Rosalie M Grivell, Chad Andersen, Jodie M Dodd

Congenital diaphragmatic hernia (CDH), is an uncommon but severe condition in which there is a developmental defect in the fetal diaphragm, resulting in liver and bowel migrating to the chest cavity and impairing lung development and function for the neonate. This condition can be diagnosed during pregnancy and as such, is potentially amenable to in-utero prenatal intervention. Neonatal surgical repair is possible, but even with early surgical repair and improving neonatal management, neonatal morbidity and mortality is high. Prenatal interventions described to date have included maternal antenatal corticosteroid administration and fetal tracheal occlusion, with both methods aiming to improve lung growth and maturity. However surgical procedures have potential maternal complications, as the uterus and amniotic sac are breached in order to gain access to the fetus. To compare the effects of prenatal versus postnatal interventions for CDH on perinatal mortality and morbidity, longer-term infant outcomes and maternal morbidity, and to compare the effects of different prenatal interventions with each other. We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (31 August 2015) and reference lists of retrieved studies. All published (including those published in abstract form), unpublished, and ongoing randomised controlled trials comparing prenatal and postnatal interventions for fetuses with CDH. Quasi-RCTs were eligible for inclusion but none were identified. Trials using a cross-over design are not eligible for inclusion. Two review authors evaluated trials for inclusion and methodological quality without consideration of their results according to the stated eligibility criteria and extracted data independently. Data were checked for accuracy. We identified 11 studies for potential inclusion. Of those, we included three studies involving 97 women. Two additional studies are ongoing.Two trials examined in-utero fetal tracheal occlusion with standard (postnatal) care in fetuses with severe diaphragmatic hernia. Whilst the trials utilised fetal interventions that were similar, there were important differences in how access was gained to the fetus and in the timing and mode of delivery. Therefore, we did not combine these trials in meta-analysis and the results are examined in separate comparisons. One trial examined the effect of antenatal corticosteroids versus placebo. Overall, the methodological quality of the trials was variable and no data were available for a number of this review's secondary outcomes. In-utero fetal occlusion by maternal laparotomy versus standard postnatal management (one trial, 24 women)For the primary infant outcome (perinatal mortality), there were no data suitable for inclusion in the analysis. There was no difference between groups in terms of long-term infant survival (risk ratio (RR) 1.06, 95% confidence interval (CI) 0.66 to 1.69). In-utero fetal occlusion by minimally invasive fetoscopy versus standard postnatal management (one trial, 41 women)The primary infant outcome (perinatal mortality) was not reported. Minimally invasive fetoscopy was associated with a small reduction in the mean gestational age at birth (mean difference (MD) -1.80 weeks, 95% CI -3.13 to -0.47), but there was no clear difference in the risk of preterm birth before 37 weeks (RR 1.75, 95% CI 0.78 to 3.92). Long-term infant survival (three to six months) (RR 10.50, 95% CI 1.48 to 74.71) was increased with the intervention when compared with standard management, and there was a corresponding reduction in pulmonary hypertension (RR 0.58, 95% CI 0.36 to 0.93) associated with the intervention. There was no difference between groups in terms of preterm ruptured membranes (< 37 weeks) (RR 1.47, 95% CI 0.56 to 3.88) or maternal infectious morbidity (RR 3.14, 95% CI 0.14 to 72.92), and there were no maternal blood transfusions. Antenatal corticosteroids versus placebo (one trial, 32 women)We also included one trial (involving 32 women) examining the effect of antenatal corticosteroids versus placebo. There was no clear difference in the incidence of perinatal mortality (our primary infant outcome) between the group of women who received antenatal corticosteroids and the placebo control (RR 1.24, 95% CI 0.50 to 3.08). Data (mean only) were reported for two of our secondary outcomes (mechanical ventilation and days of hospital admission) but standard deviations (SDs) were not provided. For the purposes of this review and to permit further analysis we have estimated the SDs based on the reported P values reported in the trial report, although our estimation does assume that the SD is the same in both the intervention and control groups. There were no differences between the antenatal corticosteroid group and the placebo control in terms of days of mechanical ventilation (MD 18.00 days, 95% CI -14.77 to 50.77) or days of hospital admission (MD 17.00 days, 95% CI -13.93 to 47.93) . There is currently insufficient evidence to recommend in-utero intervention for fetuses with CDH as a part of routine clinical practice. We identified three small studies, with only one study adequately reporting on the primary outcome of this review - perinatal mortality, and there were few data pertaining to many of this review's secondary outcomes.WIth regard to the administration of antenatal corticosteroids, there remains a gap in current research, and a large multicentre trial with adequate statistical power should be undertaken to answer this unresolved question. More studies are needed to further examine the effect of in-utero fetal tracheal occlusion on important neonatal outcomes and long-term infant survival and health. Long-term follow-up is of particular importance, and should include morbidity and mortality measures. Further studies should examine the benefits of an in-utero intervention on subgroups with moderate and severe congenital diaphragmatic hernia. Indeed, there are three ongoing studies, being conducted by European, North and South American fetal medicine centres, which will contribute to this gap. Ongoing research and any implementation into clinical practice should include standardisation of the procedure, inclusion criteria and long-term childhood follow-up.