View
2
Download
0
Category
Preview:
Citation preview
www.aana.com/aanajournalonline AANA Journal August 2020 Vol. 88, No. 4 273
Scant evidence exists regarding use of tranexamic acid (TXA) in high-risk obstetrics. The aim of this review was to evaluate the efficacy of prophylactic TXA in high-risk patients for postpartum hemorrhage. The review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. Only studies examin-ing the effects of TXA compared with placebo in miti-gating postpartum hemorrhage were included. The primary outcomes were blood loss intraoperatively and postoperatively. The secondary outcomes were the frequency of additional uterotonic therapy and postoperative hemoglobin concentration. Three trials consisting of 203 patients were included. Compared with placebo, there was a low quality of evidence that
TXA may reduce blood loss intraoperatively (mean difference, −361.41; 95% CI, −573.13 to −149.69; P = .0008) and postoperatively (mean difference, −177.95; 95% CI, −296,65 to −59.25; P = .003). We also found a moderate quality of evidence that TXA decreased the number of uterotonic agents used (risk ratio, 0.26; 85% CI, 0.16 to 0.41; P < .00001) but did not affect postoperative hemoglobin level (mean difference, 0.41; 95% CI, −0.08 to 0.90; P = .10). Prophylactic TXA may decrease blood loss and reduce the number of rescue uterotonics in high-risk patients undergoing cesarean delivery.
Keywords: Blood loss, high-risk pregnancy, placenta previa, postpartum hemorrhage, tranexamic acid.
Prophylactic Tranexamic Acid in High-Risk Patients Undergoing Cesarean Delivery: A Systematic Review and Meta-Analysis of Randomized Controlled Trials
Nathan E. Stortroen, DNP, CRNATito D. Tubog, DNAP, CRNAScott K. Shaffer, DNAP, CRNA
Surprisingly, the available technology (eg, ultrasonography, 3-dimensional imaging) and advanced medical therapies (eg, bilateral uter-ine artery ligation, uterine tamponade, total hysterectomy) have done little to change the
morbidity and mortality associated with postpartum hem-orrhage (PPH) in high-risk parturients. Postpartum hem-orrhage, following vaginal or cesarean delivery, continues to be one of the leading causes of maternal death through-out the world.1,2 In the most recent Pregnancy Mortality Surveillance System report, 11.5% of pregnancy-related deaths in the United States were due to hemorrhage within 42 hours.3 Additionally, the medical practice of birth via cesarean delivery continues to increase greatly in both high-income and low-income regions around the globe.4,5 In fact, the cesarean delivery rate is continuously rising and in 2017 increased to 32%.6 Traditionally, PPH associ-ated with cesarean delivery is approximated to be twice that of vaginal delivery with an estimated blood loss of 1,000 mL.4,7,8 However, in 2017 the American College of Obstetricians and Gynecologists redefined PPH as cumula-tive blood loss equal or greater than 1,000 mL with cor-responding clinical presentations of hypovolemia 24 hours after the birth process regardless of delivery method.6,9
Risk factors for PPH include uterine atony, trauma in the birth canal, retained placental tissue, and abnor-malities in coagulation.6 Other factors such as prolonged labor, chorioamnionitis, and a high risk of bleeding are linked to PPH.6 Women with a diagnosis of placenta previa make up a large portion of the high-risk partu-rient category, occurring in 4 of every 1,000 births.10 Placenta previa is defined as any portion of the placenta, partially or completely, covering the cervical os.10 This well-known obstetric condition is associated with sub-stantial blood loss leading to the requirement of cesarean delivery.2,11,12
Mechanical and pharmacologic methods are instituted as part of the active management of the third stage of labor. The 3 common strategies used in the active man-agement of the third stage of labor are the use of oxyto-cin, uterine massage, and umbilical cord traction.6 In ad-dition, several interventions have been used to decrease blood loss associated with PPH, including intrauterine balloon tamponade, bilateral uterine artery ligation, and the use of antifibrinolytic agents, namely tranexamic acid (TXA). In a study of 20,060 women, TXA significantly reduced mortality due to bleeding (risk ratio [RR], 0.81; 95% CI, 0.65-1.00; P = .045).13
274 AANA Journal August 2020 Vol. 88, No. 4 www.aana.com/aanajournalonline
The mechanism of action of TXA is to competitively bind with fibrin, thus preventing plasmin and plasmino-gen binding with fibrin.14 Tranexamic acid is not new to the medical arena. Since the 2011 CRASH-2 (Clinical Randomisation of an Antifibrinolytic in Significant Haemorrhage) report,15 TXA has been used in multiple settings. Tranexamic acid therapy is found in open-heart surgeries, total joint replacements, long spine surgeries, and multiple gynecologic procedures. Tranexamic acid has been used predominately as an intravenous therapy, but its topical applications are also being heavily researched.11
The use of TXA in cesarean deliveries and for reduc-tion of PPH has been extensively studied.4,5,12,16,17 Scant evidence is available, however, regarding the use of TXA in high-risk obstetric patients. Because limited research findings are reported on this topic, the authors per-formed a systematic review and meta-analysis of available randomized controlled trials on the use of TXA in high-risk obstetrics.
MethodsWe conducted this systematic review and meta-anal-ysis in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)18 statement and the recommendations of the Cochrane Handbook for Systematic Reviews of Interventions.19
• Data Source and Search. The Patient, Intervention, Comparison, and Outcome (PICO) framework that guided the search was: Does a high-risk parturient under-going cesarean delivery (P) have decreased blood loss (O) when treated with prophylactic tranexamic acid (I) com-pared with placebo (C)?
Published studies examining the effects of TXA in reducing the incidence of PPH were extensively searched from inception until March 2019 using MEDLINE (PubMed), Cumulative Index to Nursing and Allied Health Literature (CINAHL), Google Scholar, the Cochrane Review Database, Embase, Scopus, ScienceDirect, Web of Science, and gray literature. The following search terms were used alone or in combination using the appropriate boolean mechanics: tranexamic acid, high-risk pregnancy, placenta previa, abruption placenta, postpartum hemor-rhage, multiple births, and blood loss. With use of the an-cestry approach, the reference lists of the eligible studies were screened for exclusion and inclusion.
• Study Selection Criteria. The inclusion criteria were identified a priori. The following inclusion criteria were defined: (1) randomized controlled trials (RCTs) evaluating the efficacy of TXA administered during a ce-sarean delivery regardless of route and dose; (2) studies comparing TXA with placebo; (3) trials in which the participants were considered as having a high-risk preg-nancy as defined by the individual study authors; and (4) trials available in full text and written in English. Studies presented in abstracts in professional conferences and published in https://clinicaltrials.gov were considered as long as these trials met the inclusion criteria to avoid publication bias. The authors independently assessed the titles and abstracts for eligibility of the studies. A discus-sion among the authors resolved any discrepancies and disagreements on included studies.
• Data Extraction. A piloted and standardized data extraction template was used for data collection. The fol-lowing information was obtained from each trial: number
Figure 1. PRISMA Flow Diagram of Search Strategy and Study SelectionAbbreviations: PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analysis; TXA, tranexamic acid.
www.aana.com/aanajournalonline AANA Journal August 2020 Vol. 88, No. 4 275
Stu
dy,
ye
ar/
cou
ntr
yIn
terv
enti
on
/typ
e o
f an
esth
esia
Co
mp
aris
on
/ty
pe
of
anes
thes
iaS
tud
y o
utc
om
es
Blo
od
lo
ssU
tero
ton
ic a
gen
ts
Tim
ing
of
asse
ssm
ent
Met
ho
ds
of
calc
ula
tio
nS
tan
dar
d/
rou
tin
eaA
dd
itio
nal
b
Abb
as e
t al
,2 20
19/E
gypt
n =
31
Tran
exam
ic a
cid,
1 g
IV,
befo
re s
kin
inci
sion
and
B
UA
L af
ter
deliv
ery
of in
fant
Ces
area
n de
liver
y un
der
gene
ral
n =
31
BU
AL
afte
r de
liver
y of
infa
nt
Ces
area
n de
liver
y un
der
gene
ral
Tota
l blo
od
loss
Nee
d fo
r ut
erot
onic
s
Nee
d fo
r bl
ood
tran
sfus
ions
Cha
nge
in H
b fr
om p
reop
to
post
op
Intr
aope
rativ
e
Pos
top
Blo
od lo
ss in
trao
pera
tivel
y w
as
asse
ssed
by
addi
ng v
olum
e of
con
tent
s of
suc
tion
bott
le a
nd t
he d
iffer
ence
in
wei
ght
betw
een
the
dry
and
soak
ed
oper
atio
ns s
heet
s an
d to
wel
s
Pos
top
bloo
d lo
ss w
as e
valu
ated
by
addi
ng v
olum
e of
con
tent
s of
pel
vic
drai
n in
spec
ted
ever
y 12
h a
nd o
n re
mov
al o
f th
e dr
ain
and
the
diff
eren
ce
in w
eigh
t be
twee
n dr
y an
d so
aked
va
gina
l pad
s 4
h po
stop
erat
ivel
y
20 IU
oxy
toci
n in
500
mL
of N
S
at 1
25 m
L/h
No
spec
ified
do
sage
was
re
port
ed
Sha
dy &
S
alla
m,11
20
17/E
gypt
n =
40
Tran
exam
ic a
cid,
1 g
IV,
afte
r pl
acen
tal d
eliv
ery
and
addi
tiona
l 2 g
app
lied
topi
cally
on
plac
enta
l bed
n =
40c
10 IU
of
oxyt
ocin
IV
aft
er d
eliv
ery
of
plac
enta
n =
40c
Tran
exam
ic a
cid,
1
g IV
, bef
ore
skin
in
cisi
on a
nd 1
0 IU
of
oxy
toci
n af
ter
deliv
ery
of p
lace
nta
Tota
l blo
od
loss
Add
ition
al
uter
oton
ics
Nee
d fo
r bl
ood
tran
sfus
ions
Cha
nge
in H
b fr
om p
reop
to
24 h
pos
top
Intr
aope
rativ
e:
from
del
iver
y of
pl
acen
ta t
o en
d of
sur
gery
Pos
top:
fro
m
end
of s
urge
ry
to 4
h p
osto
p
Intr
aope
rativ
ely,
blo
od lo
ss w
as
mea
sure
d by
add
ing
volu
me
of
cont
ents
of
suct
ion
cann
iste
r (w
hich
w
as c
hang
ed a
fter
pla
cent
al d
eliv
ery
to
avoi
d a
mix
ture
of
bloo
d an
d am
niot
ic
flui
d) a
nd b
lood
fro
m p
ariti
es a
nd t
he
diff
eren
ce in
wei
ght
betw
een
dry
and
soak
ed s
heet
s an
d to
wel
s
Pos
top
bloo
d lo
ss e
valu
ated
by
wei
ghin
g th
e so
aked
pad
s m
inus
the
dr
y w
eigh
t of
the
pad
s
10 IU
of
oxyt
ocin
IV a
nd
tran
exam
ic a
cid,
2
g, s
oake
d in
to
wel
, whi
ch
was
app
lied
to
the
plac
enta
l be
d fo
r 5
min
Oxy
toci
n,
met
hyle
rgom
etrin
e,
carb
opro
st,
mis
opro
stol
Suj
ata
et
al,24
201
6/In
dia
n =
30
Tran
exam
ic a
cid,
10
mg/
kg,
befo
re s
kin
inci
sion
n =
30
NS
bef
ore
skin
in
cisi
on
Nee
d fo
r ad
ditio
nal
uter
oton
ics
in
firs
t 24
h
Tota
l blo
od
loss
Nee
d fo
r bl
ood
tran
sfus
ion
Hb
24 h
aft
er
surg
ery
Pos
top
Blo
od lo
ss =
{[(H
b P
reop
– H
b 48
h
post
op)/H
b P
reop
] × [(
0.36
69 ×
H3 )
=
(0.0
3219
× W
) + 0
.604
1]} +
{(V
× 1
8)/
Hb
Pre
op}
5 IU
IV o
f ox
ytoc
in d
ilute
d to
5 m
L w
ith N
S
give
n ov
er 3
0 s
and
infu
sion
of
20 IU
oxy
toci
n in
450
mL
NS
ove
r 3
h,
follo
wed
by
10
IU o
xyto
cin
in
500
mL
of N
S
over
nex
t 5
h
Oxy
toci
n,
met
hyle
rgom
etrin
e,
carb
opro
st,
mis
opro
stol
Tab
le 1
. Su
mm
ary
of
Ran
do
miz
ed C
on
tro
lled
Tri
als
Exa
min
ing
Tra
nex
amic
Aci
d F
or
Ces
area
n D
eliv
ery
in P
atie
nts
Wit
h H
igh
Ris
k o
f P
ost
par
tum
Ble
edin
gA
bbre
viat
ions
: BU
AL,
bila
tera
l ute
rine
arte
ry li
gatio
n; H
, hei
ght
of p
atie
nts
in m
eter
s; H
b, h
emog
lobi
n; IU
, int
erna
tiona
l uni
t; IV
, int
rave
nous
; NS,
nor
mal
sal
ine;
pos
top,
pos
tope
rativ
e;
preo
p, p
reop
erat
ive;
V, t
otal
vol
ume
of b
lood
tra
nsfu
sed;
W, w
eigh
t of
pat
ient
; 18
(in b
lood
loss
for
mul
a), h
emog
lobi
n co
ncen
trat
ion
of p
acke
d re
d bl
ood
cells
.a U
tero
toni
c ag
ents
giv
en r
outin
ely
or a
s a
stan
dard
pro
toco
l aft
er t
he d
eliv
ery
of p
lace
nta.
b Add
ition
al u
tero
toni
c ag
ents
adm
inis
tere
d as
nee
ded
afte
r th
e ro
utin
e or
sta
ndar
d ut
erot
onic
s w
ere
adm
inis
tere
d.c T
he s
tudy
had
a 2
-arm
com
paris
on.
276 AANA Journal August 2020 Vol. 88, No. 4 www.aana.com/aanajournalonline
of participants; ASA physical status; total blood loss; the hemoglobin (Hb) and hematocrit levels preoperatively and postoperatively; the dose, route, and timing of administra-tion of TXA; the number of patients requiring additional uterotonic agents; and the number of transfusions. All the data outcomes were collected by a single investiga-tor (N.S.) and were verified by another author (T.D.T.). Discrepancies of data were resolved by a third author (S.S.).
• Risk of Bias Assessment. Two authors appraised the included RCTs and assessed the methodological quality of each study using the Risk of Bias algorithm outlined in the Cochrane Handbook for Systematic Reviews of Intervention.19 The evaluators assessed the quality of the article based on random sequence generation; allocation concealment; blinding of participants, personnel, and outcome assessors; incomplete outcome data; selective reporting; and other sources of bias. Two independent authors (N.S., T.D.T.) rated the article as “high risk,” “low risk,” or “unclear risk” of bias, and each evaluator was instructed to identify the reasons for each rating. Another author (S.S.) resolved any discrepancies or disagreements.
• Quality of Findings. The authors rated the overall quality of evidence using the Grading of Recommenda-tions, Assessment, Development, and Evaluation
(GRADE) approach.20 The GRADE method rates out-comes as high, moderate, low, or very low. Since all evi-dence included in this review involved RCTs, the baseline quality of evidence was graded as high. Subsequently, an outcome was downgraded by 1 level for serious concerns and 2 levels for very serious concerns about the risk of bias assessment, inconsistency, imprecision, indirect-ness, and high probability of publication bias.
• Statistical Analysis. The Review Manager (RevMan 5.3; The Nordic Cochrane Centre) was used for meta-analysis in this study.21 The primary outcomes were total intraoperative blood loss and total postoperative blood loss. Intraoperative blood loss was measured from the de-livery of the baby to the delivery of the placenta, whereas postoperative blood loss was measured from the deliv-ery of the placenta to the time indicated by the studies’ authors. The secondary outcomes included additional uterotonic agents required, change in postoperative Hb concentration, and the number of patients requiring a blood transfusion. For dichotomous outcomes, effect sizes were estimated by calculating the pooled RR with 95% CI. For continuous data, outcomes were calculated using the mean difference (MD) with 95% CI. The ran-dom-effects model was used to pool the estimates of both
Figure 3. Postoperative Blood Loss: Forest Plot of 95% Confidence Interval (CI) for Mean Difference (MD)2,11,24
Abbreviations: IV, inverse variance; Random, random-effects model.
Figure 4. Requirement of Additional Uterotonic Agents: Forest Plot of 95% Confidence Interval for Risk Ratio2,11,24
Abbreviations: Fixed, fixed-effects model; IV, inverse variance.
Figure 2. Intraoperative Blood Lossa: Forest Plot of 95% Confidence Interval for Mean Difference2,11,24
Abbreviations: IV, inverse variance; Random, random-effects model.aCompared with placebo as control.
www.aana.com/aanajournalonline AANA Journal August 2020 Vol. 88, No. 4 277
dichotomous and continuous endpoints anticipating methodological and clinical heterogeneity of data. For the binary endpoint, a significant effect needed a 95% CI not to include 1. For continuous outcomes, a significant effect required a 95% CI not to include 0.
When data were reported as median and range (inter-quartile range), the mean and standard deviation were calculated using the algorithm proposed by Wan and colleagues.22 In RCTs with multiarm groups, data were processed individually. Trials with data not suitable for meta-analysis were described qualitatively in the review.
Heterogeneity was assessed using I2 statistics as de-scribed in the Cochrane Handbook for Systematic Reviews of Intervention.19 An I2 greater than 50% was considered substantial heterogeneity. To explore the clinical and methodological heterogeneity, we designed a priori sub-group and sensitivity analyses. Clinical heterogeneity was explored by examining studies involving patients with placenta previa. A sensitivity analysis was performed by pooling estimates of only studies with a low risk of bias. If results from the sensitivity analysis were unchanged, we concluded that the risk of bias did not influence the effect estimates.
Since there are only 3 RCTs in the review, publication bias was not explored by visual inspection of the funnel plot for symmetrical configuration. For the same reason,
an Egger regression test was not conducted. The quality of findings was generated using GRADEpro GDT software.23
ResultsThe initial search yielded 109 citations and 5 additional studies identified from gray literature. After relevant pub-lications were retrieved and screened, a total of 16 RCTs were reviewed in detail. Eventually, 3 studies evaluating the effects of the use of prophylactic TXA in PPH were analyzed (Figure 1).
• Demographic Characteristics. A total of 203 patients were included in this meta-analysis.2,11,24 Investigators of 2 of the RCTs2,11 evaluated TXA in patients with placenta previa risk factors of PPH.24 In 2 studies, cesarean delivery was performed under general anesthesia.2,11 The dose of the TXA varies between studies. Authors of 2 RCTs ad-ministered 1 g of TXA,2,11 and 1 study used 10 mg/kg.24 One study used TXA topically in 1 of the 2 intervention arms; otherwise, tranexamic was administered intrave-nously.11 All trials administered TXA before skin incision, after placental delivery, or after uterotonic agents were given. The uterotonic agents varied between studies.
There was no standardized method of blood loss calcu-lation for intraoperative and postoperative blood loss. One study measured blood loss using the preoperative and 48-hour postoperative Hb levels.24 In all 3 RCTs, TXA use was reported to be free of complications. Characteristics of the included studies can be found in Table 1.
• Primary Outcomes. Two studies reported intraopera-tive blood loss.2,11 Compared with placebo, there was a significant reduction in intraoperative blood loss in pa-tients treated with TXA (MD, −361.41; 95% CI, −573.12 to −149.69; P = .0008; Figure 2).
Postoperative blood loss was reported in all 3 RCTs.2,11,24 There was a significant decrease in blood loss postopera-tively compared with placebo in patients with TXA (MD, −177.95; 95% CI, −296.65 to −59.25; P = .003; Figure 3). In both intraoperative (I2 = 82%) and postoperative (I2 = 96%) blood loss, there was substantial heterogeneity.
Clinical heterogeneity was explored by pooling studies involving patients with placenta previa. Pooled analysis showed no change in I2 statistics. In addition, sensitiv-ity analysis was performed by initially excluding studies with a high risk of bias for random sequence generation, allocation concealment, and blinding. After the analysis, pooled estimates of intraoperative and postoperative blood loss did not affect substantial heterogeneity.
• Secondary Outcomes. Three studies comprising 202 patients evaluated the effects of TXA and the number of additional uterotonic agents used.2,11,24 Only 14.8% of the patients treated with prophylactic TXA required addi-tional uterotonic agents compared with 60.4% in patients who received placebo and active control (Figure 4).
Only 2 studies evaluated the difference in postop-erative Hb levels between patients treated with TXA
Figure 5. Summary of Risk of Bias for Each Trial2,11,24
Key: Plus sign (green), low risk of bias; blank box, unclear risk of bias.
278 AANA Journal August 2020 Vol. 88, No. 4 www.aana.com/aanajournalonline
and placebo.2,11 Pooled estimates showed no difference between TXA and placebo (MD, 0.41; 95% CI, −0.08 to 0.09; I2 = 0%; P = .10).
• Risk of Bias. According to the Cochrane Risk of Bias tool, all 3 studies2,11,24 have a low risk of bias for random sequence generation, allocation concealment, blinding of participants, incomplete outcome data, and selective reporting. There was an unclear risk of bias for outcome assessors because of the studies’ failure to outline blind-ing of investigators (Figure 5).
• Quality of Findings. GRADEpro software was used to generate the quality of findings table.23 Four major outcomes were analyzed using the GRADE criteria. We found the evidence to be of low quality that TXA reduced the amount of intraoperative and postoperative blood loss compared with placebo. Our confidence in the effect estimates is limited because of the number of studies included in the meta-analysis and the heterogeneity of the studies. Conversely, we found that the quality of our evidence for the additional need of uterotonic agents and the postoperative Hb level was moderate (Table 2).
DiscussionIn this systematic review and meta-analysis, we ex-plored the efficacy of prophylactic TXA in decreasing the amount of blood loss during cesarean delivery in patients at high risk of bleeding. Our analysis showed evidence that prophylactic TXA may reduce blood loss both intraoperatively and postoperatively. Similarly, our pooled estimates suggested that TXA lowered the addi-tional requirements of uterotonic agents. Although there is a statistical significance in blood loss and supplemental uterotonic administration, the overall quality of evidence is low to moderate because of the small sample size and substantial heterogeneity across all studies.
Multiple studies have recently demonstrated the ef-ficacious use of TXA in elective cesarean deliveries and PPH.4,5,11,14 These studies have investigated the differ-ent routes of administration of TXA, including oral, intravenous, and topical. Tranexamic acid is described as a lysine analog that specifically behaves as an antifi-brinolytic.14 The action of TXA is to prevent plasmino-gen converting to plasmin, thus decreasing fibrinolysis in the patient.5 Tranexamic acid reaches peak plasma concentrations nearly immediately after administration and has a half-life of 2 hours.5 The US Food and Drug Administration has given TXA a category B designation for the parturient.11 Many researchers do have concerns related to TXA, primarily the possibility of thromboem-bolic events.2
The evidence levels generated from GRADEpro dem-onstrated that the data from this meta-analysis was low to moderate evidence. The outcomes of intraoperative blood loss and postoperative blood loss were graded as low evidence because of substantial variability between
the studies and small sample size. The outcomes postop-erative Hb level and additional use of uterotonic agents were graded as moderate evidence because of the small sample size. The reason for the small sample size is due to significantly limited data on this subject, indicating the need for future trials with larger sample sizes.
For patients with a high risk of bleeding, our analysis of intraoperative bleeding showed that TXA reduced blood loss by a mean of 361 mL. Our findings were similar to a previous meta-analysis comparing TXA com-prising 14 RCTs of cesarean delivery. Li and colleagues16 reported that blood loss from delivery of the baby to placental expulsion is lower in TXA by a mean of 141.25 mL. The findings of increased mean intraoperative blood loss in our current meta-analysis vs the previous meta-analysis warrant some discussion.
The investigators are particularly concerned with the amount of blood loss associated with placenta previa and the limiting treatment effects that traditional uterotonics have on this population. As the placenta is peeled from the uterus, the fibrinolytic system is activated, increas-ing plasminogen and fibrin degradation products.4,5,17 This cycle continues up to 10 hours post partum and can obviously lead to substantial hemorrhaging. As the rate of cesarean deliveries continues to increase, there is also a linear increase of patients presenting with pla-centa previa.2 Unfortunately, the uterus does a poor job of contracting in its lower segment and when there is any retained placenta product increasing the chance of PPH.4,17,24 When there is retained placenta, the tradition-ally used uterotonic agents become less effective.17,24
The data on postoperative blood loss in this study were similar to the findings of Li and colleagues.16 Our findings indicated that the mean postoperative blood loss is higher than the previous meta-analysis because of the types of patients and the number of studies included in our meta-analysis. The current meta-analysis also showed that patients with TXA required reduced numbers of ad-ditional uterotonic agents. This outcome was expected because the amounts of blood loss intraoperatively and postoperatively were reduced with the use TXA.
The assessment strategies to determine blood loss varied with each study; however, in all 3 RCTs2,11,24 included in this review, the quantification of blood loss used objective techniques such as weighing soaked pads and using preoperative and postoperative Hb concentra-tions. The use of quantitative blood loss is the preferred technique over the traditional visual estimated blood loss method to minimize underestimation or overestimation of blood loss in obstetrics.25
There are a number of limitations in this review. First, the number of participants in most of the RCTs resulted in a small effect size and small sample size. According to Cohen, the effect size can be categorized as any of the following: a Cohen d equal to 0.2 and below is consid-
www.aana.com/aanajournalonline AANA Journal August 2020 Vol. 88, No. 4 279
Ou
tco
mes
An
tici
pat
ed a
bso
lute
ef
fect
sa (
95
% C
I)
Rel
ativ
e ef
fect
(9
5%
CI)
No
. o
f p
arti
cip
ants
(s
tud
ies)
Cer
tain
ty
of
evid
ence
(G
RA
DE
)bC
om
men
ts
Ris
k w
ith
p
lace
bo
(c
om
par
iso
n)
Ris
k w
ith
use
o
f p
rop
hyl
acti
c tr
anex
amic
aci
d
(in
terv
enti
on
)
Intr
aope
rativ
e bl
ood
loss
(IB
L)
asse
ssed
with
am
ount
of
bloo
d lo
ss
from
del
iver
y of
bab
y to
del
iver
y of
pl
acen
ta
Mea
n IB
L w
as
−21
6.05
mL
Mea
n IB
L in
inte
rven
tion
grou
p w
as 3
61.4
1 m
L lo
wer
(ran
ge, 5
73.1
3-14
9.69
low
er)
—14
2(2
RC
Ts)
⊕⊕
Low
c,d
Use
of
prop
hyla
ctic
tra
nexa
mic
ac
id m
ay r
educ
e IB
L sl
ight
ly
Pos
tope
rativ
e bl
ood
loss
(PB
L)
asse
ssed
with
am
ount
of
bloo
d lo
ss
afte
r de
liver
y of
pla
cent
a
Mea
n P
BL
in
inte
rven
tion
grou
p w
as 1
77.9
5 m
L lo
wer
(2
96.6
5-59
.25
low
er)
—20
3(3
RC
Ts)
⊕⊕
Low
c,d
Use
of
prop
hyla
ctic
tra
nexa
mic
ac
id m
ay r
esul
t in
a s
light
red
uctio
n in
PB
L
Pos
tope
rativ
e he
mog
lobi
n (P
-Hb)
as
sess
ed w
ith h
emog
lobi
n le
vels
af
ter
surg
ery
Mea
n P
-Hb
was
0.1
1 m
g/dL
Mea
n P
-Hb
in
inte
rven
tion
grou
p w
as
0.41
mg/
dL h
ighe
r (0
.08-
0.9
high
er)
—12
2(2
RC
Ts)
⊕⊕⊕
Mod
erat
ecU
se o
f pr
ophy
lact
ic t
rane
xam
ic
acid
like
ly r
esul
ts in
a s
light
in
crea
se in
P-H
b
Add
ition
al r
equi
rem
ents
of
uter
oton
ic
agen
ts a
sses
sed
with
num
ber
of
addi
tiona
l ute
roto
nic
agen
ts u
sed
Stu
dy p
opul
atio
n:
RR
= 0
.26
(0.1
6 t0
.41)
20
2(3
RC
Ts)
⊕⊕⊕
Mod
erat
ecU
se o
f pr
ophy
lact
ic t
rane
xam
ic
acid
pro
babl
y re
sults
in a
sl
ight
red
uctio
n in
add
ition
al
requ
irem
ents
of
uter
oton
ic a
gent
s
604
per
1,00
015
7 pe
r 1,
000
(97-
248)
Mod
erat
e:
604
per
1,00
015
7 pe
r 1,
000
(97-
248)
Tab
le 2
. S
um
mar
y o
f Fi
nd
ing
s: P
rop
hyl
acti
c T
ran
exam
ic A
cid
fo
r P
ost
par
tum
H
emo
rrh
age
in H
igh
-Ris
k P
atie
nts
Un
der
go
ing
Ces
area
n D
eliv
ery
Abb
revi
atio
ns: G
RA
DE
, Gra
ding
of
Rec
omm
enda
tions
, Ass
essm
ent,
Dev
elop
men
t, a
nd E
valu
atio
n; M
D, m
ean
diffe
renc
e; R
CTs
, ran
dom
ized
con
trol
led
tria
ls; R
R, r
isk
ratio
.a T
he r
isk
in t
he in
terv
entio
n gr
oup
(and
its
95%
CI)
is b
ased
on
the
assu
med
ris
k in
the
com
paris
on g
roup
and
the
rel
ativ
e ef
fect
of
the
inte
rven
tion
(and
its
95%
CI).
b G
RA
DE
Wor
king
Gro
up g
rade
s of
evi
denc
e: H
igh
cert
aint
y: W
e ar
e ve
ry c
onfi
dent
tha
t th
e tr
ue e
ffect
lies
clo
se t
o th
at o
f th
e es
timat
e of
the
effe
ct; M
oder
ate
cert
aint
y: W
e ar
e m
oder
atel
y co
nfid
ent
in t
he e
ffect
est
imat
e: T
he t
rue
effe
ct is
like
ly t
o be
clo
se t
o th
e es
timat
e of
the
effe
ct, b
ut t
here
is a
pos
sibi
lity
that
it is
sub
stan
tially
diff
eren
t; L
ow c
erta
inty
: Our
co
nfid
ence
in t
he e
ffect
est
imat
e is
lim
ited:
The
tru
e ef
fect
may
be
subs
tant
ially
diff
eren
t fr
om t
he e
stim
ate
of t
he e
ffect
; Ver
y lo
w c
erta
inty
: We
have
ver
y lit
tle c
onfi
denc
e in
the
effe
ct
estim
ate:
The
tru
e ef
fect
is li
kely
to
be s
ubst
antia
lly d
iffer
ent
from
the
est
imat
e of
effe
ct.
c Sub
stan
tial h
eter
ogen
eity
acr
oss
stud
ies,
an
I2 st
atis
tic >
50%
. d S
mal
l sam
ple
size
.
280 AANA Journal August 2020 Vol. 88, No. 4 www.aana.com/aanajournalonline
ered “small effect size”; 0.5, as “medium effect size” and 0.8 and above, as “large effect size.”26 Studies with small sample sizes report larger effect size and may lead to reporting bias.27,28 Second, there is no common method for assessing blood loss. The variability of the calculation and assessment technique may introduce bias in the total amount of blood loss intraoperative and postoperatively. Third, there is substantial heterogeneity across all studies included in this meta-analysis. We were unable to deter-mine the explanation of the clinical and methodological heterogeneity because the number of RCTs in this review is limited. Finally, the timing of postoperative blood loss varies between studies. Although the World Health Organization defines PPH as blood loss in the first 24 hours after delivery, some studies examine the blood loss past the 24-hour mark.29
To our knowledge, this systematic review is the first meta-analysis assessing the role of TXA in high-risk obstetrics, specifically PPH related to placenta previa. Although the number of included trials was only 3, the results are promising. The use of TXA may benefit pa-tients with a history of placenta previa, placental abrup-tion, and other bleeding states found during pregnancy.
Our review findings highlight areas where future clinical trials are needed. Although our pooled estimates indicated statistical significance, we ask readers to be judicious when entertaining thoughts of integrating these results to clinical practice, mainly because of the considerable heterogeneity between studies and the small sample size. We recommend that future large-scale RCTs include patients with a history of pregnancy-induced hypertension, placental abruption, and placenta accreta. In addition, future studies should focus on recommended doses to determine the adequate concentration of TXA needed to inhibit fibrinolysis.
ConclusionOur review findings showed that TXA is effective in reducing the total amount of blood loss both intraopera-tively and postoperatively in patients with a high risk of bleeding during cesarean delivery. Our meta-analysis also showed that patients receiving TXA required a decrease in the total number of additional uterotonic agents re-quired. However, we caution the extrapolation of these findings until high-quality trials with a larger sample size are conducted and pooled for meta-analysis.
REFERENCES 1. Say L, Chou D, Gemmill A, et al. Global causes of maternal death: a
WHO systematic analysis. Lancet Glob Health. 2014;2(6):e323-e333. doi:10.1016/S2214-109X(14)70227-X
2. Abbas AM, Shady NW, Sallam HF. Bilateral uterine artery ligation plus intravenous tranexamic acid during cesarean delivery for placenta pre-via: a randomized double-blind controlled trial. J Gynecol Obstet Hum Reprod. 2019;48(2):115-119. doi:10.1016/j.jogoh.2018.10.023
3. Pregnancy Mortality Surveillance System. Centers for Disease Con-trol and Prevention. Updated February 4, 2020. Originally accessed
May 27, 2019. https://www.cdc.gov/reproductivehealth/maternalin-fanthealth/pregnancy-mortality-surveillance-system.htm
4. Maged AM, Helal OM, Elsherbini MM, et al. A randomized placebo-controlled trial of preoperative tranexamic acid among women under-going elective cesarean delivery. Int J Gynaecol Obstet. 2015;131:265-268. doi:10.1016/j.ijgo.2015.05.027
5. Wang H-Y, Hong S-K, Duan Y, Yin H-M. Tranexamic acid and blood loss during and after cesarean section: a meta-analysis. J Perinatol. 2015;35(10):818-825. doi:10.1038/jp.2015.93
6. Martin JA, Hamilton BE, Osterman MJ, Driscoll AK, Drake P. Births: final data for 2017. Ntl Vital Stat Rep. 2018;67(8):1-50.
7. Chi BH. Translating clinical management into an effective public health response for postpartum haemorrhage. BJOG. 2015;122(2):211. doi:10.1111/1471-0528.13221
8. World Health Organization (WHO). WHO Guidelines for the Manage-ment of Postpartum Haemorrhage and Retained Placenta. World Health Organization; 2009. Accessed February 26, 2019. https://apps.who.int/iris/bitstream/handle/10665/44171/9789241598514_eng.pdf;jsessionid=779DC610418A4DFA5482B5001913AA33?sequence=1
9. Shields LE, Goffman D, Caughey AB. Practice bulletin no. 183: Post-partum hemorrhage. Obstet Gynecol. 2017;130(4):e168-e186. doi: 10.1097/AOG.0000000000002351
10. Lockwood CJ, Russo-Stieglitz K. Placenta previa: epidemiology, clinical features, diagnosis, morbidity and mortality. UpToDate. Updated July 1, 2019. Originally accessed May 24, 2019. https://www.uptodate.com/contents/placenta-previa-epidemiology-clinical-features-diagnosis-morbidity-and-mortality
11. Shady NW, Sallam HF. Adjunctive IV tranexamic acid versus topi-cal tranexamic acid application of the placental bed for prevention of postpartum hemorrhage in women with placenta previa: a ran-domized controlled trial. Int J Reprod Contracept Obstet Gynecol. 2017;6(12):5205-5212. doi:10.18203/2320-1770.ijrcog20175228
12. Fan D, Xia Q, Liu L, et al. The incidence of postpartum hemorrhage in pregnant women with placenta previa: a systematic review and meta-analysis. PLoS ONE. 2017;12(1):1-15. doi:10.1371/journal.pone.0170194
13. WOMAN Trial Collaborators. Effect of early tranexamic acid admin-istration on mortality, hysterectomy, and other morbidities in women with post-partum haemorrhage (WOMAN): an interna-tional, randomised, double-blind, placebo-controlled trial [pub-lished correction appears in Lancet. 2017;389(10084):2104]. Lancet. 2017;389(10084):2105-2116. doi:10.1016/S0140-6736(17)30638-4
14. Simonazzi G, Bisulli M, Saccone G, Moro E, Marshal A, Berghella V. Tranexamic acid for preventing postpartum blood loss after cesarean delivery: a systematic review and meta-analysis of random-ized controlled trials. Acta Obstet Gynecol Scand. 2016;95(1):28-37. doi:10.1111/aogs.12798
15. CRASH Trial Collaborators; Roberts I, Shakur H, Afolabi A, et al. The importance of early treatment with tranexamic acid in bleeding trauma patients: an exploratory analysis of the CRASH-2 randomised controlled trial. Lancet. 2011;377(9771):1096-1101. doi:10.1016/S0140-6736(11)60278-X
16. Li C, Gong Y, Dong L, Xie B, Dai Z. Is prophylactic tranexamic acid administration effective and safe for postpartum hemorrhage prevention? A systematic review and meta-analysis. Medicine. 2017;96(1):e5653. doi:10.1097/MD.0000000000005653
17. Novikova N, Hofmeyr GJ. Tranexamic acid for preventing postpar-tum haemorrhage. Cochrane Database System Rev. 2010;7:CD007872. doi:10.1002/14651858.CD007872.pub2
18. Moher D, Liberati A, Tetzlaff J, Altman D. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med. 2009;6(7):e1000097.
19. Higgins JP, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0. Cochrane Collaboration; 2011. https://training.cochrane.org/handbook
20. Ryan R, Hill S. How to GRADE. Cochrane Consumers and Com-munication Group. Version 3.0. December 2016. Updated July 24, 2018. URL updated April 22, 2020. https://figshare.com/articles/How_to_GRADE/6818894
www.aana.com/aanajournalonline AANA Journal August 2020 Vol. 88, No. 4 281
21. Review Manager (RevMan). Version 5.3. The Nordic Cochrane Centre, Cochrane Collaboration; 2014. Accessed April 22, 2020. https://training.cochrane.org/online-learning/core-software-cochrane-reviews/revman/revman-5-download
22. Wan X, Wang W, Liu J, Tong T. Estimating the sample mean and stan-dard deviation from the sample size, median, range and/or interquar-tile range. BMC Med Res Methodol. 2014;14:135. doi:10.1186/1471-2288-14-135
23. GRADEpro GDT: GRADEpro Guideline Development Tool. McMaster University; 2015 (developed by Evidence Prime Inc). Accessed April 22, 2020. htttps://gradepro.org
24. Sujata N, Tobin R, Kaur R, Aneja A, Khanna M, Hanjoora VM. Randomized controlled trial of tranexamic acid among parturients at increased risk for postpartum hemorrhage undergoing cesarean delivery. Int J Gynaecol Obstet. 2016;133(3):312-315. doi:10.1016/j.ijgo.2015.09.032
25. Quantification of blood loss: a WHONN Practice Brief Number 1. JOGNN. 2014;44(1):158-160. doi:10.1111/1552-6909.12519
26. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Lawrence Erlbaum Associates; 1988.
27. Chalmers TC, Celano P, Sacks HS, Smith H Jr. Bias in treatment assign-ment in controlled clinical trials. N Engl J Med. 1983;309(22):1358-1361. doi:10.1056/NEJM198312013092204
28. Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias: dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA. 1995;273(5):408-412. doi:10.1001/jama.1995.03520290060030
29. World Health Organization (WHO). WHO recommendations for the prevention and treatment of postpartum haemorrhage. World Health Organization; 2012. Accessed February 20, 2019. http://apps.who.int/iris/bitstream/10665/75411/1/9789241548502_eng.pdf
AUTHORSNathan E. Stortroen, DNP, CRNA, is employed by Texas Wesleyan Uni-versity, Fort Worth, Texas, and is director of clinical education for the university’s Graduate Program in Nurse Anesthesia.
Tito D. Tubog, DNAP, CRNA, is employed by Texas Wesleyan Uni-versity, where he is the associate program director of nurse anesthesia.
Scott K. Shaffer, DNAP, CRNA, is an assistant professor at Texas Wesleyan University.
DISCLOSURESThe authors declared no financial relationships with any commercial entity related to the content of this article. The authors did not discuss off-label use within the article.
Recommended