Key words
tocolysis - atosiban - hexoprenaline - efficacy
Schlüsselwörter
Tokolyse - Atosiban - Hexoprenalin - Wirksamkeit
Introduction
Preterm birth is defined as a live birth occurring before 37 weeks of gestation. Higher
body mass index (BMI), advanced maternal age and prior use of assisted reproductive
techniques such as
in vitro fertilization (IVF) cause higher rates of preterm birth, especially in
industrial nations. Other factors for preterm birth are a short cervix, infection,
previous preterm birth or
uterine overdistension. Preterm infants have higher risk of morbidity, mortality
and neurodevelopmental impairment [1], [2], [3].
Around 10.6% of all deliveries worldwide are premature [4]. The number of premature deliveries in Germany has remained almost the same in recent
years; in 2008,
it was 8.0% and in Austria it was 7.9% in 2016 [5].
Several tocolytic agents exist; however, only few have been approved by the FDA or
EMA [6]. Approved agents include β2-adrenoceptoragonists (ritodrine,
terbutaline, hexoprenaline), calcium channel blockers (e.g., nifedipine), magnesium
sulfate, nitrates (e.g., nitroglycerin), oxytocin receptor antagonists (e.g., atosiban),
and prostaglandin
inhibitors (e.g., indomethacin) [6]. It is recommended that tocolytic agents are given for 48 hours to allow the administration
of antenatal corticosteroids to
reduce the risk of infant respiratory distress syndrome (IRDS) and intraventricular
hemorrhage (IVH) [7] or permit the transfer of the mother to a tertiary
hospital [8]. Hexoprenaline and atosiban have been deemed safe for the fetus [9]. In general, atosiban is known to have
significant fewer side effects than beta-mimetics with similar efficacy [10], [11]. The most common side effects of
hexoprenaline are palpitations, maternal tachycardia and chest pain [10].
Repeated or prolonged tocolysis is controversial; it is not recommended in most guidelines
or not even mentioned. In some cases, maintenance tocolysis could be indicated to
allow the fetus to
achieve a higher gestational age [12]. The German S2k-guideline does not recommend long-term tocolysis as it does not
significantly reduce the preterm birth
rate or fetal morbidity and mortality. Long-term tocolysis may be useful in cases
with very early gestational age or prolapse of the amniotic sac. In such cases, a
risk-benefit assessment
should be done together with the expectant mother [5]. In our institution, long-term tocolysis was used during the study period if considered
necessary.
The aim of this study is to compare the efficacy of atosiban and hexoprenaline for
both short- and long-term tocolysis in women with preterm labor occurring between
24 + 0 and 34 + 0 weeks of
gestation. Secondary outcome parameters are maternal side effects and neonatal
short- and long-term neurological outcomes.
Methods
For this retrospective cohort study, all women with threatened preterm labor between
24 + 0 and 34 + 0 weeks of gestation who gave birth at the hospital of the Medical
University Graz between
2007 and 2017 were identified. Gestational age was determined according to the
first day of the last period or by crown–rump-length measurement in cases of uncertainty.
Inclusion and exclusion criteria
Women with threatened preterm labor defined by the presence of spontaneous and regular
contractions (≥ 4 contractions over 20 minutes) and cervical dilation from 0 to 3 cm
for primigravida,
and from 1 to 3 cm for multigravida, and a cervical length of less than 25 mm
were included in the study [5], [13].
Exclusion criteria were preterm premature rupture of membranes (PPROM, examined with
AmniSure [Qiagen GmbH, Germantown, USA]), signs of Triple I (intrauterine inflammation,
infection or
both), contraindications for the tocolytic agent, preeclampsia/HELLP syndrome,
maternal concomitant disease, multiple gestation, severe peripartum bleeding, major
fetal malformation or
intrauterine death before the start of tocolysis.
Tocolytics
The risk of preterm birth was examined using a fetal fibronectin test (PartoSure,
Parsagen Diagnostics GmbH, Boston, USA). After taking the womenʼs history, vaginal
screening for infection
was performed followed by obstetric ultrasound to assess the fetus and the cervix.
In cases with threatened preterm labor, tocolysis was given for 48 hours according
to the physicianʼs
preference, with tocolysis consisting either of atosiban (oxytocin receptor antagonist)
or hexoprenaline (beta-mimetic) to allow lung maturation with corticosteroids (two
doses of 12 mg
betamethasone) [14]. Primary contraindications for hexoprenaline were hypersensitivity against the substance,
liver or kidney insufficiency, general heart
disease or (pre)eclampsia and for atosiban were also hypersensitivity, uterine
bleeding, uterine infections and abnormal fetal heart rate [15]. Atosiban and
hexoprenaline were given as continuous intravenous infusion with an initial loading
dose followed by higher continuous doses (18 mg/hour for atosiban and 18 µg/hour for
hexoprenaline),
reduced after three hours to 6 mg/hour for atosiban and 4.3 µg/hour for hexoprenaline
over a period of 48 hours in total. If one tocolytic agent failed, a switch to the
other tocolytic agent
was possible. Failure was determined as persistence of subjective and objective
contractions and/or cervical opening. Long-term tocolysis was defined if a second
course of tocolysis was
given when contractions reappeared after 48 hours until 34 + 0 weeks of gestation
at the latest. Tocolytic efficacy was present when women did not deliver within 48
hours without switching
the tocolytic drug.
Maternal and neonatal outcome parameters
Maternal and neonatal outcome parameters were compared between the two tocolytic agents.
Maternal parameters included demographic data, mode of delivery and side effects of
the tocolytic
agents such as flushing, palpitations, nausea/vomiting or maternal tachycardia.
Neonatal parameters included arterial and venous umbilical cord pH, Apgar scores at
1, 5 and 10 minutes, birth
weight, hospital stay (days) in the neonatal intensive care unit (NICU), incidence
of infant respiratory distress syndrome (IRDS) of any grade, intraventricular hemorrhage
(IVH) of any
grade, periventricular hemorrhage (PVH), periventricular leukomalacia (PVL) of
any grade, retinopathy of prematurity (ROP) of any grade and, in addition, neonatal
mortality.
All neonates born before 34 + 0 weeks of gestation or with a birthweight of less than
2000 g were admitted routinely to the NICU which is located next to our delivery ward.
At the age of two years, infants with very low birth weight (< 1500 g) or early onset
impairment were routinely followed up with a cognitive and neurodevelopmental examination
and Bayley
II/III test at a corrected age of 2 years.
Statistical analysis
SPSS version 23 was used for statistical analysis. Quantitative data are expressed
as means and standard deviation, or median (minimum-maximum), depending on the skewness
and normality of
data distribution. Continuous variables were compared using t test or Mann–Whitney
U test, as appropriate. Categorical variables were compared between the two groups
using χ2 test
after Pearsonʼs and Fisherʼs exact test. Statistical significance was defined
as a p value of < 0.05.
Results
During the study period 761 women fulfilled the inclusion criteria and received tocolysis
for preterm labor with 374 women receiving hexoprenaline and 387 receiving atosiban.
There were no
significant differences between the two groups with regard to demographic parameters
except for gestational diabetes ([Table 1]).
Table 1 Maternal characteristics of women presenting with preterm labor at the time of hospital
admission (n = 761).
|
H (n = 374)
|
A (n = 387)
|
p value
|
|
Data are presented as numbers and percentages. H: hexoprenaline; A: atosiban
|
|
Age (years)
|
29.4 (15.8 – 41.7)
|
29.8 (15.8 – 45.9)
|
0.328
|
|
Body mass index (kg/m2)
|
22.7 ± 3.8
|
22.6 ± 4.6
|
0.694
|
|
Gravidity
|
2.2 (1 – 11)
|
2.1 (1 – 9)
|
0.475
|
|
Parity
|
1.7 (1 – 6)
|
1.6 (1 – 7)
|
0.419
|
|
Gestational age at the start of tocolysis (weeks)
|
30.3 (20.1 – 40.9)
|
29.9 (21.9 – 38.6)
|
0.084
|
|
Gestational diabetes
|
20 (5.3%)
|
33 (8.5%)
|
0.000
|
Short-term maternal outcome
[Table 2] shows the obstetric outcomes and maternal side effects of the two tocolytic agents.
No significant differences were observed for any obstetric
parameter between the groups. 15% of women in both groups were overdue after
successful tocolysis and had to be induced. Mean gestational age at delivery was 34 weeks
in the Hexoprenaline
group and 35 weeks in the Atosiban group. 55% and 60% of women in the Hexoprenaline
and Atosiban group, respectively, had a vaginal delivery; the remaining women had
an operative vaginal
delivery or cesarean section. In the Hexoprenaline group more women had a vaginal
delivery and primary cesarean section and fewer had a secondary cesarean section compared
to the Atosiban
group (p = 0.035). In terms of complications, hexoprenaline caused significantly
more palpitations (p = 0.000) and maternal tachycardia (p = 0.018). Other than this,
the rate of side effects
was low, with similar incidences for hypotension, flushing, nausea or vomiting
in both groups.
Table 2 Maternal outcomes associated with tocolytic therapy (n = 761).
|
H (n = 374)
|
A (n = 387)
|
p value
|
|
Data are presented as numbers and percentages. H: hexoprenaline; A: atosiban
|
|
Gestational age at delivery
|
34.5 (± 4.4)
|
35.2 (± 4.0)
|
0.047
|
|
Induction of labor
|
58 (15.5%)
|
59 (15.2%)
|
1.000
|
|
Mode of delivery
|
|
|
|
|
|
223 (59.6%)
|
214 (55.3%)
|
0.035
|
|
|
18 (4.8%)
|
24 (6.2%)
|
|
|
49 (13.1%)
|
44 (11.4%)
|
|
|
84 (22.5%)
|
105 (27.1%)
|
|
Palpitations
|
26 (7.0%)
|
7 (1.8%)
|
0.000
|
|
Hypotension
|
4 (1.1%)
|
1 (0.3%)
|
0.268
|
|
Flushing
|
1 (0.3%)
|
0 (0%)
|
0.553
|
|
Nausea and vomiting
|
14 (3.7%)
|
13 (3.4%)
|
0.144
|
|
Maternal tachycardia
|
13 (3.5%)
|
4 (1.0%)
|
0.018
|
|
Unspecific side effects
|
18 (4.8%)
|
6 (1.6%)
|
0.005
|
Efficacy of atosiban versus hexoprenaline and long-term tocolysis
Tocolytic efficacy is shown in [Table 3]. Atosiban showed significantly better efficacy in the first 48 hours of tocolysis
and the initial course of
tocolysis (p = 0.000) with 63% of women not giving birth in the Atosiban group
and 56% of women in the Hexoprenaline group across all gestational ages. A switch
to another tocolytic agent
was carried out in 15% of cases, with a further 114 women not being delivered
within 48 hours ([Table 2]).
Table 3 Tocolytic efficacy of atosiban and hexoprenaline in women presenting with preterm
labor between 24 and 34 weeks of gestation (n = 761).
|
|
|
p value
|
|
Data are presented as numbers and percentages. H: hexoprenaline; A: atosiban
|
|
Short-term tocolysis
|
H (n = 374)
|
A (n = 387)
|
|
|
Tocolytic efficacy
|
|
|
|
|
|
208 (55.6%)
|
245 (63.3%)
|
0.000
|
|
|
37 (9.9%)
|
25 (6.5%)
|
0.500
|
|
Long-term tocolysis
|
H (n = 203)
|
A (n = 258)
|
|
|
|
140 (69.0%)
|
180 (69.7%)
|
0.466
|
When calculating the significance based on the start of tocolysis according to gestational
age, [Table 4] shows that there were no significant differences in
efficacy between the tocolytic agents. However, it seems that hexoprenaline showed
a higher efficacy with regard to short-term tocolysis at later gestational ages while
atosiban was more
efficient when threatened preterm labor occurred at an earlier gestational age.
Table 4 Tocolytic efficacy of atosiban and hexoprenaline in women presenting with preterm
labor according to gestational age at the start of tocolysis.
|
Tocolytic efficacy depending on GA at start of tocolysis
|
Gestational age (weeks + days)
|
Hexoprenaline
|
Atosiban
|
p value
|
|
Data are presented as numbers and percentages. H: hexoprenaline; A: atosiban
|
|
No failure at 48 h
|
24 + 0 – 26 + 6
|
29 (56.9%)
|
44 (72.1%)
|
0.112
|
|
27 + 0 – 28 + 6
|
27 (56.3%)
|
46 (66.7%)
|
0.332
|
|
29 + 0 – 31 + 6
|
77 (64.7%)
|
83 (62.4%)
|
0.793
|
|
32 + 0 – 34 + 6
|
66 (55.5%)
|
64 (59.8%)
|
0.590
|
|
No failure at 7 days
|
24 + 0 – 26 + 6
|
29 (56.8%)
|
44 (72.1%)
|
0.368
|
|
27 + 0 – 28 + 6
|
27 (56.2%)
|
46 (66.6%)
|
0.295
|
|
29 + 0 – 31 + 6
|
77 (64.7%)
|
83 (62.4%)
|
0.342
|
|
32 + 0 – 34 + 6
|
66 (55.5%)
|
64 (60.3%)
|
0.113
|
66% (n = 258/387) of women in the Atosiban group and 54% (n = 203/374) in the Hexoprenaline
group received long-term tocolysis of similar efficacy up to 34 + 0 weeks of gestation
(p = 0.466).
Short-term neonatal outcomes
Neonatal outcome parameters are presented in [Table 5]. Birthweight was significantly lower in the Hexoprenaline group (p = 0.001), although
these neonates
were born slightly earlier than the neonates in the Atosiban group. No differences
between the groups were observed for any other neonatal parameter or for neonatal
mortality or
morbidity.
Table 5 Short-term neonatal outcomes according to the maternal tocolytic therapy for preterm
labor.
|
H (n = 374)
|
A (n = 387)
|
p value
|
|
Data are presented as numbers and percentages. H: hexoprenaline; A: atosiban; IRDS:
infant respiratory distress syndrome (of any grade); IVH: intraventricular hemorrhage
(of any
grade); PVH: periventricular hemorrhage (of any grade); PVL: periventricular
leukomalacia; ROP: retinopathy of prematurity (of any grade)
|
|
Weight (g)
|
2361(± 852)
|
2487 (± 827)
|
0.001
|
|
Length (cm)
|
47.0 (± 4.0)
|
47.7 (± 4.0)
|
0.078
|
|
Head circumference (cm)
|
33.1 (± 2.4)
|
33.2 (± 2.6)
|
0.217
|
|
Length percentile
|
36.6 ± 26.4
|
36.8 ± 23.3
|
0.975
|
|
Weight percentile
|
44.5 ± 24.9
|
43.3 ± 23.2
|
0.757
|
|
Head circumference percentile
|
43.8 ± 26.0
|
40.8 ± 25.5
|
0.116
|
|
Apgar 1 score
|
7.8 ± 1.8
|
8.2 ± 1.5
|
0.028
|
|
Apgar 5 score
|
9.2 ± 1.3
|
9.4 ± 1.1
|
0.157
|
|
Apgar 10 score
|
9.5 ± 1.0
|
9.6 ± 1.0
|
0.241
|
|
Umbilical artery pH
|
7.28 ± 0.09
|
7.29 ± 0.07
|
0.076
|
|
Umbilical vein pH
|
7.35 ± 0.07
|
7.36 ± 0.07
|
0.037
|
|
NICU admission (days)
|
13 (8.2)
|
12 (7.3)
|
0.201
|
|
Mortality
|
0
|
0
|
not applicable
|
|
IRDS
|
82 (45.3%)
|
84 (47.7%)
|
0.241
|
|
IVH
|
18 (9.7%)
|
13 (7.0%)
|
0.805
|
|
PVH
|
2 (1.1%)
|
1 (0.5%)
|
0.795
|
|
PVL
|
9 (4.9%)
|
13 (7.0%)
|
0.541
|
|
ROP
|
89 (23.8%)
|
125 (32.3%)
|
0.651
|
|
Abnormal neurological behavior
|
20 (12.3%)
|
12 (7.0%)
|
0.094
|
Long-term neonatal outcomes
At 2-years follow-up, preterm-born infants were assessed for long-term neurodevelopmental
outcomes. Data were available for 104 infants. Half of the infants from the Hexoprenaline
group
(49%, n = 30/61) and the Atosiban group (51%, 22/43) showed no neurodevelopmental
impairment. Severe impairment was recorded in 26% in both groups. Neurodevelopmental
assessment by Bayley
test was available for 61 infants. 50% of infants in the Hexoprenaline group
and 65% of infants in the Atosiban group did not show any impairment; only 5% and
9% of infants, respectively,
showed severe impairment ([Table 6]).
Table 6 Long-term neurodevelopmental outcomes of infants according to maternal tocolytic
therapy for preterm labor.
|
|
|
p value
|
|
Data are presented as numbers and percentages. H: hexoprenaline; A: atosiban
|
|
Clinical investigation/neuromotor impairment
|
H (n = 61)
|
A (n = 43)
|
|
|
|
30 (49%)
|
22 (51%)
|
1.000
|
|
|
15 (25%)
|
10 (23%)
|
|
|
16 (26%)
|
11 (26%)
|
|
Bayley test
|
H (n = 38)
|
A (n = 23)
|
|
|
|
19 (50%)
|
15 (65%)
|
0.408
|
|
|
17 (45%)
|
6 (26%)
|
|
|
2 (5%)
|
2 (9%)
|
Discussion
The present retrospective cohort study compares the tocolytic efficacy of atosiban
and hexoprenaline as tocolytic agents in pregnancies with threatened preterm labor
and found that atosiban
showed a better efficacy in the first 48 hours of tocolysis while there was no
significant difference in efficacy with regard to maintenance tocolysis up to 34 + 0
weeks of gestation.
These findings are contrary to those of a Cochrane review by Flenady et al., where
no benefits were reported for atosiban compared to beta-mimetics with regard to the
prolongation of
pregnancy or neonatal outcomes [16].
The significance of atosibanʼs efficacy disappeared when efficacy was calculated according
to gestational age at the time when tocolysis was initiated. It seems that atosiban
is especially
beneficial in early pregnancy but this may not be the case at later gestational
ages. There is evidence that the number of oxytocin receptors in the myometrium increases
at later gestational
ages but also in cases with preterm labor [17], which could be a possible explanation for our findings.
41% of all tocolysis in Austria is maintenance tocolysis administered for longer than
48 hours [18]. The reason for the discrepancy compared to other countries
could be a lack of knowledge about the harms and uses of tocolysis, especially
in terms of long-term outcomes. At our institution, long-term tocolysis is prescribed
on a case-by-case basis
with the decision made by the attending obstetrician together with the expectant
mother. Prematurity is associated with higher morbidity and mortality rates; therefore,
it seems obvious that
prolongation of the pregnancy should improve the outcome. However, there are only
a few studies which demonstrate that maintenance tocolysis improves maternal and neonatal
(long-term) outcomes
and does not merely prolong pregnancy [12], [18], [19]. Short-term tocolysis should be given for
48 h to allow the administration of antenatal corticosteroids to improve the neonatal
outcome [20], and it is often recommended that maintenance tocolysis
should be avoided [8], [21], [22], [23]. In our series receiving
maintenance tocolysis, maintenance tocolysis appeared to successfully prevent
preterm labor in many patients without incurring any severe maternal or neonatal side
effects. Based on our
series, however, due to the retrospective analysis it cannot be definitively stated
whether long-term tocolysis was responsible for a delay in delivery in women with
ongoing contractions after
48 hours.
It is well-known that hexoprenaline causes more side effects such as palpitations,
maternal tachycardia or chest pain in comparison to atosiban, and our study confirms
these observations
[8], [11], [16], [24]. However, both tocolytic agents seem to
be quite well tolerated. Nevertheless, atosiban is recommended as the first choice
tocolytic agent, especially in women with multiple gestations or maternal disease
[9]. According to Berger et al.ʼs review of the literature, the prophylactic administration
of aspirin as a tocolytic agent should also be further investigated, as it both reduces
the
risk of preeclampsia and fetal growth restriction and leads to a reduction of
preterm labor [25].
Neonatal morbidity and mortality rates were similar in both groups. Birth weight of
the neonates in the Atosiban group was slightly higher but these neonates were also
born at slightly later
gestational ages. This is in line with a multicenter randomized controlled trial
in Belgium and the Netherlands (APOSTEL III) that compared neonatal outcomes after
using atosiban or nifedipine
for tocolysis [26]. Van Vliet et al. highlighted the improvement in primary neonatal outcomes as the
main goal of tocolysis, not merely prolongation of the
pregnancy. Our results are similar to the findings of the APOSTEL III trial with
regard to neonatal outcomes and maternal side effects.
A cohort study by Pinto Cardoso et al. showed a reduced risk of neonatal death as
well as a reduced risk of severe intraventricular hemorrhage after using tocolysis
in general, that was
similar to our findings. Atosiban may affect neuroprotection and prepares the
fetus for labor [27] while corticosteroids stabilize arterial blood pressure and
have an anti-inflammatory effect [28], [29]. Both Gilard et al. and Di Renzo et al. discussed the occurrence of IVH and white
matter lesions in preterm infants; they believe that the use of steroids and magnesium
sulfate and not the tocolytic drug itself are responsible for decreased numbers of
cases with IVH and
other complications.
Prematurity is the most common cause of increased morbidity and mortality in infancy.
Follow-ups can identify risk factors for cognitive, neurological and behavioral disorders
and establish
new treatment options [30]. Sixty-one children in the Hexoprenaline group and 43 children in the Atosiban group
were available for follow-up. The number of
cases in our follow-up cohorts are small because only infants with a birth weight
of up to 1500 g were invited for assessment of their 2-year neurodevelopmental outcome.
25% of these children
showed severely impaired neuromotor development on clinical examination. In two
thirds (hexoprenaline) and half of the infants (atosiban), respectively, a Bayley
test was performed to identify
any impairment more precisely. At this follow-up, 50% of the infants in the Hexoprenaline
group and 65% in the Atosiban group showed no impairment, while 5% and 9%, respectively,
showed severe
impairment. These findings are in accordance with the current literature. The
APOSTEL III trial was followed in 2020 by the first published follow-up trial that
compared neonatal outcomes
after tocolysis with nifedipine or atosiban. 46% of the children of the first
trial participated in the follow-up study, which was comparable with our follow-up
data. Abnormal development was
found in 38% in the Atosiban group but there were no differences with regard to
neurodevelopment, executive function, behavior or health [31].
The strength of our study is its large sample size and the absence of differences
with regard to maternal characteristics. Since the tocolytic agents used at our institution
are not
necessarily available in other industrialized countries, only very little data
is available which specifically compares these two drugs. In addition, long-term follow-up
data of low
birthweight infants was available in our study.
Our study has some major limitations. The first is its retrospective design without
randomization to treatment groups. Furthermore, long-term follow-up is not available
for all infants,
possibly leading to data bias.
Conclusion
In conclusion, there is a significant difference in efficacy between the two tocolytic
agents, with atosiban showing a higher efficacy during the first course of administration,
especially
when administered at an early gestational age, although this effect was not seen
for long-term administration. There was no significant difference between the two
tocolytic agents in regard to
maternal or neonatal outcomes; however, hexoprenaline is associated with a higher
rate of known maternal side effects.
