Anemia and Intrauterine Transfusion

The incidence of fetal anemia secondary to rhesus D (RhD) sensitization has decreased since the introduction of RhD immunoglobulin prophylaxis in the late 1960s, to rates of approximately 1 in 1000 pregnancies. However, other red blood cell (RBC) antigens, parvovirus B19 infection, maternal-fetal hemorrhage, and homozygous thalassemia also cause fetal anemia and combine to reach a rate of approximately 6 cases in 1000 live births. Although spectral analysis examining bilirubin levels in serial amniotic fluid samples was originally used to detect fetal anemia and determine the timing of therapy, most treatment centers currently rely on noninvasive Doppler studies of the middle cerebral artery (MCA). An increased peak MCA blood flow velocity more than 1.5 multiples of the median is an accurate threshold in detecting moderate-to-severe fetal anemia requiring intervention. The value of this peak velocity threshold may be increased with each serial transfusion treatment to decrease the false positive rate. Fetal blood sampling from the umbilical vein just before starting the intrauterine transfusion (IUT) is the gold standard for determining the degree of fetal anemia. IUT is not used before 18 weeks’ GA because umbilical vein access is not reliable. For cases requiring earlier intervention, fetal intraperitoneal transfusion of red cells may initially be the intervention of choice.

IUTs are typically performed using local anesthesia and require minimal maternal sedation and analgesia. However, the anesthesiologist should be prepared for an emergent cesarean delivery at any time during the procedure if the fetus is at a viable GA. Using ultrasound image guidance, a 20- or 22-gauge needle is used to access the umbilical vein. The access point is typically near the placental cord insertion to provide stability ( Fig. 63.1 ). Puncture of an umbilical artery rather than the umbilical vein is associated with prolonged bleeding and fetal bradycardia secondary to spasm. Occasionally a free loop of the umbilical cord or intrahepatic portion of the umbilical vein may be used. The umbilical cord has no known pain receptors, but needle access of the intrahepatic portion of the umbilical vein stimulates pain receptors with passage of the needle into the fetus. Fentanyl attenuates the fetal stress response from intrahepatic fetal needle placement. In one study, fetal stress hormone changes with IUT were unrelated to site of needle placement; however, these results are difficult to interpret because hormone levels may be affected by changes due to the underlying fetal anemia and hemodynamic alterations associated with intravascular volume expansion. Given this uncertainty, fentanyl (10-20 μg/kg) is administered intramuscularly to the fetus before use of an intrahepatic approach. There is some evidence to suggest that fetal anesthesia does not alter MCA peak systolic velocity flow patterns following transfusion.

Diagram of cordocentesis procedure used in intrauterine transfusion.

Inset window shows tip of 22-gauge spinal needle entering the umbilical vein.

Redrawn from Ralston SJ, Craigo SD. Ultrasound-guided procedures for prenatal diagnosis and therapy. Obstet Gynecol Clin North Am. 2004;31:101–123.

An intramuscular muscle relaxant (e.g., rocuronium 2.5 mg/kg) can be administered to the fetus to decrease the likelihood of fetal movement that could dislodge the needle or sheer the umbilical vein. If a muscle relaxant is administered directly into the umbilical vein the dose of muscle relaxant can be reduced (e.g., rocuronium 1.0 mg/kg). The volume of type O, rhesus (Rh)-negative, irradiated, viral screened, packed RBCs transfused is estimated from the GA, estimated fetal weight, donor unit hemoglobin (Hb), and pretransfusion Hb. The rate of transfusion is typically 5 to 10 mL/min with a target hematocrit of 45% to 55%. Steady intravascular location of the needle tip can be assessed with Doppler imaging throughout the transfusion injection. Periodic sampling is used to guide the final transfusion volume. After IUT therapy, fetal Hb levels decrease approximately 0.3 g/dL/day and multiple repeat IUTs are typically required at 1- to 3-week intervals, depending on the rate of Hb decline.

Perinatal fetal loss rate is approximately 2% per IUT, and transient fetal bradycardia (8%) is a common complication. Other complications including emergency cesarean delivery, intrauterine infection, preterm PROM, and preterm delivery occur in approximately 3% of IUT procedures. Although survival rates are significantly less for hydropic fetuses, recent published overall survival rates with IUT typically exceed 95%. A long-term outcome study of 291 children (median age of 8.2 years with a range of 2-17 years of age) who underwent IUT during gestation for hemolytic disease found a 4.8% rate of neurodevelopmental impairment, including cerebral palsy (2.1%), severe developmental delay (3.1%), and bilateral deafness (1.0%). Severe fetal hydrops was independently associated with neurodevelopmental impairment.

Congenital Heart Defects

Congenital heart anomalies occur with a frequency of approximately 1/100 live births (see also Chapter 78 ). Ventricular septal defects are the most common cardiac anomaly. The majority of cardiac heart defects are not amenable to fetal intervention. Ultrasound imaging allows diagnosis of cardiac defects as early as 12 to 16 weeks gestation, but is generally performed at 18 to 22 weeks gestation, when obstetric ultrasound assessments are used to screen for other fetal abnormalities.

The majority of fetal surgical cardiac interventions focus on opening a stenotic valve or enlarging a restricted opening. These include (1) aortic balloon valvuloplasty for treatment of critical aortic stenosis and evolving hypoplastic left heart syndrome (HLHS), (2) atrial septostomy for highly restrictive or intact atrial septum seen in HLHS, (3) pulmonic valvuloplasty for pulmonary atresia or intact ventricular septum and hypoplastic right ventricles, and (4) pericardiocentesis to treat congenital cardiac tumors or aneurysms. In utero intervention attempts to halt or reverse the morbid effects of the cardiac lesion before irreversible consequences occur. Early childhood mortality from severe cardiac defects, such as HLHS, remains in the range of 25% to 35%. Survivors have significant associated abnormalities in neurologic development.

The most commonly performed procedure is an aortic valvuloplasty for aortic stenosis with evolving HLHS. Significant aortic stenosis maintains fetal circulation primarily through the low-resistance foramen ovale and diminishes left ventricular development. Selection guidelines for fetal aortic valvuloplasty focus on the presence of significant aortic stenosis, evolving HLHS, and the potential for a technically successful procedure and biventricular postnatal outcome. For this procedure, the fetus is ideally positioned with the left chest anterior, and ultrasound is used to guide the percutaneous passage of an 18- or 19-gauge needle cannula through the uterus, through the fetal chest, and into the apex of the left ventricle ( Fig. 63.2 A ). Maternal local anesthesia infiltration or neuraxial block is typically used for these procedures, and fetal resuscitation drugs must be readily available. In some cases, general anesthesia may be preferred for uterine relaxation to facilitate an external version to change fetal position and improve cannula trajectory. Before cannula insertion, intramuscular fentanyl and a paralytic agent are administered to the fetus with ultrasound guidance, as detailed in the section on “Fetal Anesthesia, Analgesia, and Pain Perception.”

(A) Depiction of ideal fetal position and insertion needle orientation. The fetal left chest is anterior, and the pathway from maternal abdomen entry point to the apex of the left ventricle is unobstructed. (B) This linear cannula course continues from left ventricle apex to the aortic valve.

Redrawn from Tworetzky W, Wilkins-Haug L, Jennings RW, et al. Balloon dilation of severe aortic stenosis in the fetus: potential for prevention of hypoplastic left heart syndrome—candidate selection, technique, and results of successful intervention. Circulation. 2004;110:2125–2131; with permission from Lippincott Williams & Wilkins.

The cannula tip is ideally positioned in the left ventricle, directly in front of the opening of the stenotic aortic valve and aligned with the left ventricular outflow tract. A coronary balloon catheter with guidewire is passed through the cannula into the stenotic valve and positioned in the aortic annulus, where it is inflated and deflated multiple times (see Fig. 63.2 B ). In certain cases, a small laparotomy is used to facilitate improved cannula alignment with the cardiac lesion. Technical success for fetal aortic valvuloplasty is approximately 75% using an angioplasty balloon over a guidewire. [CR] Technical success creates improved left ventricular function, improved aortic and mitral valve development, and birth of a live neonate in 90% of interventions. Fetal complication rates from centers in Linz, Austria ( n = 24) and Boston ( n = 70) for fetal aortic valvuloplasty include: fetal bradycardia (17% and 38%, respectively), pericardial effusion (13% and 14%, respectively), ventricular thrombosis (21% and 15%, respectively), and fetal death (13% and 8%, respectively). A recent systematic review noted complication rates following fetal aortic valvuloplasty of preterm delivery (16%), neonatal death (16%), bradycardia (52%), and hemopericardium (20%). Approximately 40% of technically successful cases result in aortic regurgitation and minimal subsequent left ventricular growth. Biventricular circulation is present at birth in approximately half of the successful cases.

In addition to treatment of aortic stenosis, other cardiac anomalies have been treated in utero. Similar surgical techniques are used for atrial septoplasty and pulmonary valvuloplasty. Outcomes from a small series of fetal atrial septostomies are promising; however, the defect created by the balloon dilation tends to close over time unless a stent is placed (which can be difficult and deployment has been successful in only 44%-62% of the time in a small case series). Pulmonary valvuloplasty for pulmonary atresia and prevention of hypoplastic right ventricle was technically successful in 7 of 11 procedures, but long-term outcome data are unavailable. In utero placement of cardiac pacing has been investigated to treat fetal cardiac arrhythmias unresponsive to conventional management with transplacentally administered antiarrhythmic drugs. Unfortunately, these initial attempts have been frequently unsuccessful.

Obstructive Uropathy

Fetal LUTO affects 2 in 10,000 live births. These obstructions can be bilateral or unilateral and occur at the ureteropelvic junction, at the ureterovesical junction, or at the level of the urethra. If the obstruction is urethral or bilateral, significant developmental consequences occur ( Box 63.1 ). Rates of perinatal mortality in these cases are as high as 90%, and survivors have more than 50% renal impairment.

Posterior urethral valves are the most common cause of congenital bilateral hydronephrosis in male fetuses. Urethral obstruction is the most common cause in females, with other possible causes including ectopic ureter, ureterocele, megacystis megaureter, multicystic kidney, or other complex pathologic processes. Ultrasound investigations of oligohydramnios resulting from decreased fetal urine output easily detect these uropathies with high sensitivity and specificity. Fetal MRI should be considered in cases of severe oligohydramnios as an additional imaging technique to determine the presence of associated fetal anomalies. Grading systems based on ultrasound imaging of renal diameter to determine the severity of hydronephrosis and assessment of urinary tract dilation are used to determine risk stratification and treatment options. Recently, a classification system for LUTO has been proposed that is based on amniotic fluid index, renal imaging, and fetal urine chemistry. The associated morbidity predicted for each type of uropathy depends on obstruction location, duration, gender, and GA at onset. Preterm delivery allows neonatal urinary tract decompression, but morbidity from pulmonary immaturity prevents early intervention and limits the efficacy of this approach.

Poor prognosis is associated with earlier presentation during gestation, more severe oligohydramnios, associated structural abnormalities, and increased concentrations of fetal urine electrolytes, osmolality, protein, and β ₂ -microglobulin. Each case should be thoroughly investigated to determine whether other anomalies are present and if the fetus is a suitable candidate for intervention. If LUTO is corrected postnatally, 25% to 30% of surviving neonates will require dialysis by age 5.

Placement of an in utero fetal vesicoamniotic shunting (VAS) for in utero treatment of LUTO allows decompression of the fetal bladder and urinary tract into the amniotic cavity. VAS prevents urine accumulation, allows normal bladder emptying and development, improves dysplastic renal histologic conditions, increases amniotic fluid volume, improves lung development, and prevents bladder wall fibrosis in animal models. Determination of which human fetuses would benefit from in utero treatment of LUTO remains uncertain. Human VAS placement started in the 1980s in an effort to improve renal development and reduce the pulmonary hypoplasia associated with oligohydramnios. These valveless, double-coiled shunts are inserted percutaneously with ultrasonographic guidance and local anesthesia. One coil remains in the urinary bladder and the other in the amniotic cavity. Prior infusion of fluid into the amniotic cavity can aid in proper shunt placement. Complications associated with these shunts include difficult placement, subsequent occlusion, and position migration (malfunction occurs in up to 60% of cases). Fetal and maternal complications include fetal trauma, iatrogenic abdominal wall injury, gastroschisis, amnioperitoneal leaking, preterm PROM, preterm labor and delivery, and infection. Neonatal survival rates after fetal VAS vary in the literature from 40% to 90%, with approximately 50% of survivors having normal renal function. A metaanalysis of LUTO treatment studies through 2015 noted a perinatal survival advantage with in utero VAS placement compared to standard care (57% vs. 39%), but ultimately there was no difference in renal function or 2-year survival. A multicenter, randomized controlled trial (the Percutaneous Shunting in Lower Urinary Tract Obstruction [PLUTO] trial) compared the perinatal mortality and renal function of fetuses with LUTO treated by either VAS or conservative noninterventional care. The trial was unable to recruit sufficient cases, with only 31 out of 150 desired patients recruited over a 4-year period. Analysis of this smaller enrollment noted a mortality benefit in the fetal treatment group at 28 days, 1 year, and 2 years of age; however substantial morbidity occurred in both groups, leading to only two children with normal renal function at 2 years of age.

Fetal cystoscopy is a recent intervention that allows prenatal visualization of the fetal urethra and ablation of the urethral obstruction. Cystoscopy facilitates diagnosis between LUTO resulting from urethral atresia and posterior urethral valves. This is an important distinction, as urethral atresia is nearly universally lethal and does not improve with VAS placement, while posterior urethral valves are amenable to fetal intervention. Ablation of posterior urethral valves may increase survival compared with expectant management. A case series and retrospective case control study demonstrated that fetoscopic laser ablation for posterior urethral valves can achieve bladder decompression and amniotic fluid normalization. In addition, fetal cystoscopy may improve 6-month survival in severe LUTO compared to no fetal intervention and result in improved renal function at birth when compared to treatment with VAS. Future prospective trials are necessary to validate these initial retrospective results.

Selective fetal intervention with shunting or cystoscopy for LUTO restores amniotic fluid volume, prevents pulmonary hypoplasia, and improves perinatal mortality. However, the effects on medium-term and long-term renal function, neurologic function, bladder function, and other morbidities remain unclear and additional clinical research is needed.

Twin-to-Twin Transfusion Syndrome

Monochorionic twins share the same placenta and typically have intertwin vascular connections that create shared blood flow between the two fetuses. A significant number of these chorionic vascular anastomoses can result in unequal placental blood flow between the two monochorionic twins that can lead to twin-to-twin transfusion syndrome (TTTS). The occurrence of TTTS is approximately 1 to 3/10,000 births. Monochorionic twinning occurs in approximately 20% to 25% of twin pregnancies, with TTTS complicating about 10% of these monochorionic gestations. TTTS usually manifests after the first trimester and is typically recognized at midgestation.

Normally, umbilical arteries carry deoxygenated blood to the surface of the placenta, where gases and nutrients are exchanged with the maternal circulation. The returning venous blood flow is “paired” with the arterial flow, and the two components lie next to each other ( Fig. 63.4 ). This fetal-placental vascular configuration (cotyledon) represents normal anatomy. A variety of abnormal unidirectional and imbalanced vascular connections are present in TTTS ( Fig. 63.5 ). In TTTS, a branch of an umbilical artery descends into the placenta and cotyledon, but instead of connecting with a paired vein, it connects with a vein that brings blood to the other twin, and this results in an arteriovenous anastomosis between the two fetuses. Although some intertwin arteriovenous vascular architecture is found in 90% to 95% of all monochorionic twins, the shared blood flow is balanced by the presence of arterioarterial and venovenous bidirectional connections, which are found in 85% to 90% and 15% to 20% of monochorionic twin placentas, respectively. The presence of arterioarterial connections is viewed as protective by allowing the overall resistance and blood flow to equalize between the twins and is associated with a significant reduction in TTTS.

(A) Normal placenta angioarchitecture (cotyledon). (B) Superficial view of bidirectional flow into and out of a cotyledon.

Redrawn from Rand L, Lee H. Complicated monochorionic twin pregnancies: updates in fetal diagnosis and treatment. Clin Perinatol. 2009;36:417–430.

Depiction of various types of vascular anastomoses present in monochorionic placentas. a-a, Arterioarterial anastomosis; a-v, arteriovenous anastomosis; N, normal angioarchitecture (cotyledon); v-a, venoarterial anastomosis.

Modified with permission from Simpson LL. Twin-twin transfusion syndrome. Am J Obstet Gynecol. 2013;208:3–18.

The complex pathophysiology of TTTS is dynamic and secondary to a variety of humoral, biochemical, hemodynamic, and functional changes in the two fetuses. The increased blood flow to the recipient twin leads to polycythemia, polyuria, and polyhydramnios and can cause hypertrophic cardiomyopathy, hydrops fetalis, and fetal death. The donor twin (often described as the “stuck” or “pump” twin) is typically hypovolemic, growth-restricted, confined against the endometrium in an oligohydramniotic sac. This twin is primarily at risk for renal failure, cardiac failure, and hydrops fetalis secondary to the high cardiac output state.

The diagnosis of TTTS requires (1) a monochorionic diamniotic pregnancy and (2) a significant discrepancy in amniotic fluid volumes with ultrasound measurements of the maximal vertical pocket (MVP) of less than 2 cm in the oligohydramniotic twin and the MVP greater than 8 cm in the polyhydramniotic twin. Twin size discordance and the presence of intrauterine growth restriction are often present in TTTS, but are not considered necessary to confirm the diagnosis. Although a variety of staging systems exist for determining the severity of TTTS, the most commonly used is the Quintero staging system ( Table 63.2 ), which is based on ultrasound findings. The progression of hypertrophic cardiomyopathy in the fetal recipient is detailed in a scoring system presented by Rychik and colleagues utilizing fetal echocardiography that provides a more detailed assessment of TTTS severity when combined with the Quintero staging system.

Fetuses with TTTS are at risk for preterm PROM, preterm delivery, and neurologic injury with white-matter lesions and long-term disability. Neurodevelopmental impairment is associated with a low GA at birth. Although outcome data stratified for the advanced stages of TTTS are limited, a higher mortality rate is likely associated with more advanced progression of disease. TTTS treatment at stage I has an 85% survival rate, while higher-staged TTTS can increase to greater than 80% mortality if untreated.

Several therapeutic management techniques have been developed to treat TTTS. Amnioreduction can help control polyhydramnios and thereby reduce the risk for both preterm labor and maternal respiratory distress. Also, placental blood flow may improve by decreasing amniotic hydrostatic pressure on the placental vasculature. Serial amnioreductions have been used to treat TTTS for more than 25 years by improving placental perfusion and decreasing the rate of preterm deliveries. A retrospective review of 223 twin sets with TTTS found that use of amnioreduction resulted in an overall birth survival rate of 78%, with survival rates of 65% for recipient twins and 55% for donor twins at 1 month of age. Another review of 112 TTTS cases reported a 61% perinatal survival rate with use of amnioreduction.

Use of an ultrasound-guided needle to create a surgical septostomy was hypothesized to improve TTTS outcome by equalizing the amniotic pressure between the two fetal sacs. In a prospective randomized trial comparing serial amnioreduction to septostomy, no survival difference was found between the techniques. Septostomy is rarely used now for TTTS treatment because it offers no outcome advantage, and the creation of a single amniotic sac can increase the risk for umbilical cord entanglement. Selective feticide with the goal of improving the chance for survival of the other twin is accomplished using techniques described previously in the section on TRAP sequence. This option is typically used only for the most severe cases of TTTS.

Selective fetoscopic laser photocoagulation (SFLP) of the vascular anastomoses between the two twins is the best therapeutic approach for treating TTTS (stages II-IV) between 18 and 26 weeks gestation. By detailed fetal ultrasound examination, the location of the placenta, cord insertions, and fetal orientation and anatomy are confirmed before the start of the procedure. Local anesthesia infiltration with monitored anesthesia care is a common anesthetic technique, though neuraxial anesthesia can also be used. Using ultrasound guidance, a 3-mm trocar or cannula is inserted percutaneously over a wire into the recipient twin’s amniotic sac and directed perpendicular to the longitudinal axis of the donor twin. The fetoscope is inserted into the trocar sheath and the laser fiber bundle passed into the fetoscope guide. Vessels that cross the membrane dividing the amniotic sacs are visualized and Doppler imaging can be used to confirm flow magnitude and direction. The abnormal connecting vessels are selectively laser coagulated, with an effort to spare the normal cotyledons. Ideally this delineates two separate placental regions with each supplying an individual fetus. After laser ablation, amnioreduction is also performed in an effort to decrease the risk for preterm labor. Nonselective laser ablation of crossing vessels is typically avoided because it is associated with more frequent rates of intrauterine fetal demise and is likely to unnecessarily ablate some normal placental vessels. Ablation of all abnormal connecting vasculature is not necessary for the procedure to be successful. Some practitioners advocate use of a sequential technique when coagulating the various types of abnormal anastomoses in an effort to create a net flux of blood from the recipient to the donor and thereby reduce the chance for hemodynamic compromise and hypotension during the procedure in the donor twin. Initially, superficial arteriovenous anastomoses that contain a donor artery are ablated, then arteriovenous anastomoses with a recipient artery, then the arterioarterial anastomoses, and finally ablation is performed at the venovenous anastomoses. A prospective multicenter trial comparing use of sequential SFLP to nonsequenced SFLP found that use of sequential SFLP significantly improved the 30-day survival of both twins and decreased fetal demise. However, use of this sequential technique is associated with longer operative times and increased case difficulty, particularly with a placenta located on the anterior uterine wall. A recent modification of SFLP called the “Solomon technique” was created to reduce recurrent TTTS. In the Solomon technique, the entire vascular equator is coagulated following the coagulation of visible vessels, in order to reduce residual anastomoses. A randomized controlled trial comparing SFLP to the Solomon technique trial noted that the Solomon technique reduces the risk of recurrence of TTTS or twin-anemia polycythemia sequence following initial treatment, but no difference in 2-year survival without neurodevelopmental disability was found.

A 2004 randomized multicenter trial compared SFLP therapy to amnioreduction for treatment of severe TTTS diagnosed between 15 and 26 weeks gestation. Rates of at least one twin survival were more frequent in the laser treatment group at both 28 days (76% vs. 56%, P < .01) and 6 months of life (76% vs. 51%, P < .01). In addition, neurologic outcomes were better in the laser treatment group. A large subgroup of survivors from this trial was followed prospectively for 6 years. No additional change in either survival rate or long-term neurologic outcome was found in comparison to the original 6-month data. This conclusion was echoed in a 2014 Cochrane review of laser treatment for TTTS. More recent work has suggested that early intervention in stage I TTTS may be beneficial compared to expectant management.

Although recent SFLP treatment studies of TTTS studies suggests a greater than 60% dual survival rate and near 90% survival of at least one twin, long-term neurologic outcome in TTTS survivors who underwent SFLP remains unclear, with rates of major neurologic abnormalities in survivors ranging from 3% to 25%. A recent review of 9 studies examining neurodevelopmental outcomes at 24 months in patients who underwent laser therapy found a mean rate of neurological injury of 14%. The mean rate of cognitive impairment was 8%, 11% for motor delay, 17% for communication delay, and 6% for cerebral palsy. Neonatal cerebral lesions correlate with neurodevelopmental impairment at 2 years of age.

The most common complication of SFLP is preterm PROM with subsequent preterm labor and delivery. Preterm PROM rates vary widely across studies, but have been reported to occur in 12% to 30% of SFLP cases for TTTS with a rate of preterm labor and delivery before 32 weeks gestation of approximately 30%. Other possible complications include placental abruption, need for a second SFLP procedure, placement of the trocar through the placenta, hemorrhage, chorioamnionitis, and possible membrane perforation resulting in limb entrapment and ischemia.

In conclusion, treatment of TTTS with SFLP results in better outcomes compared to amnioreduction alone. Additional research is needed to determine optimal timing, further technique refinement, and long-term neurologic outcome for the treatment of TTTS with SFLP.

Amniotic Band Syndrome

Amniotic band syndrome (ABS) includes a wide variety of fetal anomalies associated with fibrous bands that entangle or constrict various parts of the fetus or umbilical cord in utero ( Fig. 63.6 ). The resulting malformations include limb and digit amputations, craniofacial abnormalities, visceral defects, and body wall defects. The incidence is approximately 1 in 3000 to 1 in 15,000 live births. The primary cause and pathogenesis of ABS remains unknown, but these defects appear secondary to either a vascular insult or other cause of abnormal perfusion to the affected area. Theories behind the cause of ABS include a primary defect in embryonic development, an early amniotic membrane rupture resulting in the creation of amniochorionic bands, or a vascular disruption early in pregnancy.

Depiction of amniotic band syndrome.

The amnion membrane can become wrapped around various fetal parts (e.g., digits, limbs, neck) and result in amputations or deformities.

Redrawn from Graves CE, Harrison MR, Padilla BE. Minimally invasive fetal surgery. Clin Perinatol. 2017;44(4):729–751.

Diagnosis is determined by findings on ultrasound imaging consistent with characteristic fetal anomalies of the involved anatomy. Visualization of the bands themselves is not a necessary component of the diagnosis. Fetal MRI can aid in diagnosis as a complementary technique to ultrasound, but the efficacy of its use is currently limited to case reports and a small case series. The progression of morbid conditions from each case of ABS is difficult to predict. In cases of limb entrapment, increasing band constriction decreases both venous and arterial flow. This decreased perfusion leads to limb edema and can eventually result in amputation distal to the constriction. In an otherwise normal fetus, use of fetoscopic-guided sectioning of the band with a laser can restore distal blood flow and may improve functional outcome in some cases. Efficacy of the technique is based on review of a small number of cases in the literature, which range from 50% to 80% of limb form and function preservation provided that some arterial flow to the affected limb is present pre-intervention. Single limb involvement is also associated with a more favorable surgical outcome compared to fetuses with multiple limb involvement. Interestingly, the rate of PROM in these procedures is higher than other fetoscopic procedures. ABS can also be an iatrogenic complication following minimally invasive fetal procedures in monochorionic diamniotic twins, and should be evaluated for especially if known chorioamnion separation or septostomy occurs.

Congenital Diaphragmatic Hernia

Approximately 1 in 2500 newborn infants has a congenital diaphragmatic hernia (CDH). During early gestation, abdominal contents herniate into the thoracic cavity and compress the fetal lungs ( Fig. 63.7 ). This results in significant neonatal morbidity and mortality from pulmonary hypoplasia, respiratory insufficiency, and pulmonary hypertension. CDH survival rates when ECMO is not required have improved to greater than 70% over the past 25 years at tertiary care centers. These highly specialized centers offer care that includes early intubation, surfactant administration, ventilation techniques to minimize lung trauma, surgical closure of the defect, and ECMO therapy. Mortality rates vary greatly with the severity of pulmonary hypertension and respiratory dysfunction. If ECMO is required, survival rates vary between 50% and 80%. Fetal intervention for CDH aims to improve fetal lung development and prevent the morbidity of pulmonary hypoplasia.

Depiction of congenital diaphragmatic hernia.

A defect in the fetal diaphragm allows abdominal contents (e.g., intestines, liver) to enter the thoracic cavity, compress the heart, and prevent lung growth and maturation.

Redrawn from Graves CE, Harrison MR, Padilla BE. Minimally invasive fetal surgery. Clin Perinatol. 2017;44(4):729–751.

In utero repair of diaphragmatic hernia in lamb models reversed both the parenchymal hypoplasia and pulmonary vascular changes associated with CDH. Initial human in utero intervention focused on open repair of the fetal diaphragm with limited success. However, these initial interventions advanced fetal surgery techniques and paved the way for a minimally invasive approach to CDH intervention.

Initial open intervention experience found that a fetal abdominal patch was necessary to accommodate the added abdominal viscera without increasing intraabdominal pressure and compromise of ductus venosus blood flow. These early open interventions also determined that in utero reduction of the herniated liver compromised umbilical circulation and significantly increased fetal mortality. Additionally, the advantage of opening the uterus with a stapling device to ensure effective hemostasis was demonstrated. However, adequate control of postoperative uterine tone remained a substantial problem. In a prospective clinical trial sponsored by the National Institutes of Health, in utero CDH treatment using an open technique for fetuses without herniation of the liver into the thorax did not improve neonatal survival compared with standard postnatal treatment.

After these efforts, tracheal occlusion was advanced as a less-invasive strategy for treating fetal CDH and improving lung development. Fetal lungs secrete over 100 mL/kg/day of fluid that exits the trachea and mouth into the amniotic cavity. In fetal lamb models, tracheal occlusion restricts the normal outflow of the fetal lung fluid and provides an increase in pulmonary hydrostatic pressure. This increased pressure pushes the viscera out of the thorax, promotes expansion of the hypoplastic lung, and thereby improves lung growth and development. Fetal reversible tracheal occlusion has replaced primary repair in utero for the treatment of CDH. Initially, a tracheal foam plug was inserted in the fetal airway during open surgery, but it failed to provide adequate occlusion. Later, metallic hemoclips were placed around the trachea after meticulous neck dissection. Unfortunately, survival with this initial open tracheal occlusion procedure was poor (15%) and lower than survival in CDH fetuses receiving postnatal standard treatment (38%).

Subsequently, minimally invasive fetal endoscopic surgical techniques replaced the open technique for placement of the clips. These procedures are performed with local or neuraxial maternal anesthesia and the administration of fetal anesthesia by intramuscular injection. A variety of occlusive devices, including cuffs, plugs, valves, and balloons, were attempted at various medical centers. Currently, percutaneous endoscopic endotracheal intubation is used to place a small detachable occlusive balloon in the fetal trachea. With initial procedures the balloon was left in place until delivery (see discussion on EXIT procedures), but currently, it is deflated and removed before term with a second fetal endoscopy. The balloon removal may improve type II pneumocyte function, increase surfactant production, and allow a vaginal delivery when appropriate.

Use of ultrasound imaging to determine the ratio of fetal lung area to head circumference (LHR) and the presence of thoracic liver herniation (“liver up” vs. “liver down”) are historically the most reliable prognostic indicators of fetal CDH outcome. Evaluation of survival in fetuses with left-sided CDH and intrathoracic liver herniation managed postnatally noted no survival with an LHR of 0.7 or less and a survival rate of 72.7% or greater with an LHR of 1.4 or greater ( Table 63.3 ). The LHR increases exponentially with GA and is less useful after 28 weeks gestation. The Antenatal CDH Registry Group normalized the LHR to GA by creating an observed to expected LHR (o/e LHR) that correlates well with survival rate ( Fig. 63.8 ). More recently, fetal MRI measurement of observed to expected total fetal lung volume (o/e TFLV) correlates well with neonatal survival in fetuses with isolated CDH as an assessment of pulmonary hypoplasia. In addition to presence of liver herniation, use of a 25% threshold with o/e LHR and o/e TFLV appear to perform the best as predictors of survival with CDH.

Survival rates of fetuses with isolated left-sided congenital diaphragmatic hernia, depending on measurement of the observed to expected lung area to head circumference ratio (O/E LHR) and position of the liver.

Redrawn from Deprest JA, Nicolaides K, Gratacos E. Fetal surgery for congenital diaphragmatic hernia is back from never gone. Fetal Diagn Ther. 2011;29:6–17.

A prospective randomized trial (1999-2001) evaluated fetal endoscopic tracheal occlusion (FETO) for prenatal treatment of CDH using both clip and balloon occlusion techniques. Inclusion criteria included a 22- to 28-week GA, left-sided thoracic liver herniation, and an LHR of less than 1.4. The trial was closed early, and no benefit in survival or reduction in 90-day morbidity was found with prenatal treatment ( n = 11) compared with control ( n = 13) (survival rates of 73% vs. 77%). Additionally, rates of PROM and preterm delivery were more frequent in the fetal intervention group, with 100% of the fetoscopic intervention group having PROM. The inclusion of fetuses with an LHR of up to 1.4 may have decreased the ability to determine a significant difference, because many of these fetuses were likely to survive with standard postnatal tertiary medical care (see Table 63.3 ).

Three European medical centers and the FETO task force began a collaboration for treatment of severe cases of CDH with a high risk for mortality (LHR < 1.0 and liver herniation into the hemithorax). Because of concern for tracheal damage by very early tracheal balloon placement, the tracheal balloon was placed at 26 to 28 weeks gestation and removed around 34 weeks gestation. Data from FETO task force cases through 2008 ( n = 210, mean GA of placement at 27 weeks, LHR < 1.0, and primarily left-sided CDH) were compared with historic postnatal treatment controls (1995-2004). Use of prenatal reversible tracheal occlusion significantly improved the survival rate (47% vs. 20% in historic controls), and delivery occurred at a median GA of 35 weeks. In addition, mean operating time was minimal (<10 minutes) and more than 95% of procedures were successful on the first attempt. Some of the survival benefit with in utero treatment may represent selection bias as well as improvement in both technique and neonatal care over time.

A 2016 metaanalysis of all studies comparing survival outcome between FETO and a contemporary control group found fetal intervention improved survival in patients with isolated CDH and a LHR ≤ 1.0. Fifty-one of 110 fetuses (46.3%) who had undergone FETO survived to discharge, compared with 6 of 101 (5.9%) in the control group. Another recent study suggests that patients undergoing FETO for severe isolated CDH have morbidity outcomes similar to those fetuses with moderate lung hypoplasia. Although FETO improves lung growth, and decreases need for ECMO, left heart hypoplasia may persist until after the postnatal CDH repair.

A multicenter randomized Tracheal Occlusion to Accelerate Lung growth trial is ongoing ( http://www.totaltrial.eu ). It compares postnatal management to both late (30-32 weeks gestation) FETO intervention for moderate lung hypoplasia and earlier FETO intervention (27-30 weeks gestation) for severe lung hypoplasia. This trial uses o/e LHR criteria, and balloon occlusion removal is planned for 34 weeks gestation, with postnatal management standardized by a consensus protocol. Long-term outcome data on pulmonary and neurologic development are needed to better determine if and when FETO should be offered for individual cases of severe CDH. New data suggest that prenatal administration of sildenafil may further reduce neonatal pulmonary hypertension, with or without invasive fetal surgery.

Myelomeningocele

Spina bifida includes all defects with incomplete neural tube closure. Myelomeningocele (MMC) is the most common type of spina bifida and results in exposure of the meninges and spinal cord through a congenital defect in the vertebrae and overlying tissues. MMC occurs during the third and fourth weeks of gestation with incomplete embryonic neural plate development. Folate supplementation in the maternal diet has decreased the rate of MMC by nearly 50%, but the benefit reached a plateau without eliminating the disorder and MMC remains at a rate of approximately 1 in 3000 live births. Improved prenatal screening with α-fetoprotein measurement and ultrasonography allows the possibility of pregnancy termination, and it is estimated that 25% to 40% of MMC pregnancies are terminated. MMC can result in lifelong morbidity and disability, including loss of motor and sensory function based on lesion level, bowel and bladder dysfunction, sexual dysfunction, hydrocephalus, Arnold-Chiari type II malformation, tethered cord, and impaired cognition. If uncorrected in utero, surgical closure of the spinal defect must be performed within a few days after birth. Children with lumbosacral MMC defects often require ventriculoperitoneal shunting for hydrocephalus. Even with successful shunting, complications of central hypoventilation, vocal cord dysfunction, and swallowing difficulties can persist because of an associated Arnold-Chiari malformation. The mean intelligence quotient in patients with MMC who underwent ventriculoperitoneal shunting is 80 (low normal) and decreased compared to those who do not require a shunt. Neonates with spina bifida have a 14% mortality rate by 5 years of age and a 35% mortality rate in those with known brainstem dysfunction and Arnold Chiari type II malformation.

The cause of MMC remains unknown. The abnormalities and deficits of MMC are hypothesized to be the result of two separate mechanisms. The primary cause is anatomic malformation with abnormal development of the spinal cord and associated tissues. Secondary damage is likely created by exposure of these open neural elements to the amniotic fluid and direct trauma. Consequently, the ability to close the defect in utero and isolate the neural tissue from contact with the intrauterine environment has the potential to improve outcome in contrast to delaying closure until after birth.

This hypothesis has been supported by animal models that found improved neonatal neurologic function with fetal closure of the defect in utero. Ultrasound assessment demonstrates that central and peripheral neurologic injury is progressive during gestation. Motor deficits and cognitive dysfunction are correlated with lesion level and higher levels are associated with a greater degree of morbidity. Improved motor function has been observed at 2 years of age in children with MMC who underwent cesarean delivery before labor onset compared to those delivered vaginally or by cesarean delivery after onset of labor. Therefore, women with fetuses with MMC and planned postnatal repair typically undergo cesarean delivery before onset of labor or rupture of membranes in an effort to minimize any additional injury to the open neural elements.

The rationale for fetal intervention for MMC is to improve functionality and quality of life. Prenatal repair for MMC is most commonly performed with an open fetal surgery technique requiring both maternal laparotomy and hysterotomy, although a few centers have started performing the repair with an endoscopic fetal technique. Considerations for anesthesiologists involved in open MMC repair (see section on “Management of Open Procedure”) include participating in comprehensive, multidisciplinary, preoperative maternal evaluation and counseling; preparing for potential intraoperative hemorrhage; planning an anesthetic that provides profound uterine relaxation; administering analgesia and muscle relaxant directly to the fetus; monitoring the need for potential fetal resuscitation or urgent delivery; and managing maternal analgesia and postoperative uterine and fetal monitoring.

MMC in utero surgery typically occurs between 19 and 26 weeks gestation. Initial human studies found that in utero repair reversed the hindbrain herniation of the Arnold-Chiari II malformation and decreased the requirement for ventriculoperitoneal shunt placement before 1 year of age. In addition, a 1999 series of 10 fetuses undergoing in utero MMC closure at 22 to 25 weeks gestation found improved lower extremity function among 6 out of 9 fetuses compared with expected function based on the level of lesion, with one fetal death at 25 weeks gestation from preterm delivery and associated respiratory insufficiency. A randomized, prospective clinical trial completed between 2003 and 2010 at three U.S. medical centers examined the risks, benefits, and outcomes of open in utero repair for MMC compared with standard postnatal repair among 183 patients. Open fetal repair reduced the need for ventriculoperitoneal shunting, decreased hindbrain herniation, and improved lower extremity motor function at 30 months of age. However, prenatal repair significantly increased the risk for various fetal and maternal complications, including spontaneous membrane rupture, partial or complete uterine dehiscence, and preterm birth with increased risk for respiratory distress syndrome ( Table 63.4 ). Two perinatal deaths occurred in each group. A 30-month follow-up study demonstrated that patients undergoing fetal surgery were more likely to have a level of function 2 or more levels better than expected (26.4% vs. 11.4%), to be able to walk independently (44.8% vs. 23.9%, P = .004), to perform better on the Bayley Mental Development Index, and to have improved Peabody Developmental Motor Scale. Long-term outcomes from this trial are still ongoing, but a pretrial cohort of 54 patients undergoing fetal MMC repair reported improved functional and behavioral outcomes at age 10 years, especially among children who did not require ventriculoperitoneal shunting.

Results of this trial should not be generalized to patients outside the inclusion criteria of the study and should be considered only at medical centers that can develop a practice with significant volume and depth of resources. It is recommended that this procedure be offered only at medical facilities with the expertise, multidisciplinary teams, services, and facilities to provide the intensive care required for these patients, while maintaining rigorous patient selection.

With improvement in surgical technique, and feasibility demonstrated in animal studies, endoscopic MMC repair is increasingly practiced. In theory, a minimally invasive approach may reduce maternal complications and obviate the need for maternal cesarean delivery. A phase I clinical trial of percutaneous endoscopic repair ( n = 10) using a biocellular patch over the defect followed by skin closure revealed improved hindbrain herniation and motor function, however preterm delivery was significant with the mean gestational birth age being 32 weeks. In addition, 2 of 10 procedures were aborted due to loss of uterine access, preterm PROM occurred in all 10 cases, and 1 fetus and 1 neonate died. Outcomes have improved with evolution of the fetoscopic repair technique to include a maternal laparotomy and exteriorization of the uterus prior to insertion of the fetoscope ports. After port placement, a portion of amniotic fluid is withdrawn and carbon dioxide is insufflated. A retrospective cohort trial of 28 patients undergoing fetoscopic MMC repair (22 fetoscopic, 4 conversion to hysterotomy, 2 abandoned) noted a mean delivery at 39 weeks gestation with a standardized surgical approach ( n = 10) using a two-port technique and a preterm PROM rate between 10% and 30%. Of the 22 patients who underwent a fetoscopic repair, 50% delivered vaginally. Recently a method using partial amniotic carbon dioxide insufflation without pre-removal of amniotic fluid has been described and successfully used in MMC repair.

A 2017 metaanalysis of 11 studies examining fetal MMC repair found no difference in mortality or need for ventriculoperitoneal shunt among fetal patients treated with an open or endoscopic technique. Patients treated with a percutaneous fetoscopic technique had a higher PROM rate (91% vs. 36%), a higher risk of preterm delivery (96% vs. 81%), and a higher rate of cerebral spinal fluid leakage from the MMC repair site (30% vs. 7%). Fetal repair via a fetoscopic approach following maternal laparotomy had a lower rate of premature birth. Uterine dehiscence was higher in the open surgical group (11% vs. 0%). The authors concluded that a fetoscopic technique is a promising, though not yet perfected, technique for fetal MMC repair. Minimally invasive endoscopic repair of MMC should still be considered an experimental treatment given the steep learning curve and lack of long-term outcome data.

Sacrococcygeal Teratoma

The incidence of sacrococcygeal teratoma (SCT) is approximately 1 in 15,000 to 40,000 live births. These teratomas are typically diagnosed by ultrasound in the second trimester of pregnancy and may grow rapidly (>150 cm ³ /week), with some reaching sizes of 1000 cm ³ or more. The larger tumors create a significant arteriovenous shunt and low resistance state and lead to the development of high-output cardiac failure, polyhydramnios, placentomegaly, and hydrops fetalis. In addition, rapidly growing tumors may rupture and hemorrhage. Perinatal mortality varies greatly in different published series, ranging from 16% to 63%. In addition, fetuses with SCT are at risk for preterm birth, intrapartum dystocia, tumor rupture with hemorrhage, and urinary obstruction. Cesarean delivery is frequently required.

Tumor staging is based on criteria from the American Academy of Pediatrics Surgical Section as detailed by Altman ( Fig. 63.9 ). Based on the external location, stage I tumors are suitable for fetal intervention whereas stage IV tumors are entirely internal and not considered appropriate for fetal resection. Fetal MRI can assist in tumor size, location, and staging. A tumor volume to fetal weight ratio greater than 0.1 cm ³ /g prior to 24 weeks gestation is predictive of poor fetal outcome.

Sacrococcygeal tumors are classified according to their location.

Type I tumors are predominantly external (sacrococcygeal) with minimal presacral component; type II tumors present externally but have a significant intrapelvic extension; type III tumors are present externally but the predominant mass is pelvic and extends into the abdomen; type IV tumors are presacral with no external presentation.

Redrawn from Altman RP, Randolph JG, Lilly JR. Sacrococcygeal teratoma: American Academy of Pediatrics Surgical Section Survey-1973. J Pediatr Surg . 1974;9:389–398.

Fetuses diagnosed with SCT before 30 weeks gestation or with rapidly growing tumors have a poor prognosis (<7% survival), but may benefit from in utero intervention. Radiofrequency ablation, embolization, thermocoagulation of the tumor or vascular supply, and cyst drainage have been employed to treat SCT in utero and prevent hydrops fetalis, but the benefit remains unclear.

A review of minimally invasive fetal intervention for SCT in 34 patients from 1980 to 2013 demonstrated an overall survival rate of 44% with mean delivery at 29.7 ± 4.0 weeks. Preexisting hydrops led to a worse outcome, with only 30% survival (6/20) in this cohort. A subsequent review of 33 patients examined the effectiveness of different techniques, comparing interstitial tumor ablation ( n = 22) to targeting the feeding vessels of the tumor ( n = 11). Fetuses undergoing vascular ablation seemed to have a survival advantage over the interstitial ablation group (63.6% vs. 40.9%), which the authors suggested may be because interstitial ablation often led to tumor necrosis and subsequent hemorrhage.

SCTs have been resected in utero ( Fig. 63.10 ) with successful outcomes, but the optimal timing and criteria for fetal intervention is uncertain. Fetal teratoma resection has a high risk for fetal hemorrhage; consequently, intraoperative placement of a venous catheter in a fetal hand, leg, or umbilical cord vein is important for the ability to transfuse compatible blood and crystalloid or rapidly administer resuscitation drugs.

(A) Ultrasound image of fetus with sacrococcygeal teratoma (SCT). Note sacral origin of mass. (B) Fetus with SCT undergoing in utero mass resection. Note tourniquet on left leg placed momentarily before intravenous catheter placement in saphenous vein.

Courtesy Dr. Anita Moon-Grady, Department of Pediatrics, University of California, San Francisco, CA.

In certain cases, fetal SCT may lead to maternal mirror syndrome, in which maternal physiology mimics the abnormal circulatory physiology of the hydropic fetus. The mother develops hypertension with increased peripheral and pulmonary edema from a hyperdynamic state. Mirror syndrome is a form of severe preeclampsia that may develop in association with fetal hydrops and in most cases necessitates delivery, although platelet levels and liver enzymes typically remain in normal ranges. Maternal mirror syndrome does not normally resolve immediately with correction of the fetal pathophysiology and can result in life-threatening maternal complications.

Congenital Pulmonary Lesions

Congenital pulmonary airway malformations (CPAMs), historically known as congenital cystic adenomatoid malformations, are typically benign nonfunctioning pulmonary tumors composed of cystic and solid components normally contained within a single lung lobe. These fetal lesions complicate approximately 1 in 25,000 to 35,000 live births. Other possible fetal abnormalities that require differentiation include bronchopulmonary sequestration, bronchogenic cysts, congenital lobar emphysema, neurogenic cysts, peripheral bronchial atresia, and CDH. CPAM includes five subtypes ( Table 63.5 ) that are based on the presumed tumor development site. Prenatal ultrasonography categorizes lesions as macrocystic (cysts > 5 mm in diameter) or microcystic (cysts < 5 mm in diameter), with microcystic lesions having a more solid or echogenic image. Typically, small lesions regress in the third trimester and can be surgically resected after birth or managed conservatively without surgical intervention. Large lesions can compress the great vessels, create pulmonary hypoplasia, and cause cardiac compression and mediastinal shift that often results in hydrops fetalis. Prognosis depends primarily on size and growth characteristics of the CPAM rather than lesion type. To normalize tumor volume to fetal size, ultrasound measurement of CPAM volume to fetal head circumference ratio (CVR) was evaluated as a predictor of hydrops fetalis and other postnatal outcomes. CVR is calculated from the volume of an ellipse using the formula (length × height × width × 0.52)(cm ³ )/fetal head circumference (cm). Based on retrospective data from 71 fetuses, a CVR less than 0.56 predicts no adverse postnatal outcome (100% negative predictive value), whereas a CVR greater than 0.56 had a positive predictive value for adverse postnatal outcome of 33%. In addition, a CVR greater than 1.6 is associated with greater risk for hydrops fetalis. A recent retrospective analysis of a series of 24 fetuses with a CVR greater than 1.6 found that an abnormal echocardiogram with hydrops was a significant predictor of mortality and need for fetal intervention.

Some lesions regress and leave minimal impact, whereas others grow in size. Patients with CPAM tumors associated with hydrops fetalis have a survival rate of less than 5% without intervention, and early delivery or fetal treatment is indicated in these cases. Administration of maternal betamethasone may improve survival in patients with high-risk CPAM. Large macrocystic lesions or large pleural effusions can be decompressed in utero by placement of shunt catheters between the cysts and the amniotic cavity ( Fig. 63.11 ). Use of fetal analgesics and muscle relaxants delivered intramuscularly with ultrasound guidance may decrease the fetal stress response and prevent fetal movement during critical portions of the procedure. Shunt placement can prevent or reverse hydrops formation, with definitive resection deferred until after birth. In a retrospective analysis of 75 fetuses undergoing shunt placement for pleural effusion or a macrocystic lung lesion, shunt placement resulted in a decrease in macrocystic lung lesion volume by 55 ± 21% and complete or partial resolution of the pleural effusion in 29% and 71% of fetuses respectively. Hydrops resolved in 83% of fetuses following intervention. Overall neonatal survival was 68%, and correlated with hydrops resolution, GA at birth, a unilateral pleural effusion, and percent reduction in lesion size. Another retrospective study noted a 59% survival rate for fetuses with hydrothorax treated with thoracoamniotic shunts, though repeat procedures were often necessary. In some CPAM lesions, shunts are ineffective because cysts are not in communication with each other, shunts malfunction, or they become displaced. In addition, shunt placement can cause fetal hemorrhage, preterm PROM, or chorioamnionitis. Thoracoamniotic shunt placement has also successfully decompressed large congenital pleural effusions caused by fetal chylothorax that otherwise would result in hydrops fetalis, pulmonary compression, and fetal or neonatal death.

Thoracic shunt insertion ( top left and right ) and deployment (bottom left) .

From van Mieghem T, Baud D, Devlieger R, et al. Minimally invasive fetal therapy. Best Pract Res Clin Obstet Gynaecol. 2012;26:711–725.

Some solid or mixed solid/cystic CPAMs inappropriate for shunting can be resected with open fetal lobectomy ( Fig. 63.12 ). Similar to open SCT resection, potential exists for significant fetal hemorrhage and need for in utero fetal resuscitation. Open resection of CPAM lesions associated with hydrops fetalis has resulted in a 30-day postnatal survival rate of 50% to 60%, with tumor resection allowing for compensatory lung growth and resolution of hydrops fetalis.

(A) Open resection of fetal congenital pulmonary airway malformation (CPAM) with fetal thoracotomy shown. (B) Resected CPAM mass. (C) Primary closure of fetal thoracotomy. (D) Pathologic specimen of resected CPAM mass.

Courtesy Dr. Anita Moon-Grady, University of California, San Francisco Fetal Treatment Center, San Francisco, CA.

In some instances, use of an EXIT procedure with fetal thoracotomy, mass resection, and a secured airway before delivery has been a successful approach for tumors with persistent mediastinal compression. In a series of nine fetuses with large lung masses (CVR 1.9-3.6 at term) who underwent an EXIT-to-resection procedure, all procedures were successful, with no significant operative complications.

Fetal Physiology and Monitoring

During fetal surgery, procedural and pharmacologic interventions can adversely affect fetal physiology by altering uteroplacental or fetoplacental circulation and gas exchange. Appropriate fetal monitoring facilitates early intervention. In addition to the physiologic effects of medications administered to the mother or fetus, a detailed knowledge of both maternal physiology of pregnancy, and fetal cardiovascular, neurologic, and placental physiology provides the basis for optimal fetal care. Uteroplacental and fetoplacental physiology, including uterine perfusion, placental gas exchange, and drug transfer are detailed in Chapter 62 , which also discusses the effects of maternal positioning, maternal neuraxial anesthesia, and administration of general anesthesia on the uteroplacental unit.

Fetal cardiac output depends primarily on heart rate. Compared to a neonate, the contractility of the fetal myocardium is decreased secondary to decreased myofibrillar density and it is intolerant of hypocalcemia because of an immature calcium regulatory system. Fetal myocardium is also less compliant than adult myocardium, near the peak of the Frank–Starling ventricular function curve, and fluid-filled lungs also inhibit additional ventricular filling. Consequently, modest changes in preload have minimal effect on cardiac output. The normal fetal cardiac output (sum of right and left ventricular output) is in the range of 425 to 550 mL/min/kg throughout gestation.

Fetal blood volume increases during gestation, and approximately twothirds of the fetal-placental blood volume resides within the placenta and the placenta receives about 40% of the fetal cardiac output. During the second trimester, fetal blood volume is estimated to be approximately 110 to 160 mL/kg of fetal body weight. After midgestation, fetal blood volume can be estimated based on GA using the equation, estimated fetal blood volume (mL) = 11.2 × GA – 209.4. In the developing fetus, hemoglobin F is the primary oxygen carrier. Beginning at 32 weeks’ gestation there is a gradual shift toward adult hemoglobin synthesis. Mean fetal Hb levels increase linearly from 11 g/dL at 17 weeks gestation to 18 g/dL in a term neonate.

Fetal lung epithelium produces more than 100 mL/kg/day of fluid that fills the lungs and facilitates appropriate pulmonary growth and development. Excess lung fluid exits the trachea and is either swallowed or flows into the amniotic fluid. Although fetal hepatic enzymes are less functional than those of adults, most drugs still undergo metabolism and the umbilical circulation provides initial hepatic metabolism (first-pass metabolism) before medications reach the fetal brain or heart. Although fetal liver function is immature, coagulation factors are synthesized independent of the maternal circulation and do not cross the placenta. The serum concentrations of these factors increase with GA ( Table 63.6 ), but fetal clot formation in response to tissue injury is decreased in contrast to that in adults throughout gestation and the first 6 months of life. Platelets first appear at 5 weeks gestation and increase in number with time, reaching a mean of 150 × 10 ⁹ /L by the end of the first trimester of pregnancy and values reaching normal adult range by 22 weeks gestation.

During open procedures, the fetal circulation and flow distribution can be severely impaired by fetal manipulation or direct compression of the umbilical cord, inferior vena cava, or mediastinum. Increased uterine activity, maternal hypotension, and significant maternal hypocarbia can all decrease uteroplacental perfusion. Fetal heart rate (FHR) monitoring is important during both minimally invasive and open procedures. During IUT, the transfusion needle can lacerate umbilical vessels with unanticipated fetal movement, and use of laser therapy for TTTS can disrupt surface placental vessels critical to fetal blood flow. During labor, FHR monitoring with external Doppler or a fetal scalp electrode is commonly employed to assess fetal well-being, but during fetal procedures, echocardiography, pulse oximetry, and ultrasound imaging of umbilical artery flow are the primary methods for fetal assessment. After exposure of the fetal head during an EXIT procedure, placement of an Hb saturation monitor and insertion of a fetal scalp electrode have both been used successfully for FHR monitoring. It remains unclear whether fetal oximetry or heart rate monitoring is more sensitive to decreased umbilical-fetal blood flow. In a fetal lamb model examining the effects of umbilical cord compression, Hb desaturation was detected by fetal pulse oximetry before the onset of bradycardia. However, FHR deceleration has been shown to precede fetal Hb desaturation with use of fetal pulse oximetry monitoring during labor.

Intraoperative ultrasonography allows imaging of FHR, cardiac contractility, and cardiac filling, as well as Doppler assessment of flow in the ductus arteriosus and umbilical artery flow. Both absent and reversed umbilical artery diastolic flow are associated with increased perinatal morbidity and mortality. In many cases, ultrasonography assessment of fetal well-being can be only intermittent. This is because ultrasonography may be periodically required to guide the intervention or the probe placement can interfere with the surgical procedure.

When intraoperative monitoring detects depression of fetal hemodynamics, steps should promptly be undertaken to improve uterine perfusion, ensure the uteroplacental interface is intact, and relieve any compression of the umbilical cord or placenta. These steps may include administration of medications to increase maternal blood pressure, cardiac output, and uterine relaxation. In some cases, administration of resuscitation medications directly to the fetus may be needed, or if previously determined to be viable ex utero fetal resuscitation may be necessary.

In utero, the fetus is unable to thermoregulate and depends on maternal body temperature secondary to placental circulation and surrounding amniotic fluid. Induction of general anesthesia, surgical exposure, and hysterotomy can reduce fetal temperature dramatically both directly and secondarily if maternal core temperature drops significantly. Fetal sheep studies demonstrate the fetus is unable to generate heat through thermogenesis and decreases in sheep fetal temperature can lead to tachycardia and hypertension in utero. In contrast, human reports associate maternal/fetal hypothermia with fetal bradycardia. Consequently, maintenance of maternal euthermia with use of forced air warming likely improves fetal well-being during minimally invasive procedures. During open fetal surgery, use of warmed fluid for intrauterine irrigation and monitoring of both maternal core and amniotic fluid temperatures are also important.

Fetal Anesthesia, Analgesia, and Pain Perception

The capability of the fetus to perceive pain remains controversial. The fetus exhibits pituitary-adrenal, sympathoadrenal, and circulatory stress responses to noxious stimuli as early as 16 to 18 weeks gestation. Although invasive fetal procedures elicit a stress response, this response is mediated at the level of the spinal cord, brainstem, and/or basal ganglia and does not necessarily correlate with conscious cortical perception of pain. Administration of opioids in preterm neonates blunts hormonal responses to surgery, including changes in plasma adrenaline, noradrenaline, glucagon, aldosterone, corticosterone, glucose, and lactate. Providing adequate anesthesia and analgesia is associated with both attenuation of the deleterious effects and improved outcomes. Stress responses secondary to invasive fetal procedures are blunted with opioid administration, but reduction of plasma stress hormone levels is not necessarily evidence of adequate analgesia.

Pressure, temperature, and vibration sensory nerve terminals develop in human skin between 6 and 10 weeks gestation. Skin nerve terminals required for peripheral sensory nociception likely develop between 10 and 17 weeks gestation. Noxious stimuli follow a reflex arc of afferent fibers that synapse on spinal cord interneurons, which then synapse with motor neurons. A fetus can reflexively withdraw from a noxious stimulus by 19 weeks GA without input from the cerebral cortex.

The perception of pain requires not only intact neural pathways from the periphery to the primary sensory cortex, but also higher cortical structures. Thalamocortical circuits, thalamic pain fibers likely reach the somatosensory cortex at 24 to 30 weeks gestation. Thalamic projections reach the visual subplate at 20 to 22 weeks, the visual cortex at 23 to 27 weeks, and the auditory cortical plate at about 26 to 28 weeks gestation. However, fetuses are unlikely to experience pain before 24 weeks gestation because the cortex requires additional growth, remodeling, and development to establish the extensive neural network of pathways to other central nervous system structures.

This timeline is supported by electroencephalographic (EEG) studies. Cortical EEG activity is present only 2% of the time in fetuses at 24 weeks gestation and increases to being present 80% of the time with EEG patterns becoming more distinct by 34 weeks gestation.

Both the long-term impact of untreated fetal stress and the timing of fetal pain perception remain unknown. Given this uncertainty and the more than 35-year history of safe anesthetic administration in neonates and fetuses undergoing invasive procedures, analgesia should be provided during fetal surgery. In addition to blunting pain, administration of fetal analgesia helps to prevent fetal movement and inhibit the circulatory stress response.

Opioid analgesics can be transferred to the fetus by maternal administration or direct fetal intramuscular or intravenous umbilical cord administration using ultrasound guidance. For most invasive procedures causing noxious fetal stimulation, fetal intramuscular administration of fentanyl 10 to 20 μg/kg (or other opioid in equivalent dosing) is used to provide analgesia immediately before the intervention. This can be achieved percutaneously using ultrasound guidance or under direct vision when a hysterotomy is performed. Some physicians administer prophylactic intramuscular atropine 20 μg/kg with opioids to minimize the risk for fetal bradycardia. Fetal movement can be prevented by intramuscular or umbilical vessel administration of a muscle relaxant such as rocuronium (intramuscular 2.5 mg/kg or intravenous 1.0 mg/kg) or vecuronium (intramuscular 0.25 mg/kg or intravenous 0.1 mg/kg) using ultrasound guidance. The onset of fetal paralysis varies with the specific drug and dose, but typically occurs in 2 to 5 minutes, with an approximate duration of 1 to 2 hours. In many instances, opioid, anticholinergic, and muscle relaxant are combined in a single injection. Maternal administration and placental transfer of intravenous remifentanil provides adequate fetal immobility during fetoscopic interventions that involve only the umbilical cord or placenta.

For open fetal procedures, placental transfer of maternally administered general anesthesia with volatile anesthetics provides fetal anesthesia. These anesthetics readily transfer across the placenta, with fetal concentration and the fetal-to-maternal (F/M) ratio depending on both the maternal inspired anesthetic concentration and the duration of maternal anesthetic administration. In human studies of anesthetic levels at the time of cesarean delivery (∼10-minute duration of general anesthesia), isoflurane has an F/M ratio of approximately 0.7. Although placental transfer of desflurane and sevoflurane may be similar, published human F/M data are not currently available in the literature. Nitrous oxide has an F/M ratio of 0.83 after only 3 minutes of administration.

High levels of volatile anesthetic can depress fetal myocardium and lead to increasing fetal acidosis. In animal models, volatile anesthetic concentrations often employed for uterine relaxation (>2 minimum alveolar concentration [MAC]) lead to significant reductions in maternal cardiac output with associated decrease in uterine perfusion up to 30%. A retrospective analysis of echocardiographic data from clinical cases of open fetal surgery and EXIT procedures reveals a moderate-to-severe left ventricular systolic fetal cardiac dysfunction with use of high concentrations of desflurane. In addition, case reports have described epileptiform EEG activity and generalized tonic-clonic seizures in both adults and children exposed to high levels of sevoflurane. Seizure activity has also been attributed to high-dose sevoflurane during an open fetal procedure. Therefore, high concentrations of volatile anesthetic administration, while useful for maternal uterine relaxation, may not be the optimal anesthetic for the fetus despite years of successful use. Consequently, a reduced level of volatile anesthetic (1.0-1.5 MAC) is combined with remifentanil and propofol infusions for open fetal surgery at some institutions. Remifentanil has significant placental transfer and prevents fetal movement during TTTS laser photocoagulation therapy. Some prefer to administer maternal remifentanil and nitroglycerin as part of the anesthetic for open fetal or EXIT procedures to decrease the amount of volatile anesthetic. Currently, no evidence indicates that any one anesthetic method provides an improved fetal or maternal outcome as long as appropriate uterine quiescence is maintained.

Anesthetic neurotoxicity of the developing brain is a concern for all providers administering anesthetic agents for fetal procedures. In animal models, anesthetics affect neonatal brain development and create histologic changes, as well as learning and memory deficits. Recent nonhuman primate studies found repeated exposures to clinically-relevant concentrations of sevoflurane during infancy resulted in neurocognitive impairment and behavioral changes at 1 to 2 years of age. However, specific long-term effects of anesthetics on human brain function in neonates or fetuses are currently inconclusive. Two prospective trials examining the effect of a short anesthetic exposure have suggested no long-term neurodevelopmental consequences. In 2016, the U.S. Food and Drug Administration advisory committee issued a warning that “repeated or lengthy use of general anesthetic and sedation drugs during surgeries or procedures in children younger than 3 years or in pregnant women during their third trimester may affect the development of children’s brains.”

Limited data about anesthetic exposure in the fetal population exists. One study looked retrospectively at the use of general anesthesia for cesarean section and the incidence of learning disabilities at age 5 and found no correlation. To date, no studies have examined the neurocognitive consequences of midgestation fetal anesthesia. No general anesthetic agent is known to be superior to another, and whether exposure to general anesthetics during gestation compared to the neonatal period is more beneficial or harmful is unknown. In an effort to systematically collect current data, an international registry has been established for the purpose of assessing the long-term neurodevelopmental outcomes of fetal surgery patients ( Clinical Trials.gov identifier NCT02591745). In addition, whether in utero exposure to anesthetic drugs compared with postnatal exposure has any effect on neurocognitive outcome is not known and makes data collection and analysis challenging, as many patients undergoing fetal procedures are re-exposed to general anesthesia during infancy and childhood.

Management of Minimally Invasive Procedures

The same considerations that apply to nonobstetric surgery during pregnancy should be followed for fetal procedures. For most fetal image-guided surgery procedures (see Table 63.1 ), use of monitored anesthetic care with local anesthesia infiltration into the superficial tissues and abdominal wall provides an adequate level of maternal comfort. Administration of additional opioid, benzodiazepine, or other anesthetic agent can be used for maternal analgesia and anxiolysis. Use of supplemental anesthetic drugs will also decrease the likelihood of fetal movement via placental transfer. Local anesthetic infiltration can also be used for fetoscopic procedures, which typically employ endoscope trocars that are only 2 to 5 mm in diameter. Neuraxial techniques can be beneficial and preferable when multiple insertion sites are required; maternal immobility must be ensured; a mini-laparotomy must be performed; or concern exists about adequate patient comfort or cooperation during the procedure. General anesthesia is rarely necessary for percutaneous procedures unless placental location and fetal orientation present potential technical difficulty or uterine exteriorization is needed, as is the case with fetoscopic MMC repair.

Although maternal intravenous fluid administration should be guided by standard intraoperative requirements, use of significant amounts of pressurized uterine crystalloid irrigation into the amniotic cavity during fetoscopic surgery should be avoided because it has resulted in maternal pulmonary edema.

In cases of IUT, cord blood sampling, or thoracic shunt placement, fetal movement may displace the needle or catheter and result in trauma, bleeding, or compromise of the umbilical circulation. In one study of fetoscopic surgery, maternal administration of remifentanil (0.1 μg/kg/min) reduced fetal movement and improved operating conditions compared with maternal administration of diazepam. Although maternally administered opioids and benzodiazepines can reduce fetal motion, they do not guarantee immobility for procedures directly involving the fetus. Fetal immobility can be safely achieved with direct fetal intramuscular or umbilical venous administration of muscle relaxant. For invasive fetal procedures that involve potentially noxious stimulation to the fetus, such as shunt placement, endoscopic MMC repair, or fetal cardiac interventions, an opioid should be administered to the fetus (e.g., intramuscular fentanyl 10-20 μg/kg). When general anesthesia is employed, placental transfer of a volatile anesthetic provides significant fetal anesthesia and decreases fetal movement, but supplemental opioids should also be administered if fetal analgesia is required.

Weight-based unit doses of atropine (20 μg/kg) and epinephrine (10 μg/kg) should be immediately available in individually labeled syringes for direct fetal administration by the surgeon under ultrasonography guidance. These medications require sterile transfer to the surgical field preoperatively, meticulous labeling, and accurate dosing before commencement of the procedure. The surgeon can administer the indicated medication by a variety of routes (intramuscular, intravenous, or intracardiac) depending on the procedure and urgency of the situation. If gestational development is compatible with extrauterine life, the obstetric team should be prepared to perform an emergency cesarean delivery if fetal bradycardia persists despite efforts to resuscitate in utero. The anesthesiologist should be prepared to emergently provide maternal general anesthesia and assist with neonatal resuscitation.

Management of Open Procedures

Although most women undergo cesarean delivery with neuraxial anesthesia, general anesthesia is primarily employed for fetal surgery requiring a hysterotomy. Unlike minimally invasive fetal procedures, open fetal surgery requires profound uterine relaxation and often entails additional fetal monitoring beyond intermittent ultrasonography. Open surgery involves more surgical stimulation, hemodynamic perturbation, and risk for fetal compromise and requires direct administration of drugs to the fetus. Compared to minimally invasive procedures, open fetal procedures present greater risk to the mother. The anesthesiologist and other team members should be prepared for significant maternal and fetal blood loss, the need for maternal and fetal resuscitation, and possible emergent delivery. A volatile anesthetic is commonly administered to provide maternal and fetal anesthesia, as well as uterine relaxation, which may require a concentration of more than 2 MAC. In an effort to decrease the fetal cardiac dysfunction and abnormal umbilical artery flow associated with high levels of volatile anesthetics, more recent techniques combine 1 to 1.5 MAC of volatile anesthetic with infusions of remifentanil and propofol. The perioperative considerations for open fetal surgery are detailed in Box 63.2 .

Box 63.2

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Perioperative Medicine Patient Simulation Local Anesthetics Opioids Prehospital Care for Medical Emergencies and Trauma Anesthesia for Orthopedic Surgery

Stay updated, free articles. Join our Telegram channel

Join