Spina bifida is the term used to describe a group of neural tube conditions where the fetal spinal cord does not close properly during the first month of pregnancy. With open spina bifida some of the vertebrae are not completely formed but are split or divided and the spinal cord and its coverings (the meninges) protrude through the opening. The most severe is where the spinal cord and meninges come out of the child's back (myelomeningocele). Open spina bifida is often associated with hindbrain herniation, where the cerebellum and brainstem tissue extend into the large opening in the base of the skull, and hydrocephalus (enlargement of the fluid filled cavities in the brain). Resulting disabilities include bladder and bowel incontinence, difficulties in moving about due to limb weakness, paralysis, deformity and loss of sensation. Conventional treatment of spina bifida is surgical repair within two days of birth, which may include the placement of a shunt between the ventricles of the baby’s brain and the belly (peritoneum) to relieve hydrocephalus. Spina bifida can be diagnosed with prenatal ultrasound or maternal serum alpha-feto protein and in utero treatment could improve outcomes; although it involves surgical incision into the mother’s abdomen and uterus to access the unborn baby.
This review aimed to compare the effects of in utero repair versus repair as a newborn. We included one randomised controlled trial involving 158 women who were from 19 to 27 weeks pregnant with a baby with severe spina bifida and evidence of hindbrain herniation. For neonatal mortality, there was no clear difference identified for prenatal versus postnatal repair. However, the numbers of neonates who died were low and so the review was likely underpowered to detect any difference. Prenatal repair was associated with reduced need for shunt placement and a reduction in the risk of moderate to severe hindbrain herniation after birth. No direct complications of the repair procedure were evident, including orthopaedic deformities. Prenatal repair was associated with an increased risk of the women experiencing preterm ruptured membranes and subsequent preterm birth (both before 34 and 37 weeks). Severe maternal illness (infection and need for blood transfusion) were not clearly different; although the review was underpowered to detect any difference in these important, less common outcomes. The included trial was of high quality (low risk of bias) but included a small number of pregnancies. There is currently insufficient evidence to recommend in utero repair for unborn babies with spina bifida.
This review is based one small well-conducted study. There is insufficient evidence to recommend drawing firm conclusions on the benefits or harms of prenatal repair as an intervention for fetuses with spina bifida. Current evidence is limited by the small number of pregnancies that have been included in the single conducted randomised trial to date.
Spina bifida is a fetal neural tube defect (NTD), which may be diagnosed in utero and is compatible with life postnatally, albeit often with significant disability and morbidity. Although postnatal repair is possible, with increasing in utero diagnosis with ultrasound, the condition has been treated during pregnancy (prenatal repair) with the aim of decreased morbidity for the child. The procedure that is performed during pregnancy does have potential morbidities for the mother, as it involves maternal surgery to access the fetus.
To compare the effects of prenatal versus postnatal repair and different types of repair of spina bifida on perinatal mortality and morbidity, longer term infant outcomes and maternal morbidity.
We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (31 July 2014).
All published, unpublished, and ongoing randomised controlled trials comparing prenatal and postnatal repair of meningomyelocele for fetuses with spina bifida and different types of prenatal repair.
Two review authors independently evaluated trials for inclusion and methodological quality without consideration of their results according to the stated eligibility criteria and extracted data.
Our search strategy identified six reports for potential inclusion. Of those, we included one trial (four reports) involving 158 women, which was at low risk of bias.
The one included trial examined the effect of prenatal repair versus postnatal repair. For the primary infant outcome of neonatal mortality, there was no clear evidence of a difference identified for prenatal versus postnatal repair (one study, 158 infants, risk ratio (RR) 0.51, 95% confidence interval (CI) 0.05 to 5.54), however event rates were uncommon and so the analysis is likely to be underpowered to detect differences.
Prenatal repair was associated with an earlier gestational age at birth (one study, 158 infants, mean difference (MD) -3.20 weeks, 95% CI -3.93 to -2.47) and a corresponding increase in both the risk of preterm birth before 37 weeks (one study, 158 infants, RR 5.30, 95% CI 3.11 to 9.04) and preterm birth before 34 weeks (one study, 158 infants, RR 9.23, 95% CI 3.45 to 24.71). Prenatal repair was associated with a reduction in shunt dependent hydrocephalus and moderate to severe hindbrain herniation. For women, prenatal repair was associated with increased preterm ruptured membranes (one study, 158 women, RR 6.15, 95% CI 2.75 to 13.78), although there was no clear evidence of difference in the risk of chorioamnionitis or blood transfusion, although again, event rates were uncommon.
A number of this review's secondary infant and maternal outcomes were not reported. For the infant: days of hospital admission; survival to discharge; stillbirth; need for further surgery (e.g. skin grafting); neurogenic bladder dysfunction; childhood/infant quality of life. For the mother: admission to intensive care; women's emotional wellbeing and satisfaction with care.