A Nonimmune Hydrops Case.

Abstract

A 25-year-old gravida 1, para 0 woman was referred to our center at 29 weeks, 5 days’ gestation with a thickened nuchal fold, possible ventricular septal defect (VSD), and a high risk for trisomy 21 result on cell-free DNA screening. The patient had a medical history notable for polycystic ovary syndrome and attention-deficit/hyperactivity disorder, not currently receiving pharmacotherapy. Her surgical history was noncontributory. The patient and her partner’s family histories were noncontributory. She had received regular prenatal care in another state, recently relocated to our state, and presented to establish care.The patient had high-risk cell-free DNA screening for trisomy 21 at 13 weeks’ gestation, with a positive predictive value of 53.4% based on test results and maternal age. In addition, at 16 weeks’ gestation, a cystic hygroma was noted on early fetal anatomy ultrasonography, which evolved into a thickened nuchal fold at her 20-week fetal anatomic survey at the outside institution. The 16-week fetal ultrasound scan had also demonstrated short femurs, an echogenic intracardiac focus, and a possible VSD. The patient had declined invasive genetic testing. She underwent no additional ultrasonography between her 20-week anatomic survey at the prior institution and her initial presentation at our center.Ultrasonography at 29 5/7 weeks’ gestation demonstrated multiple stigmata of presumed trisomy 21 as well as hydrops fetalis with moderate bilateral pleural effusions, ascites, and head and torso skin edema (Fig 1); an enlarged and hypoechoic liver (Fig 2); and a small stomach with polyhydramnios raising concern for possible esophageal atresia. The spleen appeared normal. The overall estimated fetal weight was at the 63rd percentile, however, the femurs were shortened below the 5th percentile and the abdominal circumference was increased at the 92nd percentile (Fig 3). Fetal echocardiography noted a small perimembranous VSD and a small, globally distributed pericardial effusion; the ventricular systolic function was normal and no arrhythmia was detected during echocardiography or any fetal monitoring. Middle cerebral artery Doppler studies were within normal limits at 1.3 multiples of the median, suggesting a lower risk for fetal anemia.At our center, pregnant women referred for suspected or confirmed fetal aneuploidy receive multidisciplinary counseling from maternal-fetal medicine, genetic counseling, radiology, pediatric cardiology, neonatology, and relevant pediatric subspecialties. In cases of potentially severe morbidity or mortality, such as with trisomy 13, trisomy 18, or hydrops, our center also offers pediatric palliative care services. In addition, in our state (Oregon) there is no gestational age limit at which a family can choose to end a pregnancy, and thus if families choose to terminate, our Fetal Therapy Clinic works closely with our institution’s Family Planning Division to coordinate care.In the setting of fetal hydrops and the cell-free DNA result, this family was counseled that although she did not have definitive findings on aneuploidy testing, the neonate will likely have trisomy 21. The team discussed that the fluid collections were attributable to nonimmune hydrops fetalis (NIHF), and as the patient had not undergone ultrasonography in nearly 8 weeks, it was difficult to discern when in gestation the hydrops had developed and the overall trajectory of progression of the hydrops. The team counseled the parents about the differential diagnosis of NIHF, which included aneuploidy in and of itself, transient abnormal myelopoiesis (TAM), or less likely, congenital chylothorax, given the relatively small size of the pleural effusions compared to the ascites. TAM, a preleukemic syndrome unique to trisomy 21, was defined and the family was informed of the reasons for our concern, namely fetal hepatomegaly, hydrops, and suspected trisomy 21. A cardiovascular cause of NIHF was considered less likely, as echocardiography showed normal cardiac function and no evidence of arrhythmia. She had no recent infectious exposures, an unremarkable personal and family history, and no ultrasonographic evidence of a tumor or thoracic abnormality. Amniocentesis for diagnostic genetic and infectious testing, as well as percutaneous umbilical blood sampling for hematologic evaluation, was considered, but the patient declined invasive testing.The patient and her family had extensive discussions with maternal-fetal medicine, neonatology, and pediatric palliative care teams regarding options at this point. Options discussed with the patient included expectant in-utero management, which carried risk of worsening hydrops, versus delivery. They were counseled that if they opted for delivery, the mortality in the NICU at this early gestational age in the setting of NIHF was high. They were also offered delivery with palliative care.The patient and her partner identified their top goals for their daughter as optimizing neonatal survival by prolonging the pregnancy as long as possible to improve postnatal survival. They expressed that they understood the fetal risks, including the risk for intrauterine fetal death and the potential maternal risks for prolongation of pregnancy with hydrops such as the development of mirror syndrome. They conveyed a goal to meet their child and avoid fetal death, if possible. As a result, they opted for expectant in-utero management with close observation and a plan to proceed with delivery in the event of worsening hydrops. She received a course of betamethasone for fetal lung maturity at 29 5/7 weeks’ gestation in alignment with these goals. The patient declined antepartum admission and inpatient management, and thus was closely monitored in the outpatient setting with formal ultrasonographic monitoring for worsening hydrops 3 times per week.At 30 weeks’ gestation, the polyhydramnios progressed, with an interval increase in the amniotic fluid index from 24 to 30 cm, and the fetal stomach was not visualized, further increasing suspicion for esophageal atresia (Fig 4). Fetal hydrops remained stable to slightly improved, with decreased abdominal ascites. Maternal status remained stable without signs or symptoms of mirror syndrome.At 31 weeks’ gestation, ultrasound findings demonstrated interval progression of hydrops compared to 2 days’ prior, with reaccumulation of ascites and mild enlargement of pleural effusions. The decision was made to proceed with delivery given worsening hydrops and the family’s expressed wishes. The patient was admitted and underwent an uncomplicated primary low transverse cesarean delivery. A viable female infant weighing 2,065 g was delivered and transferred to the neonatal resuscitation team after 60 seconds of delayed cord clamping. Resuscitation was provided according to the Neonatal Resuscitation Program guidelines. The infant had apnea and bradycardia, which resolved after intubation. The infant’s Apgar scores were 3, 4, and 8 at 1, 5, and 10 minutes after birth, respectively.The infant’s admission examination findings were notable for low-set ears, prominent horizontal palmar creases bilaterally, full abdomen with palpable liver, and significant edema of her face and torso, which made it difficult to assess for dysmorphic facial features. Her initial chest radiograph was consistent with diffuse bilateral ground glass opacities as a result of surfactant deficiency, small bilateral pleural effusions, and esophageal atresia with tracheoesophageal (TE) fistula due to enteric tube coiling in the proximal esophagus with visible air in the nondilated stomach and intestines. A complete blood cell count on admission to the NICU was notable for significant leukocytosis of 160,000/μL (160 × 109/L) with atypical white blood cells (WBC) of 42.9%, hematocrit of 41%, and platelet count of 155,000/μL (155 × 109/L). Admission coagulation studies were notable for a prolonged international normalized ratio of 1.98, elevated partial thromboplastin time of 52.5 seconds, and a low fibrinogen level of 116 mg/dL (1.16 g/L). She received fresh frozen plasma, cryoprecipitate, and red blood cell transfusions within the first 48 hours after birth. Postnatal echocardiography on the day of birth showed a small perimembranous VSD, trivial pericardial effusion, large patent ductus arteriosus, moderate secundum atrial septal defect, and qualitatively normal biventricular systolic function but severely abnormal left ventricular diastolic function.The pediatric hematology/oncology team was consulted and thought that the fetal and infant findings were most consistent with TAM. Pediatric general surgery was also consulted and multidisciplinary discussion held with the family, given indications for multiple competing urgent interventions including ligation of the TE fistula, esophageal anastomosis, and pharmacologic management of TAM. Given the hyperleukocytosis and risk for sequelae, the infant began treatment with cytarabine when she was approximately 30 hours old, before flow cytometry results were available. Her WBC began to decrease promptly and was 78,000/μL (78 × 109/L) at 3 days of age. Surgical intervention for the esophageal atresia and TE fistula was deferred due to her critical illness and instability. Flow cytometry results available 48 hours after birth confirmed the diagnosis of TAM; additional testing for GATA1 mutations was not performed.Unfortunately, the infant continued to have worsening anasarca, increased pleural effusions, and worsening respiratory failure with hypoxia despite increased ventilator settings, supplemental oxygen, and inhaled nitric oxide. Conversations about prognosis continued with her parents and the various specialists involved in her care. Placement of thoracostomy tubes was considered but due to the estimated very high risk of mortality from the combination of prematurity, hydrops, trisomy 21, TAM, and TE fistula with esophageal atresia, all were concerned that further interventions may cause pain or suffering and would not change the eventual outcome. Based on these discussions, the parents were supported in their decision to redirect goals of care to comfort only. The neonate underwent compassionate extubation while being held by her parents and died on her third postnatal day.Umbilical cord blood testing for genetic analysis confirmed the diagnosis of trisomy 21 with fluorescence in situ hybridization (FISH) analysis on interphase cells. No metaphase cells were available for karyotype, likely due to the infant’s underlying hydrops and myeloproliferative disorder. To inform recurrence risk, maternal and paternal karyotype testing was offered to discern whether the infant’s trisomy 21 was due to a sporadic nondisjunction event or a translocation that may have been inherited by a parent. The maternal karyotype was notable for a balanced robertsonian translocation between chromosomes 13 and 21 (45,XX,der[13;21][q10;q10]). This type of translocation is rare, making up only 2% of all robertsonian translocation carriers with a prevalence of 1 in 100,000 in the general population. (1) The paternal karyotype was normal male 46,XY. Given this information, it was inferred that the infant’s karyotype was 46,XX,der(13;21)(q10;10)+21.Approximately 3% of individuals with trisomy 21 have an unbalanced robertsonian translocation. Although trisomy 21 and trisomy 13 account for the majority of livebirths with aneuploidy, there is little information regarding the reproductive risks for carriers of a 13;21 robertsonian translocation with most of the data in the literature based on FISH studies of spermatozoa. Although data on female carriers of this robertsonian translocation are limited, the parents were counseled that given the maternal translocation, there is a 15% risk for trisomy in the second trimester, a 10% risk to have a liveborn infant with trisomy 21, and a 1% risk to have a liveborn infant with trisomy 13. (2)Hydrops fetalis is defined by the Society for Maternal-Fetal Medicine as the presence of at least 2 pathologic fluid collections, including pleural effusion, pericardial effusion, ascites, and skin edema. Polyhydramnios and a thickened placenta are also common ultrasonographic findings (3). NIHF, which comprises nearly 90% of all cases of hydrops in the current era, refers to cases not caused by red cell alloimmunization (4).NIHF complicates 1 in 1,700 to 3,000 pregnancies and portends a poor prognosis (4). Santo et al reported a case series of 71 pregnancies complicated by prenatally identified NIHF that continued after 20 weeks’ gestation. In that study, the live birth rate was 62%, the 28-day neonatal survival rate was 48%, and 25% survived without major morbidities at a mean follow-up age of 29 months. (5) Prematurity further complicates the prognosis; several case series of liveborn infants with NIHF have found lower gestational age at delivery to be an independent risk factor for neonatal death. (6)(7)(8)NIHF is a disease state with many potential etiologies. Although older literature considered many cases to be idiopathic, contemporary data have shown that a cause can be prenatally detected in 60% to 82% of cases. (5)(9) When postnatal detection is included, a cause can be found in nearly 85% of cases. (10) The most common etiology of NIHF, comprising 7% to 16% of cases, is a cardiovascular cause such as fetal arrhythmia, congenital cardiac anomaly leading to cardiac failure, vascular abnormalities, or cardiac tumors.Chromosomal abnormalities, particularly trisomy 21 and monosomy X, comprise the second most common cause of NIHF. Aneuploidy accounted for 13% of cases in a large systematic review, and it is a particularly common etiology when hydrops is identified in early pregnancy. (10) In some instances, hydrops in an aneuploid fetus develops in the setting of a cardiac anomaly or impaired lymphatic drainage, especially when a cystic hygroma is seen on ultrasonography. However, NIHF has also been described in trisomy 21 in the absence of structural heart defects secondary to TAM, as in our patient’s case. The pathogenesis of prenatal TAM begins with perturbation of fetal hematopoiesis in the setting of trisomy 21, followed by acquisition of somatic mutations in the GATA1 gene in fetal liver hematopoietic cells with subsequent monoclonal expansion. (11) TAM typically presents in the neonatal period, but approximately 10% of cases present prenatally as early as 26 weeks’ gestation. (12) Ultrasonographic findings of intrauterine TAM most commonly include hepatomegaly (due to extramedullary hematopoiesis and sequestration of blast cells) with hypoechoic appearance, possible splenomegaly, and possible development of NIHF. (12) One systematic review of 39 fetuses with prenatally detected trisomy 21 and TAM reported the prevalence of hydrops at 30.8%. (12) In this series, of the 13 cases with hepatomegaly and hydrops, 12 (92.3%) died either prenatally or in the early neonatal period, underscoring the presence of hydrops as a marker of a poor prognosis. (12) Percutaneous umbilical blood sampling demonstrating leukocytosis with blast predominance can inform prenatal diagnosis, however, the role of prenatal therapy in intrauterine TAM is limited: case reports of supportive intrauterine blood transfusions have yielded disparate fetal outcomes, and no cases have been reported to date in which curative chemotherapy has been initiated in the prenatal setting. (12)Single gene disorders may be causative in other cases of hydrops. Sparks and colleagues evaluated the diagnostic yield of whole exome sequencing in a case series of nonimmune hydrops. (13) In their series of 127 consecutive cases of unexplained NIHF, whole exome sequencing identified a diagnostic pathogenic variant in 29% of fetuses. Of the single-gene disorders identified on whole exome sequencing, RASopathies composed the largest proportion (30%), followed by inborn errors of metabolism, skeletal dysplasias, lymphatic disorders, neurodevelopmental disorders, cardiovascular disorders, and hematologic disorders. A variant of potential clinical significance was detected in an additional 9% of cases; these findings are much greater than previously reported diagnostic yields of whole exome sequencing and suggest a greater contribution of single-gene disorders to NIHF than previously appreciated.Other less common causes of NIHF include hematologic disorders such as α-thalassemia, congenital infections (particularly parvovirus B19 and cytomegalovirus), thoracic abnormalities (eg, congenital pulmonary airway malformation [CPAM]), twin-to-twin transfusion syndrome (TTTS), or highly vascular tumors (eg, sacrococcygeal teratomas, lymphangiomas, and neuroblastomas), or vascular malformations that lead to high output cardiac failure.One rare but significant maternal complication of NIHF is the development of mirror syndrome, in which the pregnant person develops hypertension, proteinuria, and generalized edema, essentially mirroring the edema of the fetus. Mirror syndrome is rare and likely underdiagnosed, therefore, the exact incidence is unclear but has been estimated to complicate up to 5% of NIHF cases. (14)(15) Although there have been case reports of mirror syndrome resolving after successful reversal of a treatable cause of NIHF, such as anemia or infection, in general, the development of mirror syndrome necessitates delivery regardless of gestational age. (4) Additional potential obstetric complications include polyhydramnios and its associated risks of preterm labor, premature rupture of membranes, and maternal respiratory symptoms. Although the optimal mode of delivery in pregnancies complicated by NIHF has not been established, vaginal delivery is preferred in cases in which comfort care is planned for the neonate, unless otherwise contraindicated. Cesarean delivery may be indicated when the patient desires neonatal intervention and the decision to proceed with delivery is based on nonreassuring antepartum surveillance. (4)The Society for Maternal-Fetal Medicine has developed an antepartum algorithm to evaluate for the causes of NIHF. (4) All patients with a concern for NIHF should undergo detailed ultrasonography and fetal echocardiography, and structurally normal fetuses should undergo Doppler interrogation of the middle cerebral artery to screen for fetal anemia. Invasive prenatal testing via amniocentesis should be offered to evaluate the fetal karyotype, chromosomal microarray, polymerase chain reaction for parvovirus, toxoplasmosis, and cytomegalovirus, and case-based consideration of whole exome sequencing and targeted testing for specific conditions such as lysosomal storage disorders, single-gene disorders, or thalassemia.The management of NIHF is largely contingent on the underlying etiology. For example, some causes of hydrops are amenable to in utero therapy, such as initiation of antiarrhythmic drugs for fetal arrhythmias, corticosteroids or needle drainage of CPAM, laser ablation of TTTS, or intrauterine transfusion for fetal anemia. Such cases should be urgently referred to a tertiary care center with fetal interventional capacity if the parents desire treatment. Infants of families that desire potentially life-prolonging interventions after birth should be delivered at centers with neonatal critical care services. Termination of pregnancy or palliative care should also be offered to families, particularly at lower gestational ages, given the universal prognosis of NIHF and prematurity regardless of the underlying etiology. Although the risk of intrauterine death is high, no evidence-based guidelines exist to guide antepartum surveillance or timing of delivery in pregnancies complicated by NIHF, and thus such management decisions should be individualized based on the fetus’ clinical trajectory, possibility of neonatal intervention, and patient preference.Hydrops fetalis is defined as the pathologic accumulation of fluid in at least 2 spaces, including pleural effusions, pericardial effusions, ascites, or skin edema.Nonimmune hydrops fetalis refers to cases of hydrops not caused by red cell alloimmunization, and comprises a broad differential of genetic, structural, infectious, hematologic, and metabolic etiologies.In the setting of trisomy 21, hydrops can develop secondary to lymphatic dysfunction, cardiac anomalies, or a unique hematologic derangement known as TAM, which is characterized prenatally by severe hepatomegaly and can affect up to 10% of neonates with trisomy 21.NIHF carries a high risk of intrauterine or neonatal mortality regardless of etiology, though some causes are amenable to fetal therapy, which could potentially reverse the findings of hydrops. Management decisions require multidisciplinary collaboration, thoughtful coordination, and patient-centered care.