Subscribe to eTOC

Prenatal Surgery to Correct Spina Bifida Found to Improves Fetal Outcomes


Prenatal surgery to correct myelomeningocele leads to better outcomes for fetuses than waiting to perform surgery after birth. That's the clear message from the findings of the first randomized trial comparing the two options, published Feb. 9 in the online edition of The New England Journal of Medicine. But there are trade-offs, the study showed, since mothers electing prenatal surgery had an increased risk of premature delivery and uterine scarring.

“We have data to suggest there is a new standard of care for parents who want to consider this,” said the study's lead author N. Scott Adzick, MD, surgeon-in-chief at Children's Hospital of Philadelphia and professor of pediatrics, obstetrics and gynecology at the University of Pennsylvania School of Medicine.

Myelomeningocele is the most common form of spina bifida, in which the neural tube fails to close during development, and the spinal cord extrudes into a sac filled with CSF. Folic acid supplementation before and early in pregnancy can reduce such neural tube defects, and the introduction of folic acid into the food supply has led to a steady decline in the incidence of myelomeningocele.

“That [decline in incidence] has been a major public health triumph,” said Michael F. Greene, MD, chief of obstetrics at Massachusetts General Hospital in Boston, who co-authored an editorial accompanying the study.

The number of babies born with myelomeningocele has also declined due to improvements in prenatal diagnostic ultrasound, with perhaps 20- to 40-percent of mothers choosing abortion. The incidence of myelomeningocele is now approximately 4 per 10,000 live births.

“Repair of these defects after delivery is fair, but it is still a huge source of morbidity,” said Dr. Greene, who was not involved in the surgery trial. “In recent decades, about 90 percent of infants survive, but with long-term neural handicap.” Symptoms and disability depend on the level of failure to close, and usually include impaired ambulation, incontinence, and cognitive impairments, as well as hydrocephalus, which requires lifelong placement of a CSF shunt.

Prenatal repair has been performed experimentally since 1997, and by 2003, three centers were performing the surgery, and had operated on more than 200 fetuses. Initial results were promising, but “we had equipoise,” Dr. Adzick said, and the best way to move forward was to perform a randomized trial.

DR. N. SCOTT ADZICK: “We have data to suggest there is a new standard of care for parents who want to consider this.”

Surgeons at the three centers — the Children's Hospital of Philadelphia, as well as the University of California-San Francisco and Vanderbilt University Medical Center in Nashville, TN — sought and received agreement from other centers contemplating development of a surgical program to wait until the results of the trial were in before proceeding, in order to funnel eligible patients to the study centers.

“We owe a debt to the fetal surgery community, who agreed not to do this before determining whether it was worth doing,” Dr. Adzick said.


Women referred to the centers who were interested in the trial were randomized to prenatal or postnatal surgery if they met the entrance criteria, including carrying a single fetus with gestational age between 19 and 26 weeks, with no other fetal anomalies. Because operating on obese patients is difficult in the best of circumstances, mothers with a body mass index of 35 or more were excluded.

There were two primary outcome variables. The first was a composite of fetal or neonatal death or the need for a shunt, evaluated at 12 months after delivery. The second was a composite score of motor and mental function, adjusted for lesion level. “As a rule of thumb, the anatomic level of the lesion determines what sort of function there is, particularly in the legs,” Dr. Adzick said.

The trial was designed to enroll 200 patients, but interim data analysis indicated prenatal surgery was superior enough to justify stopping the trial after 183 women were randomized. Results for 158 patients who were evaluated at 12 months were reported in the New England Journal of Medicine.

There were no maternal deaths, and two deaths of fetuses or newborns in each group. Prenatal surgery was associated with prematurity, with an average gestational age of 34.1 weeks, versus 37.3 weeks in the postnatal group. One-fifth of those in the prenatal group experienced a respiratory distress syndrome, probably caused by prematurity, Dr. Adzick said. Mothers undergoing prenatal surgery were more likely to experience a variety of adverse effects, including spontaneous membrane rupture and a thinned or dehiscent uterine wall at the surgery site.

The first primary outcome of death or need for a shunt was met in 68 percent of the prenatal group, and 98 percent of the postnatal group (p<0.001). Rates of actual shunt placement were 40 percent in the prenatal-surgery group and 82 percent in the postnatal-surgery group (p<0.001).Newborns in the prenatal group also had significantly lower rates of hindbrain herniation, brainstem kinking, and abnormal location of the fourth ventricle, all indicative of a less disturbed spinal cord.

There were 134 patients who had been followed out to 30 months, and were evaluated for the second primary outcome. There was no difference between the groups on the mental development component, with an average score of about 88 in each (100 is normal). Infants in the prenatal surgery group scored significantly better on motor function, with 42 percent walking independently, versus 21 percent in the postnatal group (p=0.01). Other related measures, including developmental motor scales and self-care ability, also favored prenatal surgery.

“The lesson is that the prenatal group had, on average, much better motor function than the post-natal group,” Dr. Adzick said. “It's very compelling.”

Dr. Greene said the message of the trial is, “If you gamble, you can win, modestly. It's not a simple message.” The outcomes for the fetus, he said, were “clearly better,” although he felt the need for a shunt, while admittedly a burden, “is not necessarily, in and of itself, a terribly important endpoint.” Unfortunately, he noted, “no matter what kind of surgery you provide, under the best of circumstances these children do not have normal mental development.”

The key question is whether it is worth the morbidity for the mother. “Operating on the uterus like that has a cost, and a risk,” Dr. Greene said. “This is not a slam-dunk, easy decision for a family faced with this problem. There are no easy choices.”

Dr. Adzick said that future work will be aimed at understanding the range of outcomes in the prenatal group, noting that some in the group appeared not to benefit. “It's going to be really important to mine the data for predictive factors,” such as massive ventriculomegaly, hydrocephalus, or club feet before birth, he said. “We'd like to be able to accurately counsel parents, because we don't want to do a big operation before birth if there is likely to be little or no benefit.”

Now that the trial is over, and other centers will be gearing up to perform the surgery, what advice does Dr. Adzick offer? “It is important to stress that this operation is incredibly difficult to do, and requires a committed, multidisciplinary team, including world-class maternal fetal ultrasound, world-class ultrafast fetal MRI,” a fetal anesthesiologist, a pediatric cardiologist, and a highly trained nursing staff,” he said. “You need the commitment of a team from start to finish.”


Adzick NS, Thom EA, Farmer DL, et al, for the MOMS Investigators. A randomized trial of prenatal versus postnatal repair of myelomeningocele. N Engl J Med 2011; E-pub 2011 Feb. 9.
    Simpson JL, Greene MF. Editorial: Fetal surgery for myelomeningocele? N Engl J Med 2011; E-pub 2011 Feb. 9.