Version May 2024
INTRODUCTION — Neural tube defects (NTDs) are the second most common major congenital anomaly (cardiac anomalies are the most common). The majority are open (ie, not covered by skin) and involve a defect of the posterior spinal column or cranium. In most cases of open spina bifida, a sac (meningocele) forms over the defect and contains neural tissue (myelomeningocele). In a minority of cases, the defect is not covered by a membrane (myeloschisis).
By comparison, in closed NTDs (spinal dysraphism or spina bifida occulta), the defect involves the posterior vertebral arches and possibly the meninges, but does not involve the spinal cord, muscles, tendons, or other tissues of the spinal column; the skin is intact, so neural tissue is not exposed to amniotic fluid. The skin may have nevi, tufts of hair, sinus tract(s), dimple(s), or hyperpigmentation marking the area over the vertebral defect [1]. (See "Closed spinal dysraphism: Pathogenesis and types".)
Open spina bifida is the focus of this topic. The spinal defect results from incomplete closure of the caudal portion of the neural tube between 22 and 28 embryological days (ie, five to six weeks of gestation). The defect usually involves the lumbar or sacral spine and can vary considerably in size. Postnatal clinical manifestations correlate with the spinal level and size of the defect, and usually include lower limb paralysis and bowel and bladder dysfunction. Open spina bifida is usually associated with a Chiari II malformation (Arnold-Chiari malformation), which includes a constellation of CNS anomalies such as hindbrain herniation, brainstem abnormalities, low-lying venous sinuses, a small posterior fossa, and some degree of hydrocephalus. The Chiari II malformation has deleterious effects on motor, cranial nerve, and cognitive functions. (See "Chiari malformations".)
This topic will discuss in utero repair of open spina bifida, including selection of appropriate candidates, options for procedures, counseling, and delivery issues. Prevention and prenatal diagnosis of NTDs and postnatal management and outcome of myelomeningocele are reviewed separately:
●(See "Preconception and prenatal folic acid supplementation".)
●(See "Neural tube defects: Overview of prenatal screening, evaluation, and pregnancy management" and "Neural tube defects: Prenatal sonographic diagnosis".)
●(See "Myelomeningocele (spina bifida): Anatomy, clinical manifestations, and complications" and "Myelomeningocele (spina bifida): Management and outcome" and "Myelomeningocele (spina bifida): Orthopedic issues" and "Myelomeningocele (spina bifida): Urinary tract complications".)
PATHOGENESIS OF EARLY NEURONAL INJURY — The pathogenesis of intrauterine neuronal injury in open spina bifida is not precisely known. It is hypothesized that the primary defect in neurulation, which exposes the neuronal elements to the intrauterine environment, is associated with neuronal loss, which may reach about 60 percent prior to 16 weeks of gestation ("first hit") [2]. A second pathogenic "hit" may occur as a result of ongoing neuronal injury from substances in the amniotic fluid and trauma [3]. A third pathogenic "hit" may occur from internal forces within the myelomeningocele sac that stretch the placode and its derivative nerves [4].
GOAL AND OUTCOME OF IN UTERO SURGERY — The technical goal of in utero spina bifida repair is to create complete, water-tight coverage of the defect, with the objective of halting progressive intrauterine neuronal injury [5,6]. Progressive in utero motor and sensory deficits have been well documented in fetuses with open spina bifida followed without prenatal intervention [7,8].
The safety and efficacy of fetal surgery for repair of open spina bifida were evaluated in the seminal Management of Myelomeningocele Study (MOMS) trial, conducted from 2003 to 2010 at three institutions in the United States with extensive experience in fetal surgery [9]. Participants were randomly assigned to undergo fetal surgery by open hysterotomy at 18 to 25 weeks gestation (n = 91) or standard postnatal repair (n = 92). The trial was stopped early because of efficacy. Findings reported in the complete trial and in subsequent follow-up studies include the following [9-14]:
●Decreased need for cerebrospinal fluid (CSF) shunting
•In the first year of life, 44 percent of infants in the fetal surgery group required CSF shunt placement for hydrocephalus compared with 84 percent in the postnatal repair group (relative risk [RR] 0.53, 95% CI 0.41-0.67) [11].
•At age 6 to 10 years, children in the fetal surgery group had a lower rate of hindbrain herniation (60 versus 87 percent), less frequently required a shunt (49 versus 85 percent), and, among those with shunts, less frequently required shunt revision (47 versus 70 percent) [12].
●Improved motor development and function
•At 30 months of age, more children in the fetal surgery group achieved independent ambulation compared with those in the postnatal repair group (45 versus 24 percent) [9,13]. Children in the fetal surgery group scored higher on standardized tests of motor development (mean psychomotor development index on the Bayley Scales of Infant Development [BSID] II tool was 64±17 versus 59±15) [9,13]. Factors associated with independent ambulation included the presence of in utero ankle, knee, and hip movement; absence of a sac over the lesion; and a myelomeningocele lesion level from L3 to S1 compared with T1 to L2 [13].
•At age 6 to 10 years, there were fewer children overall who maintained independent ambulation, though the difference between the two groups persisted (29 percent in the fetal surgery group versus 11 percent in the postnatal surgery group) [12]. Children in the fetal surgery group also scored higher on assessments of motor and self-care skills (mean composite score 92±9 versus 85±18) [12].
●Possible improvement in bladder function – Whether the rate of bladder dysfunction is lower in children who undergo fetal repair compared with those who undergo postnatal repair is unclear. If there is an improvement, it does not appear to be dramatic and therefore urologic outcomes alone should not be the sole impetus to perform fetal surgery.
•At 30 months of age, the fetal surgery group had a nonsignificant reduction in the need for clean intermittent catheterization (CIC) compared with the postnatal surgery group (38 versus 51 percent; RR 0.74, 95% CI 0.48-1.12) [10].
•By school age, more children in the fetal surgery group reported voiding volitionally compared with the postnatal surgery group (24 versus 4 percent) and fewer children in the fetal surgery group required CIC (62 versus 87 percent) [14]. Rates of bladder augmentation, vesicostomy, and urethral dilation were similar in the two groups.
●Modest improvements in quality of life and impact on family
•At age 6 to 10 years, parental assessments of health-related quality of life (HR-QOL) and impact of the child's health condition on the family were better in the fetal surgery group compared with the postnatal surgery group, though the differences were modest [12].
●Similar cognitive outcomes as expectant management – Based on the available data, fetal surgery does not appear to improve cognitive and adaptive functioning.
•At 30 months of age, children in both groups had similar scores on standardized tests of mental development (mean mental development index on the BSID-II was 89±15 for the fetal surgery group versus 86±18 for the postnatal repair group) [13].
•At age 6 to 10 years, scores on standardized developmental tests were similar in both groups within each domain of adaptive functioning (communication, daily living, and socialization) and for the overall composite score (89±10 versus 88±12) [12]. Additionally, there were no apparent differences on a variety of parent-rated assessments of attention, executive functions, and behavior.
The participants of the MOMS trial are being followed into adolescence and adulthood to evaluate the long-term effect of fetal intervention on motor function, cognitive development, bowel and bladder function, and other important outcomes, including sexual function.
CENTERS FOR FETAL SURGERY — Professional societies have recommended that centers for fetal surgery have dedicated operational infrastructure and necessary resources to allow for appropriate oversight and monitoring of clinical performance and facilitate multidisciplinary collaboration between the relevant specialties [15,16]. Within this framework, an expert group proposed three levels of care [17]. Open spina bifida repair should only be performed at a center meeting criteria for level 3, which is defined by the capability to manage any associated maternal complications and comorbidities, as well as access to neonatal and pediatric surgical intervention, including indicated surgery for neonates with congenital anomalies. It has been suggested to use a simulator for coordinating and training the team in these extremely specialized centers [18].
INCLUSION AND EXCLUSION CRITERIA FOR CONSIDERATION OF REPAIR — Most fetal surgery centers performing open spina bifida repair generally base their inclusion and exclusion criteria on those used in the MOMS trial [9], but modifications of the original exclusion criteria are common [19,20].
Inclusion criteria
●Open spina bifida with the upper boundary between T1 and S1
●Hindbrain herniation
●Gestational age 19+0 to 25+6 weeks of gestation at the time of the procedure
●Maternal characteristics:
•≥18 years of age
•Understands the requirements of the procedure and capable of consenting for their own participation and complying with travel and follow-up requirements
•Has been counseled about their pregnancy options (including pregnancy termination and no fetal surgery) and desires fetal surgery
•Meets psychosocial criteria (as determined by the psychosocial interviewer using a standardized assessment) to handle the implications of the procedure
•Provides permission for postnatal follow-up of the child
Potential exclusion criteria — Most of the following are potential exclusion criteria because individualization is required to assess patient specific risks, benefits, values, and goals.
●Pregnancy at increased risk for preterm birth because of:
•Previous preterm birth
•Placental abnormality (previa, abruption, placenta accreta spectrum)
•Short cervix in the current pregnancy
•Early fetal growth restriction
•Maternal hypertension
•History of cervical insufficiency
•Congenital uterine anomaly associated with preterm birth
However, exclusion because of an increased risk for preterm birth is often decided on a case-by-case basis, given the variability in the magnitude of preterm birth risk associated with traditional risk factors and the difficulty in determining the precise risk for specific patients. For example:
•A very short cervix (≤15 mm) has been associated with a 50 percent risk of preterm birth <32 weeks and is an absolute exclusion criterion, whereas a short cervix (typically defined as <25 mm) is a relative exclusion factor. The inverse relationship between cervical length and risk of preterm birth exists for all cervical lengths <30 mm and is considered in decision-making when the cervical length is >15 mm and <25 to 30 mm [21].
•Placenta previa is an absolute exclusion criterion because approximately 15 percent of affected pregnancies require delivery before 34 weeks of gestation.
•Not all congenital uterine anomalies are associated with preterm birth and even those that carry this association (bicornuate uterus, unicornuate uterus, uterus didelphys, septate uterus) do not result in preterm birth in all patients.
●Obesity – Although the MOMS trial excluded pregnant individuals with body mass index (BMI) ≥35 kg/m2, many centers no longer use this exclusion [19] as some studies have demonstrated that BMIs ≥35 to 40 kg/m2 have not impacted surgical outcomes [22,23]. At the authors' center, a multidisciplinary team individualizes this decision. For example, they may consider excessive central obesity a contraindication because of significant technical challenges to performing fetal surgery.
●Maternal viral disorder – The MOMS trial excluded pregnant individuals with a positive screening test for hepatitis B virus (HBV), hepatitis C virus (HCV), or human immunodeficiency virus (HIV) because of the risk of vertical transmission during fetal surgery. However, HBV positivity may not be an absolute exclusion criterion given a case report of successful active and passive fetal HBV vaccination and subsequent in utero repair of a fetus with myelomeningocele carried by a HBV-positive mother [24].
Many centers report that they offer fetal surgery for patients with positive HCV or HIV test results who have an undetectable viral load [19,25]. Ethical discussions have supported the conclusion that maternal-fetal interventions of pregnant individuals infected with HIV, HBV, or HCV but low or undetectable viral loads is permissible as such interventions are expected to improve survival and overall outcome of offspring [26]. At our center, a multidisciplinary team extensively discusses these issues with the patient to enable a well-informed shared decision.
●Presence of an additional fetal anomaly unrelated to the NTD – At our center, exclusion is individualized depending on the nature of the anomaly and its impact on prognosis.
●Abnormal fetal karyotype – The MOMS trial required documentation of a normal fetal karyotype for study inclusion and most centers continue to require normal chromosomal microarray (CMA) in accordance with accepted recommendations for genetic testing in the setting of any major congenital anomaly [27]. In a series of patients with isolated spina bifida, 2.6 percent had pathogenic copy number variants (CNVs) and up to 16 percent had CNVs [28]. While these data support the use of CMA in all cases for long-term prognostic counseling, they do not necessarily mean that these fetuses should be excluded from prenatal repair. A pathologic variant does not affect the success of the repair (gestational age at delivery, other obstetric and early neonatal outcomes) [29]. There is ongoing debate about whether denying fetal interventions to fetuses with pathologic genetic variants is ethical, as such intervention may improve overall survival and prognosis. At our center, this decision is individualized based on input from a team of genetic experts (maternal-fetal medicine and pediatric), a fetal therapy board, and the patient.
●Lack of social support by a spouse, partner, parent, or close friend for the duration of pregnancy (in part because these patients need to travel to a fetal surgery center, which is often in a city far from their home and requires temporary local residence).
●Multifetal pregnancy – While most fetal centers consider multiple gestation an exclusion criterion, emerging publications of in utero repair in dichorionic and monochorionic twin gestations are promising [30].
●Severe fetal kyphosis (≥30 degrees).
●Previous hysterotomy in the upper uterine segment (eg, previous classical cesarean birth, repair of a uterine anomaly, extensive myomectomy, or fetal surgery).
●Technical limitations precluding surgery, such as uterine fibroids, amnion-chorion membrane separation, or uterine anomaly.
●Risk for hemolytic disease of the fetus and newborn (eg, D or Kell sensitization) or risk for fetal and neonatal alloimmune thrombocytopenia in the current pregnancy.
●Low amniotic fluid volume (usually defined as amniotic fluid index <6 cm) is likely due to a fetal anomaly, poor placental perfusion or function, or membrane rupture. Low amniotic fluid volume that responds to maternal hydration is not an exclusion criterion.
●Severe depression (eg, based on Beck Depression Inventory-II questionnaire score <29).
●Maternal medical condition that is a contraindication to surgery or anesthesia.
PREOPERATIVE FETAL NEUROLOGICAL EVALUATION
Utilize imaging expertise — Surgical success depends, in part, on identifying those fetuses with open spina bifida who are appropriate candidates for in utero repair. Fetal evaluation using ultrasound and/or magnetic resonance imaging (MRI) plays an important role in this assessment and is highly dependent on the experience and skill of the imager. A multidisciplinary approach should include pediatric radiologists experienced in fetal MRI and maternal-fetal medicine/fetal intervention specialists experienced in fetal anomaly ultrasound.
Confirm open spina bifida — Ultrasound examination is usually adequate to confirm that the spinal defect is open (myelomeningocele, myeloschisis), not closed (eg, myelocystocele and lipomyelomeningocele); closed defects are not amenable to and would not benefit from in utero repair. Ultrasound diagnosis of an open versus closed defect is reliable; amniotic fluid testing for acetylcholinesterase is unnecessary for confirmation. When sonographic visualization of the defect is not adequate because of a difficult fetal lie or maternal factors (eg, obesity), MRI should be performed to confirm the diagnosis.
●One way to distinguish open defects from closed defects is to identify the Chiari II malformation (hindbrain herniation, brainstem abnormalities, low-lying venous sinuses, a small posterior fossa, and some degree of hydrocephalus). Closed defects generally are not associated with this malformation (image 1).
●Another way is to carefully evaluate the surface covering the defect, particularly at the margin of the defect where it contacts the skin of the lower back.
•Myeloschisis has no overlying sac containing neuronal elements and cerebrospinal fluid (CSF); instead, the neural placode is tethered to the level of the defect itself (image 2).
•Myelomeningoceles are covered with membrane, which tends to be very thin, whereas closed defects are covered with skin, which tends to be thick. On both ultrasound and MRI, membrane-covered lesions are at an acute angle at the point of contact with the lower back [31], whereas skin-covered lesions are at an oblique angle, often with visible subcutaneous tissue in contiguity with the sac wall (image 3).
Identify hindbrain herniation — Hindbrain herniation is an inclusion criterion for in utero repair. On ultrasound, the fetus should be imaged in the sagittal plane with its back at the maternal abdominal wall, if possible. Focusing on the craniocervical junction will reveal either anechoic CSF posterior to the cerebellum and extending down into the upper cervical canal (indicating no hindbrain herniation) or solid cerebellar tissue will be visualized dorsal to the cord, obliterating the cisterna magna and retrospinal CSF space (indicating hindbrain herniation) (image 4).
In patients with a difficult fetal lie or maternal factors (eg, obesity) that prevent successful visualization on ultrasound, MRI should be performed and is usually diagnostic. Typically, single-shot fast spin-echo (SSFSE) T2-weighted images are utilized, which provide excellent tissue contrast and bright fluid signal, and are relatively resistant to fetal motion. Care should be taken to obtain in-plane sagittal views of the brain, including a midline view; operators may find it helpful to obtain an odd number of images to ensure a single midline image.
Some centers categorize the degree of hindbrain herniation, using imaging findings including obliteration of the cisterna magna and effacement of the fourth ventricle, in addition to presence of cerebellar tissue in the cervical canal, which is generally considered the imaging criterion for hindbrain herniation whether grading or a description is used [32].
Identify the defect level — Candidates for fetal repair should have a vertebral defect beginning between T1 and S1, so identifying the upper extent of the defect is important. This can be accomplished via ultrasound or MRI; ultrasound has the advantages of showing each vertebral level in the axial plane so the upper level of the defect can be easily determined and the ability to readily distinguish vertebral levels with intact posterior elements from those that are defective posteriorly (image 5).
MRI can be more challenging, particularly in a mobile fetus, but generally the technique depends more heavily on sagittal plane imaging of the spine and relies on echoplanar imaging, which allows for distinct visualization of the bony structures.
Helpful landmarks on either modality include the 12th rib (to identify the last thoracic segment) and the normal lumbosacral lordosis (to identify L5-S1).
Quantify ventriculomegaly — We quantify ventriculomegaly because it is a factor when counseling parents about the risks and benefits of fetal versus postnatal repair. We believe in-utero repair can be beneficial even in fetuses with severe ventriculomegaly because it may preserve lower extremity movement and future independent walking, delay the need for a CSF shunt, and possibly increase the success of future endoscopic third ventriculostomy. However, some centers consider severe ventriculomegaly an exclusion criterion. A detailed analysis of postnatal follow-up data from the MOMS trial revealed that children who had severe ventriculomegaly (defined as >15 mm) on prenatal imaging were highly likely to need treatment for hydrocephalus postnatally, even if they underwent fetal repair [11]. In this analysis, 79 percent of fetuses with severe ventriculomegaly prior to fetal repair required postnatal treatment for hydrocephalus, compared with 85 percent of fetuses with severe ventriculomegaly managed expectantly and repaired postnatally. A subsequent study observed that perforation of the cavum septi pellucidi on ultrasound at the time of referral appeared to reliably identify those cases at highest risk for needing treatment of hydrocephalus by one year of age [33].
Ventriculomegaly can be quantified on ultrasound or MRI, but measurements should be standardized and consistent from scan to scan [34]. It is important to be aware that in utero MRI measurements are typically 1 mm greater than ultrasound measurements in the same fetus on the same day, likely due to the higher spatial resolution of ultrasound and more accurate edge definition. It’s also important to be aware that if imaging is performed after fetal repair, ventricles typically increase in size by 2 to 3 mm, which is thought to be due to reaccumulation of CSF in the supratentorial compartment after closure of the defect.
Assess for motor deficits — Documentation of baseline motor deficits helps to inform parental counseling and set appropriate expectations for postnatal motor function [20,35]. The lower fetal extremities should be assessed carefully for positional abnormalities at rest and with activity. Clubbed or rocker bottom feet suggests motor neuron injury and can be viewed as a form of contracture due to denervation.
Lower extremity motion should be assessed at each joint, including hip flexion, knee flexion and extension, and ankle dorsiflexion and plantarflexion (with plantarflexion referred to as "gas pedal" movement). These movements can be evaluated with ultrasound by imaging the fetus in the longitudinal plane [36]. They can be evaluated with MRI by setting up a thicker slab (such as 20-mm thick slices, rather than the typical 3- or 4-mm thick slices) and running a dynamic sequence (same slab run over and over again in the same position, rather than moving through the fetus). Single-shot T2 or balanced steady-state free precession (bSSFP) sequences can be used for this purpose, with bSSFP having the advantage of being faster overall (movie 1).
Assess for imaging-based exclusion factors — The majority of inclusion and exclusion criteria for prenatal repair of open spina bifida are historical and clinical factors, but the following findings are assessed at ultrasound (including routine echocardiography) or MRI and may exclude the patient as a candidate for prenatal surgery, as discussed above (see 'Potential exclusion criteria' above):
●Short cervix.
●Placenta previa.
●Additional fetal anomalies. Syndromes associated with spina bifida (eg, trisomy 18) may have already been detected by genetic testing or by an earlier ultrasound examination prior to referral, but the fetal imager should still pay careful attention to the rest of the fetal anatomic assessment to exclude potential syndromic etiologies. Some syndromes, such as caudal regression and omphalocele-exstrophy-imperforate anus-spine (OEIS) sequences, would typically be associated with a skin-covered NTD and would be an exclusion factor for prenatal repair. Certain stigmata of Chiari II malformation, such as dysgenesis of the corpus callosum and subependymal gray matter heterotopia, should not be mistaken as unrelated anomalies suggestive of an underlying genetic syndrome. These are part of the Chiari II malformation and would not disqualify a patient from prenatal repair. (See "Chiari malformations", section on 'Chiari II anatomy'.)
●Uterine anomaly. Mullerian anomalies can be challenging to identify prenatally, as the gravid uterus distorts the external uterine contour and limits the usefulness of the imaging signs that are typically applied to identify anomalies in the nongravid state. (See "Congenital uterine anomalies: Clinical manifestations and diagnosis".)
●Fetal kyphosis greater than 30 degrees with the apex at the open defect. This is an exclusion factor because it can preclude successful closure. Kyphosis can be measured using either ultrasound or MRI, and should be assessed during fetal motion to determine if the angulation is fixed or flexible (image 6).
PARENTAL COUNSELING
Informed consent
●Counseling needs to be comprehensive and should ideally involve all relevant disciplines, including the maternal-fetal medicine/fetal intervention specialist, pediatric radiologist, neonatologist, pediatric neurosurgeon, pediatric general surgeon, and pediatric anesthesiologist. An institutional support system that includes dedicated nurses, social workers, psychologists, and psychiatrists is also an integral part of preoperative counseling, evaluation, and care.
●Both in utero and postnatal surgical repair should be discussed with and offered to all patients. We emphasize that postnatal repair, which typically occurs within the first three days after birth, is an alternative option with reasonable outcomes and has the advantage of no added maternal risk.
●If the patient is considering in utero repair, we discuss all the available approaches, with full transparency of our center's experience. We describe the technical benefits and difficulties of the various procedures and the risks and potential benefits, and emphasize the limited information on long-term outcomes. (See 'Potential perioperative maternal and obstetrical risks' below and 'Potential fetal and postnatal pediatric risks' below and 'Goal and outcome of in utero surgery' above.)
●Proceeding with a specific prenatal surgical approach is based on shared decision-making.
●Rarely during fetoscopic surgery (laparotomy-assisted and percutaneous), technical difficulties or complications require conversion to an open hysterectomy approach. Thus, if the patient elects a fetoscopic approach, the possible need to convert to the open approach should be discussed and agreed upon.
Potential perioperative maternal and obstetrical risks
●Uterine bleeding – Bleeding can occur from the myometrial entry site, leading to maternal hemoperitoneum. Although bleeding is usually self-limited, the patient needs to be monitored closely and, if necessary, treated with blood products. The frequency of maternal blood transfusion ranges from 2.3 to 9 percent of cases [9,37,38]. The risk of substantial uterine bleeding can be minimized by ultrasound-guided access, placement of sutures around instrument ports, and direct visualization of and ability to suture bleeding ports on the exposed uterus.
●Pulmonary edema – Pulmonary edema has been reported in 6 to 13.7 percent of mothers, with a higher frequency consistently reported for the laparotomy-assisted fetoscopic approach [9,37-39]. Meticulous intraoperative fluid balance and standard postoperative care including incentive spirometry are important to prevent this complication. No reports of the need for intubation have been published. Standard treatment of pulmonary edema is administered.
●Chorioamniotic separation (CAS) – The amnion and chorion are fused and the exocoelomic cavity eliminated on ultrasound at approximately 15 weeks of gestation. Iatrogenic CAS can occur as a result of an intrauterine procedure and is identifiable on ultrasound. It may increase the risk of preterm prelabor rupture of membranes (PPROM), preterm labor, and chorioamnionitis. Rarely, it may result in an umbilical cord accident and fetal death [40,41]. The frequency of CAS ranges from in 10 to 40 percent of cases [9,37-39,42-44]. There is no treatment for CAS.
●Preterm labor – Preterm labor is common (20 percent [45]). Acute preterm labor is managed according to standard principles. (See "Inhibition of acute preterm labor".)
●Preterm prelabor rupture of membranes (PPROM) – Amniotic membrane rupture prior to 37 weeks gestation has been reported in 19 to 44 percent of prenatally repaired NTD cases [9,38,42,43]. Gestational age at PPROM is an important prognostic factor, with some studies reporting a median gestational age of 32 weeks at occurrence [39]. Standard of care treatment includes antibiotics and a course of antenatal corticosteroids. If there is evidence of an infection, delivery should be expedited. (See "Preterm prelabor rupture of membranes: Management and outcome".)
●Chorioamnionitis – Inflammation of the amnion and chorion due to a bacterial infection has been reported in 2.2 to 4.5 percent of cases [9,38]. Any invasive procedure may lead to intrauterine infection, even though the procedure is performed in aseptic conditions and peri-operative antibiotics are administered. If chorioamnionitis is diagnosed, administration of antibiotics and delivery are required to prevent further maternal complications.
●Side effects from drugs (eg, magnesium sulfate, indomethacin, nifedipine) commonly used before, during, and after fetal surgery are managed according to standard principles.
●Placental abruption – Abruption rarely occurs during surgery but is reported in 2 to 6.6 percent of cases when the postoperative period is also considered [9,38]. Abruption is managed according to standard principles. (See "Acute placental abruption: Management and long-term prognosis".)
●Postoperative surgical complications that are not unique to fetal surgery, including wound infection, urinary tract infection, and pneumonia, can occur and are managed according to standard principles.
●Other
•Amniotic fluid embolism is a theoretic risk, but has never been reported during open or fetoscopic surgery.
•Maternal death is also a theoretic possibility and should be included as a potential outcome in the counseling and consent process; however, intensive intra- and postoperative monitoring by experienced obstetric anesthesiologists and perinatologists should allow for early diagnosis and prompt management of life-threatening maternal complications.
Potential fetal and postnatal pediatric risks
●Adverse pregnancy outcome – The fetus is at risk for adverse consequences of obstetrics complications such as preterm birth, PPROM, oligohydramnios, abruption, CAS, chorioamnionitis, and bleeding inside the uterus.
The risk of preterm birth differs by the type of procedure used for in utero repair. The mean gestational at birth by procedure is: percutaneous approach - 32 weeks [46], open hysterotomy - 34 to 35 weeks [9,47], mini-hysterotomy - 35 to 36 weeks [37,48], and laparotomy-assisted fetoscopic - 37.6 weeks [38,39].
●Suboptimal result
•The procedure may fail to reduce the potential for tethering (scarring) of the neural placode within the spinal canal, which may increase the likelihood that the child will need future additional surgery. We estimate this occurs in approximately 10 percent of cases.
•The procedure may not accomplish the desired effect of a water-tight closure of the NTD, allowing cerebrospinal fluid (CSF) to leak through the repair and resulting in the need for additional surgery at birth. We estimate this occurs in less than 5 percent of cases.
•The procedure may fail to reverse hindbrain herniation.
●Fetal injury and other harms
•The procedure may result in a fetal injury leading to fetal death, but this is rare.
•The fetus may be at risk of adverse effects from general anesthesia administered to the mother. The United States Food and Drug Administration (FDA) warns that lengthy use of general anesthetic and sedation drugs during surgeries or procedures in pregnant people may affect the developing fetal brain [49]. However, these potential effects also apply to newborns, so deferring the repair to the newborn period does not necessarily reduce this possible risk. Although children who need surgery and anesthesia have worse scores for some neurodevelopmental outcomes compared with children who do not need surgery and anesthesia, these differences cannot yet be attributed to anesthesia exposure and there is still uncertainty that exposure to anesthesia has any lasting effects in children or fetuses.
•The fetus is at risk of adverse effects from medications administered directly to it. A combination of medications (eg, fentanyl, vecuronium, and atropine) is routinely administered to fetuses during open fetal surgery to provide adequate sedation and analgesia [50]. Potential fetal risks include bleeding or tissue necrosis from the injection, tachycardia or bradycardia, and infection. Medications or fetal transfusion may be administered to manage these risks, as appropriate.
PROCEDURE
General principles of preoperative and intraoperative care — Our general approach is described. Protocols vary across fetal therapy centers.
●The mother and fetus are kept warm (as close to 37°C as possible) throughout the procedure.
●Prophylactic antibiotics – We administer cefazolin preoperatively and continue for 24 hours postoperatively.
●Tocolytics – We administer indomethacin or nifedipine preoperatively and immediately postoperatively.
●Neuroprotection – We begin magnesium sulfate preoperatively in pregnancies approximately 24 weeks or more and continue for 24 hours.
●Maternal anesthesia and analgesia
•An epidural catheter is placed preoperatively to use for postoperative pain control.
•General anesthesia is initiated with propofol. A moderate to high level (1-3 minimum alveolar concentration [MAC]) of an inhalational agent (generally sevoflurane) is used to maximize uterine relaxation.
●Fetal anesthesia and analgesia – The fetus is given an intramuscular injection of fentanyl for pain control, vecuronium for immobilization, and atropine to suppress the fetal stress response (eg, bradycardia) before fetal surgery is begun.
●The fetal heart rate is monitored repeatedly during the surgery.
●A no-touch technique is used when handling the neural placode.
Approaches to in utero repair
Open hysterotomy — The open fetal procedure involves maternal laparotomy. The uterus is exteriorized to facilitate surgical access and intraoperative imaging.
The ultrasound transducer is placed directly onto the myometrium and the edge of the placenta is identified and marked using cautery. If necessary, the fetus is externally verted to an appropriate presentation and positioned underneath the intended hysterotomy site prior to opening the uterus.
A large (8 cm) hysterotomy incision is made as far from the placenta as possible, with the orientation of the uterine incision parallel to the marked edge of the placenta to minimize the risk of extension into the placenta. If the placenta is anterior, uterine entry is either fundal or posterior; if the placenta is posterior, uterine entry is anterior. It is imperative to avoid placental bed bleeding since it cannot be easily controlled. The membranes are stapled to the uterine wall at the edges of the hysterotomy. Bleeding between the membranes and uterine wall leading to a subchorionic hematoma is a potentially serious complication of hysterotomy creation, with the potential for membrane dissection and abruption. Such bleeding can be catastrophic and require delivery and maternal blood transfusion.
Once the hysterotomy has been created, the NTD is exposed in the opening. A catheter is placed in the uterus for infusing warm saline or lactated Ringer solution to maintain intrauterine fluid volume, which will prevent umbilical cord compression and fetal cooling during the procedure.
An elliptical skin incision is made around the defect and carried down to the fascia. Extraneous skin and membrane are excised, a water-tight closure is performed, and the skin is closed. Different centers have evolved modifications of the original description of the technique described in the MOMS trial, which involved directly closing the dura when possible, otherwise with a collagen-based dural graft.
The uterus is closed in two layers with absorbable sutures. Interrupted stay sutures are placed first, but not tied, and then a continuous suture is placed to close the hysterotomy. Prior to tying the continuous suture, a catheter is used to refill the amniotic cavity with saline or lactated Ringers containing nafcillin under ultrasound guidance. The fluid is replenished to a level of low normal fluid and then the stay sutures are tied. When the suture line is confirmed to be hemostatic and watertight, the abdominal wall is closed in layers in the usual fashion.
●Outcome data – The most recent available data for open hysterotomy from over 1000 cases reported to the North American Fetal Therapy Network (NAFTNet) Fetal Myelomeningocele Consortium registry show comparable or better maternal, obstetric, and fetal/neonatal complication rates and outcomes among participating centers than those previously reported in the 2011 MOMS trial [45]. Chorioamniotic separation (CAS) was seen in 10.6 percent, preterm prelabor rupture of membranes (PPROM) in 23.8 percent, and preterm labor in 20.4 percent of cases. The modified hysterotomy closure suturing technique (ie, utilizing a third imbricating layer providing serosal-to-serosal apposition) has resulted in a low rates (≤5 percent) of uterine thinning and dehiscence [42].
Mini-hysterotomy — The mini-hysterotomy technique is similar to the open hysterotomy approach (laparotomy, exteriorization of the uterus, and placental border marking), except the length of the uterine incision is much smaller.
The fetus is gently moved by external manipulation guided by ultrasound so that the spinal defect is located against the planned hysterotomy site. An electric blade is used to make a 2.5 to 3.5 cm hysterotomy that is superficial to the fetal defect and at least 2 cm away from the border of the placenta. The membranes are sutured to the inner layer of the myometrium, and a neonatal retractor is used to keep the hysterotomy open. Some surgeons have suggested utilizing of a plastic wound retractor (Alexis) at the mini-hysterotomy site [37].
Since the size of the NTD may be larger than the hysterotomy orifice, the fetus is constantly and carefully moved so that a specific portion of the lesion is accessible to the surgeons. The lesion is dissected and repaired in the same manner as described in the open hysterotomy approach.
Uterine closure is performed in two layers. Prior to tightening the last stitch, warm saline or lactated Ringer solution containing nafcillin is infused into the amniotic cavity to achieve normal amniotic fluid volume by ultrasound. The reminder of surgery is similar to the open hysterotomy approach.
●Outcome data – The mini-hysterotomy technique results in comparable benefits as the open hysterotomy approach [48], but facilitates surgery at earlier gestational ages, which is advantageous because earlier repair is a predictor of the ability to walk. It is also associated with a higher gestational age at delivery (35.3 weeks) compared with the MOMS trial (34.1 weeks) [9]. In a systematic review, compared with the standard open hysterotomy technique (n = 181), mini-hysterotomy (n = 176) was associated with a lower risk of preterm birth (21.4 versus 47.3 percent) and lower frequency of requirement for ventriculoperitoneal shunt placement (13.0 versus 29.1 percent), while maintaining comparable frequency of reversal of hindbrain herniation (78 versus 72.5 percent) and need for additional postnatal surgery (0 versus 5.8 percent) [51].
Laparotomy-assisted fetoscopy — Laparotomy-assisted fetoscopic repair combines laparotomy and percutaneous approaches and is a promising alternative to open hysterotomy repair because it appears to have a lower risk of uterine thinning and dehiscence and thus enables a higher rate of vaginal birth in the index and future pregnancies [52]. In this approach, a lower transverse or midline vertical laparotomy incision is made and the uterus is exteriorized; two to three F vascular access ports placed in the uterine wall allow fetal access with the endoscope and operating instruments [53-57]. Anchoring sutures are placed under ultrasound guidance to ensure that they are placed through the membranes and not partially through the uterine wall. Once the sutures have been placed, the first port is situated in the center of the anchoring sutures under ultrasound guidance using the Seldinger technique. The port should be placed 2 to 5 cm from the placental edge at the highest point on the uterus to obtain the best view of the lesion. The amniotic fluid is removed and replaced with heated and humidified CO2, which does not cause fetal acidemia and is safe for the fetus [44,58-60]. The fetoscope is then introduced into the gas-filled amniotic cavity and the fetus is positioned in the preferred orientation for the surgery under direct vision through the fetoscope. A second port (and a third port if needed) is then placed under direct visualization. The fetal head is guided into the maternal pelvis and the fetus is maintained in a cephalic presentation with the spine anterior for the entire procedure.
The fetal repair differs among centers but involves dissection of the neural placode, obtaining a water-tight closure with or without a patch, and layered suturing. The CO2 is then evacuated and the amniotic cavity is refilled with warm saline or lactated Ringer's solution containing nafcillin. The access ports are removed and the uterine openings are closed with two single monofilament absorbable stitches. The abdominal wall is closed in layers per routine.
●Outcome data – Retrospective studies demonstrated noninferiority to the open hysterotomy approach, with later gestational age at birth (37.6 weeks versus 34.1 weeks in the MOMS trial), less uterine scar associated complications, and high vaginal birth rates in the index pregnancy (50 percent; by comparison, vaginal birth is contraindicated after open hysterotomy) [61-63]. In the International Fetoscopic Myelomeningocele Repair Consortium cohort of 300 patients who underwent prenatal percutaneous or fetoscopic repair, the rate of preterm prelabor rupture of membranes (PPROM) was higher than in studies of open repair (55 versus 32 to 46 percent), although the gestational age at birth was approximately 34 weeks for both groups [63].
Percutaneous approaches
●Direct percutaneous technique – A direct percutaneous technique is also being developed [64,65]. This approach differs from the above techniques because it does not involve either laparotomy or hysterotomy. The uterine cavity is accessed using a percutaneous Seldinger technique. Three size 11 to 16 French trocars are placed into the uterine cavity under ultrasound guidance. Partial amniotic CO2 insufflation is performed similar to the laparotomy-assisted fetoscopic approach. The fetus is then positioned using laparoscopic instruments under endoscopic visualization. The neural placode is released and dissected; skin is further undermined to allow approximation at the midline. The leading group performing this surgery places a biocellulose patch over the placode, followed by a single layer skin closure [46]. If direct skin closure cannot be achieved, a skin substitute is placed over the biocellulose patch. The uterine puncture sites do not require closure to seal. The gas is removed, and amnioinfusion is performed with warm saline or lactated Ringers containing nafcillin to replace the volume removed. The cannulae are removed and skin sutured in a standard fashion as for any laparoscopic surgery.
•Outcome – No direct comparison between the percutaneous and open hysterotomy approaches have been published, but advocates of the percutaneous method suggest it mitigates the risks of open maternal-fetal surgery because of less anesthesia risk, improved maternal postsurgical recovery, lower rates of uterine dehiscence, and higher rates of safe vaginal birth [66]. On the other hand, the percutaneous technique does not allow plicating the chorioamniotic membranes, a limitation associated with earlier gestational age at delivery compared with the three approaches described above (32 weeks compared with ≥34 weeks with other approaches). To address this issue, a modification has been introduced that includes mini-laparotomy with placement of sutures to secure the membrane prior to percutaneous insertion of the 5 mm port (discussed below).
●Percutaneous mini-laparotomy technique – This technique is similar to the direct percutaneous repair but adds preliminary steps to allow fixation of the membranes prior to trocar placement [67]. The mini-laparotomy incision is made at the expected site of the camera port. A 2 to 3 cm skin incision is made sharply with a scalpel in the abdominal wall. The incision is carried down to the fascia with cautery. Two stay sutures are placed in the fascial layer with the tails left long, and the fascial layer is incised between the 2 stitches. The abdominal wall layers are split bluntly and held open with retractors. The peritoneum is identified, and the abdominal cavity is entered. The uterus is identified and palpated. A wound retractor (Alexis Wound Protector) is placed to provide direct visualization and surgical access to the uterus. Two sutures are placed through the uterine wall at the anticipated location of the 5 mm port site. The trocars are then placed into the uterine cavity under ultrasound guidance by the Seldinger technique, exactly thorough the sutured locations. The remainder of the surgery is similar to the description of the percutaneous approach.
•Outcome data – With this technique, PPROM occurred in 57 percent of cases at an average gestational age of 33.4 weeks, the average gestational age at delivery was 36.1 weeks, and 71 percent of patients gave birth vaginally [67]. While these results seem promising, data are available on only a small group of patients from a single institute.
POSTOPERATIVE CARE AND PREGNANCY MANAGEMENT
●Tocolysis – All patients receive postoperative tocolysis per protocol.
●Analgesia – Analgesia is provided via an epidural catheter initially and then orally. After 48 hours, pain is usually well controlled with acetaminophen.
●Prenatal care – In most centers, patients are discharged once stable on post operative day 3 to 4 and return weekly for follow-up. Patients that experience no complication and are stable can travel back home and have the remainder of the pregnancy managed locally [68]. Although some programs have required that the patient remain close to the performing center for the remainder of pregnancy, over time, more obstetricians have become comfortable managing these patients at their local offices.
●Antepartum fetal surveillance – Fetal imaging is described below. (See 'Follow-up imaging' below.)
Weekly biophysical profiles and Doppler studies (umbilical artery, middle cerebral artery, ductus venosus) are performed beginning at 34 weeks of gestation, or earlier if standard indications arise.
●Chorioamniotic separation (CAS) and preterm prelabor rupture of membranes (PPROM) – In the event of CAS, close fetal monitoring and maternal admission are considered for early identification and management of complications (eg, umbilical cord accident).
Patients who with PPROM <34 weeks of gestation are managed conservatively in the hospital, according to standard guidelines. (See "Preterm prelabor rupture of membranes: Management and outcome".)
FOLLOW-UP IMAGING
●Fetal – Monthly comprehensive ultrasound examinations are performed beginning at 30 to 32 weeks of gestation and a fetal MRI is performed six weeks postprocedure to evaluate for hindbrain herniation reversal.
Ultrasound and MRI can both be utilized to assess the success of in utero spina bifida repair. Approximately six weeks after closure of the open spina bifida, the appearance of the Chiari II malformation should improve, with ascent of the hindbrain structures from the cervical spinal canal back into the posterior fossa and return of cerebrospinal fluid (CSF) to the fourth ventricle and retrocerebellar space. Reaccumulation of fluid in the supratentorial compartment should also occur, with an increase in size of the extra-axial CSF spaces and mild increase in lateral ventricular size on the order of 2 to 3 mm compared with preoperative imaging.
If the appearance of the posterior fossa does not improve post-repair, the lower back repair site should be evaluated to look for any evidence of a persistent leak. A CSF leak may accumulate at the repair site and form a fluid collection, known as a pseudomeningocele. Even without a fluid collection, the lack of hindbrain herniation reversal should raise concern for persistent CSF leak at the repair site.
Postoperative evaluation of the clivus-supraoccipital angle, transcerebellar diameter, and lateral ventricle size by MRI and ultrasound are good predictors of the need of postnatal CSF diversion due to progressive hydrocephalous [69]. A significant increase in lateral ventriculomegaly post-repair (greater than the expected 2 to 3 mm) may indicate obstruction of the ventricular outflow, as from aqueductal stenosis. Typically, these cases are associated with severe ventriculomegaly prior to repair. It should be noted that in fetuses with Chiari II malformation and aqueductal stenosis, the expected dilatation of the third ventricle may not be present.
●Maternal – Any time the uterus is imaged after fetal intervention, the uterine surgical site should be assessed. In the MOMS trial, one-third of patients had a very thin myometrium or partial dehiscence at the hysterotomy site at the time of delivery [9]. The management of these cases needs to be individualized as dehiscence could lead to uterine rupture before or during delivery.
DELIVERY — We deliver all patients who underwent an open hysterotomy repair by cesarean birth at 36+0 to 37+0 weeks, if still undelivered [9]. Patients who underwent any of the fetoscopic methods and have no standard contraindications for vaginal birth are induced at 39+0 to 39+6 weeks [52]. Induction should be managed carefully, similar to induction of labor after cesarean birth. (See "Trial of labor after cesarean birth: Intrapartum management", section on 'Induction of labor'.)
NEWBORN CARE — At birth, all neonates should be admitted to the neonatal intensive care unit or other specialized neonatal care unit for comprehensive multidisciplinary evaluation and respiratory support, if needed. The lesion and its repair will be carefully inspected to identify cerebrospinal fluid (CSF) leakage. Transcranial ultrasonography should be considered within the first 48 hours of birth to evaluate the degree of ventriculomegaly. We advise ongoing close monitoring for the development of hydrocephalus. This monitoring should be arranged prior to hospital discharge and is optimally conducted at a comprehensive multidisciplinary clinic specialized in the care of infants with spina bifida.
SUBSEQUENT PREGNANCY — Obstetric ramifications on subsequent pregnancies depend on the repair method used.
●The open hysterotomy approach has been associated with a 14 to 21 percent rate of uterine dehiscence and 7 to 14 percent rate of uterine rupture, comparable to those reported for pregnancies after classical cesarean delivery [70-72]. Hence, patients need to be aware of any symptoms of labor, clinicians need to carefully evaluate for signs of uterine rupture, and cesarean birth should be planned for early term.
Recent unpublished data from the North American Fetal Therapy Network (NAFTNet) Fetal Myelomeningocele Consortium registry suggest that maternal, obstetric, and fetal/neonatal outcomes among participating centers are comparable to or better than those originally reported in the MOMS trial. They reported significantly lower rates of uterine dehiscence or rupture in subsequent pregnancy, as low as 5 percent. While these data need to undergo peer review, they appear to show ongoing advances since the MOMS trial [42]. Preliminary results of pregnancies after laparotomy-assisted repair show an 80 percent vaginal birth rate with no significant complications [73]. By comparison, the vaginal birth rate after mini-hysterotomy is about 50 percent and 0 percent after open hysterotomy.
●For the percutaneous approach, while rates of uterine thinning/dehiscence are reported to be low, data regarding subsequent pregnancies are still lacking.
ONGOING STUDIES AND FUTURE IMPLICATIONS — It is important to remember that no prenatal or postnatal repair method provides a cure for the condition and lifetime morbidity is significantly increased in this population. Preliminary results from ongoing research is promising. For example:
●The Cellular Therapy for In Utero Repair of Myelomeningocele (CuRe) Trial offers cellular therapy for in utero repair of spina bifida. In this trial, mesenchymal stromal cells are harvested from the placenta, suspended in culture, and seeded on an extracellular matrix. They are eventually placed in utero on a form of synthetic scaffold over the spinal lesion. Pilot studies in animal models were promising [74-76].
●Another study used cryopreserved human umbilical cord (HUC) as a meningeal patch for in utero spina bifida repair [77]. Use of HUC demonstrated superior organization for regenerative cellular growth when compared with a collagen patch, and thus appeared to be a better patch material for in utero spina bifida repair, potentially reducing tethering and improving spinal cord function.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Congenital malformations of the central nervous system".)
SUMMARY AND RECOMMENDATIONS — Open spina bifida is a major congenital anomaly associated with significant morbidity. Treatment options include in utero or postnatal repair. Repair reduces but does not eliminate morbidities associated with the condition. (See "Myelomeningocele (spina bifida): Anatomy, clinical manifestations, and complications" and "Myelomeningocele (spina bifida): Management and outcome".)
●Goal of in utero repair – The technical goal of in utero spina bifida repair is to create complete, water-tight coverage of the defect, with the objective of halting progressive intrauterine neuronal injury. (See 'Goal and outcome of in utero surgery' above.)
●Benefits of in utero rather than postnatal repair include (see 'Goal and outcome of in utero surgery' above and 'Parental counseling' above):
•Decreased need for cerebrospinal fluid (CSF) shunting
•Improved motor development and function
•Possible improvement in bladder function
•Modest improvements in quality of life and impact on family
●Pregnancy complications of in utero repair include increased risks of (see 'Potential perioperative maternal and obstetrical risks' above and 'Potential fetal and postnatal pediatric risks' above):
•Chorioamniotic separation (CAS)
•Chorioamnionitis
•Uterine bleeding
•Preterm prelabor rupture of membranes (PPROM)
•Preterm labor
•Placental abruption
•Fetal injury
•Uterine dehiscence or rupture
●Candidates for in utero repair – Most fetal surgery centers performing open spina bifida repair generally base their inclusion and exclusion criteria on those used in the Management of Myelomeningocele Study (MOMS) trial, but modifications of the original parameters are common. Three key inclusion criteria are:
•Open spina bifida with the upper boundary between T1 and S1
•Hindbrain herniation
•Gestational age 19+0 to 25+6 weeks of gestation at the time of the procedure
Exclusion criteria vary among centers but relate to baseline risk for preterm birth and technical factors. (See 'Inclusion and exclusion criteria for consideration of repair' above.)
●Preoperative fetal evaluation – Preoperative fetal evaluation should include a detailed ultrasound examination and/or MRI to confirm open spinal bifida and evaluate the defect, associated brain findings, and neurologic implications. (See 'Preoperative fetal neurological evaluation' above.)
●Approaches for in utero repair – Prenatal repair should only be offered in highly specialized level 3 fetal centers. Prenatal repair approaches include open hysterotomy, mini-hysterotomy, laparotomy-assisted fetoscopic, and percutaneous fetoscopic repair. Detailed discussion of the available approaches and associated adverse outcomes will allow more informed consent and decision-making, enabling parents to choose the management option that best fits their needs and expectations. While limited data comparing the different neurosurgical repair methods suggest they have comparable fetal neurologic outcomes, all centers do not offer the same method and the methods differ significantly in the rate of maternal complications, gestational age at delivery, and chances of vaginal birth. (See 'Centers for fetal surgery' above and 'Procedure' above.)
●Delivery – We deliver all patients who underwent open hysterotomy repair by cesarean birth at 37+0 weeks, if still undelivered. Patients who underwent any of the fetoscopic methods with no standard contraindications for vaginal birth are induced no later than 39+0 weeks. Induction should be managed carefully, similar to induction of labor after cesarean birth. (See 'Delivery' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
