Maternal–fetal surgery for fetal myelomeningocele has proven beneficial in reducing long-term morbidity in the randomized Management of Myelomeningocele trial.1
Although three experienced fetal centers were selected to participate in the Management of Myelomeningocele trial, interest was widespread by other centers after publication of the results of the study to offer this procedure. Concerns for potential poor outcomes at these startup centers led to the development of guidelines by leaders from 12 organizations interested in maternal and fetal health.2 These guidelines included recommendations on the resources and clinical expertise that should be present at a fetal center as well as the need for strict adherence to the Management of Myelomeningocele inclusion and exclusion criteria (Table 1).
More than 4 years have passed since the publication of the Management of Myelomeningocele trial with more than an estimated 300 cases reported to the Fetal Myelomeningocele Consortium sponsored by the North American Fetal Therapy Network since that time. Clearly, a relaxation of the original selection criteria can occur with exciting new therapies such as fetal myelomeningocele repair because physicians are faced with subtle individual differences in their patient's clinical scenario that may not meet the Management of Myelomeningocele inclusion criteria. Added to this pressure is the parents themselves who are driven to do everything they can to improve the life of their unborn child.
The objective of the current study was to determine the current maternal and fetal selection criteria and operative approaches used at centers performing fetal myelomeningocele surgery.
MATERIALS AND METHODS
An online survey was created based on controversial issues that were historically discussed among members of the Fetal Myelomeningocele Consortium. The consortium consists of maternal–fetal medicine specialists, pediatric surgeons, and pediatric neurosurgeons and nurses from 17 fetal therapy centers throughout the United States that have an interest in fetal myelomeningocele repair. Four of the centers had not performed their first myelomeningocele repair at the time of the survey. The group meets before the North American Fetal Therapy Network biannual meeting.
The survey was initially conceived as a quality assurance project. Results were tabulated and presented in graphic form to the members of the Fetal Myelomeningocele Consortium at the annual April 2015 meeting. At that time, the decision of the Consortium was that the data were worthy of publication. Per the Journal's request, the questionnaire was submitted post hoc to the University of Texas institutional review board and they determined that had it been submitted before undertaking the survey, it would have been designated a class II exemption for institutional review board review.
An initial e-mail invitation was sent to the principal investigators from the 17 centers. Responses were de-identified so that the survey could be answered anonymously. The survey was sent out on three occasions over a 4-week period.
Principal investigators were not asked whether they had standard guidelines at their institution. Instead they were asked to answer a series of 35 questions related to diagnostic testing, inclusion and exclusion criteria, and clinical management at their center (Appendix 1, available online at http://links.lww.com/AOG/A777). Answering all questions was not required. Clinical scenarios were proposed to assess the application of fetal and maternal inclusion and exclusion criteria for offering fetal myelomeningocele repair at their center.
Eleven of the 17 principal investigators (65%) responded to the survey for their center. Twenty-four of the 35 questions in the survey were answered by all the principal investigators. Nine of the questions were answered by 10 of 11 respondents, whereas two questions were answered by 9 of 11 respondents. In the case of three questions related to the interpretation of images to assign a grade for the Chiari malformation, a decision was made to exclude the responses from the analysis. This decision was based on a discussion among the members of the Fetal Myelomeningocele Consortium that concluded that only the presence of any Chiari malformation was an inclusion criterion. In the case of the inquiry as to how many cases had been performed at their institution after the Management of Myelomeningocele trial, four of the 11 centers chose not to respond.
All centers used magnetic resonance imaging as part of their preoperative evaluation (Table 2). Wide variation was present from center to center related to the requirement for amniotic fluid α-fetoprotein, acetylcholinesterase, and DNA microarray assessment.
Ten of 11 centers continued to use the Management of Myelomeningocele criteria of maternal body mass index (BMI, calculated as weight (kg)/[height (m)]2) less than 35 for offering fetal intervention; one center offered surgery for BMIs up to 40 only under institutional review board approval. However, centers were inconsistent in the timing of the BMI that was used. Two centers used the BMI at the start of pregnancy, four used the BMI at the time of referral, and five used the BMI at the time of evaluation. When clinical scenarios were proposed related to well-controlled insulin-dependent diabetes, a history of preterm delivery, and hepatitis C infection, there was considerable variation on the willingness of centers to offer maternal–fetal surgery for fetal myelomeningocele repair. Only one center would offer surgery for the human immunodeficiency virus-positive patient if her viral load was negative on antiretroviral therapy.
All centers continued to use the level of absence of a normally formed vertebral arch (bony dysraphism) to define inclusion if the level of the neural tube defect was between T1 and S1. One center offers surgery for S2 lesions under institutional review board approval. All but one center would not offer intervention if the myelomeningocele was thought to be skin-covered.
Because decreased shunting for hydrocephalus and improved ambulation were the major benefits demonstrated for fetuses undergoing in utero repair in the Management of Myelomeningocele trial, significant fetal ventriculomegaly or lack of leg movement on ultrasonography might be considered a contraindication to fetal myelomeningocele repair. However, 10 of 11 centers did not consider ventriculomegaly and 9 of 11 centers did not consider lack of leg movement as an exclusion criterion. One center used a threshold of up to 12 mm for ventriculomegaly to offer intervention. Centers were split on whether they would offer intervention if gray matter heterotopia was noted on magnetic resonance imaging; 70% would consider a grade I intracranial bleed as an exclusion criterion.
A fetus with a congenital defect was an exclusion criterion for the Management of Myelomeningocele trial. When asked about a clinical scenario of a fetus with a unilateral cleft lip, a myelomeningocele lesion that met the Management of Myelomeningocele criteria, normal karyotype, and normal DNA microarray, 6 of 11 centers would offer maternal–fetal surgery.
Centers varied on their preference for the type of laparotomy incision used for myelomeningocele repair. Two centers routinely utilized a high transverse incision, two routinely used a vertical incision, and the remaining seven centers chose the type of incision based on the patient's habitus. Fetal monitoring during the procedure is performed by pediatric cardiology using echocardiography in 45% of the centers; the remaining centers used ultrasonography only for monitoring. Betamethasone is administered to enhance fetal lung maturity when a viable gestational age is reached at the time of the scheduled procedure at 82% of the centers. As to the tocolytic regimen after completion of the myelomeningocele repair, all centers used indomethacin for 48 hours. Eighty-two percent used this in conjunction with magnesium sulfate and 82% of centers maintained the patient on oral nifedipine for tocolysis for the remainder of the pregnancy.
Since the publication of the Management of Myelomeningocele trial, many new centers have offered maternal–fetal surgery for the correction of fetal myelomeningocele. The current survey indicates that there is considerable variation in the diagnostic requirements for the patient to be considered a candidate for fetal repair of myelomeningocele. Although the fetal inclusion criteria established in the Management of Myelomeningocele trial continue to be followed by most centers, substantial flexibility in the maternal exclusion criteria was noted in the survey.
Preoperative evaluation of amniotic fluid was found in the survey to vary considerably. Many centers would not require acetylcholinesterase—a sensitive indicator of a cerebrospinal fluid leak. However, 10 of the 11 centers would not offer fetal myelomeningocele repair for a skin-covered lesion. This can be difficult to ascertain by ultrasonography and even magnetic resonance imaging with some lesions being partially skin-covered. An additional consideration is whether DNA microarray should be included in the amniotic fluid analysis. Limiting fetal myelomeningocele repair to less than 26 weeks of gestation may not allow for return of results in this window of time. In addition, copy number variant results of unknown clinical significance require parental assessment and are not likely to be completed in the preoperative period.
The original Management of Myelomeningocele criterion of BMI less than 35 appears arbitrary and has been questioned as too limiting. Twenty-eight percent of the U.S. pregnant population is now considered obese and the risk for fetal myelomeningocele increases by twofold with a maternal BMI greater than 30.3,4 Centers appear to have addressed this limitation by using the patient's prepregnancy BMI or expanding the BMI upper limit to 40 with institutional review board approval. Insulin-dependent diabetes is another risk factor for wound complications, but 30% of centers felt that a motivated patient with good glucose control should not be excluded from consideration for fetal myelomeningocele repair. Most interesting from an obstetric perspective is that several centers would be willing to offer the procedure even in the face of a history of previous preterm delivery whose current pregnancy was managed with progesterone prophylaxis.
Before the Management of Meningomyelocele trial, fetal exclusion criteria used by some centers included lack of leg movement on prenatal ultrasonography and ventriculomegaly greater than 17 mm.5 Such criteria were not used in the Management of Myelomeningocele trial. At the time of the survey, more than 90% of centers continued to offer maternal–fetal surgery for myelomeningocele repair in cases in which ventriculomegaly was 18 mm, lack of fetal leg movement, or both. Recently a secondary analysis of the entire Management of Myelomeningocele cohort found that when a presurgical ventricular dilation of 15 mm or greater was present, fetal myelomeningocele repair did not reduce the need for ventriculoperitoneal shunting.6 The authors concluded that fetal myelomeningocele repair should not be offered in these situations. An important advance in our knowledge will be the planned establishment of a post–Management of Myelomeningocele registry of fetal cases by the North American Fetal Therapy Network organization to help delineate other preoperative findings that point to a lack of benefit from in utero repair.
Our survey demonstrated relative consistency among centers in the perioperative care of patients. Two exceptions to the Management of Myelomeningocele protocol were evident in the current survey. Preoperative betamethasone to enhance fetal lung maturity was used by 80% of the centers currently performing fetal myelomeningocele repair.1 Intraoperative fetal echocardiography was used routinely during surgery in the Management of Myelomeningocele study. All centers used continuous ultrasound monitoring of the fetal heart rate throughout the fetal repair; however, less than half of the centers used fetal echocardiography by pediatric cardiology.
A strength of the current survey was that it was completed anonymously. One perceived limitation is that only 65% of invited principal investigators responded. However, based on the knowledge that approximately 13 centers in the United States are performing fetal myelomeningocele repair, we feel the results of the current study accurately reflect the current status of this procedure in our country. The survey has several limitations. Four of the centers in the consortium had not yet performed fetal myelomeningocele repair at the time of the survey; these centers may have contributed to the responses. Only one individual at each center was invited to complete the questions. Individual variations in practice at a center were therefore not addressed by the survey.
Currently variation in the practice patterns for offering and performing maternal–fetal surgery for myelomeningocele repair exists among centers. Scientific and professional collaboration such as the Fetal Myelomeningocele Consortium is key to the critical optimization of such rare and high-risk and high-reward fetal therapies. Ongoing evaluation of inclusion and exclusion criteria as well as operative techniques is warranted to ensure continued safety, efficacy, and beneficence.
1. Adzick NS, Thom EA, Spong CY, Brock JW III, Burrows PK, Johnson MP, et al.. A randomized trial of prenatal versus postnatal repair of myelomeningocele. N Engl J Med 2011;364:993–1004.
2. Cohen AR, Couto J, Cummings JJ, Johnson A, Joseph G, Kaufman BA, et al.. Position statement on fetal myelomeningocele repair. Am J Obstet Gynecol 2014;210:107–11.
3. Gunatilake RP, Perlow JH. Obesity and pregnancy: clinical management of the obese gravida. Am J Obstet Gynecol 2011;204:106–19.
4. McMahon DM, Liu J, Zhang H, Torres ME, Best RG. Maternal obesity, folate intake, and neural tube defects in offspring. Birth Defects Res A Clin Mol Teratol 2013;97:115–22.
5. Sutton LN, Adzick NS, Bilaniuk LT, Johnson MP, Crombleholme TM, Flake AW. Improvement in hindbrain herniation demonstrated by serial fetal magnetic resonance imaging following fetal surgery for myelomeningocele. JAMA 1999;282:1826–31.
6. Tulipan N, Wellons JC III, Thom EA, Gupta N, Sutton LN, Burrows PK, et al.. Prenatal surgery for myelomeningocele and the need for cerebrospinal fluid shunt placement. J Neurosurg Pediatr 2015;16:613–20.