Tranexamicacidforthetreatmentofpostpartum

haemorrhage:aninternationalrandomised,double
blindplacebocontrolledtrial

CLINICALTRIALPROTOCOL
ProtocolNumber:ISRCTN76912190

NUMBER DATE
FINALVERSION Version1.0 11May2009
AMENDMENT(ifany)

Version1.0

FULLTITLEOFS
SHORTTITLE:
TRIALACRONY
PROTOCOLNU
EUDRACTNUM
BACKGROUND:
Almostall(9
maternal mo
postpartum
themwilldie
importantca
Systemicant
surgical bloo
identified 2
tranexamic a
reduces tran
bleeding(RR
TXAsignifica
treatmentin
randomised
TXAinPPHc
participants.
thequalityo
toosmallto
sideeffects.
statethatTX
on which th
conducted.
AIM: The WO
hysterectom
woman with
thromboemb
OUTCOME: Ou
death(which
PRIMARY OUT
causeofdea
SECONDARYOU
(a) Death
(b) Surgical
SUMM

STUDY:

YM:
MBER:
MBER:
: Each year,
9%)ofthede
ortality accou
period. Abou
e,withanave
auseofmater
tifibrinolytica
od loss. A sys
11 randomis
acid (TXA) red
nsfused volum
R=0.67,95%CI
ntlyreducesu
intractablep
trials on the
conductedby
Althoughthe
ofthetrialsw
assesstheeff
Themostrec
XAmaybeuse
his recommen
OMAN Trial a
y and other
h clinically d
boliceffect,o
utcomes will
heveroccursf
TCOME:Thepri
thwillbedes
UTCOMES:
Intervention
MARY
Tranexam
Aninterna
WORLDMA
THEWOM
ISRCTN769
20080084
worldwide a
eathsareinlo
unting for be
ut 14 million
erageinterval
rnalmortality
agentsarewid
stematic revie
ed controlled
duces the ris
me by 1.1 un
I0.41to1.09)
uterineblood
postpartumha
effectiveness
theapplicant
erewasasign
waspoor.Non
fectsofTXAo
centlyupdate
edinthetrea
ndation is ba
aims to deter
morbidities (
diagnosed po
nbreastfedb
be collected
first).
imaryoutcom
cribed.
ns: including
Protoc
icacidforthe
ational,rando
ATERNALANTIFIB
ANTRIAL
912190
44138
bout 530,000
owandmiddle
etween one
mothers dev
lfromonsett
inhighincom
delyusedins
ew of random
d trials inclu
k of blood tr
its (95%CI 0.
).Therewasn
lossinwome
aemorrhagein
s of TXA in th
tsidentifiedt
nificantreduct
nehadadequ
ntheclinicall
edPPHtreatm
tmentofPPH
ased is low a
rmine the eff
surgical inter
stpartum hae
abieswillalso
at 42 days af
meistheprop
hysterectomy
Page1of36
colISRCTN769
etreatmentof
omised,double
BRINOLYTICTRIA
CLIN
0 women die
eincomecoun
quarter and
velop postpart
todeathofab
mecountriesw
surgerytopre
mised controlle
uding 20,781
ransfusion by
64 to 1.59). T
noevidenceo
enwithmeno
ntheUK.How
e treatment o
threetrialsof
tioninaverag
uateallocation
yimportante
mentguideline
Hifothermea
and recomme
fect of the ea
rventions, blo
emorrhage. T
obeassessed.
fter randomis
portionofwo
y, brace sut
912190
fpostpartum
eblind,place
L
NICALTRIALS.GO
e from causes
ntries.Obstet
one third of
tum haemorr
bout2to4h
whereitacco
eventclotbre
ed trials of an
randomised
a relative 39
TXA also red
ofanincrease
rrhagiaandis
wever,atpres
of PPH. A sys
theprophyla
gepostpartum
nconcealmen
endpointsofm
espreparedb
asuresfail,but
ends that fur
arly administr
ood transfusio
The use of
.
sation, at disc
menwhodie
ure (BLynch
V
haemorrhage
bocontrolled
VID: NCT00
s related to
trichaemorrh
f deaths, mo
rhage (PPH) e
ours.Obstet
untsforabou
akdown(fibri
ntifibrinolytic
d participants
9% (RR=0.61,
uces the nee
driskofthrom
srecommend
sentthereisli
stematic revie
acticuseofTX
mbloodlossin
nt andevenin
mortality,hys
ytheWorldH
tpointsoutth
rther clinical
ration of tran
on, risk of no
health servic
charge from r
orundergoh
/Cho), select
Versiondate:
e:
trial
872469
pregnancy an
hageisthelea
ost of which
each year and
richaemorrh
ut13%ofmat
nolysis)inord
agents in sur
s. The result
95%CI 0.54
ed for reoper
mboticevents
dedforconsid
ittlereliablee
ew of random
XA,including
nwomentrea
naggregateth
sterectomyan
HealthOrgani
hatthequalit
trials of TXA
nexamic acid
nfatal vascu
ces and safe
randomising h
hysterectomy
tive arterial
11May2009
nd childbirth.
adingcauseof
occur in the
d about 2% of
ageisalsoan
ternaldeaths.
dertoreduce
rgical patients
ts show that
to 0.69). TXA
ration due to
s.
derationasa
evidencefrom
mised trials of
atotalof460
tedwithTXA,
hetrialswere
ndthrombotic
zation(WHO)
tyofevidence
A in PPH are
on mortality,
lar events) in
ty, especially
hospital or at
.Theprimary
embolisation,
9
.
f
e
f
n
.
e
s
t
A
o
a
m
f
0
,
e
c
)
e
e
,
n
y
t
y
,
Page2of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

laparotomy for other reasons, manual removal of placenta, intrauterine tamponade (packing or gauzing the
uterine cavity, condomcatheter, any other method of intrauterine tamponade), artery ligation, to achieve
haemostasis
(c) Bloodtransfusionbloodorbloodcomponentunitstransfused
(d) HealthStatusmeasuredusingtheEQ5D
(e) Thromboembolicevents(myocardialinfarction,strokes,pulmonaryembolism,DVT)
(f) Otherrelevantmedicalevents
(g) Lengthofstayathospital/timespentatanintensivecareunit
(h) Needformechanicalventilation
(i) Statusofbreastfedbaby/ies
(j) Costeffectiveness
TRIAL DESIGN: A large, pragmatic, randomised, double blind, placebo controlled trial among 15,000 women with a
clinicaldiagnosisofpostpartumhaemorrhage
DIAGNOSISANDINCLUSION/EXCLUSIONCRITERIA:
M Alllegallyadultwomenwithclinicallydiagnosedpostpartumhaemorrhagefollowingvaginaldeliveryofababyor
caesareansection.TheclinicaldiagnosisofPPHmaybebasedonanyofthefollowing:
x estimatedbloodlossaftervaginaldeliveryofababy>500mLOR
x >1,000mLfromcaesareansectionOR
x bloodlosssufficienttocompromisethehaemodynamicstatusofthewoman
M The fundamental eligibility criterion is the responsible clinicians uncertainty as to whether or not to use an
antifibrinolyticagentinaparticularwomanwithpostpartumhaemorrhage.
x Women for whom the responsible doctor considers there is a clear indication for antifibrinolytic therapy
shouldnotberandomised.
x Womenforwhomthereisconsideredtobeaclearcontraindicationtoantifibrinolytictherapyshouldnotbe
randomised.
M Wheretheresponsibleclinicianissubstantiallyuncertainastotheappropriatenessofantifibrinolyticagentsina
particularwomanwithPPH.
M Therearenootherprespecifiedexclusioncriteria.
TEST PRODUCT, REFERENCE THERAPY, DOSE AND MODE OF ADMINISTRATION: A dose of tranexamic acid (1 gram by intravenous
injection)orplacebo(sodiumchloride0.9%)willbegivenassoonaspossibleafterrandomisation.Ifafter30minutes
bleedingcontinues,orifitstopsandrestartswithin24hoursafterthefirstdose,aseconddosemaybegiven.
SETTING: ThistrialwillbecoordinatedfromtheLondonSchoolofHygiene&TropicalMedicine(UniversityofLondon)
and conducted worldwide in hospitals in low, middle and high income countries. It is likely that most patient
recruitmentwillbeincountrieswithhighratesofmortalityandmorbidityfrompostpartumhaemorrhage.
DURATIONOFTREATMENTANDPARTICIPATION:Thefirstdosewillbegivenimmediatelyafterrandomisation.Ifrequired,the
seconddosewillbegivenupto24hoursafterthefirstdose.Nofurthertrialtreatmentwillbegiven.Participationwill
endatdischargefromrandomisinghospital,deathorat42dayspostrandomisationwhicheveroccursfirst.
CRITERIAFOREVALUATION:Allpatientsrandomlyassignedtooneofthetreatmentswillbeanalysedtogether,regardless
ofwhetherornottheycompletedorreceivedthattreatment,onanintentiontotreatbasis.
CLINICALPHASE: 3
PLANNEDTRIALSTART: May2009
PLANNEDDATEOFLASTPATIENTENROLMENT: 31December2014 PLANNEDDATEOFLASTOUTCOME 11February2015
Version1.0


T

S
T
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
4
5
6
A
A
A
a
b
c
d
e
TABLEO
Summary
Tableofcontent
1 INTROD
1.1 Needfo
1.2 Tranexa
1.3 Potentia
1.4 Objectiv
2 TRIALD
2.1 Overvie
2.2 Settings
2.3 Number
2.4 Recruitm
2.5 Eligibilit
2.6 Consent
2.7 Random
2.8 Treatme
2.8.1 Dosese
2.8.2 Drugma
2.8.3 Adminis
2.8.4 Othertr
2.9 Adverse
2.10 Unblind
2.11 Measur
2.12 Dataco
2.13 Monitor
2.14 Endoft
2.15 Analysis
3 TRIALO
3.1 Sponsor
3.2 Indemn
3.3 Protoco
3.4 Indepen
3.5 TrialSte
3.6 Collabo
3.7 TrialMa
3.8 Contact
3.9 Publicat
3.10 Financia
4 ABBREV
5 REFEREN
6 APPEND
Appendix1:En
Appendix2:Ou
Appendix3:Cou
a. Briefinform
b. Consentpr
c. Informatio
d. Informedc
e. Informedc
OFCONT
ts
DUCTION
oratrial
amicacidandit
alsideeffectso
ve
ESIGN
w:Pragmaticde
s
rofpatientsne
mentofcollabo
ty:Inclusioncrite
tandethicalco
misation
ent
election
anufacture,blin
strationoftrial
reatmentsforP
eEvents
ding
esofoutcome
llection
ring
trialforparticip
s
ORGANISATION
rshipandtrialm
ity
olDevelopment
ndentDataMon
eeringCommitt
ratorsrespons
anagementGro
tingtheTCCina
tion&Dissemin
alsupport
VIATIONSUSED
NCES
DICES
tryform
utcomeform
untry/sitespec
mationleafletf
rocedureoverv
nsheetforwo
consentformfo
consentformfo
Protoc
TENTS
tseffectonblee
oftranexamica
signandtheunce
eeded:Estimated
oratinginvestig
eria;Exclusioncrit
onsiderations
ndingandsupp
treatment
PPH
pants
ANDRESPONSI
management
t
nitoringComm
tee
sibilities
oup&TrialCoor
anemergency
nationofresult
D
ificdocuments
forpregnantwo
view
manandherre
orwoman
orrepresentativ
Page3of36
colISRCTN769
eding
cid
ertaintyprinciple
deventrate;Sam
ators
teria;Eligibilityg
plyoftrialtreatm
IBILITIES
ittee
rdinatingCentr
s
omen&family
epresentative
ve
912190
e;Randomisation
mplesizeandsize
raph
ment
reresponsibiliti
V
n;Followup
oftreatmenteff
es
Versiondate:
fect
11May2009
1
3
4
5
5
5
6
7
7
7
7
8
8
9
11
11
11
11
12
12
13
14
14
15
15
16
16
17
17
17
17
18
19
19
20
20
20
21
22
23
25
26
28
30
30
31
32
35
36
9
Version1.0

1

Eachyear,w
thesedeaths
isresponsibl

Postpartum
>1000mLaft
bloodlossof

Ofthe14mi
deathof2to
in hospital,
maternalmo

PPH also cau

borne viral i
figureincrea
transfused b
countries the
common.
9

Severe anae
year.
10
Seve
andtowork.

Systemicant
surgical bloo
identified 2
tranexamic a
patients, TXA
operation du
events.
18

TXAsignifica
treatment in
fromrandom
ofTXAinPPH
participants.
treated with
adequateall
clinicallyimp
treatmentgu
ofPPHifoth
andrecomm

1 IN
worldwide,abo
sareinlowan
eforbetween
haemorrhage
tercaesarean
faslittleas20
illionwomen
o4hours.
2
Al
where effect
ortalityinhigh
uses hospital
nfections. Ap
sesto5%or6
blood is consi
e risk of trans
mia is a com
reanaemiaca
11

tifibrinolytica
od loss. A syst
11 randomis
acid (TXA) red
A reduces tra
ue to bleeding
ntlyreducesu
n intractable p
misedtrialson
Hconductedb
20
Although
h TXA [weight
ocationconce
portantendpo
uidelinesprep
ermeasuresf
mendsthatfurt
NTRODU
out530,000w
ndmiddleinc
nonequarter
e (PPH) is com
section.How
00mLcanbeli
whohavePP
thoughmany
ive emergenc
hincomecoun
morbidity. M
proximately 1
6%forwomen
derably highe
sfusion transm
mon consequ
ancausedisa
agentsarewid
tematic review
ed controlled
duces the risk
ansfused volu
g (RR=0.67, 9
uterineblood
postpartum h
ntheeffective
bytheapplica
there was a
ed mean red
ealmentand
ointsofmorta
paredbytheW
fail,butpoint
therclinicaltr
Protoc
UCTION
womendiefro
comecountrie
andonethird
mmonly defin
wever,theseth
ifethreatenin
PHeachyear,
ydeathsfrom
cy care has t
ntries,accoun
Many women
1% of women
nwithinstrum
er in countrie
mitted infecti
uence of PPH
blingfatiguea
delyusedins
w of random
d trials inclu
k of blood tra
me by 1.1 un
95%CI 0.41 to
lossinwome
haemorrhage
enessofTXA
antsidentified
statistically s
uction of app
eveninaggre
ality,hysterec
WorldHealth
soutthatthe
rialsofTXAin
Page4of36
colISRCTN769
omcausesrel
es.
1
Haemorrh
dofobstetric
ned as blood
hresholdsdo
ngforawoma
about2%die
PPHoccurou
the potential
tingforabout
require bloo
n with sponta
mentaldeliver
es that do no
on is low, bu
and affects a
andseriously
urgerytopre
ised controlle
uding 20,781
ansfusion by
nits (95%CI 0
1.09). There
enwithmeno
in the UK.
19

inthetreatm
dthreetrialso
ignificant red
proximately 1
egatethetrial
ctomyandthr
Organization
equalityofev
PPHarecond
912190
latedtopregn
hage,whichu
deaths.
2

loss of >500m
nottakeinto
anwithsevere
e,withanave
utsidehealthc
to save lives
t13%ofmate
od transfusion
neous vagina
riesorcaesare
ot screen all b
ut adverse rea
about 11% of
reduceawom
eventclotbrea
ed trials of an
randomised
a relative 39%
.64 to 1.59).
was no evide
rrhagiaandis
However, at
entofPPH.A
oftheprophyl
duction in ave
00 mL] the q
lsweretoosm
romboticside
(WHO)state
idenceonwh
ducted.
V
nancyandchi
usuallyoccurs
mL after vagi
accountpre
eanaemiaorc
erageinterval
arefacilities,
s.
4,5
PPH is al
ernaldeaths.
6

n which some
l deliveries re
eansections.
7
blood for tran
actions related
f the 14 milli
manscapacity
akdown(fibri
ntifibrinolytic
participants
% (RR=0.61, 9
TXA may also
ence of an inc
srecommend
present there
Asystematicr
lacticuseofTX
erage postpar
uality of the
malltoassess
effects.Them
thatTXAmay
ichthisrecom
Versiondate:
ildbirth.Nearl
inthepostpa
inal delivery o
existinghealt
cardiacdiseas
fromonseto
asignificantn
so an import
etimes can tr
equire transfu
7
Theriskofin
nsfusion.
8
In
d to blood tra
on women w
ytolookafte
nolysis)inord
agents in sur
s. The result
95%CI 0.54 to
o reduce the
creased risk o
dedforconsid
e is little relia
eviewofrand
XA,including
rtum blood lo
trials was po
stheeffectso
mostrecently
ybeusedint
mmendationis
11May2009
lyall(99%)of
artumperiod,
of a baby, or
thstatus,and
se.
3

ofbleedingto
numberoccur
tant cause of
ansmit blood
usion, but the
nfectionfrom
high income
ansfusion are
with PPH each
rherchildren
dertoreduce
gical patients
ts show that
o 0.69). In all
need for re
of thrombotic
derationasa
able evidence
domisedtrials
atotalof460
oss in women
or. None had
ofTXAonthe
updatedPPH
thetreatment
sbasedislow
9
f
,
r
d
o
r
f
d
e
m
e
e
h
n
e
s
t
l

c
a
e
s
0
n
d
e
H
t
w
Page5of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

1.1 NEEDFORATRIAL

The WOMAN Trial will provide a reliable scientific basis for recommendations as to whether or not tranexamic acid
shouldbeusedinthetreatmentofPPH.IfTXAreducesmortalityinwomenwithPPH,thiswouldbeofconsiderable
significance worldwide. There is a global commitment to the Millennium Development Goal (MDG) of reducing
maternal deaths by threequarters by the year 2015, a commitment that requires a reduction of the maternal
mortalityratioby5.5%eachyear.Becausematernalhaemorrhageaccountsforoveraquarterofdeaths,aneffective
treatment for PPH would contribute importantly to the MDG of reducing maternal mortality. TXA might also reduce
theneedforhysterectomy,decreasetheriskofanaemiaandavoidtheneedforbloodtransfusion.Bloodisascarce
resource in many countries with a risk of transfusion transmitted infections. If TXA was effective in the hospital
setting, further research could be conducted to evaluate its use in thecommunity, possibly includingthe use oforal
ratherthanintravenousadministration.

The results of this trial will be disseminated by publication in peer reviewed medical journals, conference
presentations,andinanupdatedversionoftheCochranesystematicreviewoftreatmentsforpostpartumbleeding.
Thereisevidencethathospitalsparticipatinginmulticentretrialsaremorelikelytoimplementthetrialresults.
21
For
thisreason,alargeinternationalmulticentretrialliketheWOMANtrialcanbeexpectedtohaveasubstantialimpact
onclinicalpractice.Thelargenetworkofcollaboratingsiteswillensurethattheresultsaredisseminatedworldwide.

1.2 TRANEXAMICACIDANDITSEFFECTONBLEEDING

In the haemostatic process, coagulation occurs rapidly at the site of a damaged vessel building a tight net of fibrin,
while at the same time, the fibrinolytic system removes the fibrin deposits that could cause permanent vascular
occlusiononcevascularrepairhastakenplace.
22
Thecoagulationandfibrinolyticsystemarebelievedtobeinastate
ofdynamicbalancewhichmaintainsanintactvascularsystem.Tranexamicacidisapotentantifibrinolyticagentthat
exerts its effect by blocking lysine binding sites on plasminogen molecules and has the potential to enhance the
effectivenessofthepatientsownhaemostaticmechanisms.Consequently,clotbreakdown(fibrinolysis)isinhibited
andexcessiveorrecurrentbleedingisreduced.

During delivery, when the placenta separates from the uterine wall, a sequence of physiologic and haemostatic
changesoccurthatreducebleeding:strongmyometrialcontractions,increasedplateletactivity,amassivereleaseof
coagulantfactorsandaparallelincreaseinthefibrinolyticactivity.
23
Asaresult,thereisatheoreticalrationaleforthe
useofantifibrinolyticagentsinthetreatmentofpostpartumhaemorrhage.
18,24,25

1.3 POTENTIALSIDEEFFECTSOFTRANEXAMICACID

As TXA inhibits the breakdown of fibrin deposits already formed, it might theoretically increase the risk of
thromboembolism.However,thesystematicreviewofTXAinsurgerydidnotshowstatisticallysignificantincreasesin
therisksofanyofthethromboemboliceventsassessed.
14

Page6of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

Events
EffectofTXA
RR 95%CI
MyocardialInfarction 0.96 0.481.90
Stroke 1.25 0.473.31
Deepvenousthrombosis 0.77 0.371.61
Renalfailure 0.73 0.163.32

During pregnancy, women have an increased risk of thromboembolic events, compared with nonpregnant women.
Theabsoluteriskofsymptomaticvenousthrombosisduringpregnancyhasbeenestimatedtobebetween0.5and3.0
per 1,000 women based on studies using radiographic documentation.
2628
Studies using objective criteria for
diagnosis have found that antepartum deep vein thrombosis (DVT) is as common as postpartum thrombosis and
occurswithequalfrequencyinallthreetrimesters.
26

A populationbased cohort study estimated an incidence of thromboembolic events to be 200 per 100,000 woman
years.
29
DVTwasthreetimesmorecommonthanpulmonaryembolismandthromboemboliceventswerefivetimes
more likely in the postpartum period than during the pregnancy. This was particularly evident with pulmonary
embolism which was 15 times more likely to occur in the postpartum period than during the pregnancy.
Thromboemboliceventswillbecollectedroutinelyaspartofthedatacollectionprocessforthistrial.

TXA passes into breast milk in very low concentrations, approximately one hundredth of the concentration in the
maternal blood. An antifibrinolytic effect in the infant is very unlikely at this low concentration.
30
The
thromboemboliceffectsonbreastfedbabieswillbeassessedinthistrial.

TXAisnotanewdrugandisgenerallywelltolerated.Adverseeventsareuncommonandusuallymanifestasnausea
ordiarrhoea,oroccasionallyasorthostaticreactions.
18

1.4 OBJECTIVE

The WOMAN trial will provide reliable evidence as to whether the antifibrinolytic agent tranexamic acid reduces
mortality, hysterectomy and other morbidities in woman with clinically diagnosed postpartum haemorrhage.
Thromboemboliceffectsonbreastfedbabieswillbeassessed.

Version1.0

2

2.1 OVER

This trial is a
administratio
haveclinicall
eitherTXAo

Pragmatic d
treatment ac
approach to
substantially
graph 1). A
anymedical
inappropriat
couldbeoffe
trialtobeclo
providedwit

Randomisati
possible. The
consecutively
randomised,
given.

Followup:N
fromtheme
death(which
daysafterra

2.2 SETT

Thepragmat
Participating
Eligible wom
participating
womentobe

2.3 NUM

Twomainfa
thetreatmen

Estimatedev
mortality aft
somepartso
2 TR
RVIEW
a large, pragm
on of tranexa
lydiagnosedp
rplacebo.The
esign and the
ctually is in ro
trial eligibilit
yuncertainas
woman shou
ornonmedic
e for this pa
eredtothepa
osertowhati
thinformation
ion: Women
e Entry form
ynumberedt
theoutcome
Noextratests
dicalrecords
heveroccursf
ndomisation.
TINGS
ticnatureoft
g hospitals or
men may hav
ghospitaland
erecruitedat
MBEROFPA
ctorsdeterm
nteffect.
ventrate:Rev
ter PPH world
ofAfrica.The
RIALDES
matic, random
mic acid on d
postpartumh
eeligibilitycri
e uncertainty
outine everyd
y is well esta
towhichoft
uld not be enr
calreasonsre
rticular indivi
atientinorou
sappropriate
naboutthetr
eligible for in
(Appendix 1
treatmentpac
einhospitaln
arerequired
sixweeks(42
first).Anyadv

thistrialwilla
r maternal he
e delivered t
beenadmitte
eachsite.
ATIENTSNE
inethenumb
viewofthelit
dwide, varying
frequencyof
Protoc
SIGN
mised, double
death, hystere
haemorrhage
teriaarebase
y principle: T
day practice. T
blished.
31
A
thetrialtreat
rolled if the r
easonablycer
dual (in com
utsidethetria
einnormalme
ialtreatment
nclusion shou
) will be used
ck,takenfrom
needstobeco
forthetrialb
2days)afterra
verseeventsw
llowforther
ealth facilities
their babies a
edfollowingt
EDED
berofpatients
teratureandd
g from about
occurrenceo
Page7of36
colISRCTN769
blind, placeb
ectomy and o
andwhofulfi
edontheunce
The pragmati
The eligibility
patient can b
mentswould
esponsible cli
tainthatone
parison with
l).Usingthe
edicalpractice
toassistthem
uld be random
d to assess e
maboxofeig
ollectedeven
utashortOu
andomisation
whichbecome
recruitmentof
s will be sele
at the partici
hedeliveryof
sneededina
datafromhos
0.6% in the
ofperipartum
912190
o controlled t
other relevant
ltheeligibility
ertaintyprinc
c design will
criteria are b
be enrolled if,
bemostapp
inician or the
ofthetreatm
either no tre
uncertaintyp
e.Clinicians,w
mintheirjudg
mised, and th
eligibility and
htpacks,shou
ifthetrialtre
tcomeform(A
norondischa
eknowntoth
fwomenfrom
ected from hi
pating hospit
fababy.Ther
trial.Thesea
spitalreports
United Kingd
hysterectom
V
trial to quant
t outcomes.
ycriteria,will
ciple.
allow us to f
based on the
and only if, t
ropriatefort
woman (or h
mentsthatmi
eatment or so
principleshou
womenandth
gement.
he study treat
collect basel
uldbechosen
eatmentisint
Appendix2)m
rgefromther
heinvestigato
mawidevarie
gh, middle a
tal or may h
reisnolimitto
aretheestima
showsthatth
om to 2.6% i
iesalsovaries
Versiondate:
tify the effect
15,000 adult
lberandomis
find out how
uncertainty p
the responsib
hatparticular
her represent
ightbealloca
ome other tr
ldallowthep
heirrepresent
tment started
ine informati
n.Onceapat
terruptedori
mustbecomp
randomisingh
orwillberepo
etyofhealthc
nd low incom
ave delivered
othemaximu
atedeventrat
herearewide
n South Afric
s,fromabout
11May2009
s of the early
women, who
sedtoreceive
effective the
principle. This
ble clinician is
rwoman(see
tative) are for
tedwouldbe
eatment that
processofthis
tativeswillbe
d, as soon as
ion. The next
ienthasbeen
snotactually
pleteddirectly
hospitaloron
ortedupto42
carefacilities.
me countries.
d outside the
umnumberof
teandsizeof
evariationsin
a and 20% in
t0.02%inthe
9
y
o
e
e
s
s
e
r
e
t
s
e
s
t
n
y
y
n
2
.
.
e
f
f
n
n
e
Page8of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

UnitedKingdomto2%inNigeriaor14%inCongoBrazzaville.Basedontheseranges,abaselineeventrateof2.5%for
mortalityand2.5%forhysterectomymightreasonablybeexpected.

Samplesizeandsizeoftreatmenteffectthatshouldbedetectable:
Assumingacontrolgroupeventrateof2.5%formortalityand2.5%forhysterectomywith1%ofwomenhavingbotha
hysterectomy and then dying, a study with 15,000 women would have over 90% power (two sided alpha=5%) to
detect a clinically important 25% reduction from 4% to 3% in the primary endpoint of mortality or hysterectomy. A
surveyofbaselineeventratesamonghospitalsthathaveexpressedinterestintakingpartshowsthatbaselineevent
rates of this magnitude are realistic and that higher baseline event rates might reasonably be expected. Experience
from the CRASH1 and CRASH2 clinical trials suggests that the anticipated rates of loss to followup (less than 1%)
wouldnotimpactimportantlyonstudypower.

2.4 RECRUITMENTOFCOLLABORATINGINVESTIGATORS

The trial will recruit collaborating sites from all countries worldwide and will continue to add sites to ensure the
samplesizeisachieved.Suitablecollaboratingsitesandinvestigatorswillbeassessedonthelevelofobstetricservice
they provide and their ability to conduct the trial. In advance of the trial starting at a site the Principal Investigator
mustagreetoadheretoGoodClinicalPracticeGuidelinesandallrelevantregulationsintheircountry.Inaddition,all
relevantregulatoryandethicsapprovalswillneedtobeinplace.

2.5 ELIGIBILITY

Immediatelyafterdeliveryofthebaby/ies,allusualcareshouldbegivenforthepreventionofPPH.Somebleedingis
expectedafterdelivery.However,ifbleedingcontinuesandadiagnosisofPPHismade,allusualtreatmentsshould
be given and at the same time the assessment for inclusion in the trial should be made. It is important to consider
inclusionasearlyaspossible.

Inclusioncriteria:
Alllegallyadultwomenwithclinicallydiagnosedpostpartumhaemorrhagefollowingvaginaldeliveryofababyor
caesarean section; women may have delivered their babies at a participating hospital or outside a participating
hospital,withhospitaladmissionfollowingdelivery:
x wheretheresponsibleclinicianissubstantiallyuncertainastowhetherornottouseTXA
x whenconsenthasbeengivenaccordingtoapprovedprocedures

TheclinicaldiagnosisofPPHmaybebasedonanyofthefollowing:
x estimatedbloodlossaftervaginaldeliveryofababy>500mLOR
x >1000mLfromcaesareansectionOR
x estimatedbloodlossenoughtocompromisethehaemodynamicstatusofthewoman

Exclusioncriteria:
x Women for whom the responsible clinician considers there is a clear indication for TXA should not be
randomised.
x WomenforwhomtheresponsibleclinicianconsidersthereisaclearcontraindicationforTXAshouldnotbe
randomised(e.g.aknownthromboemboliceventduringpregnancy).

The fundamental eligibility criterion is the responsible clinicians uncertainty as to whether or not to use an
antifibrinolyticagentinaparticularwomanwithpostpartumhaemorrhage.
Version1.0

The TXA sum

ensure they
eachwoman

Graph1:Eli


2.6 CON

This trial w
haemorrhage
care.Eligible
be altered a
processinth
toICHGCPa

AdvanceInfo
of all pregn
antenatal/de
and obtain a
mmary of pro
have adequa
n.
igibility
SENTAND
ill be carried
e is an emerg
ewomenhav
s a result of
hissituationre
andtherequir
ormation:Th
nant women
eliveryperiod
advance cons
oduct charact
ate informatio
ETHICALCO
d out worldw
gency situatio
vealifethreat
their blood lo
equirescarefu
rementsinthe
emajorityof
to cause u
.Also,itisno
sent. Therefo
Protoc
teristics
30
and
on when con
ONSIDERAT
wide and wil
on and clinica
teningconditi
oss or labour
ulconsideratio
eDeclaration
womendeliv
undue concer
otpossibleto
ore, where po
Page9of36
colISRCTN769
d an Investiga
sidering the r
TIONS
l include wo
al activities w
ion.Furtherm
pains or by
onbearingin
ofHelsinki.
verwithoutco
rn by provid
identifyinad
ossible, a sum
912190
ators Brochu
riskbenefit a
omen soon a
will be directe
more,theirph
drugs admini
mindapplica
omplicationsa
ding detailed
dvancethose
mmary of the
V
re will be pr
nd the appro
fter delivery
ed towards th
hysical,menta
stered during
bleregulatory
anditwould
information
womenwho
e trial inform
Versiondate:
ovided to inv
opriateness of
of a baby.
he provision o
alandemotio
g the labour.
yrequirement
notbeinthe
n about this
willgoonto
mation will be
11May2009
vestigators to
f the trial for
Postpartum
of emergency
nalstatemay
The consent
ts,adherence
bestinterest
trial in the
developPPH,
e provided to
9
o
r
m
y
y
t
e
t
e
,
o
Page10of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

pregnant women (Appendix 3a). Refusal to be considered for participation will be documented in the womans
medicalrecordsandherdecisionrespected.

Followingdeliveryofherbaby,andonceawomanhasbeendiagnosedwithPPH,acriticalclinicalemergencysituation
exists. The risk of death is highest early after delivery. The process by which information will be given and consent
obtainedwilldependontheneedforurgentclinicalinterventionandherphysical,mentalandemotionalstate.Also,
theavailabilityandabilityofapersonalrepresentativetomakeadecisiononthewomansbehalfwillhavetobetaken
intoconsideration.Theapproachwhichwillallowthewomantohavethemostinputintothedecisionmakingprocess
withoutendangeringherlifewillbeutilised:

a) The woman is fully competent: The woman will be approached with the agreement of the primary carer (the
midwife or doctor) at the time of diagnosis. Factors which may impair her decision making process including pain,
altered level of consciousness due to drugs given and degree of blood loss, will be taken into consideration. An
Information Sheet (Appendix 3c) will be provided and the study will be discussed with her and a written consent
obtained(Appendix3d).Ifthewomanisunabletoreadorwrite,thentheinformationsheetmaybereadtoherand
shemaythenmarktheconsentformwitheitheracrossorthumbprint.Inthisevent,awitnessNOTassociatedwith
thetrial,mustprovideafullsignatureconfirmingthemark.

b) The womans mental capacity is impaired and either a Personal or Professional representative is available:
Informationshouldbegiventothewomantakingherlevelofmentalimpairmentintoconsideration.Oralrefusalby
thewomanshouldberespectedandsheshouldnotbeenrolled.
a. If a Personal Representative (PeR) who is knowledgeable about the womans values and beliefs is available, an
InformationSheetwillbeprovided.Opportunityforquestionsshouldbegivenandwrittenconsentobtained.If
thePeRisunabletoreadorwrite,thentheinformationsheetmaybereadtohim/herandamarkwitheithera
cross or thumbprint made on the consent form. In this event, a witness NOT associated with the trial, must
provideafullsignatureconfirmingthemark.
b. IfaPersonalRepresentativeisnotavailableandthewomanisunabletoprovidevalidinformedconsent,thenan
independentdoctor/midwife/othersitestaffallowedtofulfilthisrole(ideallytheprimarycarerifs/heisnot
partofthetrialteam)maybeaskedtoconsentasaProfessionalRepresentative(PrR).Informedconsentgiven
byarepresentativeshallrepresentthewomanspresumedwill.

c)ThewomansmentalcapacityisimpairedandneitheraPersonalnorProfessionalrepresentativeisavailable:In
situations where the woman is facing a clinical emergency and no PeR/PrR is available, the investigator and ONE
independent person (doctor or midwife) who is not participating in this trial may enrol the woman into the trial by
certifyinginwritinginthewomansmedicalrecordsthat:
x thewomanisfacingalifethreateningpostpartumhaemorrhage;
x thewomanisunabletogiveherconsentasaresultofhermedicalcondition;
x itisnotfeasibletocontactthewomansPeR/PrRtoobtainconsentwithinthewindowperiod;and
x neitherthewomannorthewomansPeR/PrRnoranymemberofthefamilyhasinformedtheinvestigatorof
anyobjectionstothewomanbeingusedasaparticipantinthistrial.
For women enrolled under such emergency consent procedure, the woman or her PeR or PrR should be informed
about the trial as soon as it is possible and asked to consent for continuation of any trial procedure. A summary
overviewoftheconsentprocedureisprovidedinAppendix3b.

Therequirementsoftherelevantethicscommitteewillbeadheredtoatalltimes.

Page11of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

2.7 RANDOMISATION

Randomisation codes will be generated and secured by an independent statistical consultant from Sealed Envelope
Ltd(UK).ThecodeswillbemadeavailabletoBreconPharmaceuticalsLimited(UK)explicitlyforthetreatmentpacksto
becreatedinaccordancewiththerandomisationlist.Eligibilitywillbedeterminedfromtheroutinelycollectedclinical
information and no trialspecific tests are required. Women eligible for inclusion should be randomised to receive
eitheractive(tranexamicacid)orplacebo(sodiumchloride0.9%)treatmentandthetrialtreatmentstartedassoonas
possible.

Baselineinformationwillbecollectedonthetrialentryformandthenextlowestconsecutivelynumberedpackwillbe
takenfromaboxofeighttreatmentpacks.Whenthetreatmentampouleisconfirmedasbeingintact,atthispointthe
patientisconsideredtoberandomisedontothetrial.TheentryformdatawillbesenttotheTrialCoordinatingCentre
assoonaspossible.Onceapatienthasbeenrandomised,theoutcomeofthewomanshouldbeobtainedevenifthe
trialtreatmentisinterruptedorisnotactuallygiven.

2.8 TREATMENT

Tranexamicacidwillbecomparedwithmatchingplacebo(sodiumchloride0.9%).

2.8.1 DOSESELECTION

In randomised trials of antifibrinolytic agents in surgery, TXA dose regimens vary widely. Loading doses range
from2.5mg/kgto100mg/kgandmaintenancedosesfrom0.25mg/kg/hourto4mg/kg/hourgivenoverperiods
of one to twelve hours. Studies examining the impact of different doses of tranexamic acid on bleeding and
transfusion requirements showed no significant differences between a high dose and a low dose. Studies in
cardiacsurgeryhaveshownthata10mg/kginitialdoseofTXAfollowedbyaninfusionof1mg/kg/hourproduces
plasmaconcentrationssufficienttoinhibitfibrinolysisinvitro.Horrowetal(1995)examinedthedoseresponse
relationship of TXA and concluded that 10 mg/kg followed by 1 mg/kg/hour decreases bleeding in cardiac
surgery, but larger doses did not provide any additional haemostatic benefit.
32
Trials of the use of TXA for the
preventionofobstetrichaemorrhageusedTXAatadoseof1gramwithoutmajorcomplications.
20

In the emergency situation, the administration of a fixed dose is more practicable since weighing women with
PPHwouldbedifficult.Therefore,afixeddoseof1gramofTXAinitiallyfollowedby1gramifbleedingcontinues,
whichiswithinthedoserangewhichhasbeenshowntoinhibitfibrinolysisandprovidehaemostaticbenefit,has
beenselectedfortheWOMANtrial.Onthebasisofexperienceinsurgery,thedoseselectedwouldbeefficacious
forlargerpatients(>100kg)butalsosafeinsmallerpatients(<50kg),astheestimateddose/kgthatthepatients
inthelattergroupwouldreceivehasbeenappliedinothertrialswithoutsignificantadverseeffects.

2.8.2 DRUGMANUFACTURE,BLINDINGANDSUPPLYOFTRIALTREATMENT

The active trial drug tranexamic acid (Cyklokapron Injection) will be purchased on the open market in the UK.
TXA is manufactured by Pfizer Ltd under Marketing Authorisation Number: PL 00032/0314. The Marketing
Authorisationguarantees that theproduct has been manufactured and released inaccordance with the United
KingdomsGoodManufacturingRegulations.

Page12of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

Placebo (sodium chloride 0.9%) will be manufactured specially to match the tranexamic acid by South Devon
Healthcare NHS Trust, Kemmings Close, Paignton, Devon, TQ4 7TW, under UK Manufacturers authorisation
Number:MS13079/MA(IMP)13079.

Ampoules and packaging will be identical in appearance. The blinding process and first stage Qualified Person
(QP) release will be done byBrecon Pharmaceuticals Limited, Wye Valley Business Park, HayonWye, Hereford
HR35PG,underUKManufacturersauthorisationNumberMIA11724/MIAIMP11724.Theblindingprocesswill
involve complete removal of the original manufacturers label and replacement with the clinical trial label
bearing the randomisation number which will be used as the pack identification. Other pack label text will be
identical for both TXA and placebo treatments and will be in compliance with requirements for investigational
medicinalproducts.TreatmentpackscontainingTXAandplacebowillbepackedinbalancedblocksof8(4TXA:
4Placebo)intoaboxinrandomorder.

BreconPharmaceuticalsLimitedwillalsoberesponsibleformaintainingtheProductSpecificationFile(PSF)until
final database lock and unblinding of the trial data. Quality control checks to assure blinding process will be
performed on a random sample of final QP released drug packs. High Performance Liquid Chromatography
analyses(HPLC)separationofknowntranexamicacidwillbeassessedagainstblindedsamplestoconfirmwhich
ampoulecontainstheplaceboandactivetreatments.Thetestedsampleswillbeunblindedtoassureaccuracyof
blinding.

The Trials Coordinating Centre (TCC) will be responsible for assuring all relevant approvals are available at the
TCCbeforereleaseofthetrialtreatmenttoasite.

2.8.3 ADMINISTRATIONOFTRIALTREATMENT

Eachtreatmentpackwillcontain:
x 4x500mgampoulesoftranexamicacidorplacebo
x 2xsterile10mLsyringeand21FGneedle
x Stickers(forattachingtodataformsandpatientmedicalrecords)

TREATMENT AMPOULES
DOSE(TRANEXAMIC
ACIDORPLACEBO)
ADMINISTRATIONINSTRUCTION
DOSE1 2 1gram
To be administered by intravenous injection at an
approximate rate of 1mL/minute to all randomised
womenassoonaspossibleafterrandomisation.
DOSE2 2 1gram
If after 30 minutes bleeding continues, or if it stops and
restarts within the 24 hours after the first dose, a second
dose may be given. To be administered by intravenous
injectionatanapproximaterateof1mL/minute.
Thetrialtreatmentinjectionsshouldnotbemixedwithbloodfortransfusion,orinfusionsolutionscontaining
penicillinormannitol.

2.8.4 OTHERTREATMENTSFORPPH

There is a wide spectrum of first and second line treatments of postpartum haemorrhage. As the trial will be
conducted worldwide, each participating site should follow its own clinical guidelines for the treatment of
postpartum haemorrhage. Information on other treatments given will be collected on the outcome form.
Page13of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

Tranexamic acid or placebo would be an additional treatment to the routine management of postpartum
haemorrhage.

2.9 ADVERSEEVENTS(AEs)

TXA has a well documented safety profile. No increase in thromboembolic risks associated with its use has been
shown to date. However, as discussed in Section 1.3 an expected complication of pregnancy is an increased risk of
thromboembolicevents.Thistrialwillcollectdataonallthromboemboliceventsassecondaryoutcomes,andallsuch
eventsareroutinelyreportedtotheindependentdatamonitoringcommittee(DMC)forunblindedreview.

Definitions:
Adverseevent(AE)
Anyuntowardmedicaloccurrenceaffectingatrialparticipantduringthecourseofaclinicaltrial

SeriousAdverseEvent(SAE)
Aseriousadverseevent(experience)isanyuntowardmedicaloccurrencethatatanydose
x resultsindeath;
x islifethreatening;
x requiresinpatienthospitalisationorprolongationofexistinghospitalisation;
x resultsinpersistentorsignificantdisability/incapacity;or
x isacongenitalanomaly/birthdefect.

AdverseReaction(AR)
Anadverseeventwhenthereisatleastapossibilitythatitiscausallylinkedtoatrialdrugorintervention

SeriousAdverseReaction(SAR)
SAEthatisthoughttobecausallylinkedtoatrialdrugorintervention

SuspectedUnexpectedSeriousAdverseReaction(SUSAR)
AnunexpectedoccurrenceofaSAR;thereneedonlybeanindexofsuspicionthattheeventisapreviouslyunreported
reactiontoatrialdrugorapreviouslyreportedbutexaggeratedorunexpectedlyfrequentadversedrugreaction.

ReportingofAdverseEventsforthistrial:Death,lifethreateningcomplicationsandprolongedhospitalstayarepre
specified outcomes to be reported in this trial and also to the independent data monitoring committee. This clinical
trialisbeingconductedinacriticalemergencycondition,usingadrugincommonuse.Itisimportanttoconsiderthe
naturalhistoryofthecriticalmedicaleventaffectingeachwomanenrolled,theexpectedcomplicationsofthisevent
andtherelevanceofthecomplicationstoTXA.

Adverse events to be reported using an adverse event reporting form will be limited to those NOT already listed as
primaryorsecondaryoutcomes,yet,whichmightreasonablyoccurasaconsequenceofthetrialdrug.Eventsthatare
part of the natural history of the primary event of PPH or expected complications of PPH should not be reported as
adverseevents.

In addition, if a woman is discharged from the randomising hospital before day 42 and is readmitted to hospital,
requires medical care for any reason or is known to have died, an adverse event form should be completed
irrespectiveofthecause.

Page14of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

If a Serious Adverse Event occurs, this should be logged by calling the Trial Coordinating Centre Emergency Helpline
and a written report submitted within 24 hours. The TCC will coordinate the reporting of all SAEs to all relevant
RegulatoryAgencies,EthicsCommitteesandlocalinvestigatorsasperlocallegalrequirements.

2.10 UNBLINDING

In general there should be no need to unblind the allocated treatment. If some contraindication to antifibrinolytic
therapy develops after randomisation, e.g. clinical evidence of thrombosis, the trial treatment should simply be
stopped and all usual standard care given. Unblinding should be done only in those rare cases when the clinician
believes that clinical management depends importantly upon knowledge of whether the patient received
antifibrinolyticorplacebo.Inthosefewcaseswhenurgentunblindingisconsiderednecessary,a24hourtelephone
service will be available and details provided in the Investigators Study File and wall posters. The caller will be told
whether the patient received antifibrinolytic or placebo. An unblinding report form should be completed by the
investigator.

2.11 MEASURESOFOUTCOME

Afterapatienthasbeenrandomised,outcomeinhospitalwillbecollectedevenifthetrialtreatmentisinterruptedor
isnotactuallygiven.NoextratestsarerequiredbutashortsinglepageOutcomeFormwillbecompleted6weeks(42
days)afterrandomisation,atdischargefromtherandomisinghospitaloratdeath(whicheveroccursfirst).

PrimaryOutcome:Theprimaryoutcomeistheproportionofwomenwhodieorundergohysterectomy.Theprimary
causeofdeathwillbedescribed.

Secondaryoutcomes:
(a) Death
(b) Surgical Interventions including hysterectomy; brace suture (BLynch/Cho); selective arterial embolisation;
laparotomy for other reasons; manual removal of placenta; intrauterine tamponade (packing or gauzing the
uterine cavity, condomcatheter, any other method of intrauterine tamponade); artery ligation, to achieve
haemostasis.
(c) Bloodtransfusionbloodorbloodcomponentunitstransfused
(d) HealthRelatedQualityoflife(HRQoL)willbemeasuredbytheproxyversionoftheEQ5Datdischargefromthe
randomising hospital or in hospital at 42 days after randomisation. The EQ5D includes single item measures of
mobility, selfcare, usual activities, pain/discomfort and anxiety/depression. Each item is coded using 3 levels
(1 = no problems; 2 = some problems; 3 = severe problems). The instrument includes a global rating of current
healthusingavisualanaloguescale(VAS)rangingfrom0(worstimaginable)to100(bestimaginable).TheEQ5D
isagenericmeasureofhealthstatusthatprovidesasimpledescriptiveprofileandasingleindexvaluethatcan
beusedintheclinicalandeconomicevaluationofhealthcare.
(e) Thromboembolicevents(myocardialinfarction,strokes,pulmonaryembolism,deepveinthrombosis)
(f) Medicaleventsincludingrenalfailure,AdultRespiratoryDistressSyndrome,hypertensivedisordersofpregnancy
(includingHELLPSyndrome,eclampsia,toxaemiaofpregnancy)andotheradverseeventsreported
(g) Lengthofstayathospital/timespentatanintensivecareunit
(h) Receiptofmechanicalventilation
(i) Status of baby/ies: The health status of the baby/ies will be ascertained and information collected on any
thromboemboliceventsinbreastfedbabies
Page15of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

(j) Costeffectiveness analysis: An economic analysis will be relevant if TXA clearly demonstrates efficacy in
achieving its clinical aims. In this case, the study will be undertaken in the form of a costeffectiveness analysis
withtheaimofestimatingtheincrementalcosteffectivenessratiocomparingtheuseofTXAwithnormalclinical
practice.Analysiswillbebasedonadjustedlifeyearsgained.AfurtheranalysiswillexploretheuseoftheEQ5D
data to quality adjust survival. In this study, the economic analysis is clearly bounded as virtually all significant
resourceusewilloccurintheinitialperiodofhospitalisation.Assuch,neitheralongtermresourceanalysisnor
ananalysisofoutofhospitalcostswillberequired.ThetrialuseofTXAislikelytomirroritsuseinnormalclinical
practice, hence the costeffectiveness estimated in the trial (adjusted for protocol driven costs) will closely
approximatecosteffectivenessinactualclinicalpractice.Dataonphysicalresourceconsumption(e.g.lengthand
nature of hospital stay) will be collected for each patient and a common unit cost at a country level will be
applied.Asensitivityanalysiswillbeundertakentoassesstherobustnessoftheeconomicanalysisinresponseto
variations in key variables such as drug prices. In all cases, the economic analysis will be integrated with the
clinicaltrialprocedurestooptimiseefficiencyandminimiseinconveniencetopatients.

2.12 DATACOLLECTION

ThistrialwillbecoordinatedfromLSHTMandconductedinhospitalsinlow,middleandhighincomecountries.Most
recruitment will be in countries with high rates of mortality and morbidity from postpartum haemorrhage. Data will
becollectedateachsitebylocalinvestigatorsandtransmittedtotheTCC.Onlydataoutlinedontheentry,outcome
andadverseeventformswillbecollectedforthistrial.

Relevantdataonanentryformwillbecollectedbeforerandomisationtoassesseligibilityandtheformcompletedif
randomised.Theoutcomeformshouldbecompletedatdeath,discharge fromtherandomisinghospitalor6weeks
(42 days) after randomisation whichever occurs first. This data should be collected from the womans and her
baby/iesroutinemedicalrecordsasnospecialtestsarerequired.

If the woman (or her PeR or PrR) withdraws a previously given informed consent or refuses to consent for
continuation in the trial, or if the woman dies and no consent is available from either a PeR/PrR, her data will be
handledasfollows:
x Datacollectedtothepointofwithdrawalofconsentwillbeusedaspartoftheintentiontotreatanalysis
x Allrelevantadverseeventsidentifiedwillbereportedasrequiredtoallrelevantauthorities

Toallowforvariationinavailabletechnologyfordatatransferavarietyofmethodswillbeusedinthistrial.Datawill
becollectedbytheinvestigatoronpapercasereportforms(CRFs)andtransmittedtotheTCCeitherasapaperform
(by fax or email) or by entering the data directly into the trial database. Data can also be transmitted by entry onto
electronicdatafileswhichcanbeemailedoruploadedtotheTCCsecurewebserver.Incaseswhereelectronicdata
files are used, data stored on the investigators computer(s) and data during transfer will be secured by encryption.
Thedatawillbeusedinaccordancewithlocallawandethicscommitteeapproval.

2.13 MONITORING

GCPsection5.18.3statesinregardtomonitoring,Thedeterminationoftheextentandnatureofmonitoringshould
bebasedonconsiderationssuchastheobjective,purpose,design,complexity,blinding,sizeandendpointsofthetrial.
In general there is a need for onsite monitoring, before, during, and after the trial; however in exceptional
circumstancesthesponsormaydeterminethatcentralmonitoringinconjunctionwithproceduressuchasinvestigators
trainingandmeetings,andextensivewrittenguidancecanassureappropriateconductofthetrialinaccordancewith
GCP.Statisticallycontrolledsamplingmaybeanacceptablemethodforselectingthedatatobeverified.
Page16of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

Thistrialisalarge,pragmatic,randomisedplacebocontrolledtrial.Theintervention(tranexamicacid)hasmarketing
authorisationinmanycountriesandhasbeeninclinicaluseforover40years.Itssafetyprofileiswellestablishedand
nosignificantseriousadverseeventsassociatedwithitsusehavebeenidentified.Thetrialwillroutinelycollectdata
onadverseeventswhichmaytheoreticallybeassociatedwiththisproductandtheconditionunderinvestigation,and
these will be reviewed routinely by the independent Data Monitoring Committee (DMC). Other than consent, the
administrationofthetrialdrugusingaroutineclinicalprocedureandcollectingroutineclinicalinformationfromthe
medical records, there are no complex procedures or interventions for the participants or investigators in this trial.
Clinical management for underlying conditions will remain as per each hospitals standard protocol. Based on these
factors, the probability of harm or injury (physical, psychological, social or economic) occurring as a result of
participationinthisresearchstudyhasbeenassessedaslowrisktoparticipantsineachofthesecategories.Basedon
thelowrisksassociatedwiththistrial,aMonitoringPlantoassureappropriateconductofthetrialwillbedeveloped
which will incorporate 100% central monitoring in conjunction with procedures such as investigator training and
meetingsandwrittenguidance. Inaddition,alldatawillbesubjecttostatisticalmonitoringandat least10%ofdata
willbesubjectedtoonsitemonitoring.
Investigators/institutionsarerequiredtoprovidedirectaccesstosourcedata/documentsfortrialrelatedmonitoring,
audits, ethics committee review and regulatory inspection. All trial related and source documents must be kept for
fiveyearsaftertheendofthetrial.

2.14 ENDOFTRIALFORPARTICIPANTS

The trial ends either at death, discharge, or six weeks postrandomisation, whichever occurs first. If during the
treatment phase a woman develops an adverse event the trial drug should be stopped, woman treated in line with
localproceduresandthenfollowedup.

ThetrialmaybeterminatedearlybytheTrialSteeringCommittee(TSC).TheDMCmaygiveadvice/recommendation
fortheearlyterminationofthetrialbuttheTSCisresponsibleforthefinaldecision.

2.15 ANALYSIS

The main analyses will compare all those allocated antifibrinolytic treatment versus those allocated placebo, on an
intention to treat basis, irrespective of whether they received the allocated treatment or not. Results will be
presentedasappropriateeffectestimateswithameasureofprecision(95%confidenceintervals).Subgroupanalyses
for the primary outcome will be based on type of delivery (vaginal or caesarean section); administration or not of
prophylacticuterotonics;andonwhethertheclinicaldecisiontoconsidertrialentrywasbasedprimarilyonestimated
bloodlossaloneoronhaemodynamicinstability.Interactiontestwillbeusedtotestwhethertheeffectoftreatment
(ifany)differsacrossthesesubgroups.Betweensitesheterogeneityineffectivenesswillbeexplored.Allanalyseswill
be conducted in STATA. A detailed Statistical Analysis Plan setting out full details of the proposed analyses will be
finalisedbeforethetrialdatabaseislockedforfinalanalysis.

Version1.0

3

3.1 SPON

TheWOMAN
coordinated
beoutlinedi

3.2 INDE

LSHTMaccep
any nonneg
annualbasis

3.3 PROT
The Protoco
agreeing to
SteeringCom

CHIEFIN
Professor
TrialsCoo
LondonS
KeppelSt
Email:Ian
TRIALM
MsHalee
TrialsCoo
LondonS
KeppelSt
Email:Ha
STATIST
Professor
MedicalS
LondonS
KeppelSt
Email:Dia

3 TR
NSORSHIPA
NTrialisspon
bytheTrialC
nrelevantag
EMNITY
ptsresponsib
gligent harm s
andLSHTMa
TOCOLDEV
l Committee
the final pro
mmittee.
NVESTIGATO
rIanRoberts
ordinatingCen
SchoolofHygi
treet,London
n.Roberts@Ls
MANAGEMEN
emaShakur,S
ordinatingCen
SchoolofHygi
treet,London
aleema.Shaku
ICIAN
rDianaElbour
StatisticsUnit
SchoolofHygi
treet,London
ana.Elbourne@
RIALOR
ANDTRIAL
nsoredbythe
CoordinatingC
reements.The
ilityattached
suffered by an
assuresthatit
VELOPMENT
consists of th
otocol. Subseq
R
ntre
ene&Tropica
WC1E7HT,U
shtm.ac.uk
NT
eniorLecture
ntre
ene&Tropica
WC1E7HT,U
r@Lshtm.ac.u
rne
ene&Tropica
WC1E7HT,U
@Lshtm.ac.uk

Protoc
GANISA
MANAGEM
LondonScho
Centre(TCC).
eresponsibilit
toitssponso
nyone as a re
willcontinue
T
he following i
quent change
alMedicine
UK
r
alMedicine
UK
uk
alMedicine
UK
k
Page17of36
colISRCTN769
ATIONA
MENT
olofHygiene
TheTCCma
tiesoftheTCC
orshipofthet
esult of partic
renewalofth
nvestigators w
es to the fina
CLIN
Profe
Divisi
Unive
Liverp
Crow
Email
DrMe
Depa
World
Aven
Email
Profe
Infect
Londo
Keppe
Email
912190
ANDRES
&TropicalM
ydelegatere
Cwillbeover
rialand,assu
ipating in this
heindemnity
who will be r
al Protocol w
ICALEXPERT
essorZarkoAlf
onofPerinata
ersityofLiverp
poolWomen's
nStreet,Liver
l:zarko@Liv.a
etinGlmezog
rtmentofRep
dHealthOrga
ueAppia20,C
l:gulmezoglum
essorCarineR
tiousDiseases
onSchoolofH
elStreet,Lond
l:Carine.Rons
V
SPONSIB
Medicine(LSHT
esponsibilities
seenbytheT
uch,wouldbe
s trial. The in
forthedurati
esponsible fo
will require th
TS
firevic
alandReprod
pool
sHospital
rpoolL87SS,
ac.uk
glu
productiveHe
anization
CH1211Gene
m@who.int
onsmans
sEpidemiolog
Hygiene&Tro
donWC1E7H
smans@Lshtm
Versiondate:
BILITIES
TM)anditsre
tothirdpart
TrialManagem
eresponsible
demnity is re
onofthistria
or the develop
he agreement
ductiveMedic
UK
ealthandRese
eva27,Switze
gyUnit
opicalMedicin
HT,UK
m.ac.uk
11May2009

esponsibilities
ieswhichwill
mentGroup.
forclaimsfor
enewed on an
al.
pment of and
t of the Trial
cine
earch
erland
ne
9
s
l
r
n
d
l
Page18of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

3.4 INDEPENDENTDATAMONITORINGCOMMITTEE(DMC)

Membership:
NAME AFFILIATION EXPERTISE
ProfessorSirIainChalmers JamesLindInitiative,Oxford,UK
Largescalerandomisedcontrolledtrials;
Obstetriccare
ProfessorPisakeLumbiganon
ProfessorofObstetrics&Gynaecology;
Convenor,ThaiCochraneNetwork;
FacultyofMedicine,KhonKaen
University,Thailand
Obstetriccare
DrGildaPiaggio StatistikaConsultoria,SoPaulo,Brazil
Statistician(Extensiveexperienceof
reproductivehealthandresearchatthe
WorldHealthOrganization)

Mortalityandseveremorbidityisexpectedwithinthetargetpopulation.Toprovideprotectionforstudyparticipants,
an independent DMC has been appointed for this trial to oversee the safety monitoring. The DMC will review on a
regular basis accumulating data from the ongoing trial and advise the Trial Steering Committee regarding the
continuingsafetyofcurrentparticipantsandthoseyettoberecruited,aswellasreviewingthevalidityandscientific
meritofthetrial.

TheDMCcomposition,name,titleandaddressofthechairmanandofeachmember,willbegivenintheDMCCharter
which will be in line with that proposed by the DAMOCLES Study Group.
33
Membership includes expertise in the
relevantfieldofstudy,statisticsandresearchstudydesign.TheDMCCharterincludes,butisnotlimitedto,defining:
(a) thescheduleandformatoftheDMCmeetings
(b) theformatforpresentationofdata
(c) themethodandtimingofprovidinginterimreports
(d) stoppingrules

StandardOperatingProcedures:TheDataMonitoringCommittee(DMC)hastheresponsibilityfordecidingwhether,
whilerandomisationisinprogress,theunblindedresults(ortheunblindedresultsforaparticularsubgroup),should
be revealed to the TSC. The DMC Charter states that they will do this if, and only if, two conditions are satisfied:
(1) the results provide proof beyond reasonable doubt that treatment is on balance either definitely harmful or
definitelyfavourableforall,orforaparticularcategoryof,participantsintermsofthemajoroutcome;(2)Theresults,
ifrevealed,wouldbeexpectedtosubstantiallychangetheprescribingpatternsofclinicianswhoarealreadyfamiliar
with any other trial results that exist. Exact criteria for proof beyond reasonable doubt are not, and cannot be,
specified by a purely mathematical stopping rule, but they are strongly influenced by such rules. DMC Charter is in
agreement with the PetoHaybittle
34,35
stopping rule whereby an interim analysis of major endpoint would generally
need to involve a difference between treatment and control of at least three standard errors to justify premature
disclosure. An interim subgroup analysis would, of course, have to be even more extreme to justify disclosure. This
rulehastheadvantagethattheexactnumberandtimingofinterimanalysesneednotbeprespecified.Insummary,
thestoppingrulesrequireextremedifferencestojustifyprematuredisclosureandinvolveanappropriatecombination
ofmathematicalstoppingrulesandscientificjudgment.

Page19of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

3.5 TRIALSTEERINGCOMMITTEE
Membership:
NAME AFFILIATION EXPERTISE
ProfessorAdrianGrant
(Chair)
Director,HealthServicesResearchUnit,
UniversityofAberdeen
HealthServicesResearch;Randomised
ControlTrials
ProfessorIanRoberts
(PrincipalInvestigator)
LondonSchoolofHygiene&Tropical
Medicine
Epidemiology;RandomisedControlTrials;
Conductoflargescaleinternationaltrials
DrMetinGlmezoglu
DrMetinGlmezoglu
WorldHealthOrganization,Geneva
Obstetrician;CoordinatingEditorofthe
WHOReproductiveHealthLibrary;
RandomisedControlTrials
DrKaosarAfsana BRACHealthProgramme,Bangladesh
Reproductive&SexualHealth&Rights;
RuralandUrbanMaternal,Neonataland
ChildHealthProgrammeinBRAC
DrOladapoOlayemi
UniversityCollegeHospital,Ibadan,
Nigeria
ConsultantObstetrician;perspectiveon
obstetricsinadevelopingcountry
ProfessorBeverleyHunt KingsCollege,London
ProfessorofThrombosis&Haemostasis,
RandomisedControlTrials

TheroleoftheTrialSteeringCommittee(TSC)istoprovideoverallsupervisionofthetrial.Inparticular,theTSCwill
concentrate on the progress of the trial, adherence to the protocol, patient safety and consideration of new
information.TheTSCmustbeinagreementwiththefinalProtocoland,throughoutthetrial,willtakeresponsibility
for:
(a) majordecisionssuchasaneedtochangetheprotocolforanyreason
(b) monitoringandsupervisingtheprogressofthetrial
(c) reviewingrelevantinformationfromothersources
(d) consideringrecommendationsfromtheDMC
(e) informingandadvisingtheTrialManagementGrouponallaspectsofthetrial

Thesteeringcommitteeconsistsofexperiencedobstetricexperts,clinicaltrialistsaswellasaReproductive&Sexual
Health&Rightsrepresentative.Facetofacemeetingswillbeheldatregularintervalsdeterminedbyneed,butnoless
thanonceayear.ATSCCharterwillbeagreedatthefirstmeetingwhichwilldetailhowitwillconductitsbusiness.

Whenoutcomedataareavailablefor1,000trialparticipants,theTSCwillreviewtherateofrecruitmentintothetrial
and the overall event rates. The TSC will consider the extent to which the rate of recruitment and the event rates
correspondtothoseanticipatedbeforethetrialandwilltakewhateveractionisneededinlightofthisinformation.

3.6 COLLABORATORSRESPONSIBILITIES

CoordinationwithineachparticipatinghospitalwillbethroughalocalPrincipalInvestigatorwhoseresponsibilitywill
bedetailedinanagreementinadvanceofstartingthetrialandwillinclude:
x Ensureallnecessaryapprovalsareinplacepriortostartingthetrial
x Delegatetrialrelatedresponsibilitiesonlytosuitablytrainedandqualifiedpersonnel
x Train relevant medical and nursing staff who see obstetric patients and ensure that they remain aware of the
stateofthecurrentknowledge,thetrialanditsprocedures(therearewallcharts,pocketsummariesandasetof
slidestoassistwiththis)
x Agreetocomplywiththefinaltrialprotocolandanyrelevantamendments
x Ensurethatallwomenwithpostpartumhaemorrhageareconsideredpromptlyforthetrial
Page20of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

x Ensureconsentisobtainedinlinewithlocalapprovedprocedures
x EnsurethatthepatiententryandoutcomedataarecompletedandtransmittedtotheTCCinatimelymanner
x EnsuretheInvestigatorsStudyFileisuptodateandcomplete
x EnsureallAdverseEventsarereportedpromptlytotheTCC
x Accountabilityfortrialtreatmentsattheirsite
x EnsurethetrialisconductedinaccordancewithICHGCPandfulfilsallnationalandlocalregulatoryrequirements
x Allowaccesstosourcedataformonitoring,auditandinspection
x Be responsible for archiving all original trial documents including the data forms for five years after the end of
thetrial

3.7 TRIAL MANAGEMENT GROUP (TMG) AND TRIAL COORDINATING CENTRE (TCC)
RESPONSIBILITIES

x TheTrialManagementGroupwillconsistoftheProtocolCommitteemembers(Section3.3)plusatrialmanager,
datamanagerandtrialadministrator.
x The TCC will act on behalf of the Sponsor and will be responsible to the TMG to ensure that all Sponsors
responsibilitiesarecarriedout.Theresponsibilitieswillinclude(butnotlimitedto):
o ReporttotheTrialSteeringCommittee
o MaintaintheTrialMasterFile
o Identifytrialsites
o Confirmallapprovalsareinplacebeforereleaseofthetrialtreatmentandthestartofthetrialatasite
o Providetrainingaboutthetrial
o Providestudymaterials
o Datamanagementcentre
o 24houradviceandunblindingservice
o Givecollaboratorsregularinformationabouttheprogressofthestudy
o Respondtoanyquestions(e.g.fromcollaborators)aboutthetrial
o Ensuredatasecurityandqualityandobservedataprotectionlaws
o Safetyreporting
o EnsuretrialisconductedinaccordancewiththeICHGCP
o Statisticalanalysis
o Publicationoftrialresults

3.8 CONTACTINGTHETCCINANEMERGENCY

Forurgentenquiries,adverseeventreportingandunblindingqueriesinvestigatorscancontactthe24hourtelephone
serviceprovidedbytheTCC.AcentraltelephonenumberisgivenintheInvestigatorsStudyFileandposters.

3.9 PUBLICATIONANDDISSEMINATIONOFRESULTS

AlleffortswillbemadetoensurethatthetrialprotocolandresultsarisingfromtheWOMANtrialarepublishedinan
establishedpeerreviewedjournal.Atleastonepublicationofthemaintrialresultswillbemade.Allpublicationswill
follow relevant external guidance such as the Uniform Requirements for Submission of Manuscripts to Biomedical
Journals issued by the International Committee of Medical Journal Editors (ICMJE) (2008 update) and the CONSORT
statement.
36,37
Links to the publication will be provided in all applicable trial registers. Dissemination of results to
patients will take place via the media, trial website (www.thewomantrial@Lshtm.ac.uk) and relevant patient
organisations.Collaboratinginvestigatorswillplayavitalroleindisseminatingtheresultstocolleaguesandpatients.
Page21of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

The success of the trial will be dependent entirely upon the collaboration of midwives, nurses and doctors in the
participating hospitals and those who hold key responsibility for the trial. Hence, the credit for the study will be
assigned tothe key collaborator(s) from a participating site as it is crucial that those taking credit for the work have
actuallycarrieditout.Theresultsofthetrialwillbereportedfirsttotrialcollaborators.

3.10 FINANCIALSUPPORT

LSHTMisfundingtherunincostsforthistrialandupto2,000patientsrecruitment.Fullfundingisbeingsoughtfrom
publicfundingorganisations.Fundingforthistrialcoversmeetingsandcentralorganisationalcostsonly.Pfizer,the
manufacturer of tranexamic acid, have provided the funding for the trial drug and placebo used for this trial. The
designandmanagementofthestudyareentirelyindependentofthemanufacturersoftranexamicacid,whichisnota
newproduct.

Largetrialsofsuchdrugs,involvingmanyhospitals,areimportantforfuturepatients,butarepracticableonlyifthose
collaborating in them do so without payment (except for recompense of any minor local costs that may arise).
Agreementforrepaymentoflocalcostswillbemadeinadvance.

Version1.0

4

4 AB
AE
AR
CONSORT
CRF
DMC
DVT
FG
GCP
HELLP
HPLC
HRQoL
ICHGCP
ICMJE
kg
LSHTM
MDG
mg
mL
PeR
PPH
PrR
PSF
QP
SAE
SAR
SUSAR
TCC
TMG
TSC
TXA
UK
VAS
WHO
BBREVIA

Adverse
Adverse
CONsol
CaseRe
DataM
DeepV
French
GoodC
Hemoly
HighPe
Health
Interna
Interna
Kilogra
London
Millenn
Milligra
Millilitr
Persona
Postpar
Profess
Produc
Qualifie
Serious
Serious
Suspect
TrialsC
TrialM
TrialSte
Tranexa
United
VisualA
WorldH
Protoc
ATIONS
eEvent
eReaction
lidatedStanda
eportForm
MonitoringCom
VeinThrombos
Gauge
ClinicalPractic
yticanemia,E
erformanceLi
RelatedQuali
ationalConfer
ationalCommi
m
nSchoolofHy
niumDevelop
am
re
alRepresenta
rtumHaemor
sionalReprese
tSpecification
edperson
sAdverseEven
sAdverseRea
tedUnexpect
CoordinatingC
anagementG
eeringComm
amicAcid
Kingdom
AnalogueScor
HealthOrgani
Page22of36
colISRCTN769
USED
ardsofRepor
mmittee
sis
ce
ElevatedLiver
quidChromat
ityofLife
enceonHarm
itteeforMed
ygiene&Trop
mentGoal
ative
rrhage
entative
nFile
nt
ction
tedSeriousAd
Centre
Group
ittee
re
isation
912190
tingTrials
enzymesand
tography
monisationof
icalJournalEd
icalMedicine
dverseReactio
V
LowPlatelet
GoodClinical
ditors
on
Versiondate:
count
Practice
11May20099
Version1.0

5

1. World He
Estimates d
Organisation
http://www.
2009).

2. AbouZahr
EditedbyMu
theWorldBa

3.LalondeA,
GynaecolOb

4.Ronsmans

5.Kongnyuy
ofMalawi:a

6.KhanKS,W
review.Lanc

7. Ekeroma
1997;104(3):

8.WorldHea

9.TaylorC,C
HazardsofT

10.AbouZah

11. World
information.

12. Mousa
2007(1):CD0

13. Nayama
countries:A

14. Henry DA
useforminim

15. WHO Re
2007.

16. Prendivi
CochraneDa

17. Gulmezo
Lumbiganon
misoprostol

18.DunnCJ,

5 RE
ealth Organiz
eveloped by
n.Availablefro
.who.int/repro
C. Antepartu
urray,JLopez
ank,1998:165
,DavissBA,A
bstet2006;94(
sC,GrahamW
EJ,MlavaG,v
nanalysisofc
WojdylaD,Sa
cet2006;367(9
AJ, Ansari A
:27884.
althOrganisat
CohenH,Jone
ransfusionAn
rC.Globalbu
Health Or
WHO/MCH/M
HA, Alfirevic
03249.
M, Moulaye
vitalact.Pros
A, Carless PA,
misingperiope
ecommendati
lle WJ, Elbou
atabaseSystR
oglu AM, Vill
P,UngerC,P
inthemanage
GoaKL.Trane
EFERENC
ation UNCF,
WHO, UNIC
om:
oductiveheal
um and Postp
z,A.Boston.H
5187.
AcostaA,Hers
(3):24353.
WJ.Maternalm
vandenBroe
causesandch
ayL,Gulmezo
9516):106674
A, Stirrat GM
tion.GlobalD
esH,AsherD,
nnualReport2
urdenofmate
ganization.
MSM/92.2.htt
c Z. Treatm
AA, Djibrill B
spectivestudy
, Moxey AJ, O
erativealloge
ons for the p
urne D, McDo
Rev2000(2):CD
lar J, Ngoc N
rendivilleW,
ementofthe
examicacid:a
Protoc
CES
United Natio
CEF, UNFPA,
lth/publicatio
artum Haemo
HarvardSchoo
chderferK.Po
mortality:who
kN.Facilityb
aracteristicso
ogluAM,Van
4.
M. Blood tran
atabaseonBl
BrantL,Chap
2007.Editedb
ernaldeathan
The prevale
tp://whqlibdo
ent for prim
B, Garba M,
yinareferenc
O'Connell D, S
neicbloodtra
prevention of
onald S. Activ
D000007.
NT, Piaggio G
PinolA,Elbou
thirdstageof
areviewofits
Page23of36
colISRCTN769
ons Populatio
and The W
ns/maternal_
orrhage. Heal
olofPublicHe
ostpartumhe
o,when,wher
basedmaterna
ofmaternalde
LookPF.WH
nsfusion in o
oodSafety.E
pmanC,Davie
bySHoTSCom
nddisability.B
ence of an
oc.who.int/hq
mary postpar
Idi N, Bouker
ceNigerianma
Stokes BJ, Mc
ansfusion.Coc
f postpartum
ve versus ex
G, Carroli G,
urneD,ElRefa
flabour.Lance
suseinsurger
912190
on Fund, Wo
World Bank. G
_mortality_20
th Dimension
ealthonbeha
morrhagetod
re,andwhy.L
aldeathrevie
eaths.Women
Oanalysisof
obstetrics an
ditedbyWHO
esT,GrayA,M
mmitee.Londo
BrMedBull20
aemia in w
/1992/(WHO
rtum haemor
rrou M. [Hae
aternity].Gyn
cClelland B, La
chraneDataba
haemorrhage
pectant man
Adetoro L, A
aeyH,Schulz
et2001;358(9
ryandotherin
V
rld Bank. Ma
Geneva. Edite
05/mme_200
ns of Sex and
lfoftheWorl
day:ICM/FIGO
Lancet2006;3
ewinthreedis
nsHealthIssu
causesofma
d gynaecolog
O.Geneva.,20
MilkinsC,Nor
on(UK);2008.
003;67:111.
women: a
O_MCH_MSM_
rrhage. Coch
emostatic hyst
necolObstetF
aupacis A, Fer
aseSystRev2
e. Geneva: W
agement in t
AbdelAleem
K.WHOmult
9283):68995.
ndications.Dr
Versiondate:
aternal Morta
ed by the W
05.pdf., (Acce
Reproduction
ldHealthOrg
Oinitiative20
368(9542):118
strictsinthec
ues2009;19(1
aternaldeath:
gy. Br J Obs
0012002:13
rfolkD,Tinega
.,2007.
tabulation
_92.2.pdf),19
rane Databa
terectomies i
Fertil2006;34(
rgusson D. An
2007(4):CD00
World Health
the third stag
H, Cheng L,
icentrerando

rugs1999;57(
11May2009
ality in 2005.
World Health
ssed 11 May
n. 1st edition.
anisationand
042006.IntJ
89200.
centralregion
):1420.
:asystematic
stet Gynaecol
2.
ateH.Serious
of available
992.
ase Syst Rev
in developing
(10):9005.
ntifibrinolytic
1886.
Organisation,
ge of labour.
Hofmeyr G,
omisedtrialof
6):100532.
9
.
h
y
.
d
J
n
c
l
s
e
v
g
c
,
.
,
f
Page24of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

19. National Collaborating Centre for Womens and Childrens Health. Intrapartum care of healthy women and their
babiesduringchildbirth.ClinicalGuidance.RCGOPress,September2007.

20.FerrerPR,I.Sydenham,E.Blackhall,K.Shakur,H.AntiFibrinolyticAgentsinObstetricHaemorrhage:ASystematic
Review.BMCPregnancyChildbirthmanuscriptID4090955672420008Submitted.

21. Ketley D, Woods KL. Impact of clinical trials on clinical practice: example of thrombolysis for acute myocardial
infarction.Lancet1993;342(8876):8914.

22.PrenticeCR.Basisofantifibrinolytictherapy.JClinPatholSuppl(RCollPathol)1980;14:3540.

23.HellgrenM.Hemostasisduringnormalpregnancyandpuerperium.SeminThrombHemost2003;29(2):12530.

24. Bonnar J, Guillebaud J, Kasonde JM, Sheppard BL. Clinical applications of fibrinolytic inhibition in gynaecology. J
ClinPatholSuppl(RCollPathol)1980;14:559.

25.BolteAC, Bouma L, vanGeijn HP. Medical therapies for primarypostpartumhemorrhage. International Congress
Series.Gynaecology,Obstetrics,andReproductiveMedicineinDailyPractice2005;1279:364368.

26. GhermanRB, Goodwin TM, Leung B, Byrne JD, Hethumumi R,Montoro M. Incidence,clinicalcharacteristics, and
timingofobjectivelydiagnosedvenousthromboembolismduringpregnancy.ObstetGynecol1999;94(5Pt1):7304.

27. Lindqvist P, Dahlback B, Marsal K. Thrombotic risk during pregnancy: a population study. Obstet Gynecol
1999;94(4):5959.

28.TogliaMR,WegJG.Venousthromboembolismduringpregnancy.NEnglJMed1996;335(2):10814.

29. Heit JA, Kobbervig CE, James AH, Petterson TM, Bailey KR, Melton LJ, 3rd. Trends in the incidence of venous
thromboembolism during pregnancy or postpartum: a 30year populationbased study. Ann Intern Med
2005;143(10):697706.

30.SummaryofProductCharacteristicsforCyklokapron.
http://emc.medicines.org.uk/emc/assets/c/html/displaydoc.asp?documentid=1489(accessed11
th
May,2009).

31. Peto R, Baigent C. Trials: the next 50 years. Large scale randomised evidence of moderate benefits. Bmj
1998;317(7167):11701.

32.HorrowJC,VanRiperDF,StrongMD,GrunewaldKE,ParmetJL.Thedoseresponserelationshipoftranexamicacid.
Anesthesiology1995;82(2):38392.

33. DAMOCLES Study Group. A proposed charter for clinical trial data monitoring committees: helping them to do
theirjobwell.Lancet2005;365(9460):71122.

34.HaybittleJL.Repeatedassessmentofresultsinclinicaltrialsofcancertreatment.BrJRadiol1971;44(526):7937.

35. Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV,Mantel N,McPherson K, Peto J, Smith PG. Design
andanalysisofrandomizedclinicaltrialsrequiringprolongedobservationofeachpatient.II.analysisandexamples.Br
JCancer1977;35(1):139.

36. Uniform Requirements for Manuscripts Submitted to Biomedical Journals: Writing and Editing for Biomedical
Publication(UpdatedOctober2008).(Accessedathttp://www.icmje.org/on11
th
May,2009).

37.MoherD,SchulzKF,AltmanDG.TheCONSORTstatement:RevisedRecommendationsForImprovingtheQualityof
ReportsofParallelGroupRandomizedTrials.Lancet2001;357(9263):11911194.

Version1.0

6

Ap
Ap
Ap

6 AP
ppendix1:Ent
ppendix2:Ou
ppendix3:Co
a) Briefi
b) Conse
c) Inform
d) Inform
e) Inform
PPENDIC
tryform
tcomeform
untry/sitespe
informationle
entprocedure
mationsheetf
medconsentf
medconsentf

Protoc
CES
ecificdocume
eafletforpreg
eoverview
forwomanan
formforwom
formforrepre
Page25of36
colISRCTN769
nts
gnantwomen
ndherreprese
man
esentative
912190
&family
entative
VVersiondate:11May20099
Page26of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

APPENDIX1ENTRYFORM(page1)

Page27of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

APPENDIX1ENTRYFORM(page2)

Page28of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

APPENDIX2OUTCOMEFORM(page1)

Page29of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

APPENDIX2OUTCOMEFORM(page2)

Page30of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

APPENDIX3a
Briefinformationleafletforpregnantwomen&family

Page31of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

APPENDIX3b
Consentprocedureoverview

Page32of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

APPENDIX3c
Informationsheetforwomanandherrepresentative,page1
(HOSPITALLETTERHEAD)
INFORMATIONSHEETFORTHEPATIENTANDHERREPRESENTATIVE(S)
THEWOMANTRIAL
TITLE OF RESEARCH: TRANEXAMIC ACID FOR THE TREATMENT OF POSTPARTUM HAEMORRHAGE: AN
INTERNATIONALRANDOMISED,DOUBLEBLIND,PLACEBOCONTROLLEDTRIAL
TRIALSITENUMBER:[IDfromdatabase]
LEAFLETVERSION:VERSION1.0DATED:11MAY2009
This hospital is taking part in an international research study to find ways to improve the
treatmentofwomenwhohaveseverebleedingafterdeliveryoftheirbaby.
(1) Wewouldliketoinviteyoutotakepartinthisstudy
(2) When you were very unwell you were included in this study and we would like you to
continuetotakepart
(3) Asarepresentativeofthepatientweareaskingyoutomakeadecisiononherbehalf
(Pleasecircletheoptionthatapplies)

TheResearchDoctorhasalreadycheckedtomakesureyou/thepatientismedicallysuitablefor
thisresearchandyouarebeingaskedtomakeadecisionaboutwhetheryou/thepatientcanbe
includedinthisstudy.Thissheetgivesinformationaboutthestudy,includingthereasonswhythe
studyisbeingdone,andtherisksandbenefitsoftakingpart.

PLEASEREADTHEINFORMATIONBELOWCAREFULLYANDASKTHEDOCTORORMIDWIFE
LOOKINGAFTERYOUANYQUESTIONSYOUMAYHAVE.

1) Whatisthepurposeofthestudy?
In this hospital, women who have a very severe bleeding after childbirth (also called
postpartum haemorrhage) are given the best available treatments. The aim of this
researchstudyistoseeifthereisabettertreatmentforwomenwhohaveseverebleeding
afterchildbirth.Wehopethatthetreatment(tranexamicacid)willhelpthebloodtoclot
sooner, and so lessen the amount of blood lost and reduce the need for a blood
transfusion and other treatments. But it is also possible that the study treatment may
cause clots where they are not needed, and because the drug is not routinely used after
childbirth,we do not know all thelikely side effects. Wehope to find that the treatment
willdoalittlemoregoodthanharmbutwedontyetknowthis.

2) Whyisthisresearchbeingdone?
Postpartumhaemorrhagecanbeaveryseriousconditionandsometimesrequiressurgery
to control the bleeding. Many thousands of women worldwide die each year from this
condition and it is important to find better ways of controlling excessive bleeding after
childbirth.

Tranexamic acid is often used to reduce bleeding after major operations such as heart
operations. Some women who have heavy menstrual bleeding (periods) also use
tranexamic acid. The WOMAN study is being done to see if TXA can reduce bleeding in
womenwithpostpartumbleeding.
Page33of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

Informationsheetforwomanandherrepresentative,page2

3) Whyhaveyoubeeninvited?
Youhavebeendiagnosedwithpostpartumhaemorrhagebyyourdoctor.Yourdoctorhas
checkedthatyouaresuitableforthestudy,butitisuptoyouwhetherornotyoudecideto
takepart.

4) Who is doing the study and who can you call if you have any questions or
problems?
Dr_____________________________isinchargeofthisstudyatthishospital.Thestudy
is coordinated by doctors and a trial team at The London School of Hygiene & Tropical
Medicine(UniversityofLondon).Ifyouhaveanyquestionsyoucancontactthedoctorat:
Address:
Telephone:
Youarealsofreetovisitthetrialwebsitetokeepuptodatewiththeprogressofthetrial:
www.thewomantrial.Lshtm.ac.uk

5) Apatientcannotbeinthisstudyif:
x Thedoctorthinksthereisaparticularreasonwhytranexamicaciddefinitelyshouldnotbegiven
x Thedoctorthinksthereisaparticularreasonwhytranexamicaciddefinitelyshouldbegiven
x Theyarenotanadult

6) Whatwillhappen/hashappenedduringthisstudy?
You will be given all the usual emergency treatments for severe bleeding after childbirth,
including fluids to replace the blood that you have lost. You will also be given a dose of
either the tranexamic acid or a placebo (a liquid which doesnt contain tranexamic acid).
Thisdosewillbegivenasaninjectionintoyourvein.Ifafterabout30minutesyouarestill
bleeding,orifthebleedingstopsandstartsagainwithin24hoursafterthefirstdose,you
maybegivenaseconddoseofthesame.Youwillnotreceivemorethantwoinjectionsfor
thestudy.

We do not know whether giving tranexamic acid on top of all the other treatments will
helpornot,sohalfthewomeninthestudywillreceivetranexamicacidandtheotherhalf
will receive a placebo. The choice of which treatment you receive is completely random
and you will have an equal chance of receiving either one. Neither you nor the doctor
treating you will know which treatment you receive. This information is kept on a
confidential list at an independent location in London. The study involves no extra tests
but your doctor/midwife will send brief details about your treatment and recovery to the
Coordinating Centre in London. They will also send information about the health of your
baby/ies. If after discharge from hospital and up to 42 days after treatment you develop
any medical problems, please let the doctor named on this form know. This information
will be used in strict confidence by the people working on the study and will not be
releasedunderanycircumstances.

7) Whatarethepossiblerisksofbeinginthestudy?
Tranexamic acid is NOT a new drug and it is widely used to reduce bleeding in conditions
such as major heart surgery. There is no conclusive evidence of serious side effects with
shorttermuse.Butthestudytreatmentmaycauseclotswheretheyarenotneededand,

Page34of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

Informationsheetforwomanandherrepresentative,page3

because the drug is not routinely used after childbirth, we do not know all the likely side
effects. Your doctor will report to the trial organisers any unexpected problems you may
have.

8) Whatarethepossiblebenefitsofbeinginthestudy?
We hope that tranexamic acid may help reduce blood loss. The knowledge that we gain
fromthisstudywillhelpwomenwithpostpartumhaemorrhageworldwideinthefuture.

9) Whatinformationdowekeepprivate?
Allinformationaboutyouandthereasonforbleedingafterchildbirthwillbekeptprivate.
Theonlypeopleallowedtolookattheinformationwillbethedoctorswhoarerunningthe
study, the staff at the Coordinating Centre and the regulatory authorities who check that
thestudyisbeingcarriedoutcorrectly.TheTrialCoordinatingCentremaywanttocollect
or copy some trial documents which will have your name and will include the signed
Consent Form. This will help them to ensure that the trial is being carried out correctly.
Your details will remain confidential and will be held in secure storage at the Trial
Coordinating Centre. Your confidential information will be kept separately from the trial
dataandwillbedestroyedwithinfiveyearsofthetrialending.Wewillpublishtheresults
ofthestudyinamedicaljournalsothatotherdoctorscanbenefitfromtheknowledge,but
your personal information will NOT be included and there will be no way that you can be
identified.

10) Canyouchangeyourmindaboutbeinginthestudy?
You can always withdraw from the study at any time. You just need to say for example
Ive decided I dont want to be in this study now. We hope that you will let us use
information about how you got on, but if you do not want us to use it please tell the
doctor.

11) Whatelsedoyouneedtoknow?
x In the event that something does go wrong and you are harmed during the study, the
London School of Hygiene & Tropical Medicine who are organising the study would be
responsibleforclaimsforanynonnegligentharmsufferedasaresultofparticipatingin
thisstudy.
x Wewillaskyoutosignaseparateconsentformandgiveyouacopytokeepandyoucan
alsokeepthisinformationsheet.
x ThisstudyhasbeenreviewedandapprovedbyaResearchEthicsCommittee.

12) Whathappensafterwards?
If after you leave this hospital you develop any problems at any time up to 42 days after
you had your baby, we would definitely want to know about it. You will be given a card
with the contact details of the research doctor at this hospital, which you should keep
safelyandshowtoanyonewhomaybetreatingyouforanyillness.

If you would like to have a copy of the final results of this study, please let the research
doctorknowands/hewillensureyoureceiveacopywhenitispublished.

Page35of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

APPENDIX3d
InformedConsentFormforwoman

Page36of36
Version1.0 ProtocolISRCTN76912190 Versiondate:11May2009

APPENDIX3e
InformedConsentFormforrepresentative

TRIALS C00RBINATINu CENTRE
Room
Lonuon School of Bygiene Tiopical Neuicine
Keppel Stieet Lonuon WCE BT
Tel
Fax
Email thewomantiialLshtmacuk
wwwwomantiialLshtmacuk

ISRCTN76912190