontrolledtrial,intention-to-treat Studyaim:compareEVARandopenrepairintreatingRAAAonmortalityand severecomplications Country:TheNetherlands Setting:threelargehospitalvascularcentresinAmsterdam Participants Numberrandomised:totaln=116(eEVARn=57;openrepairn=59) Age(meanyears,95%CI):eEVAR=74.5(72.3to77.5);openrepair=74.5(72.2to 76.8) Gender(M/F):eEVAR=49/8;openrepair=50/9 Inclusioncriteria:maleandfemalesover18years;clinicaldiagnosisofRAAA; aneurysmaccompaniedbyacutehaemorrhageoutsideoftheaorticwallofCTA; suitableforeEVARandopenrepair Exclusioncriteria:extensionoftheaneurysmtojuxtarenalorsuprarenalaorta; kidneytransplant;horseshoekidney;allergytointravenouscontrast;connective tissuedisease;severehaemodynamicinstabilityprohibitingCT eEVARanatomicalsuitabilityrequirements:suitableinfrarenalanchoringsegment, minimumlengthoftheinfrarenalsegmentofatleast10–15mm,infrarenaldiameter of20–32mm,noobstructingcalcifications,tortuosityofthrombosis,suitableiliac anchoringsegment,ipsilateraliliacdiameterof8–18mm,contralateraliliacdiameter of10–20mm,atleastoneiliacarteryshouldbeabletoaccommodateanendograft CVDriskfactors(n(%)):diabetes(EVARn=2(4%),openrepairn=1(2%)); hypertension(EVARn=13(23%),openrepairn=10(17%));smoker(EVARn=23 (40%),openrepairn=20(34%));hyperlipidaemia(EVARn=13(23%),openrepair n=19(32%));renaldisease(EVARn=1(2%),openrepairn=2(3%));pulmonary disease(EVARn=7(12%),openrepairn=3(5%));carotiddisease(EVARn=16 (28%),openrepairn=10(17%));cardiacdisease(EVARn=16(28%),openrepair n=14(24%))TypeofRAAA:infrarenal Interventions eEVARdescription:aortouniiliacendograftandcontralateraliliacoccludingdevice, followedbyafemoro-femoralcrossoverbypassgraft Openrepairdescription:midlinelaparotomyandexclusionofruptureaneurysmby eitherpolyestertubeorbifurcatedgraft;conductedundergeneralanaesthesia Outcomes Compositedeathandseverecomplicationsat30daysafterintervention;long-term mortalityrates(6monthsafterrandomisation);lengthofhospitalandICUstay; durationofintubation/ventilation;useofbloodproducts;forEVAR,occurrenceof endoleaks Notes Studyperiod:April2004toFebruary2011;threemaintrialcentres,allother(seven) regionalhospitalstransferredparticipantstooneofthetrialcentres Riskofbias Bias Authors’ Supportforjudgement judgement Randomsequencegeneration Lowrisk ‘Therandomizationsequencewasgeneratedbyanindependent (selectionbias) clinicalresearchunitusingALEAsoftwareforrandomizationin clinicaltrialswitha1:1allocationusingrandomblocksizesof4or 6,stratifiedforeachparticipatingcentre’. Allocationconcealment (selectionbias) Lowrisk ‘Allocationwasconcealedusingsequentiallynumberedopaque sealedenvelopes’. Blindingofparticipantsandpersonnel Lowrisk Surgicalteamnotpossibletoblind,butunlikelytoinfluence (performancebias);alloutcomes outcomes Blindingofoutcomeassessment Lowrisk Doubledatabaseentry;endpointadjudicationcommitteeblinded; (detectionbias);alloutcomes independentsafetycommitteeblinded Incompleteoutcomedata(attrition Lowrisk AllpatientsaccountedforinCONSORTdiagram;bothtreatment bias);alloutcomes groupshadsimilardropoutratesandreasons Selectivereporting(reportingbias) Unclearrisk AlthoughallpredescribedoutcomeslistedintheReimerink publicationarereported,qualityoflifeandcost-effectivenessas describedinthestudyprotocol,werenotincluded,suggesting furtherpublicationstoemerge Otherbias Lowrisk None CVD,cardiovasculardisease;eEVAR,emergencyendovascularaneurysmrepair;F,female;ICU,intensivecareunit;M,male;RAAA, rupturedabdominalaorticaneurysm. BadgerSA,etal.BMJOpen2016;6:e008391.doi:10.1136/bmjopen-2015-008391 5 Open Access Table4 ThedetailsoftheImmediateManagementofthePatientwithRupture:OpenVersusEndovascular(IMPROVE) study Methods Studytype:multicentre,randomisedcontrolledtrial,openlabel,intention-to-treat Studyaim:toassesswhetherEVARvsopenrepairreducesearlymortalityforpatientswith suspectedRAAA Country:UKandCanada Setting:30hospitalvascularunitsandspecialistcentres Participants Numberrandomised:totaln=613(eEVARn=316;openrepairn=297) Age(meanyears(±SD)):eEVAR=76.7(7.4);openrepair=76.7(7.8) Gender(M/F):eEVAR=246/70;openrepair=234/63 Inclusioncriteria:menandwomenovertheageof50years;clinicaldiagnosisofRAAAor rupturedaortoiliacaneurysm,madebyaseniortrialhospitalclinician Exclusioncriteria:previousaneurysmrepair;ruptureofanisolatedinternaliliacaneurysm, aortocavaloraortoentericfistulae;recentanatomicalassessmentoftheaorta;connective tissuedisorder;ifinterventionwasconsideredfutile eEVARanatomicalsuitabilityrequirements:noabsoluterequirementswillbesetforthe study,butproximalneckmorphologywithadiameterexceeding32mmoralengthlessthan 10mmmaybeconsideredunfavourable,andiliacarterydiametersshouldbeintherangeof 8–22mm CVDriskfactors(n(%)):notgiven TypeofRAAA:‘rupturedAAAorrupturedaortoiliacaneurysm’ Interventions eEVARdescription:endovascularsupracoeliacaorticballoonocclusionwillbeusedto supportlessstablepatients;mostinterventionsperformedwithaortouniiliacgraft,butsome patientsreceivedbifurcatedgrafts,withsubsequentfemoro-femoralcrossovergraftwith contralateraliliacocclusion;controlofaortaachievedusinglocal/regionanaesthesia,with generalanaesthesiausedlaterinprocedureifnecessaryOpenrepairdescription:CTscanis optional;aneurysmsrepairedbycross-clampingtheproximalaortaandinsertingaprosthetic inlaygraft;performedundergeneralanaesthesia Outcomes 30-daymortality,24handin-hospitalmortality,costs,reinterventionsatprimaryadmission timeandplaceofdischarge;cost-effectivenessandmortalityat12monthsareplannedfor futurereporting Notes ParticipantsrecruitedSeptember2009toJuly2013;flowdiagramshows623randomised, but10wereexcludedafterDataMonitoringCommitteereviewedparticipants,613usedin analysis;only275(87%)ofEVARand261(88%)ofopenrepairhadconfirmedRAAA Riskofbias Bias Authors’ Supportforjudgement judgement Randomsequencegeneration Lowrisk ‘Andindependentcontractorprovidedtelephonerandomisation,with (selectionbias) computer-generatedassignationofpatientsina1:1ratio,usingvariableblock sizeandstratifiedbycentre’. Allocationconcealment Lowrisk ‘Andindependentcontractorprovidedtelephonerandomisation,withcomputer (selectionbias) generatedassignationofpatients…’ Blindingofparticipantsand Lowrisk Surgicalteamnotpossibletoblind,butunlikelytoinfluenceoutcomes personnel(performancebias); alloutcomes Blindingofoutcome Lowrisk Dataverificationperformedcentrallyatthetrialcorelaboratory,unclearif assessment (detectionbias); blinding,butunlikelytoinfluenceoutcomes alloutcomes Incompleteoutcomedata Lowrisk Allpatientsaccountedfor,withbothtreatmentgroupshavingsimilardropout (attritionbias);alloutcomes rates/reasoning Selectivereporting(reporting Unclearrisk Allpredescribedoutcomesreportedon,butnotallappropriateoutcomes bias) reportedyet,suggestingfurtherpublicationstoemerge Otherbias Lowrisk None CVD,cardiovasculardisease;eEVAR,emergencyendovascularaneurysmrepair;F,female;M,male;RAAA,rupturedabdominalaortic aneurysm. Details about the trial design, characteristics of partici- exclusion, major and minor short-term complications, pants, diagnosis of RAAA, eEVAR and open repair pro- long-term all-cause mortality and long-term complica- cedures were recorded. Data were collected on the tions, quality of life, and economic analysis. If data were primary outcome short-term mortality (30-day or missing from publications of the included studies, in-hospital) and secondary outcomes: aneurysm attempts weremade tocontact studyauthors. 6 BadgerSA,etal.BMJOpen2016;6:e008391.doi:10.1136/bmjopen-2015-008391 Open Access Figure2 Short-termmortality (30-dayorin-hospital)of emergencyendovascular aneurysmrepair(eEVAR)versus openrepair. Assessment of riskof bias in included studies Acute Aneurysm Trial (AJAX) and Hinchliffe trials only Included studies were evaluated for quality, independ- used aortouniiliac devices, while the Immediate ently by two review authors (SAB and RB), using The Management of the Patient with Rupture: Open Versus Cochrane Collaboration’stool for assessing riskof bias.36 Endovascular (IMPROVE) used both methods, with a This tool provides judgements made on the domains of greater proportion being aortobiiliac.38–48 However, due sequence generation, allocation concealment methods, to a paucity of information, subgroup analysis was not blinding, incomplete outcome data, selective outcome possibleatthistime. reporting and other relevant biases. Evaluations of low risk, unclear risk or high risk were given for each of the Sensitivityanalysis six domains for each included study. Any disagreements Although all the participants in the IMPROVE trial had betweenreviewauthorswereresolvedthroughdiscussion. a clinical diagnoses of RAAA, on start of the interven- tion, it was found that only 536 (87%) of the 613 rando- Measuresof treatment effect mised participants had, in fact, a ruptured AAA.46 48 Of Analysis was planned on an intention-to-treat basis, and the remaining 77 participants, 22 had symptomatic non- therefore all randomised patients from the included ruptured AAA, or other final diagnoses. Sensitivity ana- studies were to be included in analysis. The outcomes lyses were planned to evaluate the effects of this trial on that are dichotomous in nature were to be compiled the outcomes. into meta-analysis and ORs with 95% CIs were to be cal- culated. For continuous data, meta-analysis would provide mean differenceswithSDs. RESULTS A total of 18 new citations were identified in the search. Assessment of heterogeneity Four were considered not relevant as they were reviews, A test for heterogeneity examines the null hypothesis and two were added to the excluded studies.49 50 One that all studies are evaluating the same effect. We study, Endovasculaire ou Chirurgie dans les Anevrysmes obtained p values comparing the test statistic with a χ2 aorto-iliaques Rompus (ECAR), fits the inclusion criteria, distribution. A measure (I2) that describes the percent- but is currently ongoing, with no published data at this age of total variation across studies due to heterogeneity time.51–53 Ten citations, from three studies, were newly rather than by chance was included. Avalue of 0% indi- included.38–48 Five prospective comparative studies were cates no observed heterogeneity, and larger values show excluded.29 49 50 54 55 The flow of included and excluded increasing heterogeneity.37 studiesisillustratedbymeansofaPRISMAdiagram(figure 1).Thesearchstringresultsarepresentedintable1. Data synthesis The three included studies had a total of 761 partici- Dataextractedindependentlybytworeviewauthors(SAB pants.38–48 All three studies were RCTs comparing and RB) was compiled and entered into RevMan by one eEVAR to open surgery repair in patients with a clinical author (RB). Comparisons of data using meta-analyses diagnosis of RAAA on outcomes that included mortality were undertaken using fixed-effects models, unless the and complications. AJAX and IMPROVE aimed to evalu- testforheterogeneityyieldedavalue>50%,inwhichcase ate longer term mortality and complications, but at the arandom-effectsmodelwasimplemented. time of this review, there was no long-term data pub- lished from the IMPROVE trial. Also, the IMPROVE trial Subgroup analysis and investigation of heterogeneity evaluated cost-effectiveness by comparing cost-per- Weplannedtoperformsubgroupanalysistoevaluatethe patient between the two trial arms. Currently none of impact of patients treated with aortouniiliac devices and the included studies have evaluated aneurysm exclusion, those treated with aortobiiliac devices. The Amsterdam minorcomplications or qualityof life. Figure3 Myocardialinfarction ofemergencyendovascular aneurysmrepair(eEVAR)versus openrepair. BadgerSA,etal.BMJOpen2016;6:e008391.doi:10.1136/bmjopen-2015-008391 7 Open Access Figure4 Strokecomplicationofemergencyendovascularaneurysmrepair(eEVAR)versusopenrepair. AJAX and IMPROVE were both multicentre studies, Blinding(performance bias and detection bias) with AJAX taking place in Amsterdam, the Netherlands, Owing to the nature of the intervention, it is not pos- and IMPROVE in the UK, with one studysite in Canada. sible to blind the surgeons and difficult to blind patients Hinchliffe et al was a single-centre trial within England. and the research team to the treatment allocation. All included participants had a clinical diagnosis of However, we determined that a lack of blinding is RAAA,butintheIMPROVEstudyonly536outofthe613 unlikely to influence the outcomes of interest, and all (87%) randomised participants actually had RAAA, the three have been listed as low risk for performance and remaining 77 had symptomatic non-ruptured AAA (22), detection bias. Attempts were made to reduce the riskof noAAA(10),orasymptomaticAAAoranotherfinaldiag- bias by the study authors: in the AJAX study, an end nosis (45). For the AJAX study, all randomised partici- point adjudication committee and independent safety pantswereconsideredsuitableforbotheEVARandopen committee, both blinded to treatment allocation, were repair,whichintheHinchliffeandIMPROVEstudiessuit- utilised. The Hinchliffe study kept surgeons blinded to abilityforeEVARwasdeterminedafterrandomisation.In dimensions of the aorta until randomisation was com- theHinchliffeandAJAXstudies,aortouniiliacgraftswere pleted, and the IMPROVE study utilised a trial core used in the endovascular trial arm, while the IMPROVE laboratory tocentrally verifyoutcomes. trialusedboththisand,morepredominately,aortobiiliac grafts. The data derived from the three studies are out- Incomplete outcome data (attrition bias) and selective linedintables2–4withqualityofdatadescribed. reporting (reporting bias) Five studies were excluded from this review. Three All three included studies adequately accounted for all studies were prospective trials, treating patients present- ing with RAAA with eEVAR.29 54 55 However, their com- participants, giving thorough explanation of all dropout rates and reasons. All three studies were at a low risk of parison to open repair was made through retrospective, ‘historical controls’ or with open repair cohorts. One attrition bias. Both the AJAX and IMPROVE trials are lacking important outcome data specified in their proto- study was a prospective comparison between eEVAR and cols, which include quality of life and costs for both open repair in patients with RAAA, but the study was non-randomised.49 A final study was a non-randomised studies, and complications and long-term survival for the studyof 55 consecutive patients presenting with RAAA.50 IMPROVE study. It is assumed that these outcomes will be reported in future publications, but until that data A portion of the participants in the study were collected are reported, the two studies are at an unclear risk of retrospectivelyand a portion prospectively. reporting bias. The Hinchliffe study reported on all out- comesspecified. Allocation (selection bias) Both the AJAX and IMPROVE studies adequately reported random sequence generation, but the Otherpotential sources of bias Hinchliffe study did not give a description of how their The AJAX and IMPROVE studies appear to be free of sequence was produced, and is therefore at unclear risk. other sources of bias, but the Hinchliffe study could be For allocation concealment, all three studies clearly at riskof bias due to an underpowered study population. explained adequate concealment methods. In the publication, it was reported that the study Figure5 Cardiaccomplication(moderateorsevere)ofemergencyendovascularaneurysmrepair(eEVAR)versusopenrepair. 8 BadgerSA,etal.BMJOpen2016;6:e008391.doi:10.1136/bmjopen-2015-008391 Open Access Figure6 Renalcomplications(moderateorsevere)ofemergencyendovascularaneurysmrepair(eEVAR)versusopenrepair. required 100 participants to be adequately powered, yet Renal complications (moderate or severe) theyonly included 32 patients. The AJAX and Hinchliffe studies reported renal compli- cations, and the random-effects model used to analyse the association found no clear difference between the EFFECTS OF INTERVENTIONS interventions, OR 0.70 (95% CI 0.09 to 5.24; p=0.73; Mortality (30-dayor in-hospital) figure6). All three studies were included in the meta-analysis for mortality (30-day or in-hospital). The fixed-effects Severe bowel ischaemia model found no clear evidence to support a difference Only a single study evaluated severe bowel ischaemia in mortality between eEVAR and open repair, OR 0.93 (95% CI 0.69 to 1.25; p=0.63; figure 2). When the (AJAX), which had an OR of 0.39 (95% CI 0.07 to 2.11), but as only one study was included, no overall IMPROVE study was removed for sensitivity analysis, due association could be determined (figure7). to randomisation of patients that did not have RAAA, the OR moved closer to one and the CI became wider Spinal cord ischaemia as the IMPROVE study had a larger study population Spinal cord ischaemia was also only evaluated in the than the other two included studies, OR 0.98 (95% CI AJAX study, which had only one event. With an OR of 0.49 to1.95; p=0.95). 3.16 and a 95% CI of 0.13 to 79.17, very little can cur- Complications (30-day) rently be concluded regarding this outcome (figure 8). Myocardial infarction Reoperation Only the Hinchliffe study reported myocardial infarc- tion, but only one event was reported so the CI is very The occurrence of reoperation, specific to the aneurysm wide, OR 3.62 (95% CI 0.14 to 95.78; figure 3). As only repair, was reported in two studies (AJAX and one study was included for this outcome, no overall asso- Hinchliffe). The fixed-effects model did not find clear ciation could be determined, and thewide CI makes any evidence to support a difference between the interven- conclusion spurious. tions, OR 0.89 (95% CI 0.39 to 2.01; p=0.78), again with a rather wide CI (figure9). Stroke Both the AJAX and Hinchliffe studies reported stroke Amputation events, but with very few events, and opposing findings, The AJAX trial was the only study to evaluate amputa- the fixed-effects model, with a non-significant OR has a tion, and with the few events, all in the open repair very wide CI that is difficult to derive any meaningful intervention group, the OR of 0.14 (95% CI 0.01 to conclusion from (OR 0.71, 95% CI 0.12 to 4.31; p=0.71; 2.78) had awide CI and no association to be concluded, figure4). as onlya single study was included (figure10). Cardiac complications (moderate or severe) Respiratory failure Cardiac complications were evaluated in the AJAX and Respiratory failure was evaluated in the Hinchliffe study, Hinchliffe studies. The fixed-effects meta-analysis found alone. With only a single event in the eEVAR arm, the a non-significant OR, but again with awide CI (OR 1.12, CI was very wide, OR 3.62 (95% CI 0.14 to 95.78) with 95% CI 0.38 to 3.30; p=0.84; figure 5). no overall association (figure 11). Figure7 Bowelischaemiaofemergencyendovascularaneurysmrepair(eEVAR)versusopenrepair. BadgerSA,etal.BMJOpen2016;6:e008391.doi:10.1136/bmjopen-2015-008391 9
Description: