Tocolytics for delaying preterm birth: a network meta-analysis (0924)

Background: Preterm birth is the leading cause of death in newborns and children. Tocolytic drugs aim to delay preterm birth by suppressing uterine contractions to allow time for administration of corticosteroids for fetal lung maturation, magnesium sulphate for neuroprotection, and transport to a facility with appropriate neonatal care facilities. However, there is still uncertainty about their effectiveness and safety.

Objectives: To estimate relative effectiveness and safety profiles for different classes of tocolytic drugs for delaying preterm birth, and provide rankings of the available drugs.

Search methods: We searched Cochrane Pregnancy and Childbirth's Trials Register, ClinicalTrials.gov (21 April 2021) and reference lists of retrieved studies.

Selection criteria: We included all randomised controlled trials assessing effectiveness or adverse effects of tocolytic drugs for delaying preterm birth. We excluded quasi- and non-randomised trials. We evaluated all studies against predefined criteria to judge their trustworthiness.

Data collection and analysis: At least two review authors independently assessed the trials for inclusion and risk of bias, and extracted data. We performed pairwise and network meta-analyses, to determine the relative effects and rankings of all available tocolytics. We used GRADE to rate the certainty of the network meta-analysis effect estimates for each tocolytic versus placebo or no treatment.

Main results: This network meta-analysis includes 122 trials (13,697 women) involving six tocolytic classes, combinations of tocolytics, and placebo or no treatment. Most trials included women with threatened preterm birth, singleton pregnancy, from 24 to 34 weeks of gestation. We judged 25 (20%) studies to be at low risk of bias. Overall, certainty in the evidence varied. Relative effects from network meta-analysis suggested that all tocolytics are probably effective in delaying preterm birth compared with placebo or no tocolytic treatment. Betamimetics are possibly effective in delaying preterm birth by 48 hours (risk ratio (RR) 1.12, 95% confidence interval (CI) 1.05 to 1.20; low-certainty evidence), and 7 days (RR 1.14, 95% CI 1.03 to 1.25; low-certainty evidence). COX inhibitors are possibly effective in delaying preterm birth by 48 hours (RR 1.11, 95% CI 1.01 to 1.23; low-certainty evidence). Calcium channel blockers are possibly effective in delaying preterm birth by 48 hours (RR 1.16, 95% CI 1.07 to 1.24; low-certainty evidence), probably effective in delaying preterm birth by 7 days (RR 1.15, 95% CI 1.04 to 1.27; moderate-certainty evidence), and prolong pregnancy by 5 days (0.1 more to 9.2 more; high-certainty evidence). Magnesium sulphate is probably effective in delaying preterm birth by 48 hours (RR 1.12, 95% CI 1.02 to 1.23; moderate-certainty evidence). Oxytocin receptor antagonists are probably effective in delaying preterm birth by 48 hours (RR 1.13, 95% CI 1.05 to 1.22; moderate-certainty evidence), are effective in delaying preterm birth by 7 days (RR 1.18, 95% CI 1.07 to 1.30; high-certainty evidence), and possibly prolong pregnancy by 10 days (95% CI 2.3 more to 16.7 more). Nitric oxide donors are probably effective in delaying preterm birth by 48 hours (RR 1.17, 95% CI 1.05 to 1.31; moderate-certainty evidence), and 7 days (RR 1.18, 95% CI 1.02 to 1.37; moderate-certainty evidence). Combinations of tocolytics are probably effective in delaying preterm birth by 48 hours (RR 1.17, 95% CI 1.07 to 1.27; moderate-certainty evidence), and 7 days (RR 1.19, 95% CI 1.05 to 1.34; moderate-certainty evidence). Nitric oxide donors ranked highest for delaying preterm birth by 48 hours and 7 days, and delay in birth (continuous outcome), followed by calcium channel blockers, oxytocin receptor antagonists and combinations of tocolytics. Betamimetics (RR 14.4, 95% CI 6.11 to 34.1; moderate-certainty evidence), calcium channel blockers (RR 2.96, 95% CI 1.23 to 7.11; moderate-certainty evidence), magnesium sulphate (RR 3.90, 95% CI 1.09 to 13.93; moderate-certainty evidence) and combinations of tocolytics (RR 6.87, 95% CI 2.08 to 22.7; low-certainty evidence) are probably more likely to result in cessation of treatment. Calcium channel blockers possibly reduce the risk of neurodevelopmental morbidity (RR 0.51, 95% CI 0.30 to 0.85; low-certainty evidence), and respiratory morbidity (RR 0.68, 95% CI 0.53 to 0.88; low-certainty evidence), and result in fewer neonates with birthweight less than 2000 g (RR 0.49, 95% CI 0.28 to 0.87; low-certainty evidence). Nitric oxide donors possibly result in neonates with higher birthweight (mean difference (MD) 425.53 g more, 95% CI 224.32 more to 626.74 more; low-certainty evidence), fewer neonates with birthweight less than 2500 g (RR 0.40, 95% CI 0.24 to 0.69; low-certainty evidence), and more advanced gestational age (MD 1.35 weeks more, 95% CI 0.37 more to 2.32 more; low-certainty evidence). Combinations of tocolytics possibly result in fewer neonates with birthweight less than 2500 g (RR 0.74, 95% CI 0.59 to 0.93; low-certainty evidence). In terms of maternal adverse effects, betamimetics probably cause dyspnoea (RR 12.09, 95% CI 4.66 to 31.39; moderate-certainty evidence), palpitations (RR 7.39, 95% CI 3.83 to 14.24; moderate-certainty evidence), vomiting (RR 1.91, 95% CI 1.25 to 2.91; moderate-certainty evidence), possibly headache (RR 1.91, 95% CI 1.07 to 3.42; low-certainty evidence) and tachycardia (RR 3.01, 95% CI 1.17 to 7.71; low-certainty evidence) compared with placebo or no treatment. COX inhibitors possibly cause vomiting (RR 2.54, 95% CI 1.18 to 5.48; low-certainty evidence). Calcium channel blockers (RR 2.59, 95% CI 1.39 to 4.83; low-certainty evidence), and nitric oxide donors probably cause headache (RR 4.20, 95% CI 2.13 to 8.25; moderate-certainty evidence).

Authors' conclusions: Compared with placebo or no tocolytic treatment, all tocolytic drug classes that we assessed (betamimetics, calcium channel blockers, magnesium sulphate, oxytocin receptor antagonists, nitric oxide donors) and their combinations were probably or possibly effective in delaying preterm birth for 48 hours, and 7 days. Tocolytic drugs were associated with a range of adverse effects (from minor to potentially severe) compared with placebo or no tocolytic treatment, although betamimetics and combination tocolytics were more likely to result in cessation of treatment. The effects of tocolytic use on neonatal outcomes such as neonatal and perinatal mortality, and on safety outcomes such as maternal and neonatal infection were uncertain.