- Email: [email protected]
Indicated Preterm Birth for Fetal Anomalies
Indicated Preterm Birth for Fetal Anomalies Sabrina D. Craigo, MD Between 2% and 3% of pregnancies are complicated by fetal anomalies. For most anomalies, there is no advantage to late preterm or early-term delivery. The risks of maternal or fetal complication are specific for each anomaly. Very few anomalies pose potential maternal risk. Some anomalies carry ongoing risks to the fetus, such as an increased risk of fetal death, hemorrhage, or organ damage. In a limited number of select cases, the advantages of late preterm or early-term birth may include avoiding an ongoing risk of fetal death related to the anomaly, allowing delivery in a controlled setting with availability of subspecialists and allowing direct care for the neonate with organ injury. The optimal gestational age for delivery cannot be determined for all pregnancies complicated by fetal anomalies. For most pregnancies complicated by anomalies, there is no change to obstetrical management regarding timing of delivery. For those that may benefit from late preterm or early-term delivery, variability exists such that each management plan should be individualized. Semin Perinatol 35:270-276 © 2011 Elsevier Inc. All rights reserved. KEYWORDS fetal anomaly, premature birth, timing of delivery
I
n the current era, the antenatal detection of fetal anomalies is a goal of pregnancy management. Screening programs exist for specific abnormalities (eg, neural tube defects, common trisomies), and payers usually recognize prenatal diagnosis as part of routine prenatal care. Enormous resources are used in an effort to identify significant fetal anomalies before birth. The goals of prenatal diagnosis are to identify major structural anomalies, detect aneuploidy and genetic disorders, identify conditions potentially amenable to prenatal treatment, and offer all patients options for pregnancy management. A limited number of conditions may be amenable to antenatal treatment, such as in utero surgery, fetal transfusion, or medical therapy. More often, prenatal identification of fetal abnormalities allows providers to arrange site of delivery, mode of delivery, and timing of delivery.
Prevalence Between 2% and 3% of pregnancies are complicated by fetal anomalies. Of these, approximately 1 in 6 are associated with chromosomal abnormalities.1 Despite the widespread use of
Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, Tufts Medical Center, Boston, MA. Address reprint requests to Sabrina D. Craigo, MD, Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, Tufts University School of Medicine, Tufts Medical Center, 800 Washington Street, Box 360, Boston, MA 02111. E-mail: [email protected]
270
0146-0005/11/$-see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1053/j.semperi.2011.05.011
prenatal ultrasound, only 44%-74% of anomalies are diagnosed before birth.2-4 The sensitivity of anomaly detection, particularly in a low-risk population, varies widely, both by center and by anomaly. Regardless of these limitations, patient expectations and demand for prenatal diagnosis remain high. After a fetal anomaly is diagnosed, 18%-41% of patients terminate the pregnancy.1,3,4 Rates of termination also vary by specific anomaly. For example, termination for brain abnormalities is more common than termination for cardiac anomalies. An additional 2% of pregnancies complicated by fetal anomaly will end in fetal death or stillbirth.1 Approximately 1% of pregnancies complicated by fetal anomalies that are nonchromosomal and have continued into the third trimester will require planning for site, mode, and timing of delivery. A subset of these patients will spontaneously deliver prematurely, removing the option to plan timing of delivery. For most anomalies, there is no advantage to delivering at a late preterm or early-term gestational age. Overall, the number of pregnancies for which late preterm or early-term birth will be recommended because of fetal anomalies is only a fraction of 1% of all births. The incidence of fetal anomalies among twin gestations is increased over that of singletons at 5%-6%.5-7 However, the incidence among dizygotic twins is similar to that of singletons, whereas the rate among monozygotic twins is increased more than 2-fold.6,7 Certain anomalies appear to be more common in twins, specifically some open neural tube defects
Indicated preterm birth for fetal anomalies and cardiac anomalies.8,9 The antenatal detection of anomalies in twins can be problematic: women with twin pregnancies may be imaged more often, increasing the likelihood an anomaly will be detected, but imaging may be difficult or suboptimal because of fetal positions and crowding. When anomalies are identified in twins, options for management vary by the specific anomaly but include termination, selective termination, or expectant management. Of the twin pregnancies complicated by a fetal anomaly and managed expectantly, approximately 50% deliver before 34 weeks.10,11 Delivery planning is complicated by the presence of the normal co-twin, because early delivery will place the co-twin at risk. The complicated decision making depends on multiple variables, including gestational age, chorionicity, status of the anomalous twin, potential benefit of intervention, and potential risk to the co-twin.
Risks of the Complication There are more than 100 categories of major fetal anomalies, with multiple variations within each category and the possibility of multiple anomalies in the same fetus. Significant parental anxiety is uniformly attached to these diagnoses. In contrast, the risks of maternal or fetal complication are specific to the anomaly. Very few fetal anomalies carry any potential maternal risk. Most anomalies have no ongoing fetal risk, but some specific anomalies do carry ongoing risks to the fetus, such as an increased risk of fetal death, risk of hemorrhage, and risk of ongoing or worsening organ damage. There are limited circumstances in which a late preterm birth or early-term birth will be recommended for the fetal risks associated with anomalies.
Advantages and Risks of Early Delivery For those cases in whom late preterm birth or early-term birth may be indicated, the advantages of early delivery include avoiding an ongoing risk of fetal death in utero, allowing delivery in a controlled setting with availability of subspecialists, and allowing direct care for the neonate with organ injury. Some anomalies are associated with an increased risk of fetal demise. Although delivery does not guarantee there is no risk of neonatal death, the risk of fetal demise is removed. The mother experiences a live birth, and the neonate can be cared for directly. Some neonates are best managed with the availability of subspecialists immediately after delivery, such as pediatric surgeons or pediatric otolaryngology specialists. A planned early-term birth allows for coordination of the multidisciplinary team required to optimally care for the perinatal patients. Some fetuses show sonographic or other evidence of progressive organ injury or involvement related to the anomaly. Examples include fetuses with cardiac anomalies developing hydrops or arrhythmias near term, or fetuses with bilateral renal disease developing worsening obstruction and oligohydramnios. A planned late preterm or early-term birth may allow direct care of the new-
271 born infant and potential cessation or even reversal of the organ injury (removal of the obstruction, cardiac support or surgery).
Risks of Continued Pregnancy near Term Fetal Risks Although most fetal anomalies carry no risk in utero, some are associated with an increased risk of fetal death. Fetal anomalies account for 20% of cases of stillbirth, whereas fetal chromosomal abnormalities are found in 8%-13%.12,13 Antepartum fetal testing is often recommended for pregnancies complicated by fetal anomaly. Benefits of this strategy are unknown, but if abnormal fetal testing is found and is persistent, delivery will be recommended. If used, such testing strategies should not begin until the late preterm gestational age, to avoid preterm deliveries for a nonspecific test. Pregnancies complicated by fetal anomalies are at increased risk for spontaneous preterm birth. Those anomalies associated with polyhydramnios are at highest risk. Twin pregnancies with one anomalous fetus have a greater rate of preterm birth than twins without anomalies.10,14 There is an ongoing risk of fetal organ injury related to some anomalies. Space-occupying lesions in the chest compromise lung development. Cardiac or renal function may deteriorate with ongoing obstruction. A sacrococcygeal tumor can enlarge and lead to fetal cardiovascular compromise. However, not all ongoing injuries are similar in severity or implications. There are no recommendations for early delivery for a fetal ovarian cyst that is thought to be hemorrhaging or torsing. Perforation of a fetal bowel obstruction is not an indication for early delivery. In contrast, a cardiac lesion that leads to a new onset of hydrops may be an indication for delivery. The difference is not just in the organ system but in the associated risk of death, permanence of the organ injury, and possibility of reversal with appropriate intervention.
Maternal Risks Most fetal anomalies carry no maternal risks other than anxiety. In a few selected circumstances, there are specific maternal risks of continuing pregnancy near term. These include maternal risk of uterine rupture for the small number of patients who undergo invasive fetal intervention during pregnancy. Currently, there are fewer than 25 reported cases of open fetal surgery per year in the United States, and the number is decreasing as interventions shifts to endoscopic techniques and limited indications remain. Open fetal surgery involves a hysterotomy; therefore, current recommendations are for delivery at 36 weeks after documentation of fetal lung maturity.15 After endoscopic intervention during pregnancy, timing of delivery is not affected unless there is a need for the ex utero intrapartum technique (EXIT).16 After a shunt is placed, recommendations for delivery should be case specific. The risk of uterine rupture is not thought to be increased, but the mean gestational age at delivery after thoracoamniotic shunt or
S.D. Craigo
272 thoracentesis is 35 weeks, related to preterm labor and preterm rupture of membranes.17 Another potential maternal risk is “mirror syndrome,” which can develop if fetal hydrops is persistent, and is often associated with pre-eclampsia. Although a rare occurrence, if mirror syndrome is suspected at any gestational age and the hydrops is not thought to be reversible, delivery is indicated. If hydrops is newly diagnosed in the late preterm or earlyterm gestational period, it carries significant risk to the fetus and potential risk to the mother if ongoing, so delivery is generally recommended. Fetal adrenal tumors may also carry a risk of maternal hypertension, but would not be an indication for late preterm delivery in the absence of maternal affects.
Neonatal Risks For those pregnancies with fetal anomalies and an indication for late preterm or early-term delivery, the neonatal risks of continued pregnancy near term are primarily the risk of unscheduled or unplanned delivery. The need for coordinated multidisciplinary care is rarely more acute than with delivery of a fetus with a significant fetal anomaly requiring intervention immediately after birth. Although staffing issues are often a consideration when delivering fetuses with anomalies (the pediatric surgeon and pediatric staff would prefer to care for a fetus with gastroschisis during a weekday shift), the urgency with which a specialist is required varies by the specific anomaly. This has been acutely demonstrated in Europe, where several neonatal deaths occurred after endoscopic tracheal occlusion for diaphragmatic hernias.16 The deaths occurred when mothers presented in labor before the tracheal plug could be removed, and staff with expertise in removing the occlusive devices were not present at delivery. For a subset of anomalies, coordinated subspecialty care immediately after birth is critical, so planned, prelabor delivery is the best course of action. Accomplishing this is not straightforward. Some recommend delivery after fetal lung maturity is documented by amniocentesis, while other experts would recommend planned delivery in the early-term period (37-38 weeks). There is little to no data to guide this decision making, but either approach would be reasonable when coordinated care is important. EXIT is a procedure that maintains fetal oxygenation through the existing fetal placental circulation at delivery to allow a procedure to be performed immediately on the fetus. Examples of fetal abnormalities for which EXIT may be recommended include fetal neck masses or congenital diaphragmatic hernia after tracheal occlusion (Table 1). During EXIT, the fetal airway would be established and the tracheal clip or balloon removed. The technique is used at the time of cesarean section, with general anesthesia to ensure uterine hypotonia. Steps are taken to maintain uterine volume in attempt to prevent abruption. Efforts are also made to prevent maternal hypotension so as to maintain adequate gas exchange. The EXIT procedure requires significant coordination and teams assigned to care for mother and fetus. Maternal risks
Table 1 Conditions Considered Potential Indications for Ex Utero Intrapartum Treatment (EXIT) Reversal of tracheal occlusion Fetal masses of neck, lung, or mediastinum Tracheal or laryngeal atresia Suspected need for ECMO (CDH, HLHS, AS with intact septum) Significant micrognathia Modified from Bianchi et al.18 AS, aortic stenosis; CDH, congenital diaphragmatic hernia; ECMO, extracorporeal membrane oxygenation; HLHS, hypoplastic left heart syndrome.
are not trivial and include hemorrhage, wound infection, and possible hysterectomy. Fetal complications can also occur. The logistical requirements and personnel needed to perform an EXIT procedure are best accomplished for a scheduled, prelabor delivery. Therefore, if an EXIT procedure is planned, late preterm or early-term delivery is recommended. The list of procedures for which EXIT has been performed is increasing, with some centers performing “fetal” surgery using the placental circulation.18,19
Optimal Gestational Age for Delivery: Recommendations by Specific Anomaly The optimal gestational age for delivery cannot be determined for the group of pregnancies complicated by fetal anomalies as a whole. Most types of fetal anomalies are not associated with complications that will put the fetus at risk as the pregnancy continues to term. In most cases, there is no change to obstetrical management regarding timing of delivery. Among the group of anomalies that potentially carries ongoing risk to the fetus and may benefit from late preterm birth or early-term birth, variability exists such that each management plan requires individualization. Table 2 lists those anomalies for which there is no expected change to gestational age at delivery, by organ system. The following paragraphs outline the possible scenarios in which late preterm or early-term delivery may be indicated because of a specific fetal anomaly.
Central Nervous System Anomalies Anomalies with a risk of worsening hydrocephaly and/or macrocephaly may benefit from delivery in the early-term period (37-38 weeks). These include Dandy Walker malformation with hydrocephaly and obstructive hydrocephaly. There is a theoretic concern that ongoing severe hydrocephaly will result in more cerebral cortical damage and more macrocrania. Whether delivery 1-2 weeks early changes any long-term outcome for this group of fetuses has not been studied, but taking the entire clinical picture into account, delivery in the early-term period is reasonable on an individual basis.20 The subgroup of anomalies with a risk of intracranial hemorrhage may also benefit from delivery in the early-term pe-
Indicated preterm birth for fetal anomalies
273
Table 2 Fetal Anomalies Not Associated With Need for Early Delivery Central nervous system Agenesis of corpus callosum Porencephaly Microcephaly Dandy Walker malformation Craniofacial Cleft lip/palate Macroglossia Neck Small goiter Chest Bronchopulmonary sequestration Tracho-esophageal atresia Cardiac Atrial septal defect Ventricular septal defect Tachyarrhythmias* Hypoplastic right ventricle Hypoplastic left ventricle Double outlet R ventricle* Abdominal wall defects Body-stalk anomaly Omphalocele Fetal tumors Liver tumor Wilm’s tumor Genitourinary Polycystic kidney disease Minimal hydronephrosis Gastrointenstinal Liver calcifications Duodenal atresia Abdominal calcification Skeletal anomalies Chondrodysplasia punctata Campomelic dysplasia Osteogenesis imperfecta Thanatophoric dysplasia Extremities Amniotic band syndrome Clinodactyly Ectrodactyly Chromosomal abnormalities Trisomy 13 Pallister–Killian syndrome Sex chromosome abnormalities
Anencephaly Encephalocele Macrocephaly
Arachnoid cysts Hydranencephaly Myelomeningocele
Hemifacial microsomia Microphthalmia
Hyper/hypotelorism Mild micrognathia
Small hygroma Stable CCAM Hydrothorax
Pulmonary agenesis Cystic lesions
AV canal defect* Pulmonary stenosis Bradyarrhythmias* Aortic stenosis* Tetralogy of Fallot Truncus arteriorsus
Cardiomyopathy Intracardiac tumors Heterotaxy Transposition* Coarctation Ebstein anomaly*
Pentalogy of Cantrell Cloacal extrophy
Bladder extrophy Gastroschisis*
Mesoblastic nephroma
Neuroblastoma
Cloaca Renal agenesis
Echogenic kidneys Ambiguous genitalia
Pyloric atresia Jejunoileal atresia Hirshprung’s disease
Imperforate anus Colonic atresia
Achondroplasia Jarco Levin syndrome Ellis-van Creveld Hypophospatasia
Hemivertebrae Achondrogenesis Diastrophic dysplasia Short rib polydactyly
Arthrogryposis Clubfoot Polydactyly
Radial aplasia Syndactyly
Trisomy 18 Turner syndrome DiGeorge syndrome
Trisomy 21 Triploidy
AV, atrioventricular; CCAM, congenital cystic adenomatoid malformation. *Consideration should be given to delivery if hydrops or a new arrhythmia develops.
riod (37-38 weeks). These include fetuses with evidence of intracranial bleeding and those with vein of Galen aneurysms. Fetuses at risk for intracranial bleeding—those with existing evidence of a bleed or those complicated by alloimmune thrombocytopenia—inherently have an ongoing risk of bleeding. For those infants with evidence of subdural, intraventricular, or cerebellar bleeding, their prognosis can be very poor, depending on the extent of existing hemorrhage. Delivery in the early-term period is reasonable for selected cases, but no change to obstetrical management is
recommended for those with poorest prognosis.21-23 For vein of Galen aneurysms, use of a scoring system may indicate which pregnancies would not benefit from intervention and which may benefit from delivery in a tertiary center before cardiovascular compromise worsens.24
Craniofacial The craniofacial anomaly that may benefit from scheduled delivery, before labor, is significant micrognathia. In some centers, this diagnosis is managed with an EXIT procedure
S.D. Craigo
274 because of the difficulties that can be encountered in establishing an airway immediately after birth. There are no criteria to distinguish severe from milder forms of micrognathia, but sonographic appearance of marked hypoplasia of the chin and presence of associated polyhydramnios should prompt consideration of scheduled delivery in the early-term period (37-38 weeks) with pediatric specialists available.
Neck A subgroup of pregnancies complicated by fetal neck anomalies requires coordinated, tertiary care immediately after birth. The diagnoses of congenital high airway obstruction sequence (ie, CHAOS), large fetal goiter, and large cystic hygroma can be considered indications for an EXIT procedure and may benefit from scheduled delivery before labor.
Chest Congenital diaphragmatic hernia (CDH) has significant variability in clinical presentation and outcome. More than onethird are further complicated by other anomalies. The size and side of the hernia, position of the liver, the size of the herniated structures, and cardiac position have been correlated with postnatal clinical course. Currently, the observed to expected lung-to-head ratio is the most widely criteria used to assess the severity of the CDH and determine which patients may benefit from in utero therapy. Endoscopic fetal tracheal occlusion by various means is offered in several centers to patients with evidence of severe pulmonary hypoplasia Tracheal occlusion leads to subsequent growth of the fetal lungs, reducing lung hypoplasia and improving survival.16 Those pregnancies with CDH not requiring antenatal treatment can be delivered in the early-term period (37-38 weeks). Some centers advocate delivery by EXIT at 37 weeks with immediate conversion to extracorporeal membrane oxygenation. If endoscopic in utero tracheal occlusion has been performed, optimal management involves endoscopic plug removal at 34 weeks. If removal of the occlusion is not accomplished before labor, delivery with an EXIT procedure at 37 weeks is reasonable, or at the very least, delivery should be coordinated at a center with expertise in removing the specific occlusive device.16 A small number of pregnancies with congenital cystic adenomatoid malformations are complicated by hydrops in the antepartum period and are resected in utero at a few centers in the United States. Therapeutic options include open fetal surgery with fetal lobectomy, aspiration or shunting of a dominant cyst, or postnatal resection. For those that have undergone in utero resection, late preterm delivery is recommended at 36 weeks. Frequently, these pregnancies are complicated by preterm premature rupture of membranes or preterm labor, so delivery planning is not required. For those presenting after 32 weeks’ gestation, delivery with EXIT procedure and immediate resection is being offered in some centers.15
Cardiac There are no cardiac anomalies that require early delivery when present in isolation. Late preterm or early-term delivery may be recommended, however, if a fetus with a cardiac anomaly develops hydrops or a complicating fetal arrhythmia. There are no data demonstrating benefit of this approach. Hydrops carries a significant risk of mortality and delivery may not improve outcome for these fetuses. However, when following patients with a diagnosis of fetal cardiac anomaly beyond 34 weeks’ gestation, the new onset of hydrops or an arrhythmia suggests deterioration, and delivery will allow direct intensive care for the baby, in attempt to improve outcome.
Abdominal Wall Defects Most abdominal wall defects carry no indication for change to obstetrical management for timing of delivery. The abdominal wall defect for which planned delivery before labor would be recommended is a giant omphalocele with liver outside the fetus. Although a benefit of cesarean delivery has not been clearly demonstrated for omphaloceles, physicians at many centers would recommend prelabor cesarean delivery in this clinical scenario in attempt to avoid liver injury. Without data demonstrating a benefit of this strategy, delivery in the late term period (37-38 weeks) would be reasonable. Controversy over management of gastroschisis has been longstanding, and to date, the optimal timing of delivery remains elusive. In the only randomized trial to compare elective delivery at 36 weeks versus awaiting spontaneous labor, no significant benefit from elective preterm delivery for gastroschisis was found.25 In contrast, others have reported improved outcome among a small number of neonates with gastroschisis delivered electively at 34 weeks, compared with historic controls.26 Another group reported outcomes of infants with gastroschisis delivered before or after 36 weeks. The study was retrospective, and the indication for delivery was not reported, but neonatal length of stay and time to full feeds were significantly worse among those infants delivered before 36 weeks.27 Similarly, at least 2 other retrospective studies of patients with gastroschisis found no benefit of delivery before 37 weeks.28,29 Efforts to determine the best time to intervene for gastroschisis on the basis of the appearance of the bowel on ultrasound have not been successful. A review of 10 observational studies found no clear evidence that bowel dilation over 10 or 18 mm was associated with worse outcomes.30 Others have suggested, on the basis of a few patients, that bowel dilation ⬎25 mm is associated with nonreassuring fetal testing and fetal demise.31 Intrauterine fetal growth restriction frequently coincides with gastroschisis. The presence of growth restriction and the increased risk of fetal death associated with gastroschisis will lead most perinatologists to recommend delivery in the early-term period. Given the frequency with which gastroschisis is diagnosed, it should also be noted that we work closely with our pediatric surgical colleagues when managing these pregnancies. Their clinical opinions on man-
Indicated preterm birth for fetal anomalies agement and interpretation of the existing data may also influence timing of delivery. Current data are not sufficient to determine the optimal timing of delivery for gastroschisis. The data favoring elective late preterm delivery is very limited, but early-term delivery (37-38 weeks) is a reasonable option. In the absence of growth restriction, awaiting spontaneous labor or delivery at term is also reasonable.
Fetal Tumors Most fetal tumors can be managed without recommendation for early delivery. Recommendations for some rare tumors alter timing of delivery. Cervical teratomas may be managed with a planned EXIT procedure, or with coordinated care involving immediate availability of a team to manage the airway of the neonate. Recommendations for fetal retinoblastomas are for delivery when fetal lung maturity is documented, to allow immediate postnatal therapy of the tumor.18 Sacrococcygeal teratomas carry an ongoing risk of cardiovascular compromise and/or bleeding, so delivery is recommended when fetal lung maturity is documented.32 Outcome has been reported as nearly uniformly fatal if placentomegaly or hydrops are present,33,34 so this is one of the few anomalies for which open fetal surgery is still offered in select centers.
Genitourinary Most genitourinary anomalies carry no indication for change to obstetrical management for timing of delivery. For a subset of anomalies in which ultrasound findings suggest overall renal function is worsening, delivery may be recommended when fetal lung maturity is documented or when oligohydramnios develops in the late preterm or early-term period. The conditions for which these recommendations apply include obstructive uropathy developing oligohydramnios, multicystic kidney with evidence of contralateral renal compromise, or a single kidney with evidence of obstruction and oligohydramnios. Delivery allows for intervention to relieve the obstruction, and risks attributable to oligohydramnios are removed. Whether this approach improves renal function long term has not been studied.
Other When gastrointestinal abnormalities, skeletal anomalies, anomalies of the extremities, or chromosomal abnormalities are present, there is no indication for change to obstetrical management for timing of delivery.
Special Considerations Twins with One Anomaly Twins with one fetus affected by anomaly should be managed on a case-by-case basis. As with singletons, recommendations will differ by anomaly. Factors that must be considered include chorionicity, gestational age, and nature of the specific anomaly. In general, after 32 weeks of gestation anomaly-specific recommendations for singletons can be followed
275 for twins, but if late preterm delivery is being considered, potential implications for the normal co-twin must be addressed. If monoamniotic twins are complicated by a fetal anomaly, there is ongoing risk of fetal death of the normal twin at all times. Without actual removal of the anomalous twin, even selective termination by cord occlusion or thermocoagulation techniques will result in a nonviable twin that can lead to cord accident of the co-twin. There has been limited success of the options for intervention.35,36 In this specific scenario, preterm delivery at 32-34 weeks is reasonable.
Lethal Anomalies Pregnancies complicated by a lethal anomaly are worthy of special consideration. Term delivery is recommended, ideally when spontaneous labor ensues. In selected cases preterm delivery may be reasonable to consider to provide compassionate care to the mother because neonatal outcome will not be changed. In all cases a primary goal of management should be to avoid interventions that pose any risk to the mother. Expectant management for obstetrical complications (eg, preterm rupture of membranes, abruption, pre-eclampsia) is contraindicated in the presence of a known lethal fetal anomaly.
Research Gaps The difficulty in determining the best timing for delivery in pregnancies complicated by fetal anomalies is related to the sheer number of different types of anomalies, each with specific implications and variations. Fortunately, most do not require any change to timing of delivery. For those that may possibly benefit from earlier delivery, research is needed regarding the role of amniocentesis for fetal lung maturity testing. Further study is also needed on the best course of management if amniocentesis is used and testing does not confirm lung maturity. Additionally, the possible role of steroids before late preterm or early-term delivery has not been determined. More specific to fetal anomalies, further research is needed to improve our ability to assess organ injury antenatally and to predict which fetuses may benefit from earlier delivery. This research will require anomaly specific focus, and likely collaborative efforts to study the more uncommon anomalies. Research efforts should also ideally address obstetrical, neonatal, and long-term outcomes for these complicated pregnancies. Only if perinatal and neonatal outcomes are studied will we determine if survival is improved when delivery is undertaken for evidence of hydrops, or if fetal death is avoided but neonatal death follows. The recent publication of the results of the MOMS (Management of Myelomeningocele) trial37 demonstrates both the difficulty in performing such research, and the importance of evaluating new therapies. Research for potential invasive therapies for lethal and nonlethal anomalies will no doubt continue in attempt to further define the role of fetal surgery, even as the indications for open fetal surgery appear to be-
276 come more limited. The role of EXIT should also be studied and defined, as the procedure itself carries maternal risks and its use influences timing of delivery.
Conclusions Prenatally diagnosed fetal anomalies complicate 1%-2% of pregnancies. Only a fraction of the anomalies will lead to recommendations for late preterm or early-term birth. For a limited number of conditions, late preterm birth, or more often early-term birth, will be indicated. The conditions that may benefit from delivery before 39 weeks include those with evidence of worsening organ injury (usually kidney, heart, or possibly brain) and those in which coordination of care for neonate immediately after birth will be critical. Opportunities for research in this area will be anomaly specific.
References 1. Dolk H, Loane M, Garne E: The prevalence of congenital anomalies in Europe. Advances Experiment. Med Biol 686:349-364, 2010 2. Ewigman BG, Crane JP, Frigoletto FD, et al: Effect of prenatal ultrasound screening on perinatal outcome. N Engl J Med 329:821-827, 1993 3. Grandjean H, Larroque D, Levi S: The performance of routine ultrasonographic screening of pregnancies in the Eurofetus Study. Am J Obstet Gynecol 181:446-454, 1999 4. Stoll C, Tenconi R, Clementi M, et al: Detection of congenital anomalies by fetal ultrasonographic examination across Europe. Community Genet 4:225-232, 2001 5. Schinzel AA, Smith DW, Miller JR: Monozygotic twinning and structural defects. J Pediatr 95:921-930, 1979 6. Drugan A, Johnson MP, Krivchenia EL, et al: Genetics and genetic counseling, in Gall SA (ed): Multiple Pregnancy and Delivery. St Louis, Mosby, 1996 pp 85-97 7. Layde PM, Erickson JD, Falek A, et al: Congenital malformation in twins. Am J Hum Genet 32:69-78, 1980 8. Little J, Nevin NC: Congenital anomalies in twins in Northern Ireland. III: Anomalies of the cardiovascular system 1974-1979. Acta Geneti Med Gemellol (Roma) 38:27-35, 1989 9. Ramos-Arroyo MA: Birth defects in twins: Study in a Spanish population. Acta Genet Med Gemellol (Roma) 40:337-344, 1991 10. Malone FD, Craigo SD, Chelmow D, et al: Outcome of twin gestations complicated by a single anomalous fetus. Obstet Gynecol 88: 1-5, 1996 11. Adra AM, Landy HJ, Nahmias J, et al: The fetus with gastroschisis: Impact of route of delivery and prenatal ultrasonography. Am J Obstet Gynecol 174:540-546, 1996 12. Pauli RM, Reiser CA, Lebovitz RM, et al: Wisconsin Stillbirth Service Program: I. Establishment and assessment of a community-based program for etiologic investigation of intrauterine deaths. Am J Med Genet 50:166-134, 1994 13. American College of Obstetricians and Gynecologists (2009) Management of stillbirth. ACOG Practice Bulletin No. 102. Obstet Gynecol 113:748-761, 2009 14. Gul A, Cebeci A, Aslan H, et al: Perinatal outcomes of twin pregnancies discordant for major anomalies. Fetal Diagn Ther 20:244-248, 2005 15. Adzick NS: Open fetal surgery for life-threatening fetal anomalies. Semin Fetal Neonat Med 15:1-8, 2010
S.D. Craigo 16. Deprest J, Nicolaides K, Doné E, et al: Technical aspects of fetal endoscopic tracheal occlusion for congenital diaphragmatic hernia. J Pediatr Surg 46:22-32, 2011 17. Deurloo KL, Devlieger R, Lopriore E, et al: Isolated fetal hydrothorax with hydrops: A systematic review of prenatal treatment options. Prenat Diagn 27:893-899, 2007 18. Bianchi DW, Crombleholme TM, D’Alton ME, et al: Fetology: Diagnosis and Management of the Fetal Patient (ed 2). New York, McGraw-Hill Medical, 2010 19. Hedrick HL, Flake AW, Crombleholme TM, et al: The ex utero intrapartum therapy procedure for high-risk fetal lung lesions. J Pediatr Surg 40:1038-1044, 2005 20. Wilkins-Haug L: Considerations for delivery of infants with congenital abnormalities. Obstet Gynecol Clin N Am. 26:399-412, 1999 21. Pretorius DH, Singh S, Manco-Johnson M, et al: In utero diagnosis of intracranial hemorrhage resulting in fetal hydrocephalus. J Reprod Med 3:136-138, 1986 22. Ghi T, Simonazzi G, Perolo A, et al: Outcome of antenatally diagnosed intracranial hemorrhage: Case series and review of the literature. Ultrasound Obstet Gynecol 22:121-130, 2003 23. Murphy MF, Bussel JB: Advances in the management of alloimmune thrombocytopenia. Br J Hematol 136:366-378, 2006 24. Yuval Y, Lerner A, Lipitz S, et al: Prenatal diagnosis of vein of galen aneurysmal malformation: Report of two cases with proposal for prognostic indices. Prenat Diagn 17:972-977, 1997 25. Logghe HL, Mason GC, Thornton JG, et al: A randomized controlled trial of elective preterm delivery of fetuses with gastroschisis. J Pediatr Surg 40:1726-1731, 2005 26. Serra A, Fitze G, Kamin G, et al: Preliminary report on elective preterm delivery at 34 weeks and primary abdominal closure for the management of gastroschisis. Eur J Pediatr Surg 18:32-37, 2008 27. Ergün O, Barksdale E, Ergün FS, et al: The timing of delivery of infants with gastroschisis influences outcome. J Pediatr Surg 40:424-428, 2005 28. Huang J, Kurkchubasche AG, Carr SR, et al: Benefits of term delivery in infants with antenatally diagnosed gastroschisis. Obstet Gynecol 100: 695-699, 2002 29. Simmons M, Georgeson KE: The effect of gestational age at birth on morbidity in patients with gastroschisis. J Pediatr Surg 31:1060-1062, 1996 30. Tower C, Ong SS, Ewer AK, et al: Prognosis in isolated gastroschisis with bowel dilatation: A systematic review. Arch Dis Child Fetal Neonat Ed 94:F268, 2009 31. Heinig J, Müller V, Schmitz R, et al: Sonographic assessment of the extra-abdominal fetal small bowel in gastroschisis: A retrospective longitudinal study in relation to prenatal complications. Prenat Diagn 28:109, 2008 32. Adzick NS, Harrison MR: The unborn surgical patient. Curr Probl Surg 35:1, 1994 33. Bond SJ, Harrison MR, Schmidt KG, et al: Death due to high-output cardiac failure in fetal sacrococcygeal teratoma. J Pediatr Surg 25:12871291, 1990 34. Flake AW, Harrison MR, Adzick NS, et al: Fetal sacrococcygeal teratoma. J Pediatr Surg 21:563-566, 1986 35. Middeldorp JM, Klumper FJCM, Oepkes D, et al: Selective feticide in monoamniotic twin pregnancies by umbilical cord occlusion and transection. Fetal Diagn Ther 23:121-125, 2008 36. Young BK, Roque H, Abdelhak Y, et al: Endoscopic ligation of umbilical cord at 19 weeks’ gestation in monochorionic monamniotic twins discordant for hypoplastic left heart syndrome. Fetal Diagn Ther 16:6164, 2001 37. Adzick NS, Thom EA, Spong CY, et al: A randomized trial of prenatal versus postnatal repair of myelomeningocele. N. Engl. J. Med. 364:9931004, 2011
I
n the current era, the antenatal detection of fetal anomalies is a goal of pregnancy management. Screening programs exist for specific abnormalities (eg, neural tube defects, common trisomies), and payers usually recognize prenatal diagnosis as part of routine prenatal care. Enormous resources are used in an effort to identify significant fetal anomalies before birth. The goals of prenatal diagnosis are to identify major structural anomalies, detect aneuploidy and genetic disorders, identify conditions potentially amenable to prenatal treatment, and offer all patients options for pregnancy management. A limited number of conditions may be amenable to antenatal treatment, such as in utero surgery, fetal transfusion, or medical therapy. More often, prenatal identification of fetal abnormalities allows providers to arrange site of delivery, mode of delivery, and timing of delivery.
Prevalence Between 2% and 3% of pregnancies are complicated by fetal anomalies. Of these, approximately 1 in 6 are associated with chromosomal abnormalities.1 Despite the widespread use of
Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, Tufts Medical Center, Boston, MA. Address reprint requests to Sabrina D. Craigo, MD, Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, Tufts University School of Medicine, Tufts Medical Center, 800 Washington Street, Box 360, Boston, MA 02111. E-mail: [email protected]
270
0146-0005/11/$-see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1053/j.semperi.2011.05.011
prenatal ultrasound, only 44%-74% of anomalies are diagnosed before birth.2-4 The sensitivity of anomaly detection, particularly in a low-risk population, varies widely, both by center and by anomaly. Regardless of these limitations, patient expectations and demand for prenatal diagnosis remain high. After a fetal anomaly is diagnosed, 18%-41% of patients terminate the pregnancy.1,3,4 Rates of termination also vary by specific anomaly. For example, termination for brain abnormalities is more common than termination for cardiac anomalies. An additional 2% of pregnancies complicated by fetal anomaly will end in fetal death or stillbirth.1 Approximately 1% of pregnancies complicated by fetal anomalies that are nonchromosomal and have continued into the third trimester will require planning for site, mode, and timing of delivery. A subset of these patients will spontaneously deliver prematurely, removing the option to plan timing of delivery. For most anomalies, there is no advantage to delivering at a late preterm or early-term gestational age. Overall, the number of pregnancies for which late preterm or early-term birth will be recommended because of fetal anomalies is only a fraction of 1% of all births. The incidence of fetal anomalies among twin gestations is increased over that of singletons at 5%-6%.5-7 However, the incidence among dizygotic twins is similar to that of singletons, whereas the rate among monozygotic twins is increased more than 2-fold.6,7 Certain anomalies appear to be more common in twins, specifically some open neural tube defects
Indicated preterm birth for fetal anomalies and cardiac anomalies.8,9 The antenatal detection of anomalies in twins can be problematic: women with twin pregnancies may be imaged more often, increasing the likelihood an anomaly will be detected, but imaging may be difficult or suboptimal because of fetal positions and crowding. When anomalies are identified in twins, options for management vary by the specific anomaly but include termination, selective termination, or expectant management. Of the twin pregnancies complicated by a fetal anomaly and managed expectantly, approximately 50% deliver before 34 weeks.10,11 Delivery planning is complicated by the presence of the normal co-twin, because early delivery will place the co-twin at risk. The complicated decision making depends on multiple variables, including gestational age, chorionicity, status of the anomalous twin, potential benefit of intervention, and potential risk to the co-twin.
Risks of the Complication There are more than 100 categories of major fetal anomalies, with multiple variations within each category and the possibility of multiple anomalies in the same fetus. Significant parental anxiety is uniformly attached to these diagnoses. In contrast, the risks of maternal or fetal complication are specific to the anomaly. Very few fetal anomalies carry any potential maternal risk. Most anomalies have no ongoing fetal risk, but some specific anomalies do carry ongoing risks to the fetus, such as an increased risk of fetal death, risk of hemorrhage, and risk of ongoing or worsening organ damage. There are limited circumstances in which a late preterm birth or early-term birth will be recommended for the fetal risks associated with anomalies.
Advantages and Risks of Early Delivery For those cases in whom late preterm birth or early-term birth may be indicated, the advantages of early delivery include avoiding an ongoing risk of fetal death in utero, allowing delivery in a controlled setting with availability of subspecialists, and allowing direct care for the neonate with organ injury. Some anomalies are associated with an increased risk of fetal demise. Although delivery does not guarantee there is no risk of neonatal death, the risk of fetal demise is removed. The mother experiences a live birth, and the neonate can be cared for directly. Some neonates are best managed with the availability of subspecialists immediately after delivery, such as pediatric surgeons or pediatric otolaryngology specialists. A planned early-term birth allows for coordination of the multidisciplinary team required to optimally care for the perinatal patients. Some fetuses show sonographic or other evidence of progressive organ injury or involvement related to the anomaly. Examples include fetuses with cardiac anomalies developing hydrops or arrhythmias near term, or fetuses with bilateral renal disease developing worsening obstruction and oligohydramnios. A planned late preterm or early-term birth may allow direct care of the new-
271 born infant and potential cessation or even reversal of the organ injury (removal of the obstruction, cardiac support or surgery).
Risks of Continued Pregnancy near Term Fetal Risks Although most fetal anomalies carry no risk in utero, some are associated with an increased risk of fetal death. Fetal anomalies account for 20% of cases of stillbirth, whereas fetal chromosomal abnormalities are found in 8%-13%.12,13 Antepartum fetal testing is often recommended for pregnancies complicated by fetal anomaly. Benefits of this strategy are unknown, but if abnormal fetal testing is found and is persistent, delivery will be recommended. If used, such testing strategies should not begin until the late preterm gestational age, to avoid preterm deliveries for a nonspecific test. Pregnancies complicated by fetal anomalies are at increased risk for spontaneous preterm birth. Those anomalies associated with polyhydramnios are at highest risk. Twin pregnancies with one anomalous fetus have a greater rate of preterm birth than twins without anomalies.10,14 There is an ongoing risk of fetal organ injury related to some anomalies. Space-occupying lesions in the chest compromise lung development. Cardiac or renal function may deteriorate with ongoing obstruction. A sacrococcygeal tumor can enlarge and lead to fetal cardiovascular compromise. However, not all ongoing injuries are similar in severity or implications. There are no recommendations for early delivery for a fetal ovarian cyst that is thought to be hemorrhaging or torsing. Perforation of a fetal bowel obstruction is not an indication for early delivery. In contrast, a cardiac lesion that leads to a new onset of hydrops may be an indication for delivery. The difference is not just in the organ system but in the associated risk of death, permanence of the organ injury, and possibility of reversal with appropriate intervention.
Maternal Risks Most fetal anomalies carry no maternal risks other than anxiety. In a few selected circumstances, there are specific maternal risks of continuing pregnancy near term. These include maternal risk of uterine rupture for the small number of patients who undergo invasive fetal intervention during pregnancy. Currently, there are fewer than 25 reported cases of open fetal surgery per year in the United States, and the number is decreasing as interventions shifts to endoscopic techniques and limited indications remain. Open fetal surgery involves a hysterotomy; therefore, current recommendations are for delivery at 36 weeks after documentation of fetal lung maturity.15 After endoscopic intervention during pregnancy, timing of delivery is not affected unless there is a need for the ex utero intrapartum technique (EXIT).16 After a shunt is placed, recommendations for delivery should be case specific. The risk of uterine rupture is not thought to be increased, but the mean gestational age at delivery after thoracoamniotic shunt or
S.D. Craigo
272 thoracentesis is 35 weeks, related to preterm labor and preterm rupture of membranes.17 Another potential maternal risk is “mirror syndrome,” which can develop if fetal hydrops is persistent, and is often associated with pre-eclampsia. Although a rare occurrence, if mirror syndrome is suspected at any gestational age and the hydrops is not thought to be reversible, delivery is indicated. If hydrops is newly diagnosed in the late preterm or earlyterm gestational period, it carries significant risk to the fetus and potential risk to the mother if ongoing, so delivery is generally recommended. Fetal adrenal tumors may also carry a risk of maternal hypertension, but would not be an indication for late preterm delivery in the absence of maternal affects.
Neonatal Risks For those pregnancies with fetal anomalies and an indication for late preterm or early-term delivery, the neonatal risks of continued pregnancy near term are primarily the risk of unscheduled or unplanned delivery. The need for coordinated multidisciplinary care is rarely more acute than with delivery of a fetus with a significant fetal anomaly requiring intervention immediately after birth. Although staffing issues are often a consideration when delivering fetuses with anomalies (the pediatric surgeon and pediatric staff would prefer to care for a fetus with gastroschisis during a weekday shift), the urgency with which a specialist is required varies by the specific anomaly. This has been acutely demonstrated in Europe, where several neonatal deaths occurred after endoscopic tracheal occlusion for diaphragmatic hernias.16 The deaths occurred when mothers presented in labor before the tracheal plug could be removed, and staff with expertise in removing the occlusive devices were not present at delivery. For a subset of anomalies, coordinated subspecialty care immediately after birth is critical, so planned, prelabor delivery is the best course of action. Accomplishing this is not straightforward. Some recommend delivery after fetal lung maturity is documented by amniocentesis, while other experts would recommend planned delivery in the early-term period (37-38 weeks). There is little to no data to guide this decision making, but either approach would be reasonable when coordinated care is important. EXIT is a procedure that maintains fetal oxygenation through the existing fetal placental circulation at delivery to allow a procedure to be performed immediately on the fetus. Examples of fetal abnormalities for which EXIT may be recommended include fetal neck masses or congenital diaphragmatic hernia after tracheal occlusion (Table 1). During EXIT, the fetal airway would be established and the tracheal clip or balloon removed. The technique is used at the time of cesarean section, with general anesthesia to ensure uterine hypotonia. Steps are taken to maintain uterine volume in attempt to prevent abruption. Efforts are also made to prevent maternal hypotension so as to maintain adequate gas exchange. The EXIT procedure requires significant coordination and teams assigned to care for mother and fetus. Maternal risks
Table 1 Conditions Considered Potential Indications for Ex Utero Intrapartum Treatment (EXIT) Reversal of tracheal occlusion Fetal masses of neck, lung, or mediastinum Tracheal or laryngeal atresia Suspected need for ECMO (CDH, HLHS, AS with intact septum) Significant micrognathia Modified from Bianchi et al.18 AS, aortic stenosis; CDH, congenital diaphragmatic hernia; ECMO, extracorporeal membrane oxygenation; HLHS, hypoplastic left heart syndrome.
are not trivial and include hemorrhage, wound infection, and possible hysterectomy. Fetal complications can also occur. The logistical requirements and personnel needed to perform an EXIT procedure are best accomplished for a scheduled, prelabor delivery. Therefore, if an EXIT procedure is planned, late preterm or early-term delivery is recommended. The list of procedures for which EXIT has been performed is increasing, with some centers performing “fetal” surgery using the placental circulation.18,19
Optimal Gestational Age for Delivery: Recommendations by Specific Anomaly The optimal gestational age for delivery cannot be determined for the group of pregnancies complicated by fetal anomalies as a whole. Most types of fetal anomalies are not associated with complications that will put the fetus at risk as the pregnancy continues to term. In most cases, there is no change to obstetrical management regarding timing of delivery. Among the group of anomalies that potentially carries ongoing risk to the fetus and may benefit from late preterm birth or early-term birth, variability exists such that each management plan requires individualization. Table 2 lists those anomalies for which there is no expected change to gestational age at delivery, by organ system. The following paragraphs outline the possible scenarios in which late preterm or early-term delivery may be indicated because of a specific fetal anomaly.
Central Nervous System Anomalies Anomalies with a risk of worsening hydrocephaly and/or macrocephaly may benefit from delivery in the early-term period (37-38 weeks). These include Dandy Walker malformation with hydrocephaly and obstructive hydrocephaly. There is a theoretic concern that ongoing severe hydrocephaly will result in more cerebral cortical damage and more macrocrania. Whether delivery 1-2 weeks early changes any long-term outcome for this group of fetuses has not been studied, but taking the entire clinical picture into account, delivery in the early-term period is reasonable on an individual basis.20 The subgroup of anomalies with a risk of intracranial hemorrhage may also benefit from delivery in the early-term pe-
Indicated preterm birth for fetal anomalies
273
Table 2 Fetal Anomalies Not Associated With Need for Early Delivery Central nervous system Agenesis of corpus callosum Porencephaly Microcephaly Dandy Walker malformation Craniofacial Cleft lip/palate Macroglossia Neck Small goiter Chest Bronchopulmonary sequestration Tracho-esophageal atresia Cardiac Atrial septal defect Ventricular septal defect Tachyarrhythmias* Hypoplastic right ventricle Hypoplastic left ventricle Double outlet R ventricle* Abdominal wall defects Body-stalk anomaly Omphalocele Fetal tumors Liver tumor Wilm’s tumor Genitourinary Polycystic kidney disease Minimal hydronephrosis Gastrointenstinal Liver calcifications Duodenal atresia Abdominal calcification Skeletal anomalies Chondrodysplasia punctata Campomelic dysplasia Osteogenesis imperfecta Thanatophoric dysplasia Extremities Amniotic band syndrome Clinodactyly Ectrodactyly Chromosomal abnormalities Trisomy 13 Pallister–Killian syndrome Sex chromosome abnormalities
Anencephaly Encephalocele Macrocephaly
Arachnoid cysts Hydranencephaly Myelomeningocele
Hemifacial microsomia Microphthalmia
Hyper/hypotelorism Mild micrognathia
Small hygroma Stable CCAM Hydrothorax
Pulmonary agenesis Cystic lesions
AV canal defect* Pulmonary stenosis Bradyarrhythmias* Aortic stenosis* Tetralogy of Fallot Truncus arteriorsus
Cardiomyopathy Intracardiac tumors Heterotaxy Transposition* Coarctation Ebstein anomaly*
Pentalogy of Cantrell Cloacal extrophy
Bladder extrophy Gastroschisis*
Mesoblastic nephroma
Neuroblastoma
Cloaca Renal agenesis
Echogenic kidneys Ambiguous genitalia
Pyloric atresia Jejunoileal atresia Hirshprung’s disease
Imperforate anus Colonic atresia
Achondroplasia Jarco Levin syndrome Ellis-van Creveld Hypophospatasia
Hemivertebrae Achondrogenesis Diastrophic dysplasia Short rib polydactyly
Arthrogryposis Clubfoot Polydactyly
Radial aplasia Syndactyly
Trisomy 18 Turner syndrome DiGeorge syndrome
Trisomy 21 Triploidy
AV, atrioventricular; CCAM, congenital cystic adenomatoid malformation. *Consideration should be given to delivery if hydrops or a new arrhythmia develops.
riod (37-38 weeks). These include fetuses with evidence of intracranial bleeding and those with vein of Galen aneurysms. Fetuses at risk for intracranial bleeding—those with existing evidence of a bleed or those complicated by alloimmune thrombocytopenia—inherently have an ongoing risk of bleeding. For those infants with evidence of subdural, intraventricular, or cerebellar bleeding, their prognosis can be very poor, depending on the extent of existing hemorrhage. Delivery in the early-term period is reasonable for selected cases, but no change to obstetrical management is
recommended for those with poorest prognosis.21-23 For vein of Galen aneurysms, use of a scoring system may indicate which pregnancies would not benefit from intervention and which may benefit from delivery in a tertiary center before cardiovascular compromise worsens.24
Craniofacial The craniofacial anomaly that may benefit from scheduled delivery, before labor, is significant micrognathia. In some centers, this diagnosis is managed with an EXIT procedure
S.D. Craigo
274 because of the difficulties that can be encountered in establishing an airway immediately after birth. There are no criteria to distinguish severe from milder forms of micrognathia, but sonographic appearance of marked hypoplasia of the chin and presence of associated polyhydramnios should prompt consideration of scheduled delivery in the early-term period (37-38 weeks) with pediatric specialists available.
Neck A subgroup of pregnancies complicated by fetal neck anomalies requires coordinated, tertiary care immediately after birth. The diagnoses of congenital high airway obstruction sequence (ie, CHAOS), large fetal goiter, and large cystic hygroma can be considered indications for an EXIT procedure and may benefit from scheduled delivery before labor.
Chest Congenital diaphragmatic hernia (CDH) has significant variability in clinical presentation and outcome. More than onethird are further complicated by other anomalies. The size and side of the hernia, position of the liver, the size of the herniated structures, and cardiac position have been correlated with postnatal clinical course. Currently, the observed to expected lung-to-head ratio is the most widely criteria used to assess the severity of the CDH and determine which patients may benefit from in utero therapy. Endoscopic fetal tracheal occlusion by various means is offered in several centers to patients with evidence of severe pulmonary hypoplasia Tracheal occlusion leads to subsequent growth of the fetal lungs, reducing lung hypoplasia and improving survival.16 Those pregnancies with CDH not requiring antenatal treatment can be delivered in the early-term period (37-38 weeks). Some centers advocate delivery by EXIT at 37 weeks with immediate conversion to extracorporeal membrane oxygenation. If endoscopic in utero tracheal occlusion has been performed, optimal management involves endoscopic plug removal at 34 weeks. If removal of the occlusion is not accomplished before labor, delivery with an EXIT procedure at 37 weeks is reasonable, or at the very least, delivery should be coordinated at a center with expertise in removing the specific occlusive device.16 A small number of pregnancies with congenital cystic adenomatoid malformations are complicated by hydrops in the antepartum period and are resected in utero at a few centers in the United States. Therapeutic options include open fetal surgery with fetal lobectomy, aspiration or shunting of a dominant cyst, or postnatal resection. For those that have undergone in utero resection, late preterm delivery is recommended at 36 weeks. Frequently, these pregnancies are complicated by preterm premature rupture of membranes or preterm labor, so delivery planning is not required. For those presenting after 32 weeks’ gestation, delivery with EXIT procedure and immediate resection is being offered in some centers.15
Cardiac There are no cardiac anomalies that require early delivery when present in isolation. Late preterm or early-term delivery may be recommended, however, if a fetus with a cardiac anomaly develops hydrops or a complicating fetal arrhythmia. There are no data demonstrating benefit of this approach. Hydrops carries a significant risk of mortality and delivery may not improve outcome for these fetuses. However, when following patients with a diagnosis of fetal cardiac anomaly beyond 34 weeks’ gestation, the new onset of hydrops or an arrhythmia suggests deterioration, and delivery will allow direct intensive care for the baby, in attempt to improve outcome.
Abdominal Wall Defects Most abdominal wall defects carry no indication for change to obstetrical management for timing of delivery. The abdominal wall defect for which planned delivery before labor would be recommended is a giant omphalocele with liver outside the fetus. Although a benefit of cesarean delivery has not been clearly demonstrated for omphaloceles, physicians at many centers would recommend prelabor cesarean delivery in this clinical scenario in attempt to avoid liver injury. Without data demonstrating a benefit of this strategy, delivery in the late term period (37-38 weeks) would be reasonable. Controversy over management of gastroschisis has been longstanding, and to date, the optimal timing of delivery remains elusive. In the only randomized trial to compare elective delivery at 36 weeks versus awaiting spontaneous labor, no significant benefit from elective preterm delivery for gastroschisis was found.25 In contrast, others have reported improved outcome among a small number of neonates with gastroschisis delivered electively at 34 weeks, compared with historic controls.26 Another group reported outcomes of infants with gastroschisis delivered before or after 36 weeks. The study was retrospective, and the indication for delivery was not reported, but neonatal length of stay and time to full feeds were significantly worse among those infants delivered before 36 weeks.27 Similarly, at least 2 other retrospective studies of patients with gastroschisis found no benefit of delivery before 37 weeks.28,29 Efforts to determine the best time to intervene for gastroschisis on the basis of the appearance of the bowel on ultrasound have not been successful. A review of 10 observational studies found no clear evidence that bowel dilation over 10 or 18 mm was associated with worse outcomes.30 Others have suggested, on the basis of a few patients, that bowel dilation ⬎25 mm is associated with nonreassuring fetal testing and fetal demise.31 Intrauterine fetal growth restriction frequently coincides with gastroschisis. The presence of growth restriction and the increased risk of fetal death associated with gastroschisis will lead most perinatologists to recommend delivery in the early-term period. Given the frequency with which gastroschisis is diagnosed, it should also be noted that we work closely with our pediatric surgical colleagues when managing these pregnancies. Their clinical opinions on man-
Indicated preterm birth for fetal anomalies agement and interpretation of the existing data may also influence timing of delivery. Current data are not sufficient to determine the optimal timing of delivery for gastroschisis. The data favoring elective late preterm delivery is very limited, but early-term delivery (37-38 weeks) is a reasonable option. In the absence of growth restriction, awaiting spontaneous labor or delivery at term is also reasonable.
Fetal Tumors Most fetal tumors can be managed without recommendation for early delivery. Recommendations for some rare tumors alter timing of delivery. Cervical teratomas may be managed with a planned EXIT procedure, or with coordinated care involving immediate availability of a team to manage the airway of the neonate. Recommendations for fetal retinoblastomas are for delivery when fetal lung maturity is documented, to allow immediate postnatal therapy of the tumor.18 Sacrococcygeal teratomas carry an ongoing risk of cardiovascular compromise and/or bleeding, so delivery is recommended when fetal lung maturity is documented.32 Outcome has been reported as nearly uniformly fatal if placentomegaly or hydrops are present,33,34 so this is one of the few anomalies for which open fetal surgery is still offered in select centers.
Genitourinary Most genitourinary anomalies carry no indication for change to obstetrical management for timing of delivery. For a subset of anomalies in which ultrasound findings suggest overall renal function is worsening, delivery may be recommended when fetal lung maturity is documented or when oligohydramnios develops in the late preterm or early-term period. The conditions for which these recommendations apply include obstructive uropathy developing oligohydramnios, multicystic kidney with evidence of contralateral renal compromise, or a single kidney with evidence of obstruction and oligohydramnios. Delivery allows for intervention to relieve the obstruction, and risks attributable to oligohydramnios are removed. Whether this approach improves renal function long term has not been studied.
Other When gastrointestinal abnormalities, skeletal anomalies, anomalies of the extremities, or chromosomal abnormalities are present, there is no indication for change to obstetrical management for timing of delivery.
Special Considerations Twins with One Anomaly Twins with one fetus affected by anomaly should be managed on a case-by-case basis. As with singletons, recommendations will differ by anomaly. Factors that must be considered include chorionicity, gestational age, and nature of the specific anomaly. In general, after 32 weeks of gestation anomaly-specific recommendations for singletons can be followed
275 for twins, but if late preterm delivery is being considered, potential implications for the normal co-twin must be addressed. If monoamniotic twins are complicated by a fetal anomaly, there is ongoing risk of fetal death of the normal twin at all times. Without actual removal of the anomalous twin, even selective termination by cord occlusion or thermocoagulation techniques will result in a nonviable twin that can lead to cord accident of the co-twin. There has been limited success of the options for intervention.35,36 In this specific scenario, preterm delivery at 32-34 weeks is reasonable.
Lethal Anomalies Pregnancies complicated by a lethal anomaly are worthy of special consideration. Term delivery is recommended, ideally when spontaneous labor ensues. In selected cases preterm delivery may be reasonable to consider to provide compassionate care to the mother because neonatal outcome will not be changed. In all cases a primary goal of management should be to avoid interventions that pose any risk to the mother. Expectant management for obstetrical complications (eg, preterm rupture of membranes, abruption, pre-eclampsia) is contraindicated in the presence of a known lethal fetal anomaly.
Research Gaps The difficulty in determining the best timing for delivery in pregnancies complicated by fetal anomalies is related to the sheer number of different types of anomalies, each with specific implications and variations. Fortunately, most do not require any change to timing of delivery. For those that may possibly benefit from earlier delivery, research is needed regarding the role of amniocentesis for fetal lung maturity testing. Further study is also needed on the best course of management if amniocentesis is used and testing does not confirm lung maturity. Additionally, the possible role of steroids before late preterm or early-term delivery has not been determined. More specific to fetal anomalies, further research is needed to improve our ability to assess organ injury antenatally and to predict which fetuses may benefit from earlier delivery. This research will require anomaly specific focus, and likely collaborative efforts to study the more uncommon anomalies. Research efforts should also ideally address obstetrical, neonatal, and long-term outcomes for these complicated pregnancies. Only if perinatal and neonatal outcomes are studied will we determine if survival is improved when delivery is undertaken for evidence of hydrops, or if fetal death is avoided but neonatal death follows. The recent publication of the results of the MOMS (Management of Myelomeningocele) trial37 demonstrates both the difficulty in performing such research, and the importance of evaluating new therapies. Research for potential invasive therapies for lethal and nonlethal anomalies will no doubt continue in attempt to further define the role of fetal surgery, even as the indications for open fetal surgery appear to be-
276 come more limited. The role of EXIT should also be studied and defined, as the procedure itself carries maternal risks and its use influences timing of delivery.
Conclusions Prenatally diagnosed fetal anomalies complicate 1%-2% of pregnancies. Only a fraction of the anomalies will lead to recommendations for late preterm or early-term birth. For a limited number of conditions, late preterm birth, or more often early-term birth, will be indicated. The conditions that may benefit from delivery before 39 weeks include those with evidence of worsening organ injury (usually kidney, heart, or possibly brain) and those in which coordination of care for neonate immediately after birth will be critical. Opportunities for research in this area will be anomaly specific.
References 1. Dolk H, Loane M, Garne E: The prevalence of congenital anomalies in Europe. Advances Experiment. Med Biol 686:349-364, 2010 2. Ewigman BG, Crane JP, Frigoletto FD, et al: Effect of prenatal ultrasound screening on perinatal outcome. N Engl J Med 329:821-827, 1993 3. Grandjean H, Larroque D, Levi S: The performance of routine ultrasonographic screening of pregnancies in the Eurofetus Study. Am J Obstet Gynecol 181:446-454, 1999 4. Stoll C, Tenconi R, Clementi M, et al: Detection of congenital anomalies by fetal ultrasonographic examination across Europe. Community Genet 4:225-232, 2001 5. Schinzel AA, Smith DW, Miller JR: Monozygotic twinning and structural defects. J Pediatr 95:921-930, 1979 6. Drugan A, Johnson MP, Krivchenia EL, et al: Genetics and genetic counseling, in Gall SA (ed): Multiple Pregnancy and Delivery. St Louis, Mosby, 1996 pp 85-97 7. Layde PM, Erickson JD, Falek A, et al: Congenital malformation in twins. Am J Hum Genet 32:69-78, 1980 8. Little J, Nevin NC: Congenital anomalies in twins in Northern Ireland. III: Anomalies of the cardiovascular system 1974-1979. Acta Geneti Med Gemellol (Roma) 38:27-35, 1989 9. Ramos-Arroyo MA: Birth defects in twins: Study in a Spanish population. Acta Genet Med Gemellol (Roma) 40:337-344, 1991 10. Malone FD, Craigo SD, Chelmow D, et al: Outcome of twin gestations complicated by a single anomalous fetus. Obstet Gynecol 88: 1-5, 1996 11. Adra AM, Landy HJ, Nahmias J, et al: The fetus with gastroschisis: Impact of route of delivery and prenatal ultrasonography. Am J Obstet Gynecol 174:540-546, 1996 12. Pauli RM, Reiser CA, Lebovitz RM, et al: Wisconsin Stillbirth Service Program: I. Establishment and assessment of a community-based program for etiologic investigation of intrauterine deaths. Am J Med Genet 50:166-134, 1994 13. American College of Obstetricians and Gynecologists (2009) Management of stillbirth. ACOG Practice Bulletin No. 102. Obstet Gynecol 113:748-761, 2009 14. Gul A, Cebeci A, Aslan H, et al: Perinatal outcomes of twin pregnancies discordant for major anomalies. Fetal Diagn Ther 20:244-248, 2005 15. Adzick NS: Open fetal surgery for life-threatening fetal anomalies. Semin Fetal Neonat Med 15:1-8, 2010
S.D. Craigo 16. Deprest J, Nicolaides K, Doné E, et al: Technical aspects of fetal endoscopic tracheal occlusion for congenital diaphragmatic hernia. J Pediatr Surg 46:22-32, 2011 17. Deurloo KL, Devlieger R, Lopriore E, et al: Isolated fetal hydrothorax with hydrops: A systematic review of prenatal treatment options. Prenat Diagn 27:893-899, 2007 18. Bianchi DW, Crombleholme TM, D’Alton ME, et al: Fetology: Diagnosis and Management of the Fetal Patient (ed 2). New York, McGraw-Hill Medical, 2010 19. Hedrick HL, Flake AW, Crombleholme TM, et al: The ex utero intrapartum therapy procedure for high-risk fetal lung lesions. J Pediatr Surg 40:1038-1044, 2005 20. Wilkins-Haug L: Considerations for delivery of infants with congenital abnormalities. Obstet Gynecol Clin N Am. 26:399-412, 1999 21. Pretorius DH, Singh S, Manco-Johnson M, et al: In utero diagnosis of intracranial hemorrhage resulting in fetal hydrocephalus. J Reprod Med 3:136-138, 1986 22. Ghi T, Simonazzi G, Perolo A, et al: Outcome of antenatally diagnosed intracranial hemorrhage: Case series and review of the literature. Ultrasound Obstet Gynecol 22:121-130, 2003 23. Murphy MF, Bussel JB: Advances in the management of alloimmune thrombocytopenia. Br J Hematol 136:366-378, 2006 24. Yuval Y, Lerner A, Lipitz S, et al: Prenatal diagnosis of vein of galen aneurysmal malformation: Report of two cases with proposal for prognostic indices. Prenat Diagn 17:972-977, 1997 25. Logghe HL, Mason GC, Thornton JG, et al: A randomized controlled trial of elective preterm delivery of fetuses with gastroschisis. J Pediatr Surg 40:1726-1731, 2005 26. Serra A, Fitze G, Kamin G, et al: Preliminary report on elective preterm delivery at 34 weeks and primary abdominal closure for the management of gastroschisis. Eur J Pediatr Surg 18:32-37, 2008 27. Ergün O, Barksdale E, Ergün FS, et al: The timing of delivery of infants with gastroschisis influences outcome. J Pediatr Surg 40:424-428, 2005 28. Huang J, Kurkchubasche AG, Carr SR, et al: Benefits of term delivery in infants with antenatally diagnosed gastroschisis. Obstet Gynecol 100: 695-699, 2002 29. Simmons M, Georgeson KE: The effect of gestational age at birth on morbidity in patients with gastroschisis. J Pediatr Surg 31:1060-1062, 1996 30. Tower C, Ong SS, Ewer AK, et al: Prognosis in isolated gastroschisis with bowel dilatation: A systematic review. Arch Dis Child Fetal Neonat Ed 94:F268, 2009 31. Heinig J, Müller V, Schmitz R, et al: Sonographic assessment of the extra-abdominal fetal small bowel in gastroschisis: A retrospective longitudinal study in relation to prenatal complications. Prenat Diagn 28:109, 2008 32. Adzick NS, Harrison MR: The unborn surgical patient. Curr Probl Surg 35:1, 1994 33. Bond SJ, Harrison MR, Schmidt KG, et al: Death due to high-output cardiac failure in fetal sacrococcygeal teratoma. J Pediatr Surg 25:12871291, 1990 34. Flake AW, Harrison MR, Adzick NS, et al: Fetal sacrococcygeal teratoma. J Pediatr Surg 21:563-566, 1986 35. Middeldorp JM, Klumper FJCM, Oepkes D, et al: Selective feticide in monoamniotic twin pregnancies by umbilical cord occlusion and transection. Fetal Diagn Ther 23:121-125, 2008 36. Young BK, Roque H, Abdelhak Y, et al: Endoscopic ligation of umbilical cord at 19 weeks’ gestation in monochorionic monamniotic twins discordant for hypoplastic left heart syndrome. Fetal Diagn Ther 16:6164, 2001 37. Adzick NS, Thom EA, Spong CY, et al: A randomized trial of prenatal versus postnatal repair of myelomeningocele. N. Engl. J. Med. 364:9931004, 2011