Fetal surgery for myelomeningocele?

The possibility of surgical repair of fetal anomalies in utero has long tantalized obstetricians and pediatric surgeons, especially since the development of high-resolution real-time ultrasonography. The presumption has been that earlier in utero repair would provide superior outcomes for the offspring than would postnatal surgery. However, postnatal treatment is much simpler and is not traumatic to the mother. Successful fetal surgery for any condition thus requires an accurate identification of which fetuses are at highest risk for a poor outcome if repair is delayed until after birth. Despite the promise of fetal surgical procedures, such repair of structural malformations has not been shown to be better than postnatal repair at improving the outcome in randomized, controlled trials (e.g., fetal tracheal occlusion for congenital diaphragmatic hernia1). Myelomeningocele is not lethal to a fetus, but postnatal repair is associated with a less-thandesirable long-term neurologic outcome. Early studies of open fetal surgery to repair myelomeningocele, which involved relatively small, nonrandomized cohorts and the use of a relatively large hysterotomy to access the fetus, reported serious maternal and fetal complications.2-4 Thus, questions remained regarding the overall risk–benefit ratio of the procedure. In this issue of the Journal, Adzick et al.5 report the results of the Management of Myelomeningocele Study (MOMS), which was designed to assess outcomes of prenatal surgery, as compared with postnatal repair. Women in this trial had a singleton fetus of 19.0 to 25.9 weeks of gestation with myelomeningocele located between T1 and S1 and evidence of hindbrain herniation; all fetuses had a normal karyotype. Of 1233 women who were initially screened, 183 (15%) were randomly assigned to undergo either prenatal surgery through the use of hysterotomy and open fetal repair or repair after delivery. In the intention-to-treat analyses, prenatal surgery resulted in a significant reduction in the composite primary end point of infant death or the need for placement of a cerebrospinal-fluid shunt at 1 year, a finding entirely driven by the substantially lower frequency of shunt placement in the prenatal-surgery group (40% vs. 82%). Prenatal surgery, as compared with postnatal surgery, also resulted in significant improvement in a composite score based on the Bayley Mental Development Index and assessment of motor development at the age of 30 months (a coprimary outcome), although the earlier surgery had no significant effect on mental development alone. There was less hindbrain herniation in the prenatal-surgery group than in the postnatal-surgery group (with no hindbrain herniation in 36% and 4% of the infants, respectively, and severe herniation in 6% and 22%, respectively). The percentages of children who were able to walk independently at the age of 3 years were 42% in the prenatal-surgery group and 21% in the postnatal-surgery group. However, the prenatal-surgery group had higher rates of maternal and certain fetal complications: spontaneous membrane rupture (46% in the prenatal-surgery group vs. 8% in the postnatal-surgery group), oligohydramnios (21% vs. 4%), preterm birth (79% vs. 15%), and more complications associated with prematurity, such as the respiratory distress syndrome. More than a third of mothers in the prenatal-surgery group