This is a repository copy of Prospective Implementation of Enhanced Recovery After Surgery Protocols to Radical Cystectomy.

White Rose Research Online URL for this paper:http://eprints.whiterose.ac.uk/130131/

Version: Accepted Version

Article:

Pang, K.H., Groves, R., Venugopal, S. et al. (2 more authors) (2018) Prospective Implementation of Enhanced Recovery After Surgery Protocols to Radical Cystectomy. European Urology, 73 (3). pp. 363-371. ISSN 0302-2838

https://doi.org/10.1016/j.eururo.2017.07.031

eprints@whiterose.ac.ukhttps://eprints.whiterose.ac.uk/

Reuse

This article is distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. This licence only allows you to download this work and share it with others as long as you credit the authors, but you can’t change the article in any way or use it commercially. More information and the full terms of the licence here: https://creativecommons.org/licenses/

Takedown

If you consider content in White Rose Research Online to be in breach of UK law, please notify us by emailing eprints@whiterose.ac.uk including the URL of the record and the reason for the withdrawal request.

1

Prospective implementation of Enhanced Recovery After Surgery (ERAS)1

protocolstoRadicalCystectomy2

3

KarlH.Pang1,4,RuthGroves2,SureshVenugopal3,AidanP.Noon4andJamesW.F.4

Catto1,4,$5

6

1. Academic Urology Unit, University of Sheffield, Sheffield, UK; 2. Department of7

Anaesthetics,SheffieldTeachingHospitalsNHSTrust,Sheffield,UK;3.Departmentof8

Urology, Chesterfield Royal and North Derbyshire Hospital, Derbyshire, UK; 4.9

DepartmentofUrology,SheffieldTeachingHospitalsNHSTrust,Sheffield,UK.10

11

$Correspondenceto;12

JamesCatto,13

AcademicUnitsofUrologyandMolecularOncology,14

GFloor,TheMedicalSchool,15

UniversityofSheffield,16

BeechHillRoad,17

Sheffield,S102RX,18

UnitedKingdom19

Tel:+44(0)114226122920

Fax:+44(0)114271226821

Email:j.catto@sheffield.ac.uk22

23

24

25

Abstractwordcount:289(limit300)26

WordCount:1978(limit2500)27

Keywords:UrothelialCancer,BladderCancer,Radicalcystectomy,ERAS28

29

30

2

Abstract31

32

Background:Multimodal enhanced recovery after surgery (ERAS) regimens have33

improvedoutcomesfromcolorectalsurgery.34

Objective: We report the application of ERAS to patients undergoing radical35

cystectomy(RC).36

Design, Setting and Participants: Prospective collection of outcomes from37

consecutivepatientsundergoingRCatasingleinstitution.38

Intervention:Twenty-sixcomponentsincludingprehabilitationexercise,sameday39

admission, carbohydrate fluid loading, targeted intra-operative fluid resuscitation,40

regional local anesthesia, cessation of NG tubes, omitting oral bowel preparation,41

avoidingdrainuse,earlymobilization,chewinggumuseandaudit.42

OutcomeMeasurementsandStatisticalAnalysis:Primaryoutcomeswerelength43

of stay and readmission rate. Secondary outcomes included intra-operative blood44

loss,transfusionrates,survivalandhistopathologicalfindings.45

Results and Limitations: 453 consecutive patients underwent RC, including 39346

(87%)with ERAS. Length of staywas shorterwith ERAS (median (IQR): 8 (6-13)47

days)thanwithout(18(13-25),p<0.001).PatientswithERAShadlowerbloodloss48

(ERAS:600(383-969)mlsvs.1050(900-1575)mlsfornon-ERAS,p<0.001),lower49

transfusion rates (ERAS: 8.1% vs. 25%, Chi sq. p<0.001) and fewer readmissions50

(ERAS: 15% vs. 25%, Chi sq. p=0.04) than those without. Histopathological51

parameters(e.g.tumorstage,nodecountandmarginstate)andsurvivaloutcomes52

didnotdifferwithERASuse (allp>0.1).Multivariableanalysis revealedERASuse53

was(p=0.002)independentlyassociatedwithlengthofstay.54

3

Conclusions:TheuseofERASpathwayswasassociatedwithlowerintra-operative55

blood lossandfasterdischarge forpatientsundergoingRC.Thesechangesdidnot56

increasereadmissionratesoralteroncologicaloutcomes.57

Patientsummary:Recoveryaftermajorbladdersurgerycanbeimprovedbyusing58

enhanced recovery pathways. Patients managed by these pathways have shorter59

lengthofstays,lowerbloodlossandlowertransfusionrates.Theiradoptionshould60

beencouraged.61

62

4

Introduction63

Radical cystectomy (RC) with pelvic lymph node dissection (PLND) is the gold standard64

treatment for muscle invasive BC [1], plays a key role in managing local failure after65

radiotherapy [2] and is an option for high risk local non-muscle invasive BC [3]. RC is a66

morbidprocedurethatoftenperformedinolderpatientswithco-existingcardiopulmonary67

disease. Many patients develop post-operative complications, including 13% (grade 3-5)68

thatrequirefurtherintervention[4].Consequently,patientswhocouldbenefitfromRCdo69

notalwaysreceivethisoption[5,6].Whilstcentralizationofmajorcancerservicesincreases70

radicaltreatmentsandsubsequentoutcomes[7],themorbidityfromRCstilllimitsitsuse.71

72

In colorectal surgery, the use of multimodal Enhanced Recovery after Surgery (ERAS)73

regimenshasreducedpost-operativemorbidityandlengthofstay[8,9].ERASintroducesa74

numberofpre-,peri-andpost-operativestepstoimprovethepatientpathway[10].Many75

ERAScomponentsaregeneric toabdominalsurgeryandsohavebeen implemented inRC76

without prospective evidence [11]. However, RC includes surgery to the urinary and77

gastrointestinaltractsandsonotallERAScomponentsmaybesuitable.78

79

TherehavebeenseveralreportsofERASinRCcohorts[10-14]andoneRCT[15].ThisRCT80

foundERASimprovedqualityoflifeandreducedmorbidityinpatientsundergoingRC,but81

did not shorten post-operative length of stay (LOS). Here we report the prospective82

adoptionofERASinalargeUKcentre,wheretheopioidreceptorantagonistAlvimopan[16]83

isnotavailableandhealthcaredesigndoesnotincentivizerapiddischarge.84

85

Materialsandmethods86

Patients87

5

Consecutive patients undergoing RC and urinary reconstruction were enrolled in a88

prospective institutionaldatabase.FromFebruary2007toOctober2016,a25pointERAS89

regimenwasimplemented.Theregimen(table1)wasderivedfromavailableevidenceand90

practicewithin colorectal surgery[10].Datawere collectedprospectivelyandallpatients91

undergoingRCwere included in thestudy.TheuseofERAS reflected thedateofsurgery.92

Duringthetransitionperiod,patientswereidentifiedasusingtheERASpathwayiftheyhad93

pre-operative carbohydrate loading, were allowed fluids until 2 hours prior to surgery,94

planned toavoidNGT,useda smaller incisionhadearlypost-operativemobilizationwith95

dietontheward.96

97

ERASProtocol98

Pre-operative:Counsellingintheoutpatientsettingwasperformedbythesurgeon(JWFC),a99

cancernursespecialist, ananaesthetist (RG)whenneeded, anda stoma therapist.Typical100

consultations included wide ranging treatment discussions and lasted 30-45 minutes.101

Patientswereadvised tomaintainanormaldietuntil thenightbefore surgery, to reduce102

cigarettesmokingandalcoholintake,andweregivenaninformationbookletregardingtheir103

expected recovery. Increasing exercise activity (prehabilitation) was stressed as an104

important aspect of recovery and patients asked towalk 1 hour per day (once or twice)105

between their initial consultation and surgery. Patients whose anaesthetic fitness was106

uncertainwerereviewedbyanAnaesthetistandcardiopulmonaryexercise(CPEX)testing107

usedinselectivecases.Pre-morbiditieswereoptimizedwherepossible.Anemiawastreated108

with intravenous iron transfusion. Prior to surgery, patients attended clinic for stoma109

marking, to obtain 6 carbohydrate dinks (e.g. PreOp TM, Nutricia) and to collect a single110

injectionoflowmolecularweightheparin(LMWHe.g.dalteparin5,000ius/c).Patientsself-111

administered dalteparin the evening before surgery and undertook carbohydrate fluid112

6

loading for the18hoursprior to surgery.Patientswereallowedoral fluidsup to2hours113

pre-operativelyandfood6hourspre-operatively.114

115

Per-operative: At induction, a pre-planned anaesthetic protocolwas used (supplementary116

table1).Importantelementsincludedlimitedfluidadministrationtargetedtolosses,theuse117

ofvasopressorstomaintainbloodpressures,theavoidanceofnasogastrictubes(NGT)and118

hypothermia (e.g. using Bair Hugger TM). Typically, only 500-1000mls intravenous119

crystalloid was administered prior to bladder removal. Intra-operative steps taken to120

reducethe impactofsurgery includedtheuseofsmall incisions(typically10cm)orrobot121

assistedlaparoscopy,theuseofvesselsealers(e.g.LigasureTM impact),clipsandfastidious122

haemostasis.Post-operativeanalgesiacommencedwiththeinsertionofrectussheathlocal123

anaestheticblocks(usually60mlsof0.125%bupivacaine)andtunnelledcannulae(lateral124

and superior to the incision prior towound closure) for a 48 hour bupivacaine infusion.125

Closurewasperformedusinga2/0PDSrectussheathsutureand4/0monocrylsubcuticular126

skinsuture.Antibioticprophylaxis(1.2gintravenousco-amoxiclav)wasadministeredfor24127

hoursinmenandfor48hoursinwomen(duetohighercontaminationfromvaginalflora).128

DVTprophylaxiswasadministeredfrom6-12hourspriortosurgeryandforatleast28days129

aftersurgeryoruntildischarge(whicheverwaslonger).130

131

Radicalcystectomy:Inmales,cystoprostatectomywasperformedinanantegrademannerto132

include the seminal vesicles. In females, anterior pelvic exenteration included the uterus,133

fallopian tubes and anterior vaginalwall. Ovarieswere spared,whenpossible in younger134

women and in those with low stage disease. Lymphadenectomy was performed after135

bladderremovalandincludedtheobturator,internalandexternaliliacchainstothelevelof136

7

the ureteric crossingof themid common iliac vessels. Ureteroileal anastomosiswas by a137

BrickertechniqueandtheStudertechniqueusedforaneobladder.138

139

Post-operative:Managementwasundertakenusingapre-specifiedERASregimen(table1).140

During the regimen’s introduction, an ERASnurse audited compliance.On post-operative141

day(POD)#1patientswereallowedchewinggum,oneclearboiledsweet/candyperhour142

and30mls clearnon-fizzyoral fluidsperhour, as comfort allowed. Intakewas reduced in143

patientsfeelingnauseousoruncomfortable.Patientsweresatoutofbedandencouragedto144

walk10-20meters.Additionalanalgesiawasallowedthroughondemandpatientcontrolled145

analgesic (PCA) intravenous opiates. On POD#2 patients aimed to walk 100meters and146

were allowed to drink clear fluids as tolerated. Nausea or vomiting were treated with147

reduced fluid intake and rest, rather than NGT. NGT were administered for repeated148

vomitingwithepigastricdiscomfortorinthepresenceofileus/obstruction.Lightdietwas149

introducedwhenthepatientpassedflatusorhadabowelmovement.Patientswithoutflatus150

orbowelmovementonPOD#3,hadaglycerinesuppositoryadministeredperrectum.Total151

parenteralnutrition(TPN)wasstartedonpatientsnottoleratingdietbyPOD#7,orsooner152

if post-operative complications were apparent. Abdominal and pelvic CT scan was153

undertakenonPOD#5 ifpatientswerenotprogressingaccording toexpectationor in the154

presenceofsignsof intra-abdominalcomplications.Dischargeoccurredwhenthepatients155

werecomfortable,self-caringwiththeirstoma,mobile,andwhentheyhadresumedfulldiet156

withbowelmotion.157

158

Statisticalanalysis159

Primary outcomes were LOS and post-discharge readmission rates. Secondary outcomes160

included intra-operative blood loss, intra- and post-operative blood transfusion rates,161

8

operative duration, overall and bladder-cancer specific survival. For analysis, BMI was162

stratifiedasunderweight(BMI<18.5),healthy(BMI18.5–24.9),overweight(BMI25–29.9)163

and obese (BMI ≥30) [17]. Pre-operative anaemiawas defined as hemoglobin <12g/dl in164

bothsexesandrenalimpairmentasestimatedGFR<40mls/min,asperournationalregistry165

database.Multivariableanalysis fora<7-dayLOSwasperformedusing logisticregression166

withfactorssignificantfromunivariableanalysis.TotestERASthroughanylearningcurve,167

casesweredividedintoquartilesbytime,andvariablesanalysedusinglogisticregression.168

169

Results170

Patientsandrecoverycomponents171

453 consecutive patients underwent radical cystectomy (table 2, figure 1). The median172

(IQR)agewas70years(64-76)and14%ofpatientswere³80yearsold.Ninety-eightwere173

female(22%)and50(11%)receivedaneobladderreconstruction.Aroundonequarterof174

patients had renal impairment (eGFR <40mls/min in 107 (24%)) prior to surgery, 100175

(22%)hadhydronephrosisorwereanephric,themedian(IQR)BMIwas29(26.0-32.8)and176

177 (39%) had Charlson Comorbidity index (CCI) of 4 or higher. Twenty-eight patients177

underwentrobotassistedsurgery,ofwhich25hadintracorporealreconstruction.Fifty-nine178

patients received neoadjuvant chemotherapy (NAC), 18 received adjuvant chemotherapy179

and 29 palliative chemotherapy. 135 patients had invasive cancer at TUR,were younger180

than80yearsofage,hadnormalrenalfunctionandagoodperformancestatus(CCI0-3)).181

Assuch,theuseofNACinthesesuitablecaseswas57/135(42%)anddidnotdifferbyERAS182

use (42% vs. 44% (non-ERAS)). Histological outcomeswere similar in patientswith and183

withoutERASrecovery(supplementarytable2,figures2aandb).Inparticular,thelymph184

nodecount(mean±st.dev:10.7±4.7forERASvs.10.3±5.8non-ERAS,Ttestp=0.6)and185

circumferentialmarginstatus(positivein2.5%(ERAS)vs.1.7%,Chisq.p=0.4)weresimilar.186

9

187

ERAScomponentswereusedin393(87%)patients(figure1a).Directadmissionfromhome188

to surgery occurred in 376 (83%), rectus sheath local anaesthetic infusions used in 241189

(53%), NGT avoided in 382 (84%), pre-operative oral bowel preparation avoided in 390190

(86%)anddrainsnotused in20 (4.4%)patients.Carbohydrate fluid loadingwasused in191

364(80%)anddrinkinguntil2hourspriortoanaesthesiaallowedin284(63%).Patients192

with ERASwere older (median (IQR) 71 years (65-76)) than thosewithout (60 (61-70),193

Mann-Whitney U test p<0.001), more commonly female (23% vs. 13%) and less often194

underwentneobladderreconstruction(6.4%vs.42%),butotherwisethetwogroupswere195

similar(table2).196

197

Lengthofstayandreadmission198

Length of stay differed significantly for patientswith ERAS (median (IQR) 8 (6-13)days)199

andwithoutERAS (18 (13-25)) recovery (supplementary figure1andp<0.001).Over the200

series, LOS reduced from a median of 17 days to 6 days (figure 1b) and varied with a201

number of factors (table 3). Longer stayswere seen in females (12 days vs. 9 formales,202

p=0.004),withneobladderreconstruction (19daysvs.9 for ileal conduit,p=0.001), those203

withanabnormalBMI(p=0.001),inthosereceivingabloodtransfusion(14daysvs.10for204

notransfusion,p=0.03)andinthosewithcomorbidities(P=0.001)(seetable3fordetails).205

Shorter stayswere seenwith robot-assisted surgery (7 days vs. 10 for open, p=0.03). In206

univariableanalysis(table3)malegender(p<0.001),ilealconduitdiversion(p<0.001),low207

BMI(p=0.01),normalrenalfunction(p<0.001),lowCCI(p<0.001),notransfusion(p=0.03),208

nodrain(p=0.04)andallcomponentsoftheERASregimen(p<0.001)wereassociatedwith209

a LOS of <7days. Multivariable analysis revealed that female gender (logistic regression210

p<0.001),neobladderreconstruction(p=0.02),BMI(p<0.001),comorbidity(CCI:p<0.001)211

10

and non-ERAS use (grouped into a single parameter, p=0.002) were independently212

associatedwithaLOSof>7days.Inonlyneobladdercases(25ERASand25non-ERAS),sub-213

groupanalysisrevealedthatERASusewasstillassociatedwithshorterLOS(median(IQR)214

15(8-20)(ERAS)vs.24(18-28)days(non-ERAS),Mann-WhitneyUtestp<0.001).215

216

Readmission occurred in 21% of patients (88/417with readmission outcomes). Twenty-217

twopatients(25%)stayed1dayand24(27%)morethan10days.Mostreadmissionswere218

within 30 days of discharge (60/88 (68%)). Patientswith ERAS had fewer readmissions219

(15%) than those without ERAS (25%, Chi sq. p=0.04). Readmission rates declined over220

time to 11% for the last 100 cases (figure 1d). We did not demonstrate differences in221

readmission length of stay by ERAS use (supplementary figure 2). ERAS use was222

significantly associatedwith shorter LOS and lower readmission rates, once adjusted for223

covariates(includinglearningcurve,logisticregressionp<0.05).224

225

Secondaryoutcomes226

Intra-operative blood loss (median (IQR))was lower for ERAS (600 (383-969)mls) than227

non-ERAS(1050(900-1575)mls)patients(Mann-WhitneyUtestp<0.001).Consequently,228

transfusion rates were lower for ERAS (n=32 (8.1%)) than for non-ERAS (n=15 (25%))229

patients (Chi sq p<0.001). Blood loss reduced across the series from an average of 1,237230

(first50 cases) to557mls (last50 cases, figure1c).Themedian (IQR)operativeduration231

was lower in the ERAS (2.9 (2.5-4.0)) vs non-ERAS (5.0 (4.5-6.0)) (Mann-Whitney U test232

p<0.001,supplementary figure2).ERASusewassignificantlyassociatedwithlowerblood233

loss (logistic regressionp<0.01),butnot fasteroperative times (p=0.5), onceadjusted for234

learningcurve.235

236

11

Mortality237

Atmedian(IQR)followupof19(8.3-37)months,335(77%)patientswerealiveandunder238

surveillance(17missing).Therewere77deaths(17%)fromBC(median(IQR)of15(7.2-239

22)months after surgery) and24 fromother causes (median (IQR)19 (6.1-34)months).240

The 30-day mortality rate was 1.7% (1 case) for non-ERAS and 0.3% (1 case) for ERAS241

patients(Chisq.p=0.14).Therewere3(5%)deaths in thenon-ERASand8(2.1%) inthe242

ERAS cohort within 90-day of cystectomy. Of the 90-day deaths, 8/11 (73%) were from243

metastatic BC. In univariable and Multivariable analysis, neither 30-day nor 90-day244

mortalityratesdifferedwithERASuse(ChiSq.andLogisticregressionp>0.60).Therewas245

no difference in overall or bladder cancer specific survival when stratified by ERAS use246

(figure2candd).247

248

249

Discussion250

Sinceintroductionintocolorectalsurgery,enhancedrecoveryprogramshaveimprovedthe251

outcomesformanypatientsundergoingadiversearrayofsurgicalprocedures(reviewedin252

[10]). The ERAS Society (www.erassociety.org) has protocols within several surgical253

specialities, including RC. Sincemany RC patients develop complications during recovery254

[4], these patients may benefit more than most from refinements in post-operative255

management. Our data support the use of ERAS, demonstrate excellent improvements in256

post-operativerecoveryandconfirmsitsoncologicalsafety.257

258

TherehasbeenoneprospectiveRCTofERASinRCpatients[15],inwhichERASuseledto259

fewer complications, a faster improvement in return of quality of life,more rapid bowel260

recovery and shorter stays in intermediate care, but no change in LOS. These findings261

12

support and conflict with the field. For example, whilst others also found ERAS leads to262

accelerated bowel recovery and fewer complications, many report shorter hospital stays263

[11,12,18].WithintheUSA,Daneshmandetal.reportedERASusing110patientsandfound264

itsusereducedmedianLOSto4days[12].WiththeUK,Arumainayagametal.foundERAS265

reducedmedian LOS by around4 days [19].LOS can reflect healthcare design aswell as266

rehabilitation.IntheUK,patientsdonotpayforhealthcareandmostaredischargedhome.267

As such, there can be reluctance for rapid discharge. In the US, expensive hospital stays268

incentivise discharge home or to cheaper skilled nursing facilities (occurred in 16%269

Daneshmandetal.cohort).WithintheGermanhealthcaresetting,reducingtheLOSisnotan270

economicpressureandsomaynothavechangedintheERASpopulation.271

272

Withinourseries,ERAS improvedrecovery,accelerateddischargehomeandalsoreduced273

theburdenofcaretothepatientandtheirmedical/nursingteams.Fasterdischargebrings274

manybenefits,includingmorerapidaccesstoadjuvantchemotherapywhennecessary.Key275

elements to the success of ERAS involved staff, patients and infrastructure. Firstly, a276

multidisciplinary approach was vital. Surgical staff engaged with anaesthetic staff to277

plan/anticipatepatientcare,nursingstaffwereengagedinimplementingERASontheward278

andauditingpathwaycompliance,whilststoma/neobladderreconstructionnursesattended279

clinics and theward to expedite competency. Unfit patients or those at increased risk of280

complications benefitted from additional surgeon/anaesthetist interaction. Secondly, pro-281

active patient engagement was vital. This included explaining anticipated recovery282

timeframes, creating an ERAS booklet that patient’s read and completed during their283

recovery,engaginginprehabilitationexerciseregimensforthepatient(andinvolvingtheir284

nextofkin in theseexercises), andplanningdischargebeforeadmission (e.g. stockingup285

with foodbeforeadmission,planning careandsupportoncedischarged).Withregards to286

13

infrastructure,itwasimportanttoidentifythepathwayasnewanddifferenttotraditional287

care.Thishelpedstafffeelcomfortablewithrapidchangesinpractice,allowedachangein288

patientflow(samedayadmission,rapidmobilisationanddischarge),andjustifiedresource289

tostudyimplementation(auditingpathwaycomplianceduringintroductionwasveryuseful290

forthelessexperiencedmedicalandparamedicalstaff).291

292

There are important limitations to our data. Firstly, the design precludes a meaningful293

Multivariable analysis of ERAS elements asmost components were used together rather294

than in different permutations. However, our analyses do reveal the importance of the295

patient(e.g.BMIandcomorbidity),whichmakesclinicalsenseandmatchesourexperience.296

Secondly,thesedataarederivedfromasingleteamandsoincludealearningcurve.Figure1297

showsthattherateofimprovementinalloutcomesslowsafter150casesandchangesmost298

rapidly around the implementation of ERAS. Improvements in these outcomes are299

associatedwithERASuse,afteradjustmentforlearningcurveandothercovariates.Assuch,300

wefeelkeydriversforchangeincludebothalearningcurveandERASuse.Smaller,shorter301

series(andsolessimpactfromlearningcurves)supportourbelief(e.g.[12][19]).Thirdly,302

the ERAS and non-ERAS cohort are imbalanced for reconstruction choice. This reflects a303

change inpracticepromptedbydatasuggestingQOL issimilar inmanypatientswith ileal304

conduitandneobladder(unpublishedfromhttp://www.abdn.ac.uk/urology/research/otis/305

and [20]) and the increased use of RC in older, less fit patients table 2) once ERAS306

improvementsbecameapparent.Webelievelessfitpatientsneedthesimplest,leastmorbid307

surgery with the fastest recovery. A direct comparison using only neobladder cases308

confirmedthatERASusewasstillassociatedwithshorterLOSandfasteroperations.Overall309

ourrateofneobladderuseissimilarorhigherthantheUKaverage(forexample,the2009310

BAUScomplexsurgerydatabaseshows5.7%receivedaneobladderintheUK).311

14

312

Conclusion313

WefoundthatchangestotheRCpathwaymadedramaticimprovementstopatientrecovery314

withoutaffectingoncologicaloutcomes.Inparticular,enhancedrecoveryusewasassociated315

withshorterlengthofstay,lowerbloodlossandtransfusionrates,andfewerreadmissions316

aftersurgery.317

318

Takehomemessages319

Making the care of patient’s undergoing bladder removal simpler and more uniform320

improves theiroutcomes. In particular, it can be associatedwith shorter stays and fewer321

readmissionsafterdischarge.322

323

Acknowledgements324

The authors would like to acknowledge the medical and nursing staff within the325

Departments of Urology and Anaesthesia at the Royal Hallamshire Hospital, Sheffield326

Teaching Hospitals Trust. In particular, the authors acknowledge how important the327

supportoftheSamBhogalandDianeLeach(Stomaservices),criticalcaredepartmentand328

Drs.D.J.Rosario,F.C.Hamdy,M.D.Haynes,J.B.Anderson,RobAitchsonandStephenWeber329

weretothesuccessofthiswork.ThisworkwasfundedbyFellowshipsfromTheUrological330

FoundationandTheRoyalCollegeofSurgeonsofEnglandtoK.H.PangandJ.W.F.Catto.331

332

333

334

335

336

15

References337

[1] Chavan S, Bray F, Lortet-Teulent J, GoodmanMM, JemalA. International Variations in338

BladderCancerIncidenceandMortality.EurUrol.2014;66:59-73.339

[2]WitjesJA,ComperatE,CowanNC,DeSantisM,GakisG,LebretT,etal.EAUguidelineson340

muscle-invasiveandmetastaticbladdercancer:summaryofthe2013guidelines.EurUrol.341

2014;65:778-92.342

[3]KulkarniGS,HakenbergOW,GschwendJE,ThalmannG,KassoufW,KamatA, etal.An343

updated critical analysis of the treatment strategy for newly diagnosed high-grade T1344

(previouslyT1G3)bladdercancer.EurUrol.2010;57:60-70.345

[4] ShabsighA, KoretsR, VoraKC, Brooks CM, CroninAM, Savage C, et al. Defining early346

morbidity of radical cystectomy for patients with bladder cancer using a standardized347

reportingmethodology.EurUrol.2009;55:164-74.348

[5]GoreJL,LitwinMS,LaiJ,YanoEM,MadisonR,SetodjiC,etal.Useofradicalcystectomy349

forpatientswithinvasivebladdercancer.JNatlCancerInst.2010;102:802-11.350

[6] Galsky MD, Stensland K, Sfakianos JP, Mehrazin R, Diefenbach M, Mohamed N, et al.351

Comparative Effectiveness of Treatment Strategies for Bladder Cancer With Clinical352

EvidenceofRegionalLymphNodeInvolvement.JClinOncol.2016;34:2627-35.353

[7]HounsomeLS,VerneJ,McGrathJS,GillattDA.Trendsinoperativecaseloadandmortality354

rates after radical cystectomy for bladder cancer in England for 1998-2010. Eur Urol.355

2015;67:1056-62.356

[8]KehletH,WilmoreDW.Multimodalstrategies to improvesurgicaloutcome.AmJSurg.357

2002;183:630-41.358

16

[9] Lemanu DP, Singh PP, Stowers MD, Hill AG. A systematic review to assess cost359

effectiveness of enhanced recovery after surgery programmes in colorectal surgery.360

ColorectalDis.2014;16:338-46.361

[10]PatelHR,CerantolaY,ValerioM,PerssonB,JichlinskiP,LjungqvistO,etal.Enhanced362

recoveryaftersurgery:areweready,andcanweaffordnot to implementthesepathways363

forpatientsundergoingradicalcystectomy?EurUrol.2014;65:263-6.364

[11]AzharRA,BochnerB,CattoJ,GohAC,KellyJ,PatelHD,etal.EnhancedRecoveryafter365

Urological Surgery: A Contemporary Systematic Review of Outcomes, Key Elements, and366

ResearchNeeds.EurUrol.2016;70:176-87.367

[12]DaneshmandS,AhmadiH,SchuckmanAK,MitraAP,Cai J,MirandaG,etal.Enhanced368

recoveryprotocolafterradicalcystectomyforbladdercancer.JUrol.2014;192:50-5.369

[13] Dutton TJ, Daugherty MO, Mason RG, McGrath JS. Implementation of the Exeter370

enhanced recovery programme for patients undergoing radical cystectomy. BJU Int.371

2014;113:719-25.372

[14]SmithJ,PruthiRS,McGrathJ.Enhancedrecoveryprogrammesforpatientsundergoing373

radicalcystectomy.NatRevUrol.2014;11:437-44.374

[15]KarlA,BuchnerA,BeckerA, StaehlerM,SeitzM,KhoderW,et al.Anewconcept for375

earlyrecoveryaftersurgeryforpatientsundergoingradicalcystectomyforbladdercancer:376

resultsofaprospectiverandomizedstudy.JUrol.2014;191:335-40.377

[16]LeeCT,ChangSS,KamatAM,AmielG,BeardTL,FerganyA,etal.Alvimopanaccelerates378

gastrointestinal recovery after radical cystectomy: a multicenter randomized placebo-379

controlledtrial.EurUrol.2014;66:265-72.380

17

[17]NepogodievD,ChapmanSJ,Glasbey J,KellyM,KhatriC,DrakeTM,etal.Determining381

SurgicalComplications in theOverweight (DISCOVER): amulticentreobservational cohort382

study to evaluate the role of obesity as a risk factor for postoperative complications in383

generalsurgery.BMJOpen.2015;5:e008811.384

[18]AningJ,NealD,DriverA,McGrathJ.Enhancedrecovery:fromprinciplestopracticein385

urology.BJUInt.2010;105:1199-201.386

[19] Arumainayagam N, McGrath J, Jefferson KP, Gillatt DA. Introduction of an enhanced387

recoveryprotocolforradicalcystectomy.BJUInt.2008;101:698-701.388

[20] Ali AS, Hayes MC, Birch B, Dudderidge T, Somani BK. Health related quality of life389

(HRQoL) after cystectomy: comparison between orthotopic neobladder and ileal conduit390

diversion.EurJSurgOncol.2015;41:295-9.391

392

393

18

Figurelegends394

395

Figure 1. The use of ERAS following radical cystectomy. ERAS Components and396

outcomesarealignedforthe453consecutivepatients.(a).Individualelementsfromthe26397

elements of ERAS are shown for each patient including robotic assisted surgery (RARC),398

omissionofapelvicdrain,theuseoforalbowelpreparation,samedayadmissiontosurgery,399

regional local anaesthesia (rectus sheath blockade), epidural use, nasogastric tube (NGT),400

small incision foropensurgery,pre-operative carbohydrate loadinganddesignating their401

pathwayasERAStofacilitateaudit.ThelowerlineindicatestheextentofERAScompliance402

(shadesofwhite(6)todarkgrey(10)foruseofERAS).(b).Lengthofstay(days)and(c).403

bloodloss(mls)acrosstheseriesareshownasmedianandinterquartilerangesforeach10404

consecutivecases.(d).Readmissionratesforeach10consecutivecasesacrosstheseries.405

406

Figure2.OncologicaloutcomesstratifiedbytheuseofERAS.Withinthiscohortof453407

patients,therewasnodifferenceinpathological(a).Lymphnodecountor(b).Marginstatus408

or(c).Overallsurvivalor(d).Bladdercancerspecificsurvival)outcomesaccordingtothe409

useofERAS.410

411

19

Table1.ElementsoftheERASprotocolusedwithinthisreport.!

Domain Item Elements

Clinic 1.Preoperativecounselingandeducation Adviceabourmaintainingactivitylevels

Dietaryandalcoholadvice

Detailsofadmissionandrecovery

Writtenmaterialdetailingpost-oprecoveryplan

2.Prehabilitationexercise Walkingfor1hourperday

3.Preoperativemedicaloptimization Optimizationofco-morbidities

Smokingcessationadvice

Plansocialaspectsofdicharge.Whowillhelpcareforpatient?

4.Correctionofanemia OralIronsupplementsorI/VIron

Priortoadmission 5.Oralmechanicalbowelpreparation Omitted.Normaldietuntilpre-opfasting

6.Selfadministeredthromboprophylaxis SingleLMWHinjection12hourspriortosurgeryadministerdathome

7.Pre-operativecarbohydrateloading Carbohydrateloading(6cartonsofdrink(e.g.NutriciaPreOp)overthe18

hourspriortosurgery).Carefuluseindiabeticpatients

Admission 8.Pre-operativeoralintake Clearfluiduntil2hourspre-op

Solidfoodsuntil6hourspre-op

9.Pre-anaesthesiamedication Avoidanceoflong-actingsedatives

Anaesthesia 10.Standardanestheticprotocol

11.Anti-microbialprophylaxis 24hoursIVAugmentin

12.Skinpreparation Twostagepreparation:Sprayalcoholic2%chlorhexidinegluconateand

paintaqueous10%povidone-iodine

13.Thromboembolicprophylaxis Thromboemboliccompressionstockings

28dayspharmacologicalprophylaxiswithLMWHstartingdaybefore

Intra-operativepneumaticcompressionstockings

14.Regionalanalgesia Epiduralanaesthesiaomitted

Rectussheathcatheters(0.125%bupivicaine)forfirst48hrs

15.Perioperativefluidmanagement Avoidoverhydration.Vasopressorstomaintainarterialhypotension.

Administer<1lcrystalloiduntilbladderremoved.

16.Nasogastricintubation NoNGToritisremovedattheendofsurgery

17.Preventingintraoperativehypothermia Useofawarmingblanket(FullbodyBairHuggerTM3M)

20

Useofawarmingblanket(FullbodyBairHuggerTM3M)

Surgery 18.Minimallyinvasiveapproach Mini-OpenCystectomyincision

RARC

19.Resectionsitedrainage Consideromittingpelvicdrain

20.Urinarydrainage Ureteralstentsortransurethralneo-bladdercathetershouldbeused.

Stentsremovedasanoutpatientat10days.Catheterremovedafter

cystogramforneobladder

21.Woundclosure 2/0polydioxanonesuture(Ethicon)torectussheath.3/0subcuticular

Monocryl(poliglecaprone)suture(Ethicon)toskin.

Post-operative 22.Post-operativediet Chewinggumtostartat4hoursaftersurgery

Oralfluidstostarteveningofsurgery-30mls/hourofclearnon-fizzyfluids

Resumedietwhenpassingflatus,mobileandpaincontrolled.

23.PreventionofPONV Anti-emeticsasneeded

Earlyresumptionoforalfluids

24.Postoperativeanalgesia Rectussheathcatheters(0.125%bupivicaine)

Patientcontrolledopiate

I/VParacetamol/Acetaminophen1gqdsuntildietresumed

25.Earlymobilization 6HoursoutofbedonPOD1

Walk10-20monPOD1

Walk100monPOD2

Walk>100monPOD3+

26.Audit Auditcompliance.Understandproblems.Keepresourcewithinteam

LMWH:Lowmolecularweightheparin

NGT:Nasogastrictube

POD:Post-operativeday

PONV:post-operativenauseaandvomiting

iRARC:RobotassistedRadicalCystectomywithintra-corporealreconstruction

I/V:Intravenous

21

Table2.Detailsofthepatientswithinthisseries.

n % n %

Sex Male 303 77% 52 87%

Female 90 23% 8 13% 0.01

Age Median(IQR) 71 65-76 66 60.8-70.3 <0.001

Age>80 Yes 60 15% 2 3.3%

No 333 85% 58 97% 0.01

BMI Underweight<18.5 2 0.5% 0 0.0%

Healthy18.5-24.9 96 24% 18 30%

Overweight25-29.9 105 27% 14 23%

Obese>30 97 25% 13 22%

Missing 93 24% 15 25% 0.9

Pre-opHb(g/dl) Median(IQR) 131 120-142 129 118-136.5 0.6

RenalFunction Normal 285 73% 26 43%

eGFR<40mls/min 102 26% 5 8.3%

Unknown 6 1.5% 29 48% 0.2

Uppertracts Normal 294 75% 42 70%

Unilateralhydronephrosis 70 18% 8 13%

Bilateralhydronephrosis 17 4.3% 0 0.0%

Anephric/solitary 5 1.3% 0 0.0%

Unknown 7 1.8% 10 17% 0.4

CharlsonCIscore 0-3 201 51% 30 50%

4-5 117 30% 13 22%

6-7 16 4.1% 3 5.0%

>8 23 5.9% 5 8.3%

Unknown 36 9.2% 9 15% 0.6

Pre-opBCphenotype Low-riskNMI 5 1.3% 1 1.7%

High-riskNMI 165 42% 20 33%

MuscleinvasiveBC 223 57% 39 65% 0.4

Reconstruction Ilealconduit 368 94% 35 58%

Neobladder 25 6.4% 25 42% <0.001

Abbreviations:NMINon-muscleinvasive,BCBladdercancer,Hbhemoglobin

*Statisticaltests:Chisquareforcategorical&Mann-WhitneyUort-testforcontinuousdata.

ERAS Non-ERAS

p-value*

22

Table3.Factorsassociatedwithlengthofstaywithinthiscohort.

Element Number % Median OR pvalue OR pvalue

Age(continous) Median(IQR) 70(64-76) 100% 10 6 15 0.98 0.95 1.0 0.2

Tumorphenotype Low-riskNMI 6 1.3% 7 6 17.5

High-riskNMI 185 41% 10 7 16

MuscleinvasiveBC 262 58% 10 6 15 2.9 0.5 16.4 0.2

Sex Male 355 78% 9 6 15

Female 98 22% 12 7.8 16 2.2 1.3 3.7 <0.001 3.9 1.9 7.8 <0.001

Robotassisted Yes 28 6.2% 7 6 10

No 425 94% 10 6 16 2.0 0.9 4.3 0.08

Reconstruction Ilealconduit 403 89% 9 6 13

Neobladder 50 11% 19 12 25.3 6.4 2.5 16.4 <0.001 5.5 1.3 22.6 0.02

BodyMassIndex(continous) Median(IQR) 29(26-32) 76% 8 6 16 1.0 0.9 1.0 0.4

HbPre-operation(g/dl) Anemia 120 26% 7.5 6 14

Normal 175 39% 8 6 12

Missing 158 35% 13 8 19 0.9 0.6 1.5 0.7

RenalFuntion Normal 311 69% 6 8 14

eGFR<40mls/min 107 24% 11 7 15

Unknown 35 7.7% 15 12 19 2.1 1.3 3.3 <0.001 1.5 0.8 3.0 0.2

Hydronephrosis None 336 74% 10 6 15

Unilateral 78 17% 10 7 13.3

Bilateral 17 3.8% 12 7 15.5

Anephric/Solitary 5 1.1% 6 5 11.5

Unknown 17 3.8% 15 7.5 21.5 0.2 0.0 1.4 0.1

CharlsonCI 0-3 231 51% 7 6 12

4-5 130 29% 10 7 13

6-7 19 4.2% 12 7 19

>8 28 6.2% 26 22.3 31

Unknown 45 9.9% 15 12 17 32.4 4.3 242.5 <0.001 55.8 6.3 493.0 <0.001

IQR 95%CI 95%CI

LengthOfstay Univariable* Multivariable*

23

Unknown 45 9.9% 15 12 17 32.4 4.3 242.5 <0.001 55.8 6.3 493.0 <0.001

Transfusion Yes 47 12% 14 8 21

No 406 88% 10 6 14 2.2 1.1 4.4 0.03 0.7 0.3 2.2 0.6

"ERASPathway" Yes 393 87% 8 6 13

No 60 13% 18 13 25 45.5 6.2 331.3 <0.001 295 7.5 11649 0.002

Pre-Opcounselling Yes 288 64% 7 6 12

No 165 36% 13 9 20.5 5.8 3.6 9.5 <0.001

Prehabilitationexercise Yes 239 53% 7 6 12

No 214 47% 12 8 19 3.9 2.6 5.9 <0.001

Mini-Incision Yes 374 83% 8 6 13

No 79 17% 16 12 24 5.3 2.6 10.6 <0.001

NGTTube Yes 71 16% 19 13 25

No 382 84% 8 6 13 13.2 4.7 36.8 <0.001

RectussheathLA No 212 47% 13 8 20

Yes 241 53% 7 6 12 3.8 2.5 5.7 <0.001

SamedayAdmission Yes 376 83% 8 6 13

No 77 17% 16 12.5 23 31.0 7.5 127.9 <0.001

Oralbowelpreparation Yes 63 14% 16 13 24

No 390 86% 8 6 13 48.1 6.7 352.7 <0.001

Carbohydydrateloading Yes 364 80% 8 6 12

No 89 20% 16 12 22 14.2 5.7 35.9 <0.001

Fastingpre-op 2hrspre-op 284 63% 7 6 12

6hrspre-op 169 37% 13 9.0 20.5 4.9 3.1 7.8 <0.001

Drain Yes 433 96% 10 6.0 15.5

No 20 4.4% 7 5.3 11.5 2.6 1.0 6.4 0.04

Closure MassPDS0 331 73% 11 7.0 17

SheathPDS2/0 122 27% 7 6.0 12 3.0 1.9 4.9 <0.001

OralFluidsfromday1 Yes 403 89% 9 6.0 13

No 50 11% 19 14.0 25.3 11.3 3.5 37.0 <0.001

Chewinggum/candy Yes 393 87% 8 6.0 13

No 60 13% 18 13.0 25 45.5 6.2 331.3 <0.001

*Univariable:Mann–WhitneyUorKruskal–Wallistests.Multivariable:Logisticregressionforstaying±7days

Abbreviations:HbHemoglobin,NMINon-muscleinvasive

(a).

(b).

(c).

(d).

Figure1

RARC

RectusLA

Bowelprep

NGT

Admitto

surgery

Nodrain

Miniin

cision

“ERAS”

Components

Carbohydrate

Epidural

0 5

10

15

20

25

30

Lengthofstay(days)

0.0

0.2

0.4

0.6

0.8

1.0

151

101

151

201

251

301

351

401

Readmission rates(%)

0

500

1000

1500

2000

2500

3000

Bloodloss(mls)

451

Logrankp=0.9

Overallsurvival

Bladdercancerspecificsurvival

Logrankp=0.9

ERAS

Non-ERAS

Mean:10.710.3

St.dev:4.75.8

TTestp=0.6

92

5.8

2.5

95

3.3

1.7

0 20 40 60 80 100

Clear

Urothelial

Softtissue

Marginstatus(%)

Non-ERAS

ERAS

ChiSq.p=0.4

a). b).

c). d).

Followup(months) Followup(months)

Numberoflymphnodes

Figure2

SupplementaryFigure1

0

1

2

3

4

5

6

7

8

1 51 101 151 201 251 301 351 401

Opduration(hrs)

1

10

100

1 51 101 151 201 251 301 351 401

ReadmissionLOS(days)

(a).

(b).

451

451