Introduction
Fetal Therapy
Intrauterine fetal surgery is a new and rapidly evolving discipline. During fetal surgery the mother is an “innocent bystander” as her fetus is treated. However, the mother must also receive medical care in the course of the fetal therapy, and both she and the fetus are at risk for complications. Some fetal surgical procedures are performed in a minimally invasive fashion, with small sheaths introduced percutaneously through the maternal abdomen and uterus, while other fetal procedures require a full maternal laparotomy, exposure of the uterus, and a maternal hysterotomy. In some cases, the procedures are done in the middle of gestation, and the goal is for the fetus to continue to heal and develop in utero. In other cases, a near-term fetus is delivered at the end of the fetal procedure.1 As these procedures involve two high-risk populations, frank discussions of the risk and benefit are required. These situations are complex, and as such, fetal therapy is often offered only when the fetus is at risk of death. A variety of processes often end in a final common pathway of hydrops fetalis. The fetal diseases that may cause hydrops and that may be treated in utero are wide-ranging. Some examples include twin-twin transfusion syndrome, massive lung tumors, and sacrococcygeal teratoma.2 While many types of fetal neurological pathology may be detected with high-level fetal ultrasound and fetal magnetic resonance imaging, the only candidates for fetal neurosurgical therapy thus far have included fetal hydrocephalus and myelomeningocele (MMC). The fetal neurosurgical patient is different from other fetal surgical patients since the treatable fetal neurosurgical lesion is rarely life-threatening. Since the fetus is not at immediate risk of demise, the risks and benefits must be even more carefully considered before undertaking fetal neurosurgical therapy.
Hydrocephalus
Treatment of hydrocephalus in the postnatal period commonly involves diversion of the cerebrospinal fluid (CSF) from the cerebral ventricles into the peritoneal cavity. In the fetus, the CSF can be diverted into the amniotic fluid with a valved catheter that prevents retrograde flow of the amniotic fluid into the ventricles. Between 1982 and 1985, 41 fetuses with hydrocephalus secondary to a variety of conditions were treated with ventriculoamniotic shunts between 23 to 33 weeks of gestation. These shunts can be placed percutaneously, in a manner similar to intrauterine blood transfusion. While there were 34 survivors, the postnatal outcomes of the treated fetuses were not as encouraging as hoped, with more than half of the survivors left with serious neurologic handicaps. Only 35% of the survivors were found to be developing normally at follow-up.3 Between 1999 and 2003, fetal ventriculoamniotic shunting was attempted in an additional four fetuses. These shunts were placed in an “open” fashion, via a maternal laparotomy and hysterotomy. Outcomes of these fetuses were not encouraging either, and the authors concluded that without new developments in the field, fetal ventriculoamniotic shunting would not be likely to offer any benefits.4 The treatment of fetal MMC has seen much more success than that of hydrocephalus, and the remainder of this chapter focuses on fetal neurosurgical therapy for MMC.
Myelomeningocele
MMC is a neural tube defect that currently affects 5–10 pregnancies per 10,000 in the United States.5 With routine prenatal screening, these defects can be detected in utero. Early studies in animal models showed that closure of these defects in utero improved neurological outcomes.6 Since the conclusion of the Management of Myelomeningocele Study (MOMS) trial, which demonstrated the benefits of prenatal compared to postnatal repair of MMC,7 fetal repair of MMC is now being offered at centers around the world. Performing fetal neurosurgery has unique maternal and fetal considerations. As the number of fetal procedures increases, understanding these unique considerations will be important for the anesthesiologist.
MMC develops in the 3rd and 4th weeks of gestation when the embryonic neural plate fails to close along its length. This results in an open spinal canal and exposure of the spinal cord to the amniotic compartment. Babies born with MMC require closure within the first few days of life, and many are faced with lifelong disabilities including motor and sensory dysfunction, bowel and bladder dysfunction, sexual dysfunction, cognitive delay, Arnold-Chiari type II malformation, hydrocephalus, and tethered cord. The degree of neurological dysfunction is associated with the level of the vertebral defect. Higher anatomic defects are therefore more neurologically significant than lower defects.
Studies suggest that nerve damage is progressive over time. The two-hit theory of nerve damage has been postulated in which the first hit is the failure of the neural tube to form correctly, and the second hit is the damage caused by continued exposure of the nerves to the uterine environment. Early animal studies demonstrated that fetal closure of MMC resulted in improved neurological outcomes, presumably by decreasing the duration of exposure of the nerves to the uterine environment.6 The MOMS trial was a multicenter randomized prospective clinical trial designed to evaluate the safety and efficacy of in utero repair of MMC. In order to be eligible for the trial, fetuses had to be between 19 and 26 weeks gestational age, have an upper MMC border of T1–S1, be a singleton pregnancy, and have normal karyotype. Maternal body mass index (BMI) had to be < 35 kg/m2. The study was stopped early for efficacy. It demonstrated that in-utero repair decreased the need for ventriculoperitoneal (VP) shunt by 12 months of age and improved motor outcomes at 30 months of age when compared to postnatal repair.7 Longer term follow-up (median follow-up time of 10 years) of non-MOMS trial fetal MMC repair patients showed that there is long-term improvement of functional status.8 However, fetal surgery is not without risk to both the mother and fetus. Therefore, the anesthesiologist must understand the maternal and fetal physiology (summarized in Tables 10.1 and 10.2) and perioperative management unique to this type of surgery.
Physiologic State | Clinical Implication |
---|---|
Cardiovascular | |
Myocardium higher proportion of noncontractile elements | Cardiac output very dependent on heart rate, minimally responsive to preload |
Uteroplacental blood flow related to uterine perfusion pressure (the difference between uterine arterial and uterine venous pressure) and inversely related to uterine vascular resistance | Maternal hypotension, aortocaval compression, and uterine contractions decrease uterine blood flow; effects of vasopressors, vasodilators, and neuroaxial and volatile anesthetic agents on uterine blood flow can be variable because they affect both uterine arterial pressure and uterine vascular resistance |
Fetal oxygen delivery dependent on ratio of maternal to fetal blood flow, oxygen partial pressure gradient, respective hemoglobin concentrations and affinities, placental diffusing capacity, and acid-base status of fetal and maternal blood (Bohr effect) | Many factors can disrupt fetal oxygenation |
Hematologic | |
Blood volume per kg of weight higher than in adult; 2/3 of blood volume on placental side of fetoplacental unit | Need greater blood volume for euvolemia |
Fetus produces own coagulation factors, plasma concentrations of coagulation factors increase with gestational age | Fetus more prone to hemorrhage |
Temperature regulation | |
Dependent on maternal temperature, cannot thermoregulate on own | Maintaining euthermia during open fetal surgery can be difficult; need to infuse normothermic solution of lactated ringers into amniotic cavity |
Neurologic | |
Neurologic pathways for cortex are developing into the third trimester | Anesthetic requirements for fetus less than that for child |
Preoperative Assessment and Preparation
Preparation for fetal neurosurgery is a multidisciplinary endeavor, involving maternal fetal medicine specialists, pediatric surgeons, pediatric neurosurgeons, radiologists, neonatologists, anesthesiologists, psychologists, social workers, and nurses. A complete maternal workup including history and physical exam with specific attention paid to symptoms such as dyspnea, syncope, and dizziness that may indicate undiagnosed underlying morbidity and labs including complete blood count, coagulation panel, and type and cross for blood are warranted. Particular attention should be paid to the airway and spine exams.
A complete fetal workup including diagnostic imaging (both ultrasound and magnetic resonance imaging (MRI)) to determine location and extent of fetal lesion, testing to exclude other anomalies, and preparation of O-negative, leukocyte depleted, irradiated, CMV negative blood cross matched to the mother is needed. The cross match is needed as maternal antibodies to red blood cells can cross the placenta. The inclusion criteria at the authors’ institution have changed slightly since the MOMS trial, and currently, in order to be eligible for the surgery, the fetus must meet the following criteria: upper level of MMC must be between T1 and S1 with hindbrain herniation, gestational age at time of surgery between 23 weeks and 25 weeks 6 days, singleton pregnancy, normal karyotype, elevated amniotic fluid alpha fetoprotein, positive acetylcholinesterase, and lack of other significant anomaly such as cardiac disease. Maternal exclusion criteria include the following: age less than 18 years old, insulin-dependent pregestational diabetes, cerclage or incompetent cervix, placenta previa, placental abruption, short cervix < 20 mm, BMI > 40 kg/m2, previous spontaneous delivery at < 37 weeks, maternal-fetal Rh isoimmunization, Kell sensitization, or history of neonatal alloimmune thrombocytopenia, maternal HIV or hepatitis B or C, uterine anomaly, hypertension that would increase risk of preterm delivery, other maternal condition that would contraindicate surgery, inability to comply with follow-up requirements after surgery, and lack of support or failure to meet other psychosocial criteria. It is imperative that the family be extensively counseled so that they understand all the potential risks and benefits.
Anesthetic Management
Anesthetic management of open fetal surgery requires an integration of cognitive and manual skills from the worlds of both pediatric and obstetric anesthesia. As with any anesthetic, communication and teamwork are crucial. Also crucial is a full understanding of the technical aspects of the surgical procedure and the infrastructure of the fetal care center. Many conflicting needs must be balanced. Some of the issues include finding a balance between providing adequate fetal perfusion and optimal operating conditions and balancing the needs of anesthetizing two patients at the same time. Measures to combat an increased risk of pulmonary edema also make the task of providing adequate fluid resuscitation harder. These issues are highlighted as a typical anesthetic for fetal MMC repair is described.
On the day of surgery, prior to induction, the mother should have a high lumbar/low thoracic epidural placed for postoperative analgesia. The abdominal incision for the surgery is higher than the incision performed for a cesarean section. The epidural should be tested to evaluate for intrathecal or intravascular placement, but it should not be fully dosed until the end of the surgical procedure. The patients often get either high-dose volatile anesthetic or nitroglycerin to relax the uterus, and the sympathectomy caused by an epidural block will exacerbate intraoperative hypotension. Indomethacin is given for tocolysis. H2 receptor blockade and nonparticulate antacid are given for aspiration prophylaxis.
The patient is brought into the operating suite and positioned supine with left uterine displacement. Standard American Society of Anesthesiologists (ASA) monitors are placed. After preoxygenation, a rapid sequence induction is performed, followed by oral intubation. A second intravenous line and an arterial line are placed. While blood loss is often quite low, the gravid uterus can be a source of catastrophic bleeding. The invasive measurement of blood pressure is warranted as vasopressors must be titrated in a patient receiving high-dose volatile anesthetics or nitroglycerin. A Foley catheter is placed and ultrasonography is used to confirm fetal position. A manual version of the fetus may be necessary to get the MMC in the proper position.
Anesthesia is maintained with volatile agent. Desflurane is usually used for quicker titratability, but isoflurane and sevoflurane have also been used. Administration of high doses of volatile anesthetic has been alluded to previously. The need for profound uterine atony is a unique consideration for open fetal surgical cases. The lower the uterine tone, the lower the resistance to blood flowing from the maternal circulation, and the better the operating conditions. It is also postulated that lower uterine tone will decrease the risk of placental abruption. Following maternal skin incision and exposure of the uterus, the volatile agent is increased until the uterine tone is (subjectively) judged to be low enough by the surgical team. Greater than 2 minimum alveolar concentration (MAC) of volatile agent is often required before uterine incision. If needed, nitroglycerin, either as bolus or infusion, can be used to further relax the uterus. The high levels of volatile anesthetic and potential use of nitroglycerin may cause significant vasodilation and hypotension in the mother. At the authors’ institution, fluid administration is often kept to less than 500 mL to decrease chances of pulmonary edema. In the face of this fluid restriction, maternal blood pressure is augmented with a phenylephrine infusion. Additional boluses of ephedrine and phenylephrine are given as needed. Maternal blood pressure is typically maintained within 10% of maternal baseline blood pressure. Exposing a fetus to volatile anesthetic can cause myocardial depression.9, 10 There is evidence from some centers that use of supplemental intravenous anesthesia (propofol and remifentanil), in addition to a lower dose of volatile agent (about 1 MAC volatile agent), may decrease fetal cardiac dysfunction and fetal acidosis.11, 12
The edges of the placenta must be clearly identified by ultrasound before any uterine incision is made. When an appropriate site on the uterus has been identified, a small uterine incision is made and a stapling device is used to extend the hysterotomy while preventing bleeding by sealing the membranes to the decidua and myometrium. There can be significant bleeding at this point if the device misfires, given the vascularity of the uterus and the uterine atony, or if venous sinuses are encountered. Small infusion catheters are placed into the now open amniotic space to infuse warmed, lactated ringers solution to replete amniotic leakage and maintain fetal and umbilical cord buoyancy. An intramuscular injection of opioid (fentanyl 20 mcg/kg) and muscle relaxant (vecuronium 0.2 mg/kg) is administered to the fetus by the surgeons. Although the fetus is being anesthetized by the halogenated agent via placental transfer from the mother, the additional fentanyl and vecuronium can further decrease fetal stimulation and ensure immobility, optimizing surgical conditions. After the hysterotomy is complete, the fetus is positioned so that the MMC is exposed in the hysterotomy window (Figure 10.1). The neurosurgeon then repairs the MMC by exposing and separating the neural structures, closing the native dura over the spinal cord when possible, closing the paraspinal myofascial flaps, and then closing the skin, either with a primary closure (Figure 10.2a) or with an acellular human dermis graft (Figure 10.2b).