REVIEW
Prevention of preterm labour
Kwabena Appiah-Sakyi
Justin C Konje
Abstract
Preterm labour complicates 3e4% of all pregnancies and its causes are
multifactorial. The incidence is rising and it is more common in the
economically deprived populations and communities in the rich countries.
The recurrence rate is doubled after one previous preterm delivery. Sec-
ondary preventive measures have not been shown to be as effective as
primary measures. A concerted multidisciplinary effort to eliminate risk
factors for inducing preterm labour such as early marriage, smoking,
short inter-pregnancy intervals, domestic violence and under nutrition
will significantly reduce the incidence while optimisation of medical disor-
ders pre-pregnancy and during pregnancy will result in a reduced
incidence.
Keywords cervical length; enhanced antenatal care; oncofetal fibro-
nectin; premature rupture of fetal membranes; preterm birth; preterm
labour; progesterone; tocolysis; ultrasound
Introduction
Preterm labour (PTL) is defined by the World Health Organisa-
tion (WHO) as the onset of labour after the gestational age of
viability and before 37 completed weeks or 257 days of preg-
nancy. It is clinically confirmed by demonstrable uterine con-
tractions associated with documented cervical changes.
Threatened preterm labour is diagnosed when there are docu-
mented uterine contractions without cervical changes.
Every year about 15 million babies are born prematurely and
preterm birth (PTB) remains the biggest cause of neonatal death.
It is also one of the commonest causes of under- 5 deaths. About
50% of preterm births follow spontaneous onset of labour, 30%
after premature rupture of fetal membranes and the remaining
20%, iatrogenic due to maternal and fetal medical indications.
There is evidence that the preterm birth rate is increasing in all
countries where there are reliable data. In the UK, the rate is
about 7.9%.
Key reasons for the rise in the number of preterm deliveries
include a rise in multiple pregnancies from reproductive tech-
niques, widespread obesity with its associated comorbidities of
hypertension and diabetes and an increased incidence of sexually
transmitted infections. Whilst an improved understanding of
some of the underlying mechanisms and advances in technolo-
gies have culminated in the introduction of new tools for both the
Kwabena Appiah-Sakyi MRCOG is Senior Attending Physician,
Department of Obstetrics and Gynecology, Sidra Medical Research
Center, Doha, Qatar. Conflicts of interest: none declared.
Justin C Konje MD FRCOG Division Chief Research, Center of Excellence in
Reproductive Sciences, Department of Obstetrics and Gynecology, Sidra
Medical Research Center, Doha, Qatar. Conflicts of interest: none
declared.
OBSTETRICS, GYNAECOLOGY AND REPRODUCTIVE MEDICINE --:-
Please cite this article in press as: Appiah-Sakyi K, Konje JC, Prevention of preterm labour, Obstetrics, Gynaecology and Reproductive Medicine
(2015), http://dx.doi.org/10.1016/j.ogrm.2015.06.005
diagnosis of preterm labour and the management of extremely
preterm babies, many controversies remain about the optimal
methods for the prevention and care of women presenting in
preterm labour.
Globally the need to address the impact of preterm births has
become crucial as recent statistics indicate that the millennium
development goals (MDGs) number 4 of reducing the under- 5
mortality by two-thirds will not be achieved due primarily to
failure to reduce neonatal deaths from prematurity.
The burden of preterm labour
About 1 million babies die from complications of preterm birth
and in most developed countries, about 70% of neonatal deaths
are attributable to prematurity. The babies that survive are at an
increased risk of disabilities involving neurological, respiratory
and mobility functions. These incapacities exact a heavy financial
toll on the affected countries, families, healthcare and educational
systems. The cost of prematurity remains high for many devel-
oped countries with high level neonatal and ongoing care of
premature babies. In the United States for example, the annual
cost associated with preterm infants is over 26 billion dollars. In
the UK, an Oxford research group reported in 2009, that the public
cost of premature babies, in terms of their health needs, education
and time off work taken by caring parents was about 939 million
pounds per year. This figure fails to capture the psychological and
emotional burden as well as the challenging family dynamics
created in the event of an unexpected preterm birth of a baby.
Fifty percent of long-term neurological morbidity in the high
income nations is linked directly to preterm delivery. Prolonging
pregnancies even for several weeks significantly reduces
newborn risks, and gestational age is the essential determinant of
most perinatal outcomes. As an example, for a fetus delivered in
the peri-viable gestational age, as few as 5 extra days could
double chances of survival and also greatly increase neurologi-
cally intact survival.
Causes of preterm labour
The pathophysiological mechanisms that underlie PTL are poorly
understood, but some clinical and laboratory evidence suggest
that a host of multiple factors trigger pathogenic processes
leading to a final common pathway for the initiation of uterine
contractions that result in spontaneous preterm labour
(Figure 1). In about 50 % of cases, a definitive risk factor cannot
be identified. A history of previous PTL is probably the most
significant risk factor, followed by multiple pregnancies. For
example, a previous history of PTB increases the recurrence risk
to 15%, 2 previous PTBs to 30 % and then 3 previous PTBs to
45%. The risk of PTL in a multiple pregnancy is 10 times that of a
singleton pregnancy. Associations between PTB and young or
advanced maternal age, short inter-pregnancy intervals, low
BMI, black ethnicity, smoking and excessive alcohol intake have
also been established.
Apart from PPROM and idiopathic PTL, the two most
common maternal conditions associated with PTB are
pregnancy-induced hypertension and antepartum haemorrhage.
This may either be spontaneous or iatrogenic in the interest of
either the mother or to rescue the fetus from an adverse
environment.
1 2015 Elsevier Ltd. All rights reserved.
 REVIEW
Final common pathway of uterine contractions
Uterine
distension
Decidual
bleeding
Infection
Figure 1
Infections such as bacteria vaginosis (BV), human immuno-
deficiency virus (HIV), syphilis, periodontal disease, subclinical
chorioamnionitis and UTI have been associated with PTB. Non-
infectious causes such as vaginal bleeding, abnormal uterine
distention and pathological weakness and dilatation of the cervix
may all trigger PTL by stimulating the release of pro-inflammatory
markers.
Primary prevention
Primary prevention of PTB involves the provision of in-
terventions before and between pregnancies which enhance the
mothers health and reduce risks of her or the baby succumbing
to preventable adverse pregnancy conditions. In the past, this
aspect of womens health received less attention but awareness is
now growing. It comprises of interventions aimed at identifying
and improving the biochemical, behavioural and social risks of
womens health or pregnancy outcomes through prevention and
management. These interventions can be grouped under pre-
conception care, enhanced antenatal care, reducing multiple
births and infections, optimizing the management of medical
disorders and progesterone prophylaxis.
(a) Preconception care
A recent WHO commissioned report on PTB titled Born too
soon e the Global Report outlines comprehensive measures to
prevent PTB. These measures start from preconception right
through the pregnancy. Preconception care initiatives, include
education on smoking cessation, better family planning and
inter-pregnancy spacing, economic empowerment programs
which alleviate poverty, community e based interventions like
teenage HPV vaccination, micronutrient food supplementation
and partner education to reduce domestic violence.
There is growing evidence that these social policy in-
terventions do reduce the risk of PTB, the most effective being
primary and secondary education of girls, increasing the legal age
of marriage, pre-pregnancy weight optimization, screening and
treating mental health disorders and other medical conditions
OBSTETRICS, GYNAECOLOGY AND REPRODUCTIVE MEDICINE --:-
Please cite this article in press as: Appiah-Sakyi K, Konje JC, Prevention of preterm labour, Obstetrics, Gynaecology and Reproductive Medicine
(2015), http://dx.doi.org/10.1016/j.ogrm.2015.06.005
such as chronic hypertension and diabetes mellitus. Policies
relevant in developed countries include smoking cessation pro-
grams, prevention of domestic violence and increased engage-
ment with deprived and marginalized communities. There is
moderate to strong evidence of the effectiveness of a number of
these interventions.
(b) Enhanced antenatal care
This care is designed to reduce or eliminate complications in
women with documented potential risks to their pregnancies.
The women receive the basic recommended antenatal care
package by the National Institute for Clinical Effectiveness (UK)
(NICE) as well as interventions targeted at improving healthy
behaviours, promoting early identification of danger signs and
increasing the womens knowledge about pregnancy complica-
tions such as antepartum haemorrhage and early warning signs
of PTL. Regular antenatal visit is emphasised and those requiring
multidisciplinary care on account of medical co-morbidities
receive extra attention.
(c) Reducing multiple births
National policies to regulate assisted reproductive techniques
(ART) and reduce multiple pregnancies are essential. The Human
Fertilisation and Embryology Act (HFEA) in the UK in 2008 set
out a national policy to reduce multiple pregnancy rates by
encouraging all fertility centres to adopt the single embryo
transfer policy. This was followed by new guidelines from the
British Fertility Society and the Association of Clinical Embryol-
ogists on single embryo transfer, which have also been widely
accepted. Women who carry multiple pregnancies whether
conceived spontaneously or by ART require close clinical moni-
toring. Others with the diagnosis of cervical weakness in previ-
ous pregnancies need prior identification and plans made to
institute early treatment such as elective cervical cerclage or
cervical pessary.
(d) Reducing infections
Although the association between PTB and infections is still
poorly understood, it is generally acknowledged that maternal
infection plays a significant role in the pathogenesis. Goldberg
et al. reported that 80% of women presenting with PTL before 30
weeks had evidence of amniotic fluid infection compared to 30%
who deliver after 37 weeks. There is also evidence of activation
of inflammatory mediators characterized by elevated concentra-
tions of cytokines (IL-6, IL-1, IL-8, and TNF) but there is limited
clarity on how these inflammatory agents are linked with the
onset of labour.
A number of clinical trials using antenatal screening to iden-
tify and treat asymptomatic Bacteria Vaginosis and bacteriuria as
well as periodontal disease have shown conflicting findings of
benefit. While one Cochrane review of moderate quality studies
showed that antenatal screening and treatment may be of value,
another reported that current knowledge does not provide
adequate evidence as to which antimicrobial agents are most
suitable for intra-uterine infections. The ORACLE trial demon-
strated, not only the lack of benefit of prophylactic antibiotics in
women presenting with PTL, but also the potential harm of
treatment in the neonate. The most recent Cochrane review on
this subject confirmed that the only subgroup of women who
2 2015 Elsevier Ltd. All rights reserved.
 REVIEW
benefit from antenatal prophylactic antibiotics are those with
previous histories of PTL and a positive screen for Bacteria
Vaginosis.
(e) Optimising management of medical disorders associated
with PTB
Iatrogenic preterm deliveries are commonly due to complications
such as diabetes mellitus, hypertension, connective tissue dis-
orders, autoimmune, endocrine and reno-vascular diseases.
Optimization of antenatal care with early administration of low
dose aspirin, the use of combined obstetric medicine clinics and
timely drug modifications all reduce the need for early delivery
on account of poorly managed disease. For women considered
high risk for preterm labour, there are suggestions that they are
best managed in dedicated Preterm Clinics where additional ed-
ucation, screening, monitoring and treatment of complications
can be provided to reduce the risks of both spontaneous and
iatrogenic PTBs. The evidence that such clinics make a difference
is, however, still not robust.
(f) Progesterone
Progesterone is an essential steroid produced by the corpus
luteum for the maintenance of early pregnancy until 7e9 weeks
of gestation when the placenta takes over this function. The
administration of the anti -progesterone, mifepristone induces
abortion in early pregnancy. Though the relevance of proges-
terone in late pregnancy is poorly understood, it appears to help
maintain uterine quiescence by inhibiting myometrial contrac-
tion through the modulation of cytokine production and inhib-
iting the expression of contraction associated protein genes
within the myometrium.
The preventative effect of progesterone on PTB has been
extensively studied but some of the results have been discor-
dant. While a meta-analyses in 2013 concluded that progester-
one was protective against the recurrence of preterm births and
improved neonatal outcomes (i.e. reducing neonatal deaths,
necrotizing enterocolitis and respiratory distress syndrome
rates), the largest RCT did not demonstrate any benefit. The
most significant benefit of progesterone for PTBs is in women
with sonographically diagnosed short cervices. In a meta-analyses
of 775 patients with cervix <25 mm, treated with vaginal
progesterone of variable doses (90 mg, 100 mg, 200 mg), there
was a significant reduction in PTBs before 28, 33, and 35 weeks
relative to placebo (relative risk 0.50, 0.58, and 0.69,
respectively). In multiple pregnancies and those complicated by
preterm premature rupture of membranes, there is no evidence
that progesterone is effective in preventing preterm delivery.
Despite the evidence of potential benefits in selected cases,
there is no consensus on the appropriate dose, route of admin-
istration, gestation to initiate treatment and long-term effects on
infants.
Secondary prevention
Diagnostic modalities
The first step in patients presenting with possible PTL is an
appropriate diagnosis. Unfortunately the diagnosis that is often
made on the basis of clinical findings is unreliable. In two sys-
tematic reviews, only 13.3% of those who fulfilled the criteria of
OBSTETRICS, GYNAECOLOGY AND REPRODUCTIVE MEDICINE --:-
Please cite this article in press as: Appiah-Sakyi K, Konje JC, Prevention of preterm labour, Obstetrics, Gynaecology and Reproductive Medicine
(2015), http://dx.doi.org/10.1016/j.ogrm.2015.06.005
regular uterine contractions with cervical change actually went
on to deliver within a week. There is, therefore, a need for a more
accurate test to help reduce maternal anxiety and the significant
cost incurred with unnecessary interventions in those presenting
with pseudo PTL.
Between 30 and 35% of preterm deliveries follow preterm
premature rupture of fetal membranes (PPROM). PPROM is,
thus, a significant cause of preterm labour. A distinction
between the clinical presentation of both PPROM and PTL can
be difficult as these could vary from mild symptoms of physi-
ological discharge to those of active labour. The poor sensitivity
and specificity of clinical assessments therefore mean that many
patients receive needless treatments (tocolytics, steroids, even
in-utero transfer (IUT)). A number of studies have suggested
that the use of oncofetal fibronectin test (ofFN), the Actim
Partus test, transvaginal ultrasound cervical length assessment
(CL), Amniosure and Nitrazine test may help improve
diagnosis.
(a) Oncofetal fibronectin versus Actim Partus
Oncofetal fibronectin is a glycosylated glycoprotein found in
plasma and extracellular matrix. This molecule is thought to be
the glue that promotes cellular adhesion at the uterine-
placental and decidualechorionic interfaces and is released into
cervicovaginal secretions when the extracellular matrix is dis-
rupted. This is the rationale for its measurement in the diagnosis
of PTL. It has also been identified in amniotic fluid, extracts of
placental tissue and in cervicovaginal secretions prior to 20e22
weeks. Testing in the first half of pregnancy or after spontaneous
rupture of fetal membranes is therefore unhelpful in the predic-
tion of impending PTL.
There is now considerable evidence that a positive oncofetal
fibronectin test (ofFN) is associated with an increased risk of PTB
in symptomatic patients while a negative ofFN is indicative of
very low risk compared to clinical assessment. When combined
with a long cervical length (CL), a negative ofFN is so reassuring
that the European Association of Perinatal Medicine recommends
no initiation of preventative measures (tocolysis, steroids and
IUT). Sexual activity, digital examinations, vaginal bleeding, and
the presence of ruptured membranes are associated with false-
positive results whilst lubrication with gel during examination
is associated with a false-negative result.
The phosphorylated insulin elike growth factor binding
protein 1 pIGFBP-1 (used in the Actim Partus Test) has been
shown to be equally effective in identifying seemingly low risk
symptomatic women for PTL. Some studies suggest that while it
does have a slightly lower negative predictive value (NPV)
relative to oncofetal fibronectin and thus may result in a few
false negative cases, it is not affected by semen and can
therefore be used in patients presenting with a recent history of
sexual intercourse. Further evidence is awaited on its value in
combination with cervical length assessments, as well as clin-
ical trials on the comparative effectiveness of Actim Partus and
ofFN tests. Recent evidence also suggests that oncofetal fibro-
nectin has limited accuracy in women with multiple pregnan-
cies especially after 32 weeks; about 1.6% of women who tested
negative delivered within a week and could not be transferred
to a tertiary service.
3 2015 Elsevier Ltd. All rights reserved.
 REVIEW
(b) Cervical length assessments
In 2013, a Cochrane review failed to recommend cervical length
screening for all low risk women. This followed the observation
of an inverse relationship between short cervices and rates of
PTB. For a given length of short cervix, in a woman with a
singleton pregnancy and no prior history of PTB, the sensitivity
of the short cervix in the prediction of PTB was 35e40%, with a
PPV of 20e30%. Women with a prior history of PTB, however,
had an increased sensitivity to 70 % and this was even higher
where they had repeated early PTBs.
As a general rule, preterm delivery is highly unlikely where
the cervical length is greater than 3 cm and highly likely when it
is <1.5e2.0 cm. The RCOG guidance recommends that women
with a prior history of PTB would require cervical length
assessment from 14 weeks to 24 weeks, and that cervical cerclage
should be considered where the cervical length <2.5 cm before
24 weeks gestation. In some centres in the US, cervical length
assessment is subcategorized into cervical length >3.0 cm, 2.0
e3.0 cm, and <2.0 with each subgroup having different man-
agement recommendations. The subgroup with a cervical length
<2.0 and at an inevitable risk of PTB are not routinely offered
oncofetal fibronectin testing but are admitted early in anticipa-
tion of a preterm birth or for cerclage.
In symptomatic women, where the diagnosis is uncertain, the
combination of CL assessment and ofFN will help determine the
likelihood of delivery within the next 7 days. It is essential to
note that ofFN has a higher false positive rate as well as being
expensive when used alone in predicting PTB. Some suggest that
if fibronectin is negative and cervical length >2.5 cm, it is
acceptable to take no further action. A recent meta-analysis has
also shown that vaginal progesterone reduces the risk of PTB in
women with asymptomatic mid-trimester short cervices.
(c) Diagnostic test for PROM
Premature rupture of fetal membranes (PROM) is one of the
commonest causes of PTL and is a clinical diagnosis which can
sometimes be challenging. Simple point of care diagnostic tests
for PROM include (a) confirmatory speculum examination
demonstrating pooling of fluid in the vagina and coming from the
cervix (b) ferning of the dried secretions observed under a
microscopic, known as arborization (which unfortunately is fast
becoming historic as most labour wards to not have side room
microscopes) and (c) alkalinity of the fluid as determined by the
Nitrazine paper test. This is based on the fact that the normal
acidic vaginal milieu (pH of 3.8e4.2) is altered by amniotic fluid
to a more basic or neutral pH; however, vaginal discharge, cer-
vical secretions, semen and blood may produce the same
changes and result in a false positive test. The Nitrazine paper
test has a sensitivity of 90% and false positive rate of 17.0%.
Both Nitrazine and Ferning testing are unreliable at earlier
gestations.
Amniosense is a panty liner which changes colour at a pH >
5.2. Its sensitivity and specificity are 98% and 65% respectively.
Routine ultrasound examination for liquor volume is only useful
where there has been loss of a large volume of liquor, which, in
most cases will also be clinically self-evident. The diagnosis of
mild oligohydramnios is subjective and confirmatory intra-
amniotic dye injections can introduce additional pregnancy
complications. Amniosure (placental A-immunoglobulin
OBSTETRICS, GYNAECOLOGY AND REPRODUCTIVE MEDICINE --:-
Please cite this article in press as: Appiah-Sakyi K, Konje JC, Prevention of preterm labour, Obstetrics, Gynaecology and Reproductive Medicine
(2015), http://dx.doi.org/10.1016/j.ogrm.2015.06.005
-1protein assay) and Actim PROM (Insulin elike growth factor
binding protein-1, IGFBP-1) are both commercially available tests
for PROM. Early indications suggest that Amniosure may be
more accurate than Actim PROM, but further studies are still
awaited.
Oncofetal fibronectin (ofFN) has also been demonstrated to be
a sensitive test for PROM, with a negative test strongly indicative
of absence of PROM. A positive test, however, does not help
management planning, as fetal membranes may be intact.
Treatment modalities for PTL
(a) Tocolysis
Tocolytic agents may reduce the number of deliveries occurring
within the first 7 days, but they neither reduce the PTB rate, nor
improve perinatal outcome. The short-term prolongation of the
pregnancy is often beneficial for antenatal corticosteroid
administration and in-utero transfer to tertiary facilities where
required. In the UK, the tocolytics in common use include; cal-
cium channel blockers (Nifedipine); an oxytocin receptor
antagonist (Atosiban) and the cyclo-oxygenase (COX) inhibitor
(indomethacin). Even though Atosiban and Ritrodrine (beta-2
adrenergic receptor agonist) are the only agents licensed for use,
Ritrodrine is no longer recommended on account of its severe
maternal side-effects.
Nifedipine: despite being unlicensed, Nifepidine is commonly
used in the UK, because of cost, ease of administration and ef-
ficacy. Calcium channel blockers directly block the influx of
calcium ions into the cell membranes, thus decreasing the
intracellular concentration of calcium, thereby inhibiting muscle
contractions and leading to uterine quiescence. A Cochrane re-
view demonstrated benefits of Nifedipine over placebo and other
tocolytic agents including betamimetics and magnesium sul-
phate in interrupting PTL. It has been suggested that Nifepidipne
may be more effective that oxytocin receptor antagonists, how-
ever, atosiban has a lesser side-effects profile. The essential
focus of attention in future research on calcium channel blockers
is to determine the optimal effective dose and route of
administration.
Atosiban: atosiban is a selective oxytocin receptor antagonist
which works by competitively inhibiting oxytocin stimulation of
uterine muscle receptors which mediate the release of free cal-
cium into the cytoplasm to initiate contraction. Atosiban has not
been approved by the US Food and Drug Administration (FDA)
on grounds of safety but it is the only approved tocolytic that is
widely used in the UK.
A recent systematic review comparing Atosiban to no treat-
ment found that it was not superior to placebo and may be
associated with a small risk of infant death up to the age of
12months. In comparison to betamimetics and calcium channels
blockers, it had lesser side eeffects, however, superior efficacy
has not been demonstrated. With regards to administration, it is
given initially as an intravenous bolus preparation followed by a
continuous infusion over 48 hours. Despite the lack of conve-
nience in its administration, its use in Europe has been wide-
spread, limited only by its cost. This may be due to its low side-
effect profile. In theory its efficiency should increase with
4 2015 Elsevier Ltd. All rights reserved.
 REVIEW
increasing gestation as oxytocin receptor concentration and
uterine responsiveness increases.
Indomethacin: the most commonly used COX inhibitor for
tocolysis is indomethacin. Compared with placebo, Indometh-
acin shows a trend towards inhibition of PTL. No differences
were noted with neonatal outcomes of respiratory distress syn-
dromes (RDS), premature closure of ductus arteriosis, persistent
pulmonary hypertension of the newborn when compared with
placebo. Compared with ritrodrine, it has similar efficacy over 48
hours, with fewer maternal side effects. There have been re-
ported adverse effects in newborns exposed to indomethacin in-
utero. These include premature closure of the ductus arteriosis,
renal and cerebral vasoconstriction and necrotizing enterocolitis.
These complications tend to be commoner after 32 weeks
gestation.
In the US, indomethacin is used as the first line tocolytic in
most centres if the woman is <32 weeks because of the relatively
few maternal side effects. From 32 to 34 weeks, Nifedipine be-
comes the tocolytic of choice. An oral or rectal dose of 50e100
mg of indomethacin is the recommended loading dose followed
by 25 mg orally 4e6 hourly. Although not currently recom-
mended for >48 hours administration, such prolonged use
should be accompanied by at least weekly sonographic assess-
ment for oligohydramnios and narrowing of the fetal ductus
arteriosis. There is currently an ongoing US study comparing
Nifedipine and Indomethacin that should provide additional in-
formation on the efficacy and side-effects of these agents.
(b) Antenatal corticosteroids
The benefits of maternal corticosteroids to enhance fetal lung
maturation are now well established with systematic reviews
showing a reduction in neonatal deaths, respiratory distress
syndrome, necrotizing enterocolitis, neonatal ICU admissions
and intraventricular haemorrhage. These benefits have been
demonstrated without the potential dangers of increased cho-
rioamnionitis, puerperal sepsis or neonatal infections.
The RCOG recommends a single course of corticosteroids
consisting of either 2 doses of 12 mg of betamethasone given 12
hours apart or 4 doses of 6 mg dexamethasone given 6 hourly.
Corticosteroids are most effective after 24 hours and up to 7 days
after administration. In the event of an acute presentation of PTL,
a rescue therapy of corticosteroids has been demonstrated to
confer some benefit in terms of improved neonatal respiratory
function.
The optimal gestation for benefit from corticosteroids is be-
tween 240 and 346 weeks. In many units, babies between the
230 and 236 weeks may also be considered for antenatal cor-
ticosteroids, after a multidisciplinary consultation between the
obstetric and neonatal teams, together with detailed discussions
involving the parents, regarding intact neonatal survival. Results
from the Epicure study have shown reduced neonatal death and
improved neurodevelopment for deliveries between 22 and 24
weeks gestation.
Repeated doses in singleton pregnancies are discouraged
because of the association with poor fetal growth and cerebral
palsy. The RCOG green-top guideline recommends that although
there are limited data to support the use of antenatal corticoste-
roids in multiple pregnancies, the overall improvement in
OBSTETRICS, GYNAECOLOGY AND REPRODUCTIVE MEDICINE --:-
Please cite this article in press as: Appiah-Sakyi K, Konje JC, Prevention of preterm labour, Obstetrics, Gynaecology and Reproductive Medicine
(2015), http://dx.doi.org/10.1016/j.ogrm.2015.06.005
outcomes in singleton fetuses would suggest that corticosteroids
could be beneficial in multiple pregnancies. The optimal dose in
multiple pregnancies is unknown and there is some evidence that
multifetal pregnancy attenuates the antenatal effect of
corticosteroids.
(c) Antibiotics
A number of reviews and meta-analyses have concluded that
there is no benefit from the use of prophylactic antibiotics
regardless of antimicrobial agents or gestational age at initiation
in cases of PTL. The ORACLE II study confirmed that while there
is no benefit with antibiotic use in the presence of intact mem-
branes, babies exposed to antenatal augmentin had an increased
risk of cerebral palsy. In a 7-year follow-up study, combination
antibiotics seemed only to increase this risk. While the cause of
this increased incidence of cerebral palsy is unclear, there is a
suggestion that subclinical chorioamnionitis may not be fully
eliminated by prophylactic antibiotics treatment, but rather
prolong intrauterine existence in a hostile environment and thus
increase the risk of cerebral infection especially with intact
membranes.
(d) Cervical cerclage
There are three types of clinical situations requiring cerclage
insertion and these include (i) history indicated cerclage e based
on maternal factors which increase the risk of PTL, (ii) ultra-
sound indicated cerclage e based on ultrasound demonstration
of a short cervix and (iii) rescue/emergency cerclage e as a
salvage measure in case of premature cervical dilatation with
exposure of fetal membranes.
The large Medical Research Council trial in 1993, together
with two other randomised trials demonstrated that cerclage only
showed real benefit when performed in women with 3 previous
preterm deliveries. For women with 2 or fewer previous preterm
deliveries, cervical cerclage offered no significant benefit with 25
cerclages needed to prevent one delivery before 33 weeks. The
typical presentation associated with cervical weakness (i.e.
painless cervical dilatation, rupture of membranes before the
onset of contraction, or a history of cervical surgery are all risk
factors) has been found to be unhelpful in deciding which pa-
tients require a history-indicated cervical cerclage.
Women with a shortened cervix on transvaginal scan who
also have a previous history of miscarriage or PTB are the most
suited for ultrasound-indicated cerclage. RCOG guidance is that
those with a short cervix but no previous history should not have
a cerclage. Women with a significant past history but not enough
to warrant an elective cerclage could be offered serial sono-
graphic surveillance and cerclage advised once the cervix is <2.5
cm. Women with multiple pregnancies, however, even in the
presence of a short cervix do not seem to benefit from cerclage
and there is evidence it may even increase perinatal mortality by
increasing preterm deliveries and miscarriage.
According to RCOG Guidelines, rescue sutures must be indi-
vidualised based on clinical presentation, cervical dilatation and
uterine activity. A small study of 26 women randomized to either
a rescue suture or expectant management had an average of 4
weeks delay to delivery compared to those having just bed rest.
Suture placement before 20 weeks, is highly likely to result in
preterm birth before 28 weeks. After the age of viability, it is
5 2015 Elsevier Ltd. All rights reserved.
 REVIEW
often difficult to justify inserting a cervical cerclage because the
risk of iatrogenic rupture of membranes increases while real
benefits decrease.
(e) Cervical pessary
The use of a specialised cervical pessary to prevent PTB is a
recent phenomenon. Previous observation studies suggested
potential efficacy of pessaries but in 2012 a multicentre study
which randomised 385 women between 20 and 24 weeks with
cervical lengths <2.5 cm into a pessary and expectant manage-
ment group showed a spontaneous PTB <28 weeks of 2% versus
8% (OR 0.23, 95% CI 0.06e0.74) and at <34 weeks of 6% versus
27 % (OR 0.18, 95% CI 0.08e0.37). In 2013, a smaller study on
110 Chinese women did not show such significant difference in
outcome. More studies are awaited to establish its clinical effi-
cacy and current use is therefore only advocated in the context of
clinical trials. A 10 Mackeen AD, Seibel-Seamon J, Muhammad J, Baxter JK, Berghella V.
FURTHER READING
1 Born too soon, the global report on preterm birth. http://www.who.
int/pmnch/media/news/2012/preterm_birth_report/en/index1.html.
2 Steer P. The epidemiology of preterm labour. BJOG: Int J Obstet
Gynaecol 2005; 112(suppl 1): 1e3.
3 Perinatal management of pregnant women at the threshold of infant
viability (the obstetric perspective) scientific impact paper no. 41.
London: RCOG, 2014.
4 Lockwood C.J. Overview of preterm labor and birth. In S.M. Ramin
(Ed.), UpToDate. Retreived from: http://www.uptodate.com/home/
index.html.
5 RCOG green top guideline no.44. Preterm prelabour rapture of
membranes, RCOG, London. http://www.jsog.org/GuideLines/
Preterm_prelabour_rupture_of_membranes.pdf.
OBSTETRICS, GYNAECOLOGY AND REPRODUCTIVE MEDICINE --:-
Please cite this article in press as: Appiah-Sakyi K, Konje JC, Prevention of preterm labour, Obstetrics, Gynaecology and Reproductive Medicine
(2015), http://dx.doi.org/10.1016/j.ogrm.2015.06.005
6 Mulhair L, Carter J, Poston L, Seed P, Briley A. Prospective cohort
study investigating the reliability of the AmnioSense method for
detection of spontaneous rupture of membranes. BJOG 2009 Jan;
116: 313e8. http://dx.doi.org/10.1111/j.1471-0528.2008.01828.x.
Epub 2008 Jul 23.
7 Kenyon S, Pike K, Jones DR, et al. Childhood outcomes following
prescription of antibiotics to pregnant women with preterm rupture
of the membranes: 7-year follow-up of the Oracle I trial. Lancet 2008;
372: 1319e27.
8 Kenyon S, Pike K, Jones DR, et al. Childhood outcomes after pre-
scription of antibiotics to pregnant women with spontaneous pre-
term labour: 7-year follow-up of the ORACLE II trial. Lancet 2008;
372: 1319.
9 Abdel-Aleem H, Shaaban OM, Abdel-Aleem MA. Cervical pessary for
preventing preterm birth. Cochrane Database Syst Rev 2013; 5:
CD007873.
Tocolytics for preterm premature rupture of membranes. Cochrane
Database Syst 2011; 10: CD007062.
Practice points
C Primary prevention is more effective in reducing the incidence of
PTL than treatment
C Progesterone pessaries have been shown to reduce PTB in those
with a shortened cervix identified in pregnancy
C There is no evidence of benefit from progesterone in multiple
pregnancies
C Tocolytics do not stop PTL but may delay delivery to allow ste-
roids to be administered
6 2015 Elsevier Ltd. All rights reserved.