An international survey on the use of thromboprophylaxis in urological surgery

(1)

Epidemiology

An International Survey on the Use of Thromboprophylaxis in Urological Surgery

Philippe D. Violette

^a,b

, Robin W.M. Vernooij

^c,d

, Yoshitaka Aoki

^e

, Arnav Agarwal

^f

, Rufus Cartwright

^g,h

, Yoichi Arai

ⁱ

, Thomas Tailly

^j

, Giacomo Novara

^k

, Tejan Baldeh

^a

, Samantha Craigie

^a

, Rodney H. Breau

^l

, Gordon H. Guyatt

^a

, Kari A.O. Tikkinen

^m,

*

aDepartmentofHealthResearchMethods,EvidenceandImpact,McMasterUniversity,Hamilton,ON,Canada;^bDepartmentofSurgery,DivisionofUrology, McMasterUniversity,Hamilton,ON,Canada;^cJuliusCenterforHealthSciencesandPrimaryCare,UniversityMedicalCenterUtrecht,UtrechtUniversity, Utrecht,TheNetherlands;^dDepartmentofNephrology& Hypertension,UniversityMedicalCenterUtrecht,Utrecht,TheNetherlands;^eDepartmentof Urology,UniversityofFukuiFacultyofMedicalSciences,Fukui,Japan;^fFacultyofMedicine,UniversityofToronto,Toronto,ON,Canada;^gDepartmentof EpidemiologyandBiostatistics,ImperialCollegeLondon,London,UK;^hDepartmentofUrogynecology,OxfordUniversityHospitalsNHSTrust,Oxford,UK;

iDepartmentofUrology,TohokuUniversityGraduateSchoolofMedicine,Sendai,Japan;^jDivisionofUrology,DepartmentofSurgery,UniversityHospital Ghent,Ghent,Belgium;^kDepartmentofSurgical,Oncological,andGastroenterologicalSciences,UrologyClinic,UniversityofPadua,Padua,Italy;^lDivisionof Urology,TheOttawaHospitalResearchInstituteandUniversityofOttawa,Ottawa,ON,Canada;^mDepartmentsofUrologyandPublicHealth,Universityof HelsinkiandHelsinkiUniversityHospital,Helsinki,Finland

a v ai l a b l e a t w w w . s c i e n c e d i r e c t . c o m

j o u r n al h o m e p a g e : w w w . e u r o p e an u r o l o g y . c o m / e u f o c u s

Articleinfo

Articlehistory:

AcceptedMay27,2020 AssociateEditor:MalteRieken

Keywords:

Bleeding

Deepveinthrombosis Guideline

Hemorrhage Practicevariation Prevention

Pulmonaryembolism Surgery

Thromboprophylaxis Thrombosis

Urology

Venousthromboembolism

Abstract

Background: The use of perioperative thromboprophylaxis in urological surgery is commonbutnotstandardized.

Objective: Tocharacterizeinternationalpracticevariationinthromboprophylaxisusein urologicalsurgery.

Design,setting,andparticipants: Weconductedascenario-basedsurveyaddressingthe useofmechanicalandpharmacologicalthromboprophylaxisinurologicalcancerpro- cedures(radicalcystectomy[RC],radicalprostatectomy[RP],andradicalnephrectomy [RN])amongpracticingurologistsinCanada,Finland,andJapan.Thesurveypresented patientproﬁlesreﬂectingaspectrumofriskforvenousthromboembolism;therespon- dentsdescribedtheirclinicalpractice.

Outcomemeasurementsandstatisticalanalysis: Theproportionofrespondentswho routinelyused(1)mechanical,(2)pharmacological,and(3)extendedpharmacological prophylaxiswasstratiﬁedbyprocedure.Alogisticregressionidentiﬁedcharacteristics associatedwiththromboprophylaxisuse.

Resultsandlimitations: Of1051 urologistscontacted,570 (54%) participated inthe survey.Japaneseurologistswerelesslikelytoprescribepharmacologicalprophylaxis thanCanadianorFinnishurologists(p<0.001forallprocedures).CanadianandFinnish urologistsexhibitedlargevariationforextendedpharmacologicalprophylaxisforRPand RN. Finnish urologists were most likely to prescribe extended prophylaxis versus CanadianandJapanese urologists(RC 98%,84%,and26%;Open RP25%,8%,and 3%;

roboticRP11%,9%,and0%;andRN43%,7%,and1%,respectively;p<0.001foreach procedure).Lessvariationwasfoundregardingtheprescriptionofmechanicalprophy- laxis,which was mostcommonly useduntilambulationordischarge.Thelengthof hospitalstaywaslongerinJapanandmaybiasestimatesofextendedprophylaxisin Japan.

*Correspondingauthor.DepartmentofUrology,HelsinkiUniversityHospital,Haartmaninkatu4, Helsinki00029,Finland.Tel.+358-50-5250971.

E-mailaddress:[email protected](KariA.O.Tikkinen).

https://doi.org/10.1016/j.euf.2020.05.015

creativecommons.org/licenses/by-nc-nd/4.0/).

(2)

1. Introduction

Venous thromboembolism (VTE), including deep vein thrombosisandpulmonaryembolism,representsaserious andsometimesfatalcomplicationofsurgery[1].Pharmaco- logical prophylaxis decreases the relative risk of VTE in surgical patients by approximately 50%, but with an increaseintherelativeriskofpostoperativemajorbleeding of50%[2].Therefore,thedecisiontousepharmacological prophylaxispresentsatradeoffbetweenareductioninVTE andanincreaseinbleeding[3].

Anadditionalchallengeregardingtheuseofthrombo- prophylaxis in urology is international clinical practice variation[4–6].Alackofknowledgeofevidenceregarding theprocedure-andpatient-specificbaselinerisksofthrom- bosisandbleeding,criticalinmakinganinformeddecision on the use of thromboprophylaxis, may,at least inpart, explainthisvariation[1,2,7].Conflictingrecommendations from differentguidelines andpreviouslack of guidelines specifictothedifferenturologicalsurgeriesmayalsocon- tribute[3,8].Tofurtherunderstandinternationalvariation inthromboprophylaxisuseinurologicalsurgery,wecon- ductedaninternationalsurvey(InternationalSurveyonUse ofThromboprophylaxisinUrologicalSurgery[ISTHMUS]).

Our goal was to characterize the within- and between- countryvariationinthromboprophylaxisuseforcommon urologicalprocedures.

2. Patientsandmethods 2.1. Survey

Wedesignedaquestionnairethatconsistedofthreescenariosintended toelicitpracticeinprescribingbothpharmacologicalandmechanical thromboprophylaxisforradicalcystectomy(RC),openandroboticradi- calprostatectomy(RP),andradicalnephrectomy(RN;Table1).Usingthe patientriskfactormodelusedintheEuropeanAssociationofUrology (EAU)guidelineonthromboprophylaxisinurologicalsurgery[3],the surveypresentedbriefpatientproﬁlesthatreﬂectedaspectrumofVTE risks (Table 1).Table 2 presents the degree of VTE risk, based on systematicreviews[2],associatedwitheachscenario.Thisinformation wasnotincludedinthesurveyandthuswasunavailabletoparticipants.

Participantsindicatedtheirusualpracticeforeachpatientproﬁleusing responseoptionsconsistingofsingle-answermultiplechoices(Table1).

Respondentsalsoprovidedtheirage,gender,andurologistproﬁle(resident/consultant).Agroupofcliniciansandmethodologistsgeneratedthe itemsinthequestionnaire,whichwasthenpilottestedandreviewed withagroupof20board-certiﬁed urologicalsurgeonsfrom Canada, Finland,andJapanwhoassesseditsfacevalidity.

Weperformedthissurveybeforethefirstprocedure-andpatient- specificthromboprophylaxisguidelineinurology—theEAUguideline— waspublished[3].WeinvitedactivelypracticingurologistsfromCanada, Finland,andJapantocompletethesurvey.InCanada,weinvitedCana- dianurologistsattendingtheannualmeetingoftheCanadianUrological Association(CUA),heldinOttawainJune2015.InFinlandandJapan,we identifiedurologistsfromtheregistersoftheFinnishUrologicalAssoci- ation andthe Japanese UrologicalAssociation.In Canada, urologists completedthesurveyattheconclusionoftheplenarysessionsandat breakperiodsthroughouttheconference.Wecollectedsurveysimme- diatelyuponcompletionandusedthenumberofurologistsattending themeetingasthedenominatorforcalculatingtheresponserate(theuse ofthenumberofurologistsattendingthemeetingasthedenominator likely underestimates the response rate). All urologists in Finland receivedaninvitationtoparticipatebyapostalletter.InJapan,asample of500urologists(ofwhom487provedtobeeligible)wererandomly identifiedfromthenationalmembershipdirectoryandinvited,bymail, toparticipate.WemailedthequestionnairesbetweenAugustandSep- tember 2015 (first round),and mailed tworounds of remindersin October(secondround)andDecember2015(thirdround).Furthermore, Conclusions: Wefoundlargevariationinclinicalpracticeregardingpharmacological thromboprophylaxiswithinandbetweencountries.Knowledgetranslationofevi- dence-basedguidelinesmayreduceproblematicinternationalvariationinpractice.

Patientsummary: Useofmedicationstodecreasebloodclotsafterurologicalcancer surgerydifferswithinandbetweencountries.Closeradherencetourologyguidelines addressingthepreventionofbloodclotsmaydecreasethisvariationandimprove patientoutcomes.

by-nc-nd/4.0/).

Table1–Scenariostestedinthesurveywithresponseoptions.

Scenario1(cystectomy):A65-yr-oldwomanwhohasaBMIof36anda personalhistoryofVTEbutisotherwisehealthyisundergoinganopen radicalcystectomy.Ifyouwereprescribingthromboprophylaxisforthis patient,whatregimenwouldyouchoosemostcommonly?

Scenario2(prostatectomy):A58-yr-oldmanwithaBMIof23andnopersonal orfamilyhistoryofVTEisundergoingaradicalprostatectomywithout lymphadenectomy(respondentswereaskedtoselectanopenorarobotic approach).Ifyouwereprescribingthromboprophylaxisforthispatient,what regimenwouldyouchoosemostcommonly?

Scenario3(nephrectomy):An80-yr-oldmanwhohasaBMIof24andno personalorfamilyhistoryofVTEisundergoinganopenradical

nephrectomy.Ifyouwereprescribingthromboprophylaxisforthispatient, whatregimenwouldyouchoosemostcommonly?

Responseoptions:

Pharmacologicalprophylaxis:

1.Nopharmacologicalthromboprophylaxis

2.Pharmacologicalthromboprophylaxisuntilthepatientisambulating 3.Pharmacologicalthromboprophylaxisforthedurationofhospitalstay 4.Pharmacologicalthromboprophylaxisforthedurationofhospitalstay and2wkafterdischarge

5.Pharmacologicalthromboprophylaxisforthedurationofhospitalstay

and>2wkafterdischarge

Mechanicalprophylaxis:

1.Nomechanicalthromboprophylaxis

2.Mechanicalthromboprophylaxisuntilthepatientisambulating 3.Mechanicalthromboprophylaxisforthedurationofhospitalstay 4.Mechanicalthromboprophylaxisforthedurationofhospitalstayand2 wkafterdischarge

5.Mechanicalthromboprophylaxisforthedurationofhospitalstayand>2 wkafterdischarge

BMI=bodymassindex;VTE=venousthromboembolism.

(3)

wesentane-mailbeforeeachroundtoinformurologistsaboutthe upcomingmailsurvey.

InCanada,theCUAboardofdirectors,whoareindependentofstudy investigators,reviewedtheproposalandapprovedtheadministrationof surveyattheCUAannualmeetingwithoutrequirementforfurtherethics review.InFinland,theethicscommitteeoftheHelsinkiandUusimaa HospitalDistrictgrantedexemptionfrom ethicalreview (R11110).In Japan,theethicscommitteeoftheUniversityofFukui(Fukui,Japan;

#20150067)approvedtheprotocol.Thereportingofthestudyconforms totheSTROBEstatement[9].

2.2. Statisticalanalysis

Foreverypatientprofile,wecalculatedthe proportionofuseof(1) mechanical prophylaxis, (2) pharmacological prophylaxis, and (3) extendedpharmacologicalprophylaxisof2wkandtestedforstatistical significanceusingchi-squareanalysis. Asthelength ofhospitalstay variessubstantiallybetween countries(considerablylongerin Japan thaninCanadaorFinland;SupplementaryTables1and2),andtherefore theimplicationofprophylaxisduringhospitalstaydiffersacrossjur- isdictions(Supplementarymaterial),definitionsofextendedprophylaxis correspondedtodurationofhospitalstayinthethreecountries(Table1 andSupplementaryTable3).Weusedmultivariablelogisticregression adjustedfortheincludedcountriestodeterminewhethertheurologist profile(resident/consultant)wasassociatedwithanymechanicalpro- phylaxis,anypharmacologicalprophylaxis,orextendedprophylaxis.We reporttheoddsratioand95%confidenceintervalforthemultivariable logisticregressionusingathresholdpvalueof<0.05.Allanalyseswere performedinSPSSversion20(IBMCorp.,Armonk,NY,USA).

3. Results

The570urologistswhoparticipatedinthissurvey(Table3) representanoverall responserateof54%(Canada:216of 385,57%;Finland:110of179,61%;andJapan:244of487, 50%).Ofthe570eligibleindividualswhorespondedtothe survey, 566 (99%) responded to at least five out of six

scenario questions. Most participants were men (89.6%

[511/570]); the median age of the participants was 43.0yr(interquartilerange:35–54)withamajoritybeing consultants(82.5%[470/570]).

Almostallrespondentsusedmechanicalprophylaxisfor patients undergoingRC (Canada: 92%[199/216], Finland:

96%[106/110],Japan:96%[234/244],p=0.13;Table4).More respondents reported usingmechanical prophylaxis after anopenRPinJapan(98%[239/244])thaninFinland(88%

[158/179])andCanada(79%[171/216],p<0.001).Similarly, afteraroboticRP,theJapaneseurologistsreportedahigher rate ofmechanical prophylaxis(100%)thanthe Canadian (81% [175/216])andFinnishurologists(75% [83/110];p<

0.001). Japanese urologists also reported more use of mechanicalprophylaxisafteranRN(95%[232/244])com- paredwithCanadian(84%[181/216])andFinnish(80%[88/

110],p<0.001).

Useofpharmacologicalprophylaxisofanydurationafter RC was more commonin Canada and Finland (both 99%

[208/210,109/110])thaninJapan(70%[171/244],p<0.001;

Table 4). FewerJapaneserespondentsindicatedthatthey wouldusepharmacologicalprophylaxisafteranRP(open RP:14%[34/244],roboticRP:33%[81/244])comparedwith Canadian (open RP: 93%[201/216],robotic RP: 90%[194/

216])andFinnish(openRP:88%[97/110],roboticRP:94%

[103/110]) respondents (p < 0.0001 for both open and robotic RP). Similarly, more respondents reported using pharmacological prophylaxis afteranRN in Finland (95%

[105/110])andCanada(92%[199/216])thaninJapan(20%

[49/244],p<0.001).

FinnishandCanadian respondentsweremorelikelyto use extended pharmacological prophylaxis after RC (Finland:98%[108/110],Canada:84%[181/216])thanJapa- neserespondents(26%[63/244],p<0.001).Itwasuncom- mon for respondents in any country to use extended

Table3–Baselinecharacteristicsoftherespondents.

Canada Finland Japan Total

Response(%) 216/385(57) 110/179(61) 244/487(50) 570(54)

Male(%) 194(89.8) 91(82.7) 226(92.6) 511(89.6)

Female(%) 19(8.8)^a 19(17.3) 18(7.4) 56(9.8)

Age(yr),median(IQR) 38.0(33–50) 48.0(30–57.5) 45.0(38–55) 43.0(35–54)

Consultant(%) 152(70.4) 97(88.2) 221(90.6) 470(82.5)

Resident(%) 62(28.7) 13(11.8) 22(9.0) 97(17.0)

IQR=interquartilerange.

a ThreemissingvaluesregardinggenderintheCanadiansample.

Table2–VTEandmajorbleedingriskestimatesofthepresentedpatientprofiles.

Scenario Operation Patientriskstrata^a RiskofVTE(%)^b Riskofmajorbleeding(%)^c

1 Openradicalcystectomy Highrisk 11.6 0.3

2 Openradicalprostatectomywithoutlymphadenectomy Lowrisk 1.0 0.1

2 Roboticradicalprostatectomywithoutlymphadenectomy Lowrisk 0.2 0.4

3 Openradicalnephrectomy Mediumrisk 2.2 0.1

VTE=venousthromboembolism.

a PatientriskstratificationaccordingtotheEuropeanAssociationofUrologyguidelineonthromboprophylaxisinurologicalsurgery[3].

b VTEdefinedassymptomaticpulmonaryembolusordeepveinthrombosiswithin30dofsurgery[2].

cMajorbleedingdefinedasbleedingrequiringreoperationwithin30dofsurgery[2].

(4)

pharmacologicalprophylaxisafteropenRP(Japan:3%[7/

244],Canada:8%[17/216],Finland:25%[28/110];p<0001) and robotic RP (Japan: 0% [0/244], Canada: 9% [19/216], Finland:11%[12/110];p=0.02).Finnishrespondents(43%

[47/110])weremorelikelytouseextendedpharmacological prophylaxisafterRNthanCanadian(7%[15/216])orJapa- nese(1%[2/244])respondents(p<0.001).

Multivariableanalysis,adjustedforthecountry,demon- stratednodifference inresidentversusconsultant urologists’ use of mechanical, pharmacological, or extended pharmacologicalprophylaxis(SupplementaryTable4).

4. Discussion

Thisisthefirststudyinurologyto examineinternational variabilityintheuseofthromboprophylaxisbetweencoun- triesindifferentcontinents.Thislarge-scalemultinational surveyidentifiedlargevariationintheuseofpharmacolog- icalVTEprophylaxisduringhospitalstayandintheuseof extendedprophylaxisafterdischargewithinandbetween countries.CanadianandFinnishurologistsreportedsimilar frequentuseofpharmacologicalprophylaxis,butJapanese respondentsreportedfarlessuse.Reporteduseofmechan- ical prophylaxis for urological procedures had muchless variation, bothwithinandbetweencountries.Finally,we foundnodifferencesbetweenresidentandconsultanturol- ogists in the use of mechanical, pharmacological, or extendedpharmacologicalprophylaxis.

4.1. Strengthsandlimitations

The strengths of the current study include a population representative of the target populations of Canadian, Finnish, and Japanese urologists, in age and gender distribution[10–12].Our study’sstrengths alsoinclude a satisfactoryparticipationrateandveryhighcompleteness

ofquestionnaireresponses.Weassessedtheuseofthrom- boprophylaxisbyclinicalcasescenariosrelevanttocurrent clinicalpractice.

Our study has limitations. First, as this survey was intended to capture an overview of practice pattern, we are not able to draw conclusion about the causesof the practicevariation.Wedidnotaskrespondentsabouttheir perceptionoftheincidenceofVTE—perceptionofalower riskinJapancouldberesponsibleforaloweruseofphar- macological, but not mechanical, prophylaxis. Whether such differences in incidence actuallyexist hasnot been established[2].Second,howurologistsactuallypracticeina hospital setting might differ from their responses to the scenariospresentedinourstudy.Inaddition,asthrombo- prophylaxis is often prescribed by more junior doctors ratherthan bythe operatingsurgeon,we didnot collect data on thespecialization ofrespondents. It is uncertain whether this approach biased estimates. Third, we have used differentsamplingstrategiesfor Canada from those used for Finland andJapan. This pragmatic decision was basedon ourgoalofincludingarepresentativesamplein each country. Fourth, many of the respondents do not perform some of the procedures presented. They may, however, stillparticipate inthe prescriptionof thromboprophylaxis. Fifth, we assessed only a limited range of surgeriesand scenarios.Fifth, the lengthof hospitalstay was longer in Japan, which may influence the use of extended prophylaxis. Finally, generalization to jurisdic- tions beyond the three that we included in this survey remainsuncertain.

4.2. Comparisonwithotherstudies

Therearefewearliersurveysexaminingtheuseofperiop- erative VTEprophylaxis. ABritishsurvey [5], on amixed population of respondents including 29 consultant Table4–Proportion(inpercentages)ofthosewithanydurationofmechanicalprophylaxis,anydurationofpharmacologicalprophylaxis, andpharmacologicalprophylaxisofatleast2wkbyprocedureandcountry.

Openradicalcystectomy Openradicalprostatectomy

Mechanical,any%(95%CI) Pharmacological Mechanical,any%(95%CI) Pharmacological Any

%(95%CI)

Extended

%(95%CI)

Any

%(95%CI)

Extended

%(95%CI)

Canada 92(87–95) 99(96–100) 84(78–88) 79(70–86) 93(86–97) 8(4–15)

Finland 96(89–98) 99(94–100) 98(93–100) 88(67–97) 88(67–97) 25(11–47)

Japan 96(93–98) 70(61–73) 26(21–32) 98(92–99) 14(9–22) 3(1–8)

Roboticradicalprostatectomy Openradicalnephrectomy

Mechanical,any%(95%CI) Pharmacological Mechanical,any%(95%CI) Pharmacological Any

%(95%CI)

Extended

%(95%CI)

Any

%(95%CI)

Extended

%(95%CI)

Canada 81(72–88) 90(82–95) 9(4–17) 80(74–85) 92(87–95) 7(4–12)

Finland 75(62–84) 94(84–98) 11(5–21) 84(76–90) 95(89–98) 43(34–53)

Japan 100(98–100) 33(23–46) 0(0–2) 95(91–97) 20(15–25) 1(0–3)

CI=conﬁdenceinterval.

(5)

urologists,30residents,and35urologyclinicalnursespe- cialistspracticingin64UKpelviccancercenters,foundthat allunitsusedlow–molecular-weightheparin(LMWH)pro- phylaxisroutinelyfortheinpatientperiodafterRCand98%

used perioperative prophylaxis after RP. Routine use of LMWH for all patients after discharge following RC was reportedin67%andafterRPin61%ofunits(investigators didnotdifferentiatebetweenapproaches,suchasopenor robotic)[5].

AUSdatabasestudyof94709menwhounderwentRP (72%open and28%robotic) foundthat52%ofmen after openRPreceivedmechanicalonly,7%pharmacologicalonly, and11%bothmechanicalandpharmacologicalprophylaxis, and30%receivednoprophylaxis[6].Dischargeprophylaxis wasnot includedin the scopeof this study. AnotherUS study[4]surveyedthemembersoftheAmericanUrological Associationin2011.Only11% of thoseinvitedresponded (1210respondents),ofwhomapproximately70%forRCand

<60%forRPreportedtheuseofanythromboprophylaxis

“frequently”or “always”[4]. Theauthorsdidnot address postdischargeprophylaxis.Althoughsmallsamplesize[5], retrospectivedatabasedesign[6],andlowresponserate[4]

limitthestrengthofinferencefromthesestudies,consis- tentwithourfinding,thereremainslittledoubtofconsid- erablevariationinpracticeofthromboprophylaxisinuro- logicalproceduresbothwithinandbetweencountries.

4.3. Implicationsoffindings

Thepracticevariationthatwe identifiedislikely dueto severalfactors. First, the quality ofevidence supporting decisionsregardingVTEprophylaxisisnotofhighquality and thus open tovariable interpretation [2,13]. Second, thelow-qualityevidencemayalsoexplainthedivergent recommendations fromseveral major VTE guidelines in different countries [8]. For instance, prominent VTE guidelinesfromtheAmericanCollegeofChestPhysicians andNationalInstituteforHealthandCareExcellencedo notconsider specific urologicalsurgeries separately but wouldconsiderallscenariosinthesurveyas“abdomino- pelvic”[14,15].Third,urologistsinsomeregionsmayalso preferto defer decisions regardingthe use of thromboprophylaxis to colleagues from other disciplines with a differentweighingoftrade-offbetweenbleedingandVTE fromanonsurgeonpointofview.Possibly,VTEriskmay alsovaryby population, givendifferencesin bodymass indexorageatsurgerythatmaypredominateindifferent regions [14], with choices of prophylaxis appropriately reflectinglocalrisk.

Weperformedthesurveypriortothepublicationofthe systematicreviewsaddressingestimatesofabsoluterisksof symptomaticVTEandmajorbleedinginurologicalsurgery [2,13]andanEAUguidelineaddressingthromboprophylaxis [3], the first procedure and patient risk factor–specific guidelineinurology.Therefore,respondentscouldnothave beeninfluencedbytheseguidelinerecommendations,and thissurveycanbeusedasabenchmarkfortrackingchanges inpracticepatternwithfurtherdisseminationofurology- specific VTE guidelines. Healthcare professionals across

jurisdictionsmayconsider,intheirclinicaldecisionmaking, this evidence-based, procedure-specific guideline [3], which hasalso beenpublished asa freelyavailableinfo- graphic [15]. Doingso wouldrationalizethe practiceand maydecreasethissubstantialvariation.

InpatientsathighriskofVTE,forsomemajorsurgeries, there is a clear net benefit to extended pharmacological thromboprophylaxis [2,3,16]. Our resultssuggest that for high-riskVTEandlowbleedingrisk(scenario1:cystectomy for high-riskpatients,Tables1and2),practice inCanada andFinlandfollowstheEAUguidelinemoreclosely[3].

In scenario2 (RPwithoutlymphadenectomy for low- riskpatients;Tables1and2),theEAUguidelinesuggests the use of extended pharmacological prophylaxis in the open approach butrecommends against the use of prophylaxis inroboticprostatectomywithoutlymphadenec- tomy [3]. In our survey, participants from Canada and Finlandtypicallyreportedtheuseofpharmacologicalpro- phylaxisinhospitalbutnotafterdischargeirrespectiveof the approach(openvsrobotic),whereasonein sevenof Japanese urologists after an open approach and one in three after a robotic approach reported in-hospital use ofpharmacologicalprophylaxis.

In scenario 3 (open RN for medium-risk patients;

Tables 1 and 2), respondents’ practice in this situation provedtobelessalignedwithEAUguidelinerecommenda- tions [3]: Canadian and Finnish respondents tended to reporttheuseofin-hospitalprophylaxisbutnotdischarge prophylaxis, whileJapaneserespondentstypicallydidnot reportanyuseofpharmacologicalprophylaxis.

4.4. Unansweredquestionsandfutureresearch

Variationinpracticesuggestslimitedadherencetoclinical guidelines, which should represent the best practice for mostpatients.Therefore,effortstoincreasefamiliaritywith EAUguidelines[3]—rigorouslyevidence-basedandthefirst procedure-specific guideline—may improve patient care.

PerioperativeVTEguidelinesdependonaclearunderstand- ingofpatient-importanttrade-offsandthequalityofsup- portingevidence.Therefore,futureeffortstoimprovethe body of evidence, especially with the conduct of high- quality trials and observational studies, will result in improved clinical guidance and patient care[3,17,18]. In addition, future surveys are needed to monitor changes in practice regarding perioperative thromboprophylaxis resultingfrompublicationofnewguidance.

5. Conclusions

Weperformedalarge-scalemultinationalsurveyandiden- tified alargevariationin theuseof pharmacologicalVTE prophylaxis within andbetweencountries.This variation existed in the use of both pharmacological prophylaxis during hospital stay andextended prophylaxis after discharge. The variation in the reported use of mechanical prophylaxisforurologicalprocedureswasmuchless,with uniformlyhighuseover75%acrossallscenarios.Knowledge translation of evidence-based guidelines may reduce

(6)

problematicvariationinpracticegloballyandthusmayhelp optimizefuturepatientcare.

Authorcontributions:KariA.O.Tikkinenhadfullaccesstoallthedatain thestudyandtakesresponsibilityfortheintegrityofthedataandthe accuracyofthedataanalysis.

Studyconceptanddesign:Violette,Agarwal,Cartwright,Tailly,Novara, Craigie,Guyatt,Tikkinen.

Acquisitionofdata:Violette,Aoki,Agarwal,Arai,Breau,Tikkinen.

Analysisandinterpretationofdata:Allauthors.

Draftingofthemanuscript:Violette,Vernooij,Guyatt,Tikkinen.

Criticalrevisionofthemanuscriptforimportantintellectualcontent:All authors.

Statisticalanalysis:Vernooij.

Obtainingfunding:Tikkinen.

Administrative,technical,ormaterialsupport:None.

Supervision:Tikkinen.

Other:None.

Financialdisclosures:KariTikkinencertifiesthatallconflictsofinterest, includingspecificfinancialinterestsandrelationshipsandaffiliations relevanttothesubjectmatterormaterialsdiscussedinthemanuscript (eg,employment/affiliation,grantsorfunding,consultancies,honoraria, stockownershiporoptions,experttestimony,royalties,orpatentsfiled, received,orpending),arethefollowing:Violettehasreceivedanhono- rariumfromJanssen,Sanofi,andAstellaspriortoJanuary2017.Taillyhas beenaconsultantforCookMedical.Novarahasbeenanadvisoryboard member and speaker for Astellas, GlaxoSmithKline, Lilly, Menarini, Nycomed,Pfizer,PierreFabre,andRecordati.Vernooij,Aoki,Agarwal, Cartwright,Arai,Baldeh,Craigie,Breau,Guyatt,andTikkinenhaveno financial conflicts of interest. Other potential conflicts of interests:

Violette, Novara, Cartwright, and Guyatt were panel members and Tikkinenwasthe chairman ofthe European Association ofUrology (EAU)GuidelineonThromboprophylaxisinUrologicalSurgery.

Funding/Supportandroleofthesponsor:TheISTHMUSprojectwas conductedbytheClinicalUrologyandEpidemiology(CLUE)Working Group supported bythe Academy ofFinland (276046 and309387), CompetitiveResearchFundingofthe HelsinkiandUusimaaHospital District(TYH2016135,TYH2017114,TYH2018120,andTYH2019321),and SigridJuséliusFoundation.PhilippeD.Violettewasalsosupportedbythe CanadianUrologicalAssociationScholarshipFoundationandRufusCart- wrightbyaResearchTrainingFellowshipfromtheUKMedicalResearch Council.Thesponsorshadnoroleintheanalysisandinterpretationof thedataorinthemanuscriptpreparation,review,orapproval.

Acknowledgements:WewouldliketothanktheTiffanyPizioliandthe CanadianUrologicalAssociationadministrativeteam,secretaryofthe FinnishUrologicalAssociationSamiRaatikainen,andresearch nurses KatjaKiianlinnaandMerjaRignell,aswellastheJapaneseUrological AssociationBoard ofDirectors, SyunsakuTsukui,andEikoKawaifor facilitatingthesurveys.WewouldalsoliketothankJosephChin,Jon Izawa,SannaMyrskysalo,NicholasPower,JukkaSairanen,andHenrikki Santtiforadviceandsupportonestimatingthetimingofambulation, durationofhospitalstay,andrepresentativenessofthestudypopulation.

AppendixA. Supplementarydata

Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.

1016/j.euf.2020.05.015.

References

[1]AndersonDR,MorganoGP,BennettC,etal.AmericanSocietyof Hematology2019guidelinesformanagementofvenousthrombo- embolism: prevention of venous thromboembolism in surgical hospitalizedpatients.BloodAdv2019;3:3898–944.

[2]TikkinenKAO,CraigieS,AgarwalA,etal.Procedure-speciﬁcrisksof thrombosisandbleedinginurologicalcancersurgery:systematic reviewandmeta-analysis.EurUrol2018;73:242–51.

[3]TikkinenKAO,CartwrightR,GouldMK,etal.EAUguidelineson thromboprophylaxis in urological surgery.http://uroweb.org/

guideline/thromboprophylaxis/.

[4]SteriousS,SimhanJ,UzzoRG,etal.Familiarityandself-reported compliance with American Urological Association best practice recommendationsforuseofthromboembolicprophylaxisamong AmericanUrologicalAssociationmembers.JUrol2013;190:992–8.

[5]PridgeonS,AllchorneP,TurnerB,PetersJ,GreenJ.Venousthrom- boembolism (VTE)prophylaxisandurologicalpelviccancersur- gery:aUKnationalaudit.BJUInt2015;115:223–9.

[6]WeinbergA,WrightJ,DeibertC,etal.Nationwidepracticepatterns fortheuseofvenousthromboembolismprophylaxisamongmen undergoingradicalprostatectomy.WorldJUrol2014;32:1313–21.

[7]TikkinenKA,AgarwalA,CraigieS,etal.Systematicreviewsofobserva- tionalstudiesofriskofthrombosisandbleedinginurologicalsurgery (ROTBUS):introductionandmethodology.SystRev2014;3:150.

[8]ViolettePD,CartwrightR,BrielM,TikkinenKA,GuyattGH.Guide- lineofguidelines:thromboprophylaxisforurologicalsurgery.BJU Int2016;118:351–8.

[9]von Elm E, AltmanDG, Egger M, etal. The Strengthening the ReportingofObservationalStudiesinEpidemiology(STROBE)state- ment: guidelines for reporting observational studies. Lancet 2007;370:573–7.

[10] CanadianUrologicalAssociation(CUA)MembershipDirectoryfor 2017.https://www.cua.org/en/membership-directory.

[11] NationalSupervisoryAuthorityforWelfareandHealth(Valvira), Finland.Registersofsocial welfareandhealthcareprofessionals.

https://julkiterhikki.valvira.ﬁ/#.

[12] MinistryofHealth,LabourandWelfare,Japan.Surveyofphysicians, dentistsandpharmacists.https://www.mhlw.go.jp/toukei/saikin/

hw/ishi/16/index.html.

[13] TikkinenKAO,CraigieS,AgarwalA,etal.Procedure-speciﬁcrisksof thrombosisandbleedinginurologicalnon-cancersurgery:system- aticreviewandmeta-analysis.EurUrol2018;73:236–41.

[14] GouldM,GarciaD,WrenS,etal.PreventionofVTEinnonortho- pedicsurgicalpatients:antithrombotictherapyandpreventionof thrombosis,9thed:AmericanCollegeofChestPhysiciansevidence- basedclinicalpracticeguidelines.Chest2012;141,e227S–77S.

[15] NationalInstituteforHealthandCareExcellence.Venousthrom- boembolisminover16s:reducing theriskofhospital-acquired deep vein thrombosis or pulmonary embolism. NICE guideline [NG89].https://www.nice.org.uk/guidance/ng89.

[16] HeitJA,SpencerFA,WhiteRH.Theepidemiologyofvenousthrom- boembolism.JThrombThrombolysis2016;41:3–14.

[17] ClinicalUrologyandEpidemiologyWorkingGroup.Infographicof theEuropeanAssociationofUrology(EAU)guidelineonthrombo- prophylaxis inurologicalsurgery. http://clueworkinggroup.com/

2017/12/01/thromboprophylaxis-infographic/.

[18] TikkinenKariAO,GuyattGordonGH.Baselinerisksofvenous thromboembolismandmajorbleedingarecrucialinthrombopro- phylaxisdecision-making.EurUrol2020.http://dx.doi.org/10.1016/

j.eururo.2020.05.032.