Noel Tulipan

A Randomized Trial of Prenatal versus Postnatal Repair of Myelomeningocele

BACKGROUND Prenatal repair of myelomeningocele, the most common form of spina bifida, may result in better neurologic function than repair deferred until after delivery. We compared outcomes of in utero repair with standard postnatal repair. METHODS We randomly assigned eligible women to undergo either prenatal surgery before 26 weeks of gestation or standard postnatal repair. One primary outcome was a composite of fetal or neonatal death or the need for placement of a cerebrospinal fluid shunt by the age of 12 months. Another primary outcome at 30 months was a composite of mental development and motor function. RESULTS The trial was stopped for efficacy of prenatal surgery after the recruitment of 183 of a planned 200 patients. This report is based on results in 158 patients whose children were evaluated at 12 months. The first primary outcome occurred in 68% of the infants in the prenatal-surgery group and in 98% of those in the postnatal-surgery group (relative risk, 0.70; 97.7% confidence interval [CI], 0.58 to 0.84; P<0.001). Actual rates of shunt placement were 40% in the prenatal-surgery group and 82% in the postnatal-surgery group (relative risk, 0.48; 97.7% CI, 0.36 to 0.64; P<0.001). Prenatal surgery also resulted in improvement in the composite score for mental development and motor function at 30 months (P=0.007) and in improvement in several secondary outcomes, including hindbrain herniation by 12 months and ambulation by 30 months. However, prenatal surgery was associated with an increased risk of preterm delivery and uterine dehiscence at delivery. CONCLUSIONS Prenatal surgery for myelomeningocele reduced the need for shunting and improved motor outcomes at 30 months but was associated with maternal and fetal risks. (Funded by the National Institutes of Health; ClinicalTrials.gov number, NCT00060606.).

The Effect of Intrauterine Myelomeningocele Repair on the Incidence of Shunt-Dependent Hydrocephalus

Background: Intrauterine myelomeningocele repair (IUMR) was first successfully performed in 1997. Preliminary reports suggest that this procedure reduces the incidence of shunt-dependent hydrocephalus when compared to conventional postnatal therapy. However, the existing cohort of IUMR patients has not yet been systematically compared to a comparable group of conventionally treated controls. Methods: Patients 1 year old or greater who had undergone IUMR at either Vanderbilt University or the Children’s Hospital of Philadelphia (CHOP) were compared to a group of conventionally treated historical controls treated and followed at CHOP. In order to measure any differences between the groups, patients were stratified according to the level of the myelomeningocele lesion and the gestational age at the time of IUMR. Results: One hundred and four IUMR patients were compared to 189 conventionally treated controls. IUMR resulted in statistically significant reductions in the incidence of shunt-dependent hydrocephalus at both lumbar and sacral lesion levels. When lumbar lesion levels were further stratified, from L1 to L5, it appeared that the benefit of IUMR was statistically significant only at levels below L2. Other factors with a significant impact on hydrocephalus were estimated gestational age and ventricular size at the time of surgery. In particular, statistically significant differences compared to controls were seen in the younger (≤25 weeks) group but not in the older (>25 weeks) group. Conclusions: IUMR appears to substantially reduce the incidence of shunt-dependent hydrocephalus when compared to conventional treatment even when lesion level is taken into account. Patients with lesions above L3 may not share in this benefit. IUMR cannot be justified in fetuses older than 25 weeks of gestation. Additional improvements might be obtained by further reducing the average age at which fetuses are operated upon. It remains to be determined whether this benefit outweighs the potential risks of intrauterine surgery.

Reduced Hindbrain Herniation after Intrauterine Myelomeningocele Repair: A Report of Four Cases

Background: It has been theorized that fetal myelomeningocele repair may reduce ongoing intrauterine injury and perhaps allow healing and regeneration of dysplastic neural tissue. We report on the postnatal imaging studies of the first 4 patients to have undergone intrauterine myelomeningocele repair at our institution. Methods: Each of the 4 patients underwent postnatal sonographic and MRI. In addition, the postnatal ultrasounds of these 4 were compared to a group of retrospective controls. Results: MRI scans of the 4 experimental subjects revealed no evidence of hindbrain herniation while other stigmata of the Chiari-II malformation persisted. In comparison to the retrospective controls this absence of herniation was distinctly unusual. Conclusion: Intrauterine myelomeningocele repair may reduce the degree of hindbrain herniation normally seen in patients with myelomeningocele. This raises the possibility that intrauterine repair may decrease the morbidity associated with the Chiari type-II malformation including brainstem dysfunction, hydrocephalus and syringomyelia.

Myelomeningocele Repair in utero: A Report of Three Cases

Background: Evidence accumulating over the last 10 years suggests that the exposed spinal cord tissue in a myelomeningocele sustains a secondary injury as the result of prolonged exposure to the intrauterine environment. These data suggest that early closure of the myelomeningocele sac might prevent such injury and in turn improve the neurologic outcome in the affected infant. Methods: Three patients with fetuses carrying the ultrasonic diagnosis of myelomeningocele elected to enter a study of the feasibility of repairing myelomeningocele in utero. At approximately 28 weeks of gestation each patient underwent laparotomy and hysterotomy, thus exposing the myelomeningocele defect. The defect was closed in a routine surgical fashion, and the hysterotomy was then closed. Results: The 3 patients recovered from surgery without incident. Early premature contractions subsided, and they were discharged by the 5th postoperative day. At between 33 and 36 weeks of gestation, each infant was delivered via cesarean section. The observed neurologic deficits were within the range expected from the anatomic level of the defects. Two of the infants have not required ventriculoperitoneal shunting. Conclusions: This limited series of patients suggests that myelomeningocele can be repaired in utero with minimal morbidity to either the mother or her fetus. A larger study will be needed to substantiate this low morbidity and to determine the extent of any neurologic benefit of early surgery.

Effect of Intrauterine Myelomeningocele Repair on Central Nervous System Structure and Function

Background: It has been postulated that intrauterine myelomeningocele repair might improve neurologic outcome in patients with myelomeningocele. A total of 59 such procedures have been performed at Vanderbilt University. Preliminary results suggested that the degree of hindbrain herniation is reduced by intrauterine repair. In an attempt to further quantify the possible benefits of this surgery, a subset of these patients was brought back to Vanderbilt for study. Methods: A group of 26 patients who had undergone intrauterine myelomeningocele repair underwent an extensive evaluation which included manual muscle testing, MR imaging and precise determination of the anatomic level of their lesions as well as multiple other tests. The results of this analysis were compared to those in 2 groups of historical controls. Results: In this group of patients intrauterine myelomeningocele repair substantially reduced the incidence of moderate to severe hindbrain herniation (4 vs. 50%). The incidence of shunt-dependent hydrocephalus was more modestly reduced (58 vs. 92%). The average level of leg function closely matched the average anatomic level of the lesion in both the fetal surgery and control groups. Conclusion: The most dramatic effect of intrauterine repair appears to be on hindbrain herniation. A less dramatic, but significant, reduction in shunt-dependent hydrocephalus is also seen. Prospective patients should be cautioned not to expect improvement in leg function as the result of this surgery. The potential benefits of surgery must be carefully weighed against the potential risks of prematurity.

In utero Repair of Myelomeningocele: A Comparison of Endoscopy and Hysterotomy

Objective: To compare endoscopic coverage of myelomeningocele with a maternal split-thickness skin graft in utero to definitive neurosurgical closure through a hysterotomy. Methods: Four fetuses with isolated myelomeningocele underwent endoscopic coverage of the defect with a maternal split-thickness skin graft in a CO2 environment at 22–24 weeks’ gestation. Subsequently, 4 fetuses underwent standard neurosurgical closure of their myelomeningoceles at 28–29 weeks’ gestation. Results: The mean operating time for the endoscopic procedures was 297 ± 69 min. Two fetal losses occurred as a result of chorioamnionitis and placental abruption, respectively. A third baby delivered at 28 weeks’ gestation after prolonged disruption of the membranes. The 2 survivors required standard closure of the myelomeningocele after delivery. The mean operating time for the hysterotomy procedures was 125 ± 8 min. No mortality occurred, and all the infants delivered between 33 and 36 weeks with well-healed myelomeningocele scars. At present, the functional levels of all infants approximate the anatomical levels of the lesions. Conclusion: With current technology, in utero repair of congenital myelomeningocele through a hysterotomy appears to be technically superior to procedures performed endoscopically.

Intrauterine Myelomeningocele Repair Reverses Preexisting Hindbrain Herniation

Background: It has been reported that intrauterine myelomeningocele repair reduces the amount of hindbrain herniation normally seen in association with the Chiari type II malformation. It is not yet known, however, whether hindbrain herniation is prevented, or whether preexisting herniation is reversed. The following study was designed to elucidate this issue. Methods: A series of 9 patients underwent intraoperative ultrasound examinations immediately prior to intrauterine myelomeningocele repair. These same patients were then evaluated postnatally using ultrasound and/or MRI. The degree of hindbrain herniation before and after repair was compared using a grading system devised by the authors. Results: Eight patients had clear evidence of moderate to severe hindbrain herniation on intraoperative scans while one was mild. In contrast, on postnatal studies 5 of 9 patients had no evidence of hindbrain herniation, while the other 4 had only mild herniation. Conclusion: Intrauterine myelomeningocele repair appears to reverse preexisting hindbrain herniation. It is postulated that continuous flow of cerebrospinal fluid through the neural placode is the force responsible for inducing migration of the cerebellum and brain stem downward through the foramen magnum. By interrupting that flow during gestation, intrauterine myelomeningocele repair enables the cerebellum and brain stem to resume a normal, or nearly normal, configuration.

The urodynamic profile of myelodysplasia in childhood with spinal closure during gestation.

PURPOSE Spinal dysraphism is the most common cause of neurogenic bladder dysfunction in newborns. Urodynamic findings in these patients include uninhibited bladder contractions, bladder areflexia, decreased compliance and detrusor-sphincter dyssynergia. Early urodynamic studies are recommended for spina bifida to help identify bladder characteristics that may cause a risk of upper tract deterioration. We recently evaluated a new early type of intervention involving closure of the neural tube defect during gestation in 25 patients at our institution. We hypothesize that this procedure results in decreased exposure of the spinal cord to amniotic fluid, which may improve neurological function. To date we have evaluated 16 of the 25 patients with video urodynamics. We compared the results to those in the literature on patients with myelomeningocele and without prenatal intervention. MATERIALS AND METHODS We performed urodynamic testing in 16 patients with a mean age of 6.5 months, including cystometrography, fluoroscopic evaluation of filling and voiding, pelvic floor electromyography and post-void residual urine measurement. In addition, we retrospectively reviewed renal ultrasound, voiding cystourethrography, catheterization need, number of urinary tract infections and medication in these cases. RESULTS Uninhibited detrusor contractions and an areflexic bladder were identified in 6% and 43% of patients, respectively, while 19% had decreased compliance and 75% had leak point pressure greater than 40 cm. water. Mean bladder capacity was 40 cc and 31% of patients had much lower capacity than expected for age. Previous renal ultrasound and voiding cystourethrography showed evidence of upper tract dilatation and reflux in 2 cases, respectively. Intermittent catheterization and anticholinergic therapy were required by 1 patient each and 1 had a significant urinary tract infection. CONCLUSIONS Urodynamic findings in this population are comparable to those previously reported in the literature in patients with spina bifida without prenatal closure of the spinal defect. The lower incidence of urinary tract infection and reflux in our study probably represents more aggressive early urological management rather than neurological improvement. These urodynamic studies were performed early in life and future evaluation may ultimately reveal improved bladder function compared with that in others with myelodysplasia. However, at this time we recommend that patients who undergo spinal cord defect closure during gestation be evaluated and treated in the same manner as those with myelomeningocele but without fetal intervention.

Robot-Assisted Endoscopic Intrauterine Myelomeningocele Repair: A Feasibility Study

Background: Animal experiments have suggested that the intrauterine environment causes secondary injury to the congenitally dysplastic spinal cord. This in turn suggests that early closure of the myelomeningocele sac might prevent secondary injury and therefore improve neurologic outcome. This study was designed to examine the technical feasibility of performing intrauterine myelomeningocele repair using a robot-assisted endoscopic system in an animal model. Methods: Six fetal sheep underwent creation and repair of a full-thickness skin lesion using the da VinciTM system. Results: With the device’s advanced articulated instruments and three-dimensional optics, it was possible to endoscopically repair the induced skin defects. Conclusion: We conclude that, with the recent evolution in robotics and minimally invasive techniques, intrauterine endoscopic surgery has become a realistic goal that promises to reduce the associated risks of fetal surgery and extend the indications for its use.

Fetal MRI in the Evaluation of Intrauterine Myelomeningocele

Background: Accurate fetal imaging is essential to the practice of maternal-fetal medicine. While ultrasonography has been the traditional mainstay of fetal imaging, its ability to resolve critical features of central nervous system (CNS) anatomy remains limited. As interest in intrauterine therapy for myelomeningocele has increased, so has the need for more accurate, noninvasive imaging of the CNS. Fetal magnetic resonance imaging (MRI) promises to fill the gap left by ultrasound. Methods: Thirty-seven MRI scans of fetuses previously diagnosed with myelomeningocele were reviewed by 2 neuroradiologists. The ability of fetal MRI to resolve the commonest CNS stigmata of spina bifida, and the incidence and extent of interobserver error, was assessed. In 4 cases, postnatal MRIs were also available. These were compared to the corresponding fetal studies. Results: The imaging quality with the technique used in this study was excellent, even without the use of maternal or fetal sedation. There were no complications, and the imaging times were minimal. Interobserver error was minimal with respect to the evaluation of ventricular dilatation and hindbrain herniation, but moderate in the description and location of the spinal lesion. As had previously been documented with ultrasonography, a reduction was seen in hindbrain herniation when comparing pre- and postnatal MRIs. Conclusion: It is concluded that fetal MRI is an effective, noninvasive means of assessing fetal CNS anatomy. Its ability to resolve posterior fossa anatomy is superior to ultrasonography while, with respect to the evaluation of hydrocephalus and the level and nature of the spinal lesion, it may be equivalent to inferior. Inclusion of the fetal MRI into the standard diagnostic armamentarium will probably await the next major advance in speed and resolution. It is conceivable that, with further advances, MRI might supplant ultrasonography as the diagnostic tool of choice for evaluation of fetal anomalies including myelomeningocele.