Evaluation of bleeding risk in patients on anticoagulation for mechanical cardiac valve operated for benign prostatic obstruction

(1)

Disponibleenlignesur

ScienceDirect

www.sciencedirect.com

ORIGINAL ARTICLE

Evaluation of bleeding risk in patients on anticoagulation for mechanical cardiac valve operated for benign prostatic obstruction

Évaluation du risque hémorragique chez les patients opérés pour hypertrophie bénigne de prostate, sous anticoagulants pour valve cardiaque mécanique

S. Gardic

^a^,∗

, V. Misrai

^b

, A.R. Azzouzi

^c

, A. Campeggi

^d

, J.-N. Cornu

^e

, A. De La Taille

^f

, S. Lebdai

^c

,

R. Mathieu

^g

, G. Robert

^h

, A. Descazeaud

^a

aDepartmentofurology,universityhospitalDupuytren,2,avenueMartin-Luther-King, 87000Limoges,France

bDepartmentofurology,cliniquePasteur,45,avenuedeLombez,31300Toulouse,France

cDepartmentofurology,universityhospital,4,rueLarrey,49100Angers,France

dDepartmentofurology,Médipôlecenter,3,rueDocteur-Joseph-Audic,56000Vannes,France

eDepartmentofurology,universityhospital,1,ruedeGermont,76000Rouen,France

fDepartmentofurology,universityhospitalHenri-Mondor,51,avenuedu Maréchal-de-Lattre-de-Tassigny,94010Créteil,France

gDepartmentofurology,universityhospital,2,rueHenri-le-Guilloux,35000Rennes,France

hDepartmentofurology,universityhospitalPellegrin,placeAmélie-Raba-Léon, 33076Bordeaux,France

Received18September2016;accepted25May2017 Availableonline24June2017

KEYWORDS Prostate;

Vaporization;

TURP;

Anticoagulation;

Bleeding

Summary

Objective.—Toevaluate bleedingriskinpatientsonanticoagulationfor mechanicalcardiac valveoperatedforbenignprostaticobstruction(BPO).

Materialandmethod.—Fifty-eightpatientsoperatedbetween1998and2014,insevenFrench departments ofUrology were included. Forty-ﬁve patients were operated by conventional surgery(transurethralresectionoftheprostate38,opensimpleprostatectomies7),and13 patientswereoperatedbyGreenlight^TM photovaporizationoftheprostate(PVP).Inorderto assessbleedingrisk,bloodtransfusionwasconsideredastheprimaryoutcome.

∗Correspondingauthor.

E-mailaddress:solenegardic@gmail.com(S.Gardic).

http://dx.doi.org/10.1016/j.purol.2017.05.012

(2)

Results.—Fifteen(26%)patientsreceivedbloodtransfusioninthepostoperativeperiod.Mean durationofhospitalizationwas8.5days.Secondarysurgerywas requiredin12cases (21%), includingendoscopic clot removal under general anaesthesia in10 patients, and suprapu- bichaemostasisin3patients.Onepatientdied72hoursaftertransurethralresectionofthe prostatebecauseofamassivepulmonaryembolism.Twoindependentpredictorsofbloodtrans- fusionwereidentified:conventionalsurgeryuseversusPVP,andhighpreoperativePSA.Blood transfusionratewassignificantlylowerinthegroupofpatientsoperatedbyPVPcomparedto conventionalsurgery(0%versus33%,P=0.010).Inaddition,thelasersurgerywasassociated withshorterdurationofhospitalization(3.4daysversus9.9days,P=0.014).However,itwas notfoundanysignificantdifferencebetweenpatientsoperatedbyPVPcomparedtoconven- tionalsurgeryintermsofsecondarybleeding(3/13vs8/45,P=0.7),orsecondsurgery(2/13 vs10/45,P=0.5).

Conclusion.—BleedingriskofBPOsurgeryinpatientswithmechanicalcardiacvalveishigh.

ThePVPseemstodecreasesigniﬁcantlytheearlyhaemorrhagicriskcomparedtoclassicsurgical proceduresforpatientswithmechanicalcardiacvalve.

Levelofevidence.— 4.

MOTSCLÉS Prostate; Vaporisation; RTUP;

Anticoagulants; Saignement

Résumé

Objectif.—Évaluerlerisquehémorragiquedes patientssous anticoagulantspourvalvecar- diaquemécanique,opérésd’unehypertrophiebénignedeprostatique(HBP).

Matérieletméthode.—Cinquante-huit patients opérés entre 1998 et 2014 dans 7services franc¸aisd’urologieontétéinclus.Quarante-cinqavaientétéopérésparchirurgiediteconven- tionnelle(38RTUP,7adénomectomiesprostatiquesparvoiehaute)et13avaientétéopéréspar vaporisationprostatiquelaserGreenlight^TM(PVP).Notrecritèreprincipaldejugementétaitla transfusionsanguinetémoignantdurisquehémorragique.

Résultats.—Vingt-cinqpatients(26%)furenttransfusésenpostopératoire.Laduréemoyenne d’hospitalisation était de 8,5jours. Douze patients ont bénéficié d’une reprise chirurgi- cale(21 %), repriseregroupant des décaillotages par endoscopie sous anesthésie générale (10 patients),et par voiesus-pubienne pour 3 patients. Un patient décéda d’uneembolie pulmonairemassive72heuresaprèsuneRTUP.Deuxfacteursprédictifsindépendantsdetrans- fusionsanguinefurentidentifiés:lachirurgieconventionnelle(versusPVP)etuntauxélevé dePSAavantchirurgie.Letauxdetransfusionsanguineétaitsignificativementplusbasdans le groupe de patients opérés par PVP versus chirurgie conventionnelle (0 % versus 33 %, p=0,010).De plus,la chirurgiepar laser était associéeà unedurée d’hospitalisation plus courte (3,4jours versus 9,9jours, p=0,014).En termes de saignements secondaires, il n’a pas été mis en évidence de différence entre les deux groupes de patients (3/13 PVP vs 8/45chirurgieconventionnelle,p=0,7).Cemêmeconstats’appliqueauxrepriseschirurgicales (2/13vs10/45,p=0,5).

Conclusion.—Le risque hémorragiquede la chirurgiepourHBP dansce groupe depatients estimportant.Lavaporisationlasersemblediminuersigniﬁcativementlerisquehémorragique immédiatcomparé auxtechniques chirurgicales classiqueschez lespatients porteursd’une valvecardiaquemécanique.

Niveaudepreuve.— 4.

Introduction

So far, transurethral resection of the prostate (TURP) remains thetreatment of referencefor the surgical man- agementofbenignprostaticobstruction(BPO)[1,2].Laser techniques were developed aiming to reduce morbidity [3,4]. It was shown that Greenlight^TM photovaporization

of the prostate (PVP) improves the quality of haemosta- sis as compared to TURP [5,6]. The number of patients requiring anticoagulation is continuously growing: almost 2% of the French population is under oral coumarin derivatives [7]. Between 2000 and 2012, the use of oral coumarin derivatives doubled up in France. In a recent series of patients operated for BPO, about the third

(3)

of them were under oral anticoagulation prior surgery [8].

Among patients under anticoagulation, those with mechanical cardiac valves require permanent, deep and rigorous anticoagulation. Therefore, they are commonly consideredatveryhighriskforBPOsurgery.Toourknowl- edge,noprevious assay wasspeciﬁcallydedicated tothis speciﬁcgroupofpatients.Thepresentstudyaimedtoassess bleedingriskinpatientsonanticoagulationformechanical cardiacvalveoperatedforBPO.

Materials and methods

Weconductedanobservational,descriptive,retrospective, multicentric study.Seven French Departments of Urology, academicor non-academic,participatedinthisstudy.The researchperiodrangedfrom1998to2014.Inordertoassess bleedingrisk, postoperativeblood transfusionwasconsid- eredasthemainoutcomeparameter.Itwasdeﬁnedasred cellstransfusionduringthehospitalizationfollowingthesur- gical procedure. Blood transfusion was decided based on haemoglobin: less than 7g/dL if no coronarydisease was present,andlessthan10g/dLincaseofcoronarydiseaseor badclinicalstatus.

A multivariable analysis was conducted to determine independentfactorsofbloodtransfusion.

Statistical analysis:theSPSS20.0 software(IBMCorpo- ration, USA) was used for statistical analysis. Univariate andmultivariateanalysiswasperformedusingadowncon- ditional logisticregression method.A P-value of0.05 was consideredasstatisticalsigniﬁcantthreshold.

Results

Fifty-eightpatientswithmechanicalcardiacvalvesoperated forBPOwereidentiﬁedinthesevenparticipatingcentres.

The surgicaltechniqueusedfor the58remainingpatients wasPVP,openprostatectomy,andmonopolarTURP(mTURP) in13,38,and7patients,respectively.

Preoperative parameters and perioperative management of oral anticoagulation

The Table 1 summarizes general and urological data of patients. Four different type of anticoagulation management were observed during the perioperative period (Table 2). Two patients did not stop oral anticoagulation in the perioperative period. They were both operated by PVP.Theirpreoperativeinternationalnormalizedratiowere below2.5atthetimeofsurgery.

In patientsundercontinuousintravenousstandard heparin, anticoagulationwasstoppedfourhoursprior surgery and restarted 2hours after the end of the procedure. In patientsunderlowmolecularweightheparinorheparincal- cium,onlytheinjectionthemorningofthedayofsurgery wascancelled.

Directly after the surgery, the anticoagulation is re- taken.Tensionisappliedtotheurinarycatheterincaseof massivebleeding.

Table1 Preoperative parameters in 58 patients with mechanicalcardiacvalvesoperatedforBPO.

Preoperativeparameters 58patients

Meanage(y.o) 72.3(55—86,SD=8.2)

MeanASAscore 2.7

Meanprostaticvolume(mL) 57.2(11—150,SD=29.4) PreoperativePSA(ng/mL) 4.3(0.38—22,SD=4.5) Prostatictreatments

Alpha-blockers,n(%) 46(79) 5-alphareductase

inhibitor,n(%)

15(25) Phytotherapy,n(%) 10(17) Hormonotherapy,n(%) 1(2)

Qmax(mL/s) 6.8(2—15,SD=3.4) Bacteriuriaonpreoperative

urinalysis,n(%)

9(16) Preoperativeurinary

catheter,n(%)

29(50) Positionofthemechanical

valve

Aortic,n(%) 51(88)

Mitral,n(%) 5(10)

Aorticandmitral,n(%) 2(3) Coumarinderivatives

Fluindione,n(%) 47(81) Warfarin,n(%) 9(16) Acenocoumarol,n(%) 2(3) Antiplateletdrugs

Aspirin,n(%) 8(14)

Aspirin+clopidogrel,n(%) 0(0) Clopidogrel,n(%) 1(2)

SD:standarddeviation;ASA:AmericanScoreofAnaesthesiology.

Table2 Anticoagulationmanagementinthepreopera- tiveperiod.

58patients Timewithoutcoumarin

derivativesbeforesurgery (days)

5.7(0—15;SD=2.5)

MeanpreoperativeINR 1.26(1—2.10;SD=0.25) MeanpreoperativeCephalin

Timetest

1.45(1—2.07,SD=0.32) Preoperativerelayoftheoral

anticoagulant

Lowmolecularweight heparin,n(%)

25(43) Intravenousstandard

heparin,n(%)

9(16) Heparincalcium,n(%) 22(38) Oralanticoagulation

continued,n(%)

2(3) Resumeoforal

anticoagulationaftersurgery (indays)

20.2(0—80;SD=16.2)

INR:Internationalnormalizedratio;SD:standarddeviation.

(4)

Table3 Outcomeparameters.

58patients Meanminimalhaemoglobin

level(g/dL)

10.6(6—15.5,SD=2.8) Bloodtransfusion,n(%) 15(26)

Secondarybleeding,n(%) 11(19) Secondarysurgicalrevision,n

(%)

12(21) Postoperativeirrigationtime

(days)

3.5(0—14,SD=2.9) Durationofurinarycatheter

(daysafterthesurgery)

5.3(1—33,SD=5) Postoperativeurinary

infection,n(%)

6(10) Meandurationof

hospitalization(days)

8.5(1—40,SD=7.1) Claviencomplications

2/3a/3b//4/5

15/16/12/0/1 SD:standarddeviation.

Overall immediate surgical outcomes

Fifteen patients (26%) required blood transfusion in the postoperativeperiod.Meandurationofhospitalizationwas 8.5 days(Table 3). Secondary surgery was required in 12 cases(21%),includingendoscopicclotremovalundergen- eralanaesthesiain10patients,andsuprapubichaemostasis in 3 patients. One patient died 72hours after mTURP becauseofamassivepulmonaryembolism.The2patientsin whomwasnotstoppedinperioperativeperiodencountered nobleedingcomplicationinthepostoperativeperiod.

Blood transfusion risk factors

Inordertoassess transfusionrisk factors,a multivariable analysiswasdone.Variablesassociatedwithtransfusionrisk withaP-valuelowerthan0.2inunivariateanalysisincluded preoperative PSA, operating time, surgical technique, and bacteriuria on preoperative urinalysis (Table 4). All wereincludedinthemultivariateanalysis(wholedatanot shown). Two independent predictors of blood transfusion

were identified:conventional surgery use versus PVP, and high preoperative PSA (Table 4). Blood transfusion rate was significantly lower in the group of patients operated by PVPcomparedtoconventional surgery(0% versus 33%, P=0.010). In addition, the laser surgery was associated with shorter duration of hospitalization (3.4 days versus 9.9 days, P=0.014). However, it was not found any significant difference between patients operated by PVP compared to conventional surgery in terms of secondary bleeding (3/13 vs 8/45, P=0.7), or second surgery rate (2/13 vs10/45, P=0.5).Only onepatienthad adiagnosis of prostatic carcinoma: he was operated by monopolar prostaticresectionandrequiredbloodtransfusion.

Discussion

Thisstudyistheﬁrstonetofocalizeonthespeciﬁcpopula- tionofBPOpatientswithmechanicalcardiacvalverequiring adeep,rigorous andpermanentanticoagulation.Bleeding riskofBPOsurgeryinthispopulationwasfoundtobevery high.Morethanaquarterofpatientsrequiredbloodtrans- fusion,andsecondarysurgerywasperformedinmorethan 20% ofthecases.Those ratesaremuchhigherthanthose encounteredfollowingBPHsurgeryinthegeneralpopulation [1—3].

Thestudyisanobservational,retrospective,smallsized, and non-randomized one,makingit at low level ofscien- tiﬁcevidence.InadditionPVPwasthesolelasertechnique evaluated,whereasholmiumandthuliumlasersmightalso beefﬁcaciousinmanagementofpatientswithmechanical cardiacvalves.However,themethodologyusedwasconsis- tentwiththatofotherfewstudiesdealingwiththesubject oforalanticoagulationinBPOsurgery[9].Inaddition,most otherstudiesneverdistinguishaetiologiesofthecoumarin derivativestreatments,andfrequentlymixpatientsunder coumarinderivativesandantiplateletaggregators.Because oftherarityofcases,higherlevelofevidencemightnotbe reached, which reinforce the need for such observational studies.

Therateoftransfusionwasusedasasurrogatemarker ofthehaemorrhagicrisk.Inpreviousstudies,thePVPther- apy wasfound to decreasesigniﬁcantly the haemorrhagic risk compared to classic surgical procedures [10], and in

Table4 Transfusionriskfactorsinunivariateandmultivariableanalyses.

Blood transfusion, (n=15)

Noblood transfusion, (n=43)

Univariateanalysis, P-value

Multivariable analysis,P-value MeanpreoperativePSA

(ng/mL)

Corrected^aornot

7.87 (SD=6.54)

2.92 (SD=2.45)

0.004 <0.001

Meandurationofsurgery (min)

59.9(SD=32) 44.5 (SD=16.5)

0.03 0.12

Bacteriuriaonpreoperative urinalysis

4 5 0.17 0.9

Lasersurgery 0 13 0.02 <0.002

aForpatientsunderalpha-reductaseinhibitor.

(5)

particularlytodecreasetransfusionrisk inBPOsurgery. In Thangasamy et al. [11] meta-analysis,it was shown that meantransfusionratewas0.3%following PVPversus 7.1%

aftermTURP(relativerisk=0.16).Theauthorsdidnotfound anydifferenceintermsofsecondarysurgerybetweenPVP groupandmTURP.Similarly,inthepresent study,PVPwas not found to prevent from secondary bleeding. Interest- ingly,inNormand’sseries,10out624patientsoperatedby mTURPrequiredre-hospitalizationforsecondarybleeding.

The curative anticoagulation wasnot considered asarisk factorofsecondaryhaemorrhage[12].

PSAwasfoundtobeanindependentpredictivefactorof transfusion.As PSA isa surrogate markerof prostate volume[13],therefore,wehypothesisedthatprostatevolume mightinﬂuencetheriskoftransfusion,whichisconsistent withother studies[9]. However,theprostate volumewas notfoundtobeassociatedwithtransfusionriskinouruni- variateormultivariateanalyses.Thismightbeexplainedby errorsor approximations in determining prostate volume, whereasPSAmightnotbeinter-observerdependent.

The AmericanUrologicalAssociation recentlypublished a review for the management of oral anticoagulation in urological surgery. The main conclusion was that a mul- tidisciplinary management of anticoagulant for patients with mechanical cardiac valves would reduce the high morbidityandmortalityofinexpertlydiscontinuingormod- ifyingtheselifesavingtherapies[14,15].Intheabsenceof clearrecommendationstomanagesuchpatients,urologists, anaesthesiologists,and cardiologists have their ownpoint ofview.Consequently,oralanticoagulationmanagementof patientsintheperioperativeperiodwasfoundveryhetero- geneousinourseries.Inaddition,thetypeofmechanical cardiacmightvalveinﬂuencethedeepofanticoagulation:

mitralvalvesareathigherriskofthrombosis,comparedto aorticones[16].Finally,centresusingPVPmighthaveused speciﬁcprotocolstomanageanticoagulation.

Conclusion

BleedingriskandmorbidityofBPOsurgeryinpatientswith mechanicalcardiacvalveishigh.ThePVPseemstodecrease signiﬁcantlythetransfusionriskcomparedtoclassicsurgi- calproceduresforpatientswithmechanicalcardiacvalve.

Thisstudyistheﬁrstonetofocalize onthisspeciﬁcpopu- lationofpatientsrequiringadeep,rigorousandpermanent anticoagulation.

Disclosure of interest

Theauthorshavenotsuppliedtheirdeclarationofcompet- inginterest.

References

[1]ReichO, GratzkeC,BachmannA, etal.Morbidity,mortality andearlyoutcomeoftransurethralresectionoftheprostate:

aprospectivemulticenterassessmentof10,654patients.JUrol 2008;180:246—9.

[2]VarkarakisJ,BartschG,HorningerW.Long-termmorbidityand mortalityoftransurethralprostatectomy:a10-yearfollow-up.

TheProstate2004;58:248—51.

[3]Rassweiler J, Teber D, Kuntz R, Hofmann R. Complications of transurethral resection of the prostate (TURP) — inci- dence, management, and prevention. Eur Urol 2006;50:

969—79.

[4]DescazeaudA,RobertG,AzzouziAR,etal.Lasertreatmentof benignprostatichyperplasiainpatientsonoralanticoagulant therapy:areview.BJUInt2008;103:1162—5.

[5]Cornu JN, Madersbacher S. Greenlight photovaporisation of the prostate: nowready for prime time. Eur Urol 2016;69:

103—4.

[6]Oelke M, Bachmann A, Descazeaud A, Emberton M, Gravas S, Michel MC, et al. EAU guidelines on the treatment and follow-up of non-neurogenic male lower urinary tract symptoms including benign prostatic obstruction. Eur Urol 2013;64(1):118—40.

[7]Les anticoagulantsen Franceen 2014: étatdes lieux,syn- thèseetsurveillance—Rapportdel’ANSM;2014http://www.

ansm.sante.fr/var/ansmsite/storage/original/application/

26ed375830c56499badf0014eb3bb81b.pdf.

[8]DescazeaudA, Robert G, LebdaiS, BougaultA, AzzousiAR, Haillot O, et al. Impact of oralanticoagulation on morbidity oftransurethral resectionofthe prostate. WorldJ Urol 2011;29(2):211—6.

[9]DescazeaudA,RobertG, AzzousiAR,BallereauC,LukacsB, HaillotO,etal.Lasertreatmentofbenignprostatichyperplasia inpatientsonoralanticoagulant therapy:areview.BJU Int 2009;103(9):1162—5.

[10]RuszatR,WylerS,ForsterT,ReichO,StiefCG,GasserTC,etal.

Safetyandeffectivenessofphotoselectivevaporizationofthe prostate(PVP)inpatientsonongoingoralanticoagulation.Eur Urol2007;51(4):1031—8[discussion1038—41].

[11]Thangasamy IA, Chalasani V, Bachmann A, Woo HH. Pho- toselective vaporization of the prostate using 80-W and 120-W laser versus transurethral resection of the prostate for benign prostatic hyperplasia: a systematic review with meta-analysis from 2002 to 2012. Eur Urol 2012;62(2):

315—23.

[12]Normand G, Cuignet J, Briffaux R, et al. ‘‘Eschar Falls’’

after prostate transurethral resection. Prog Urol 2006;16:

461—3.

[13]Hochberg DA, Armenakas NA, Fracchia JA. Relationship of prostate-speciﬁc antigen and prostate volume in patients with biopsy proven benign prostatic hyperplasia. Prostate 2000;45(4):315—9.

[14]Culkin DJ,Exaire EJ,Green D,SolowayMS,GrossAJ, Desai MR, et al. Anticoagulation and antiplatelet therapy in uro- logicalpractice:ICUD/AUAreviewpaper.JUrol2014;192(4):

1026—34.

[15]TikkinenKA,CraigieS,AgarwalA,SiemieniukRA,Cartwright R, Violette PD, et al. Procedure-speciﬁc risks of thrombo- sisandbleedinginurologicalnon-cancersurgery:systematic reviewandmeta-analysis.EurUrol2017[pii:S0302-2838(17) 30114-8].

[16]Van Nooten GJ, Caes F, Franc¸ois K, Van Bellleghem Y, Bové T, Vandenplas G, et al. Twenty years’ single-center experience withmechanical heart valves: a critical review of anticoagulation policy. J Heart Valve Dis 2012;21(1):

88—98.