II Guideline Application
Purpose and objectives
This guideline aims to improve both the outpatient and the inpatient care of patients
at imminent risk of preterm birth in order to reduce the rate of preterm births. If
preterm birth cannot be prevented, the aim is to reduce perinatal and neonatal morbidity
and mortality. This should lead to improvements in the psychomotor and cognitive development
of children born preterm.
Targeted areas of patient care
Outpatient and/or inpatient care
Target user groups/target audience
The recommendations of this guideline are aimed at gynecologists in private practice,
gynecologists in hospitals, pediatricians in hospitals, midwives in private practice
and midwives in hospitals. Other target user groups include advocacy groups for affected
women and children, nursing staff (obstetrics/postnatal care, pediatric intensive
care), medical and scientific societies and professional associations, institutions
for quality assurance (e.g. IQTIG), healthcare policy institutions and decision-makers
at the federal and state level, funding agencies and payers.
Adoption and period of validity
The validity of this guideline was confirmed by the executive boards of the participating
medical societies, working groups, organizations and associations as well as by the
executive boards of the DGGG, the SGGG and the OEGGG and the DGGG/OEGGG/SGGG guidelines
commission in February 2019 and was thus confirmed in its entirety. This guideline
is valid from 1 February 2019 through to 31 January 2022. Because of the contents
of this guideline, this period of validity is only an estimate. The guideline may
need to be updated earlier in urgent cases. If the guideline continues to mirror current
knowledge, its period of validity may also be extended.
III Method
Basic principle
The method used to prepare this guideline was determined by the class to which this
guideline was assigned. The AWMF Guidance Manual (version 1.0) has set out the respective
rules and requirements for different classes of guidelines. Guidelines are differentiated
into lowest (S1), intermediate (S2) and highest (S3) class. The lowest class is defined
as a set of recommendations for action compiled by a non-representative group of experts.
In 2004, the S2 class was divided into two subclasses: a systematic evidence-based
subclass (S2e) and a structural consensus-based subclass (S2k). The highest S3 class
combines both approaches. This guideline is classified as: S2k
Grading of recommendations
Grading of evidence and grading of recommendations is not envisaged for S2k-level
guidelines. The individual Statements and Recommendations are differentiated by syntax,
not by symbols ([Table 2 ]).
Table 2 Grading of recommendations.
Level of recommendation
Syntax
Strong recommendation, highly binding
must/must not
Simple recommendation, moderately binding
should/should not
Open recommendation, not binding
may/may not
In addition to the level of evidence, the above listed classification of “Recommendations”
also takes account of the clinical relevance of the underlying studies and the various
measures/factors which were not included in the grading of evidence, such as the choice
of patient cohort, intention-to-treat or per-protocol outcome analyses, medical and
ethical practice when dealing with patients, country-specific applicability, etc.
Statements
Scientific statements given in this guideline which do not consist of any direct recommendations
for action but are simple statements of fact are referred to as “Statements”. It is
not possible to provide any information about the grading of evidence for these Statements.
Achieving consensus and strength of consensus
As part of the structured process to achieve consensus (S2k/S3 level), authorized
participants attending the session vote on draft Statements and Recommendations. This
can lead to significant changes in the wording, etc. Finally, the extent of consensus
is determined based on the number of participants ([Table 3 ]).
Table 3 Grading of strength of consensus.
Symbol
Strength of consensus
Extent of agreement in percent
+++
Strong consensus
> 95% of participants agree
++
Consensus
> 75 – 95% of participants agree
+
Majority agreement
> 50 – 75% of participants agree
−
No consensus
< 51% of participants agree
Expert consensus
As the name already implies, this term refers to consensus decisions taken with regard
to specific Recommendations/Statements made without a prior systematic search of the
literature (S2k) or for which evidence is lacking (S2e/S3). The term “expert consensus”
(EC) used here is synonymous with terms used in other guidelines such as “good clinical
practice” (GCP) or “clinical consensus point” (CCP). The strength of the recommendation
is graded as previously described in the chapter “Grading of recommendations”, i.e.,
purely semantically (“must”/“must not” or “should”/“should not” or “may”/“may not”)
without the use of symbols.
Addendum of the OEGGG
To 6.9.1 Mode of delivery depending on fetal presentation and position
The Austrian Society of Gynecology and Obstetrics (OEGGG) is of the opinion that there
is no clinical or scientific basis for the Recommendation that cesarean section should
be the preferred mode of delivery based on an assumed lower risk of perinatal cerebral
hemorrhage. The OEGGG is of the opinion that the mode of delivery of infants at the
limit of viability (GW 22 + 0 bis 24 + 6) must be adapted to take the individual maternal
and fetal clinical situation into account. For singletons at the limit of viability
and in cephalic presentation, the OEGGG recommends an individualized management of
delivery, which takes the maternal and fetal clinical situation into account and where
the clinical decision process also includes the option of vaginal delivery as the
mode of delivery [1 ].
To 6.6.5 Application of antenatal steroids before late preterm delivery
Based on the results of the ALPS trial [2 ] and the recommendations of the Society for Maternal Fetal Medicine (SMFM), the OEGGG
is of the opinion that the administration of antenatal steroids in GW 34 + 0 to GW
36 + 6 may be considered, in accordance with the specifications of the SMFM.
Addendum of the SGGG
To 6.6. Administration of antenatal steroids
The opinion of the SGGG on the issues in this chapter is presented in SGGG Expert
Letter No. 56, which discusses the indications for glucocorticoid therapy to promote
antenatal lung maturation and the appropriate doses when preterm birth is imminent
(only available in German: “Glucocorticoidtherapie zur antenatalen Lungenreifung bei
drohender Frühgeburt: Indikationen und Dosierung”). Reasoning: The evidence-based recommendations in Switzerland differ slightly from those given
in this guideline, particularly with regard to the administration of antenatal glucocorticoids
in gestational weeks 34 + 0 to 36 + 0 [3 ].
To 1. Definition and Epidemiology (and various other chapters: 6.9.1., 6.9.6., 6.9.7.,
8.8., 8.9.)
As regards care at the limits of viability, please refer to the recommendations for
Switzerland which were developed together with neonatologists. Reasoning: The recommendations for Switzerland diverge in many points from the recommendations
for Germany. They are currently being revised [4 ].
To 6.2. Tocolysis
With regard to tocolytic drugs, the use of beta-mimetics for tocolysis has been approved
in Switzerland and they can be used as the tocolytic drug of first choice; see also
SGGG Expert Letter No. 41 on tocolysis for preterm labor (only available in German:
“Tokolyse bei vorzeitiger Wehentätigkeit”). Reasoning: The recommendations for Switzerland differ in many points from the recommendations
for Germany [5 ].
To 8.8 Clinical management before GW 22
The option of terminating the pregnancy should be mentioned to patients with a poor
prognosis. Reasoning: The option of terminating the pregnancy by inducing the birth in cases where there
is a serious physical or psychological risk to the mother is not mentioned in the
guideline, even though it is clinically important.
IV Guideline
6 Tertiary Prevention
6.1 Bed rest
Consensus-based Statement 6.S21
Expert consensus
Strength of consensus ++
[6 ], [7 ], [8 ], [9 ], [10 ]
There is currently no data which can confirm that bed rest reduces the rate of preterm
births. However, bed rest does increase the maternal risk of thrombosis and contributes
to the development of muscular atrophy and osteoporosis.
6.2 Tocolysis
Consensus-based Recommendation 6.E18
Expert consensus
Strength of consensus +++
The aim of tocolysis must be to prolong the pregnancy by at least 48 hours. This additional
period would make it possible to administer antenatal steroids and carry out an in-utero
transfer to a perinatal center with a neonatal intensive care unit.
6.2.1 Indications
Consensus-based Recommendation 6.E19
Expert consensus
Strength of consensus +++
Tocolytic therapy should be administered if the patient has spontaneous, regular,
preterm contractions of ≥ 4/20 min with shortening of the functional cervical length
(transvaginal measurement) and/or opening of the cervix.
Consensus-based Statement 6.S22
Expert consensus
Strength of consensus +++
If the indications are present and contra-indications have been excluded, tocolysis
is indicated in the period between GW 22 + 0 and GW 33 + 6.
Consensus-based Statement 6.S23
Expert consensus
Strength of consensus +++
[11 ], [12 ]
In cases of premature labor with cervical dilation, tocolytic therapy (beta sympathomimetics,
atosiban, nifedipine, indomethacin, NO donors) can delay the birth by 48 h in 75 – 93%
of cases and by 7 days in 62 – 78% of cases.
6.2.2 Drugs
Consensus-based Recommendation 6.E20
Expert consensus
Strength of consensus ++
Because of the significantly higher rate of maternal side effects (beta sympathomimetics)
compared to other tocolytic drugs and the lack of evidence confirming its tocolytic
efficacy (magnesium sulfate), beta sympathomimetics and magnesium sulfate should no
longer be used for tocolysis.
Of all the tocolytic drugs, beta sympathomimetics have the greatest rate of maternal
(up to 80% cardiovascular) and fetal side effects as well as requiring the most monitoring
[12 ]. There is also the additional problem of lung edema which occurs in around 1/350
applications [13 ]. They should therefore no longer be used for tocolysis [14 ].
The data on the use of magnesium sulfate as a tocolytic drug is controversial. Meta-analyses
[11 ], [12 ] showed that magnesium sulfate was an effective tocolytic in terms of prolonging
the pregnancy by 48 hours compared to placebo (OR 2.46; 95% CI: 1.58 – 4.94); however,
this flies in the face of the results and statements of the 2014 Cochrane Review [15 ], which were generated using 37 studies with 3571 pregnant women. According to the
Cochrane Review, magnesium sulfate was not more effective than placebo or even no
therapy at prolonging pregnancy for more than 48 hours and does not reduce the rate
of preterm births. However, the tocolytic efficacy of magnesium sulfate depends in
the dose, which in turn has an impact on the incidence of maternal side effects. International
guidelines no longer recommend using magnesium sulfate for tocolysis [16 ], [17 ], [18 ].
Consensus-based Recommendation 6.E21
Expert consensus
Strength of consensus ++
[11 ], [12 ]
After considering their efficacy and side effects profile, calcium antagonists (nifedipine),
oxytocin-receptor antagonists (atosiban) and COX inhibitors (indomethacin) should
be used preferentially for tocolysis, even though some have not yet been approved
for use.
6.2.3 Combining several tocolytics
Consensus-based Recommendation 6.E22
Expert consensus
Strength of consensus +++
[13 ], [19 ]
Based on current data, combining different tocolytics is associated with significantly
increased rates of maternal side effects compared to administering a single tocolytic,
and as there are no data confirming any increase in efficacy, combining different
tocolytics should be avoided.
Consensus-based Recommendation 6.E23
Expert consensus
Strength of consensus +++
[20 ]
Tocolytics should not be administered in combination with oral/vaginal progesterone
(“adjunctive tocolysis”), because data on this issue is still insufficient.
6.2.4 Tocolysis for extremely preterm birth, multiple pregnancy and intrauterine
growth restriction
Consensus-based Statement 6.S24
Expert consensus
Strength of consensus +++
[21 ]
Evidence from randomized controlled studies on the benefits of tocolytics for extremely
preterm birth, multiple pregnancy and intrauterine growth restriction is lacking.
The decision whether to administer tocolytics in such cases must be made on a case-by-case
basis.
6.2.5 Long-term tocolysis
Consensus-based Recommendation 6.E24
Expert consensus
Strength of consensus +++
[22 ], [23 ], [24 ], [25 ]
According to the information currently available, long-term or maintenance tocolysis
(generally defined as tocolysis for more than 48 h) should not be used to reduce the
rate of preterm births or neonatal morbidity and mortality rates.
6.3 Progesterone for maintenance tocolysis
Consensus-based Recommendation 6.E25
Expert consensus
Strength of consensus +++
After tocolysis, pregnant women with a singleton pregnancy should not be given progesterone
to maintain the pregnancy and prevent preterm birth.
A meta-analysis carried out in 2017 which selectively included high-quality studies
on this issue found that the use of progesterone for maintenance tocolysis did not
significantly reduce the rate of preterm births before the 37th week of gestation
(OR 1.23, 95% CI: 0.91 – 1.67) [26 ].
6.4 Cervical pessary for shortened cervical length after premature labor
Consensus-based Statement 6.S25
Expert consensus
Strength of consensus +++
There is some evidence from a prospective randomized study that placement of a cervical
pessary in pregnant women previously treated for premature labor who have a shortened
cervical length as measured by transvaginal ultrasound (< 25 mm between GW 24 + 0
and GW 29 + 6; < 15 mm between GW 30 + 0 and GW 33 + 6) may reduce the rate of preterm
births.
Pratcorona et al. recently published a prospective randomized study which included
357 patients between GW 24 + 0 and GW 33 + 6 [27 ]. If patients had a shortened cervical length (≤ 25 mm between GW 24 + 0 and GW 29 + 6;
≤ 15 mm between GW 30 + 0 and GW 33 + 6) 48 hours after being treated for premature
labor, they were managed either by placing a cervical pessary or by standard protocol.
The primary study outcome, in this case, the preterm birth rate before the 34th week
of gestation, did not differ significantly between groups (10.7 vs. 13.7%; RR 0.78
[95% CI: 0.45 – 1.38]). However, the preterm birth rate before the 37th week of gestation
was significantly lower after placement of a cervical pessary (14.7 vs. 25.1%; RR
0.58 [95% CI: 0.38 – 0.90]) as was the number of patients readmitted to hospital after
previously being treated for premature labor (4.5 vs. 20.0%; RR 0.23 [95% CI: 0.11 – 0.47]).
However, these results could not be confirmed in the APOSTEL VI trial [28 ].
6.5 Administration of antibiotics for premature labor
Consensus-based Recommendation 6.E26
Expert consensus
Strength of consensus +++
Cases of premature labor without rupture of membranes must not be treated with antibiotics
with the goal of prolonging the pregnancy or reducing neonatal morbidity.
Meta-analyses found that the administration of antibiotics to cases with premature
labor and no rupture of membranes had no effect on the duration of the pregnancy,
the preterm birth rate, respiratory distress syndrome or neonatal sepsis [29 ], [30 ]. Given these findings, the potential risks of administering antibiotics when their
administration is not indicated need to be discussed.
6.6 Administration of antenatal steroids
6.6.1 Administration and dosage
Consensus-based Recommendation 6.E27
Expert consensus
Strength of consensus +++
[31 ]
Antenatal steroids must be administered to women at imminent risk of preterm birth
before GW 34 + 0, with treatment consisting of 2 × 12 mg betamethasone administered
IM at an interval of 24 h (alternatively: dexamethasone, 4 × 6 mg every 12 h).
6.6.2 Starting in which week of gestation?
Consensus-based Recommendation 6.E28
Expert consensus
Strength of consensus +++
Antenatal steroids should also be administered in cases at imminent risk of preterm
birth < GW 24 + 0 if maximum therapy in a neonatal intensive care unit is planned.
A recently published meta-analysis found 8 non-randomized studies on this issue [32 ]. The impact on neonatal mortality and morbidity of a single dose of corticosteroids
administered in the period GW 22 + 0 to GW 23 + 6 is shown in [Tables 4 ] and [5 ].
Table 4 Effects of antenatal steroids on the outcome of infants between GW 22 + 0 and GW
22 + 6 [32 ].
GW 22 + 0 – GW 22 + 6
OR
95% CI
Neonatal mortality
0.58
0.38 – 0.89
Intraventricular cerebral hemorrhage (grade III – IV) or periventricular leukomalacia
1.03
0.55 – 1.93
Chronic pulmonary disease
1.19
0.52 – 2.73
Necrotizing enterocolitis (> stage II)
0.59
0.03 – 12.03
Table 5 Effects of antenatal steroids on the outcome of infants between GW 23 + 0 and GW
23 + 6 [32 ].
GW 23 + 0 – GW 23 + 6
OR
95% CI
Neonatal mortality
0.50
0.42 – 0.58
Intraventricular cerebral hemorrhage (grade III – IV) or periventricular leukomalacia
0.75
0.55 – 1.03
Chronic pulmonary disease
0.94
0.59 – 1.51
Necrotizing enterocolitis (> stage II)
0.93
0.66 – 1.32
While neonatal mortality was significantly reduced after a single dose of corticosteroids,
it apparently had no effect on morbidity. Given the rapid recent progress in the field
of neonatal intensive care, prospective randomized studies on this issue are urgently
required.
6.6.3 Repeat administration of antenatal steroids
Consensus-based Recommendation 6.E29
Expert consensus
Strength of consensus +++
If steroids are administered to women before the 29 + 0 week of gestation because
of an imminent risk of preterm birth and steroids were administered more than 7 days
previously, a further dose of steroids may be administered after the patient has been
re-assessed if the imminent risk of preterm birth is increasing.
Zephyrin and colleagues used a Markov model to investigate how to achieve the right
balance between risks and benefits with repeat administration of antenatal steroids
[33 ]. The improved neonatal outcomes after multiple glucocorticoid administrations were
set against the risk of fetal growth restriction. After 29 + 0 weeks of gestation,
a repeat administration of antenatal steroids was associated with increasing risks
for the infant ([Fig. 1 ]). Any repeat administration of antenatal steroids should therefore be limited to
cases with a very low gestational age (< GW 29 + 0).
Fig. 1 Benefits of administering antenatal steroids according to gestational age [33 ]
6.6.4 Timing of antenatal steroid administration
Consensus-based Statement 6.S26
Expert consensus
Strength of consensus +++
The timing of and indication for administering antenatal steroids must be carefully
weighed up, as neonatal morbidity and mortality can only be reduced in the period
between 24 h and 7 days after the first administration. There is some evidence that
administering antenatal steroids already has an effect before 24 h.
There are now a number of cohort studies which show that perinatal morbidity and mortality
depend significantly on the timing of lung maturity [34 ], [35 ], [36 ]. An example of this is shown in [Fig. 2 ], which depicts the neonatal survival of infants born preterm at ≤ 26 weeks of gestation
[36 ].
Consensus-based Recommendation 6.E30
Expert consensus
Strength of consensus +++
[37 ], [38 ]
Patients with premature contractions and a cervical length of > 30 mm or 15 – 30 mm
as measured by transvaginal ultrasound and who additionally test negative for fibronectin,
phIGFBP-1 and PAMG-1 should not be given antenatal steroids just because of the contractions,
as the risk of preterm birth in the next 7 days is low (< 5%).
Consensus-based Recommendation 6.E31
Expert consensus
Strength of consensus +++
[39 ]
So-called rapid maturation, consisting of the administration of a second dose of betamethasone
after just 12 h rather than after 24 h, should be avoided as this significantly increases
the risk of necrotizing enterocolitis.
Fig. 2 Survival of very immature infants (< 26th week of gestation) according to the timing
of antenatal steroid administration [36 ].
6.6.5 Administration of antenatal steroids and late preterm birth
Consensus-based Recommendation 6.E32
Expert consensus
Strength of consensus ++
Administering antenatal steroids to patients between GW 34 + 0 and GW 36 + 5 with
an imminent risk of preterm birth should currently be avoided as there are still no
studies on the impact this can have on the childrenʼs psychomotor development later
on.
The ALPS trial found a significant reduction in neonatal respiratory distress in children
born in late preterm at GW 34 + 0 to GW 36 + 5, whose mothers were given 2 × 12 mg
betamethasone IM antenatally [2 ]. The ASTECS trial, which studied pregnant women who underwent elective cesarean
section at term, also reported a significant reduction in RDS in children born to
mothers who received 2 × 12 mg betamethasone antenatally [40 ]. However, at a school assessment carried out by teachers 10 years later, it was
found that significantly more children from the intervention group were in the lower
performance quartile and fewer children were in the top performance quartile [41 ]. No follow-up examinations of the children in the ALPS trial have been carried out
to date. Because of this, no antenatal corticoids should be administered to this group
of patients for the time being.
6.7 Emergency cerclage
Consensus-based Recommendation 6.E33
Expert consensus
Strength of consensus +++
An emergency cerclage may be placed in women with a singleton pregnancy and cervical
dilation of more than 1 cm before GW 24 + 0 with the goal of significantly prolonging
the pregnancy.
Consensus-based Recommendation 6.E34
Expert consensus
Strength of consensus +++
Women treated with emergency cerclage should receive indomethacin and antibiotics
perioperatively.
A meta-analysis published in 2015 (n = 772 women from 11 studies, n = 496 underwent
emergency cerclage placement, n = 276 were managed expectantly) found a significant
prolongation of pregnancy and reduction of perinatal mortality after placement of
an emergency cerclage for cervical dilation (duration of pregnancy: plus 5.4 weeks,
perinatal mortality reduced from 58.5% to 29.1%) [42 ]. The administration of indomethacin and cefazolin increased the percentage of women
who did not give birth within the following 4 weeks (92.3 vs. 62.5%) [43 ].
6.8 Neuroprotection
Consensus-based Statement 6.S27
Expert consensus
Strength of consensus +++
[44 ]
Periventricular/intraventricular hemorrhage (PIVH) and periventricular leukomalacia
(PVL)/diffuse cerebral white matter injury are typical forms of brain injury found
in survivors of preterm birth.
6.8.1 Magnesium
Consensus-based Recommendation 6.E35
Expert consensus
Strength of consensus +++
[45 ], [46 ]
Magnesium may be administered intravenously for fetal neuroprotection to patients
< GW 32 at imminent risk of preterm birth.
Treatment should be started with a bolus of 4 – 6 g administered over 30 min, followed
by a maintenance dose of 1 – 2 g for 12 h. The aim is to double the magnesium levels
in maternal serum. If the birth does not occur within 12 h, magnesium may be administered
again later on when preterm birth is once again imminent.
6.8.2 Delayed cord clamping
Consensus-based Recommendation 6.E36
Expert consensus
Strength of consensus +++
[47 ], [48 ], [49 ]
Cord clamping of infants born preterm should be delayed or umbilical cord milking
should be carried out.
6.9 Delivery
6.9.1 Delivery depends on fetal presentation
Consensus-based Recommendation 6.E37
Expert consensus
Strength of consensus ++
[50 ], [51 ], [52 ], [53 ], [54 ], [55 ], [56 ], [57 ], [58 ], [59 ], [60 ], [61 ], [62 ], [63 ]
Delivery by cesarean section may be considered after carefully weighing up the risk/benefits
in each individual case if the fetus is aged < GW 30 + 0 and in cephalic presentation.
Consensus-based Recommendation 6.E38
Expert consensus
Strength of consensus ++
[64 ]
Depending on the sonographically estimated fetal weight and other factors, delivery
by cesarean section should be considered to reduce neonatal morbidity and mortality
if the fetus is aged < GW 36 + 0 and in breech presentation.
6.9.2 Longitudinal uterine incision for cesarean section
Consensus-based Recommendation 6.E39
Expert consensus
Strength of consensus +++
Particularly in cases of extremely preterm birth, longitudinal uterine section may
be appropriate in individual cases as it may be the most beneficial form of delivery
for the infant.
Consensus-based Recommendation 6.E40
Expert consensus
Strength of consensus +++
[65 ], [66 ]
Because of the increased risk of uterine rupture, women who have had a previous longitudinal
c-section must be delivered by primary repeat c-section in all subsequent births.
6.9.3 Vaginal operative delivery
Consensus-based Recommendation 6.E41
Expert consensus
Strength of consensus ++
[67 ]
Because of the increased risk of intraventricular hemorrhage, fetuses under the age
of 34 + 0 weeks of gestation should not be delivered by vacuum extraction.
6.9.4 Fetal blood gas analysis
Consensus-based Recommendation 6.E42
Expert consensus
Strength of consensus +++
Fetal blood gas analysis should not be carried out for fetuses under the age of 34 + 0
weeks of gestation because of the potential risk of injury.
6.9.5 Antibiotic prophylaxis for group B streptococcus
Consensus-based Recommendation 6.E43
Expert consensus
Strength of consensus +++
[68 ]
If the GBS status of a case of preterm birth is positive or unknown, antibiotic prophylaxis
must be administered during delivery.
6.9.6 Cooperation with the Neonatology Department
Consensus-based Recommendation 6.E44
Expert consensus
Strength of consensus ++
[69 ], [70 ], [71 ]
A pediatrician/neonatologist must be involved early on in the treatment and counselling
of women with an imminent risk of preterm birth.
The treating pediatrician must be given all information about the pregnant woman which
may be important for the initial medical treatment and therapy of the preterm infant.
Such information includes any medication taken, HBsAg status, blood group, CMV antibody
status (up to the 32nd week of gestation), findings from any prenatal diagnostic workups,
and results of microbiological screening of the pregnant woman at imminent risk of
preterm birth for GBS, MRSA, MRGN as well as the results of any repeat screenings
if pregnancy is prolonged.
Consensus-based Recommendation 6.E45
Expert consensus
Strength of consensus ++
[72 ]
When an infant is born preterm (< GW 35 + 0), a physician with experience in neonatology
must be present to directly oversee the care of the newborn infant. If there is an
imminent risk of preterm birth before GW 32 + 0 and/or the estimated weight/birthweight
is < 1500 g, a specialist physician with a subspecialization in neonatology must be
on call.
6.9.7 Terminal care
Consensus-based Recommendation 6.E46
Expert consensus
Strength of consensus +++
[73 ], [74 ], [75 ]
Specially trained staff must be called in to offer palliative and terminal care to
deceased or dying newborns and their family in the perinatal phase. Terminal care
is included in perinatology training. According to the tenets of the German Medical
Association, offering terminal care with dignity is a key medical duty for physicians
which they cannot delegate.
7 Special Aspects Relating to Twin and Multiple Pregnancies
7.1 Epidemiology and etiology
Consensus-based Statement 7.S28
Expert consensus
Strength of consensus +++
[76 ], [77 ]
Women carrying a multiple pregnancy have a significantly higher risk of preterm birth.
7.2 Prevention
7.2.1 Progesterone
Consensus-based Recommendation 7.E47
Expert consensus
Strength of consensus +++
[78 ], [79 ]
Women must not be given progesterone to prevent preterm birth only because they are
carrying twins.
Consensus-based Recommendation 7.E48
Expert consensus
Strength of consensus +++
Women carrying a twin pregnancy who have a cervical length of ≤ 25 mm before GW 24 + 0
as measured by transvaginal ultrasound should receive a daily dose of 200 – 400 mg
progesterone applied intravaginally until GW 36 + 6.
An individual patient data meta-analysis (IPDMA) of six studies [79 ], [80 ], [81 ], [82 ], [83 ], [84 ] carried out by Romero et al. in 2017, which compared the application of vaginal
progesterone with placebo or no treatment in 303 asymptomatic women with twin pregnancy
and a cervical length of ≤ 25 mm in the second trimester, found a significant reduction
in preterm births before the 33rd week of gestation (31.4 vs. 43.1%; RR 0.69 [95%
CI: 0.51 – 0.93]) and improved neonatal outcomes (e.g., lower neonatal mortality rate
[RR 0.53; 95% CI 0.35 – 0.81], lower incidence of respiratory distress syndrome [RR
0.70; 95% CI: 0.56 – 0.89], fewer neonates with a birthweight < 1500 g [RR 0.53; 95%
CI: 0.35 – 0.80]) [85 ].
7.2.2 Cerclage
Consensus-based Recommendation 7.E49
Expert consensus
Strength of consensus +++
Primary or secondary cerclage should not be placed in women with twin pregnancies.
The first meta-analysis of three prospective randomized studies found a significantly
higher preterm birth rate before the 35th week of gestation for women carrying a twin
pregnancy after placement of a primary or secondary cerclage (76 vs. 36%; RR 2.15,
95% CI: 1.15 – 4.01) [86 ], [87 ], [88 ], [89 ]. Another meta-analysis has since been carried out which additionally took individual
patient data into account [90 ]. This meta-analysis found that placement of a cerclage had no negative effect on
the preterm birth rate or perinatal morbidity, at least for patients with a short
cervix, before the 24th week of gestation.
7.2.3 Cervical pessary for shortened cervical length
Consensus-based Recommendation 7.E50
Expert consensus
Strength of consensus +++
A cervical pessary can be placed in individual cases with twin pregnancy and a cervical
length of ≤ 25 mm before GW 24 + 0 as measured by transvaginal sonography.
Given the fact that prospective randomized studies have reported both positive [91 ], [92 ], [93 ] and negative [94 ], [95 ] data, the decision whether or not to carry out this procedure must be made on a
case-by-case basis.
7.2.4 Cervical pessary after preterm labor and shortened cervical length
Consensus-based Statement 7.S29
Expert consensus
Strength of consensus +++
There is some evidence from a prospective randomized study that placement of a cervical
pessary in cases with twin pregnancy previously treated for preterm labor and with
a shortened cervical length as measured by transvaginal ultrasound (< 20 mm between
GW 24 + 0 and GW 29 + 6; < 10 mm between GW 30 + 0 and GW 33 + 6) can reduce the rate
of preterm births.
In a prospective randomized study which included 132 women with twin pregnancy between
GW 24 + 0 and GW 33 + 6 [96 ], patients who were found to have a shortened cervical length (≤ 20 mm between GW
24 + 0 and GW 29 + 6; ≤ 10 mm between GW 30 + 0 and GW 33 + 6) 48 h after treatment
for preterm labor either underwent placement of a cervical pessary or received the
usual standard care. The primary study outcome – i.e., the preterm rate before the
34th week of gestation – was significantly lower in the intervention goup (16.4 vs.
32.3%; RR 0.51 [95% CI: 0.27 – 0.97]) as was the number of readmitted patients after
treatment for preterm labor (5.6 vs. 21.5%; RR 0.28 [95% CI: 0.10 – 0.80]). Moreover,
placement of a cervical pessary significantly reduced the prevalence of necrotizing
enterocolitis (0 vs. 4.6%) and of neonatal sepsis (0 vs. 6.2%).
7.2.5 Emergency cerclage
Consensus-based Recommendation 7.E51
Expert consensus
Strength of consensus +++
If the cervix has opened more than 1 cm before GW 24 + 0, emergency cerclage may be
carried out even in women with a twin pregnancy with the aim of significantly prolonging
the pregnancy.
As has already been established for women with singleton pregnancies, cohort studies
have shown that a twin pregnancy can also be prolonged if an emergency cerclage is
placed in women with an opened cervix before GW 24 + 0 [97 ], [98 ], [99 ], [100 ].
8 Preterm Premature Rupture of Membranes (PPROM)
8.1 Prevalence and Etiology
Consensus-based Statement 8.S30
Expert consensus
Strength of consensus +++
[101 ]
Around 3% of all pregnant women are affected by preterm premature rupture of membranes
(rupture of membranes before GW 37 + 0): 0.5% before the 27th week of gestation, 1%
between 27 and 34 weeks of gestation and 1% between the 34th and the 37th week of
gestation.
8.2 Risk factors
Consensus-based Statement 8.S31
Expert consensus
Strength of consensus +++
[102 ], [103 ]
A previous history of PPROM is a significant risk factor for preterm premature rupture
of membranes. The additional risk factors are similar to those for spontaneous preterm
birth.
8.3 Diagnostic workup
Consensus-based Recommendation 8.E52
Expert consensus
Strength of consensus +++
[104 ], [105 ]
In most cases, PPROM can be diagnosed by speculum examination. If there is still some
uncertainty, then biochemical tests must be carried out.
Consensus-based Recommendation 8.E53
Expert consensus
Strength of consensus +++
A digital examination must be avoided in patients with PPROM.
When examining patients with PPROM, a digital examination must be avoided where possible,
because digital examinations increase the risk of ascending infection and significantly
reduce the latency period to delivery [106 ], [107 ].
8.4 Latency period
Consensus-based Statement 8.S32
Expert consensus
Strength of consensus +++
[108 ], [109 ]
More than 50% of all patients with PPROM are delivered within one week.
8.5 Maternal and fetal risks
Consensus-based Statement 8.S33
Expert consensus
Strength of consensus +++
[110 ], [111 ], [112 ], [113 ], [114 ], [115 ]
Patients with PPROM have a risk of clinical infection. Additional risks include placental
abruption and umbilical cord prolapse.
8.6 Triple I ([Table 6 ])
Consensus-based Statement 8.S34
Expert consensus
Strength of consensus +++
Internationally, the term “Triple I” has superseded the term chorioamnionitis to differentiate
maternal fever from infection or inflammation or both.
Table 6 Classification of maternal fever and Triple I*.
Definition
* Triple I: inflammation or infection or both; ** amniotic fluid obtained by amniocentesis;
*** postpartum histopathology of the placenta [116 ].
Maternal fever
Maternal fever is present when the orally measured temperature exceeds 39.0 °C.
If the orally measured temperature is between 38.0 and 38.9 °C, the temperature should
be measured again after 30 minutes. If the temperature again exceeds 38.0 °C, then
maternal fever is present.
Suspicious for Triple I
Maternal fever of unclear origin together with at least one of the following criteria:
fetal tachycardia of more than 160 beats/min for > 10 min
maternal leukocytes > 15 000 µl without the administration of corticosteroids
purulent discharge from the cervix
Confirmed Triple I
Suspicion of Triple I and objective findings of infection, such as:
positive Gram staining of amniotic fluid**, low glucose concentrations (< 14 mg/dl),
increased number of leukocytes (> 30 cells/mm3 ), positive bacterial culture
or
histopathological findings*** of inflammation or infection of both of the placenta,
the amniotic membranes or the umbilical cord (funisitis)
8.7 Maternal and fetal risks associated with Triple I
Consensus-based Statement 8.S35
Expert consensus
Strength of consensus +++
[117 ], [118 ], [119 ], [120 ], [121 ], [122 ]
In addition to sepsis, maternal risks associated with Triple I include uterine dysfunction
with the risk of failure to progress in labor and uterine atony post partum. In cases
where delivery was by cesarean section, risks include wound infection, endomyometritis,
thrombophlebitis and pelvic abscess formation.
Consensus-based Statement 8.S36
Expert consensus
Strength of consensus +++
[123 ], [124 ]
The fetus may develop inflammatory response syndrome as part of Triple I. Affected
infants have a higher risk of sepsis post partum.
8.8 Clinical management of PPROM before GW 22
Consensus-based Recommendation 8.E54
Expert consensus
Strength of consensus +++
[125 ], [126 ], [127 ]
If PPROM occurs before the fetus has achieved viability, the risk of maternal sepsis,
fetal pulmonary hypoplasia and fetal skeletal deformities must be discussed with the
future parents.
Consensus-based Recommendation 8.E55
Expert consensus
Strength of consensus +++
Antibiotic therapy may be considered in patients with PPROM before the fetus has achieved
viability.
As almost all studies on antibiotic therapy in cases with rupture of membranes only
recruited patients after the 24 + 0 week of gestation, there are no reliable data
on the administration of antibiotics before the fetus has achieved viability. But
the risk that the patient may develop sepsis due to ascending infection suggests that
antibiotic therapy is advisable [128 ]. The same regimen as the one described for PPROM between (GW 22 + 0) GW 24 + 0 and
GW 33 + 6 GW can be used.
Consensus-based Recommendation 8.E56
Expert consensus
Strength of consensus +++
Antenatal steroid administration, tocolysis and neuroprotection with magnesium must
not be carried out in cases with PPROM before the fetus has achieved viability.
8.9 Clinical management of PPROM between (GW 22 + 0) GW 24 + 0 and GW 33 + 6
Consensus-based Recommendation 8.E57
Expert consensus
Strength of consensus +++
Recommendation: Between GW 22 + 0 and GW 23 + 6 the further course of action should
be agreed upon with the parents in accordance with the German-language guideline “Frühgeborene
an der Grenze der Lebensfähigkeit 024 – 019” [Preterm infants at the limits of viability].
8.9.1 Expectant management
Consensus-based Recommendation 8.E58
Expert consensus
Strength of consensus ++
If PPROM occurs between GW 24 + 0 and GW 33 + 6 or between GW 22 + 0 and GW 23 + 6
if maximum therapy is requested, expectant management must be considered first if
there is no immediate risk to mother or child.
If PPROM occurs between GW 24 + 0 and GW 33 + 6 or between GW 22 + 0 and GW 23 + 6
if maximum therapy is requested, the risks of ascending infection must be weighed
against the neonatal risks which can result from preterm birth ([Table 7 ]). An ascending infection with chorioamnionitis, preterm placental abruption, pathological
CTG, or umbilical cord prolapse are indications for immediate delivery of the fetus.
Otherwise expectant management is currently the international standard of care [129 ].
Table 7 Planned delivery vs. expectant management of PPROM between the 24th and the 37th
week of gestation.
Planned delivery vs. expectant management
RR
95% CI
[130 ]
Neonatal sepsis
0.93
0.66 – 1.30
Neonatal infection (positive blood culture)
1.24
0.70 – 2.21
RDS
1.26
1.05 – 1.53
Cesarean section
1.26
1.11 – 1.44
Perinatal mortality
1.76
0.89 – 3.50
Intrauterine fetal death
0.45
0.13 – 1.57
Neonatal mortality
2.55
1.17 – 5.56
Mechanical ventilation required
1.27
1.02 – 1.58
Transfer to neonatal intensive care unit
1.16
1.08 – 1.24
Chorioamnionitis
0.50
0.26 – 0.95
Endomyometritis
1.61
1.00 – 2.59
Induction of labor
2.18
2.01 – 2.36
8.9.2 Administration of antenatal steroids
Consensus-based Recommendation 8.E59
Expert consensus
Strength of consensus +++
Patients with PPROM between GW 24 + 0 and GW 33 + 6 or between GW 22 and GW 23 + 6,
if maximum therapy is requested, must be given antenatal steroids consisting of 2 × 12 mg
betamethasone administered IM at an interval of 24 h (alternatively dexamethasone,
4 × 6 mg every 12 h).
8.9.3 Administration of antibiotics
Consensus-based Recommendation 8.E60
Expert consensus
Strength of consensus ++
[131 ]
Patients with PPROM between GW 24 + 0 and GW 33 + 6 or between GW 22 and GW 23 + 6,
if maximum therapy is requested, must be given antibiotic therapy.
Consensus-based Recommendation 8.E61
Expert consensus
Strength of consensus +++
[108 ], [129 ], [131 ]
The data are not sufficient to permit any recommendations to be made about specific
therapy regimens. One option is IV administration of ampicillin over 2 days followed
by 5 days of oral amoxicillin and a single oral dose of azithromycin at the start.
Amoxicillin must not be combined with clavulanic acid.
8.9.4 Tocolysis
Consensus-based Statement 8.S37
Expert consensus
Strength of consensus +++
[132 ]
Tocolysis is not associated with any significant improvement in perinatal morbidity
and mortality rates in cases with PPROM.
8.9.5 Neuroprotection
See 6.8.1.
8.9.6 Maternal and fetal monitoring
Consensus-based Recommendation 8.E62
Expert consensus
Strength of consensus +++
Patients with PPROM must be carefully monitored for Triple I. Clinical signs include
maternal fever plus one of the following: fetal tachycardia (> 160 beats/min) or leukocytes
> 15 000/µl or purulent discharge from the cervix.
Pregnant women with preterm premature rupture of membranes should be routinely examined
for signs of infection. In addition to the above-mentioned clinical parameters, such
signs also include symptoms such as painful uterus, uterine contractions, maternal
blood pressure and heart rate [116 ]. Blood count and CRP must additionally be monitored at least once a day. However,
the benefit of daily laboratory tests is disputed [133 ]. Kunze et al. reported an AUC of just 0.66 for a combination of maternal fever,
CRP and leukocytes to predict FIRS [134 ]. Musilova et al. reported a sensitivity of 47%, specificity of 96%, positive predictive
value of 42% and negative predictive value of 96% for a CRP value of 17.5 mg/l in
maternal serum to predict intraamniotic infection or inflammation [135 ].
Daily CTG monitoring of patients with PPROM is standard clinical practice. But currently
there is no fetal monitoring method which can reliably detect intrauterine inflammation
or infection. Neither CTG nor the use of a biophysical profile (CTG plus fetal breathing
movements and other fetal movements, fetal tone and amniotic fluid volume assessment)
are suitable predictors for intrauterine infection (CTG: sensitivity 39%; biophysical
profile: 25%) [115 ].
Regular monitoring of amniotic fluid volumes is similarly of little benefit. While
a reduction in amniotic fluid volume increases the risk of umbilical cord compression
and demonstrably reduces the time to the start of labor, its predictive value for
a negative outcome is low [136 ]. The use of Doppler sonography has no proven benefits for premature rupture of membranes
[137 ].
Consensus-based Statement 8.S38
Expert consensus
Strength of consensus ++
[138 ]
The use of amniocentesis to diagnose Triple I is only useful in exceptional cases,
e.g. when the source of maternal infection is not clear.
Consensus-based Statement 8.S39
Expert consensus
Strength of consensus +++
[134 ], [139 ]
The prediction of Triple I based on biochemical parameters measured in vaginal secretions
is not useful according to current knowledge.
8.9.7 Amniotic infusion
Consensus-based Statement 8.S40
Expert consensus
Strength of consensus +++
[140 ]
The value of amniotic infusion in cases of PPROM cannot be sufficiently evaluated
based on the data currently available.
8.9.8 Antibiotic prophylaxis for Group B streptococcus
See the recommendations on GBS prophylaxis.
8.9.9 Delivery
Consensus-based Recommendation 8.E63
Expert consensus
Strength of consensus +++
[129 ], [130 ]
Patients with PPROM between GW 24 + 0 and GW 33 + 6 or between GW 22 and GW 23 + 6,
if maximum therapy is requested, can be delivered from GW 34 + 0 onwards. Indications
for immediate delivery are Triple I (suspicion of Triple I or confirmed), premature
placental abruption, pathological CTG or high risk, or umbilical cord prolapse.
Consensus-based Recommendation 8.E64
Expert consensus
Strength of consensus +++
Patients with Triple I (suspicion or confirmed) must be given antibiotics and their
infant must be delivered.
8.10 Clinical Management of PPROM between GW 34 + 0 and GW 36 + 6
Consensus-based Recommendation 6.E65
Expert consensus
Strength of consensus +++
If preterm premature rupture of membranes occurs between GW 34 + 0 and GW 36 + 6,
expectant management may be considered as an alternative to prompt delivery, with
the aim of prolonging the pregnancy until GW 37 + 0. This does not apply if Group
B streptococcus is detected in vaginal secretions.
A total of 1839 women between GW 34 + 0 and GW 36 + 6 who had preterm premature rupture
of membranes (PPROM) were recruited into the PPROMT trial between 2004 and 2013 [141 ]. Immediate induction of labor was compared with expectant management. In the study
group, 21% of infants were born after the 37th week of gestation to women managed
expectantly compared to only 3% in the control group. The prevalence of neonatal sepsis
was the same for both groups, however respiratory distress syndrome (RDS) occurred
significantly less often after expectant management. In this group, the birthweight
of the children was also significantly higher and the stay in the neonatal intensive
care unit or in hospital was shorter. However, as expected, uterine bleeding before
or during birth occurred more often in the mothers of these children as did peripartum
fever. The c-section rate was significantly lower compared to the group who had induction
of labor [141 ].
The results of the PPROMT trial were supported by the findings of the PPROMEXIL and
PPROMEXIL-2 trials [142 ], [143 ]. But if Group B streptococcus colonization was diagnosed, the prevalence of early
onset sepsis was significantly higher among affected neonates (15.2 vs. 1.8%; p = 0.04)
[144 ].
According to a meta-analysis of this issue which included 12 studies, expectant management
was still not found to be associated with an increased prevalence of neonatal sepsis.
Following immediate induction of labor, the rates for RDS, neonatal mortality, required
ventilation, endomyometritis and cesarean section were significantly higher while
the incidence of chorioamnionitis was lower [130 ]. A patient-level meta-analysis came to similar conclusions [145 ].
Consensus-based Recommendation 8.E66
Expert consensus
Strength of consensus +++
Clinical monitoring and antibiotic therapy in cases with PPROM between GW 34 + 0 and
GW 36 + 6 must follow the recommendations for (GW 22 + 0) GW 24 + 0 – GW 33 + 6. Antenatal
steroids, tocolysis or neuroprotection with magnesium must not be administered.
9 Psychosomatic Care and Supportive Therapy
Consensus-based Recommendation 9.E67
Expert consensus
Strength of consensus ++
Pregnant women admitted to hospital for premature labor and women who had a preterm
birth should be offered psychosomatic care and supportive therapy.
In addition to worries about the health consequences of a preterm birth (which are
difficult to estimate), therapeutic measures, which can include immobilization, medication
to stop contractions and the administration of corticosteroids, may be experienced
as stressful. If there are additional stresses (a previous experience of loss, prior
mental health problems, partnership difficulties, etc.), then the incidence of anxiety
and depression is higher [146 ], [147 ], [148 ]. Particularly for large families, admission of the mother to hospital represents
substantial organizational pressures for the family.
There are a number of psychometric tests which are used to detect psychological and
social stress factors, such as HADS, the Babylotse Plus screening questionnaires,
etc. [149 ].
Affected couples should be offered acute psychological crisis intervention, followed
by offers of supportive talks and psychotherapy where necessary. This also supports
parent-child bonding.
The support offered by self-help groups such as the German federal association “Das
Frühgeborene Kind” [The Preterm Infant] [150 ] can help affected parents, and parents should be informed about such options.
Affected families should be actively offered options in the context of the Frühe Hilfe
network. This is a German network that creates local and regional support systems
offering coordinated services to parents and children, which aims to improve familial
and social development opportunities for children and parents, both in the early stages
and over the long term [151 ].
The “Babylotse” program, which arranges the transfer of families from the regular
healthcare system to the Frühe Hilfe network and other social care systems has proven
to be particularly useful. The core aspect of this program is the role it plays in
guiding parents to find and use the most suitable options from among the numerous
local choices available.
All of these measures are services which provide compassionate support to the patient
and her family and which are offered in addition to the care provided by the attending
midwife.
