eMedicine Specialties > Pediatrics: Cardiac Disease and Critical Care Medicine > Neonatology

Hydrops Fetalis

Author: Ashraf H Hamdan, MB, BCh, MSc, MD, MRCP, Clinical Assistant Professor of Pediatrics, Vanderbilt University Medical Center
Contributor Information and Disclosures

Updated: Aug 20, 2008

Introduction

Background

Hydrops fetalis (ie, fetal hydrops) is a serious fetal condition defined as abnormal accumulation of fluid in 2 or more fetal compartments, including ascites, pleural effusion, pericardial effusion, and skin edema. In some patients, it may also be associated with polyhydramnios and placental edema.

Hydrops fetalis has been a well-recognized fetal and neonatal condition throughout history. Until the latter half of the 20th century, it was believed to be due to Rhesus (Rh) blood group isoimmunization of the fetus. More recent recognition of factors other than isoimmune hemolytic disease that can cause or be associated with fetal hydrops led to the use of the term nonimmune hydrops to identify those cases in which the fetal disorder was caused by factors other than isoimmunization.

In the 1970s, the major cause of immune hydrops (ie, Rh D antigen) was conquered with the use of immunoglobulin (Ig) prophylaxis in at-risk mothers. This conquest was quickly followed by recognition that the nonimmune causes of hydrops were, in fact, more common than had been suspected.

Pathophysiology

Several hypotheses regarding the pathophysiologic events that lead to fetal hydrops have been suggested. The basic mechanism for the formation of fetal hydrops is an imbalance of interstitial fluid production and the lymphatic return. Fluid accumulation in the fetus can result from congestive heart failure, obstructed lymphatic flow, or decreased plasma osmotic pressure. The fetus is particularly susceptible to interstitial fluid accumulation because of its greater capillary permeability, compliant interstitial compartments, and vulnerability to venous pressure on lymphatic return.

Compensatory mechanisms for maintaining homeostasis during hypoxia that results from underlying disease include increased efficiency of oxygen extraction, redistribution of blood flow to the brain and heart, and volume augmentation to enhance cardiac output. Unfortunately, these mechanisms increase venous pressure and ultimately produce interstitial fluid accumulation and characteristic hydropic changes in the fetus. Increased venous pressure contributes to edema and effusion by increasing the capillary hydrostatic pressure and decreasing the lymphatic return.

Furthermore, the hepatic synthesis of albumin may be impaired because of decreased hepatic perfusion and increased extramedullary hematopoiesis. Because albumin acts as the predominant oncotically active plasma protein, hypoalbuminemia increases transcapillary fluid movement at times of circulatory compromise.

Hydrops has been produced in the ovine fetus by anemia, tachyarrhythmia, occlusion of lymphatic drainage, and obstruction of cardiac venous return. Hypoproteinemia and hypoalbuminemia are common in human hydrops, and reduced intravascular oncotic pressure has been speculated to be a primary cause for the disorder.

However, in the sheep model, a 41% reduction in total serum protein accompanied by a 44% decline in colloid osmotic pressure failed to produce fetal hydrops. Furthermore, a study in humans showed that, despite a significant negative correlation between the fetal serum albumin and the degree of fetal hydrops, most of the fetuses with hydrops had albumin levels within the reference range.¹ These results suggest that hypoalbuminemia is unlikely to cause the primary onset of hydrops.

A closer look at the animal studies provides the clues necessary to understand the pathophysiology of hydrops. In one study, profound anemia was induced in fetal sheep; the hydrops that resulted was unrelated to hematocrit levels, blood gas levels, acid-base balance, plasma proteins, colloid oncotic pressure, or aortic pressure.² A difference was found in central venous pressure (CVP), which was much higher in persons with hydrops. The hematocrit level was reduced by 45% in a study of particular notation; however, CVP was maintained unchanged, and no fetus developed hydrops under these conditions.

Induced fetal tachyarrhythmia has led to fetal hydrops in several studies. Key to the development of fetal hydrops in these studies was an elevation in CVP; the anemia was only of indirect importance. CVP was markedly elevated, with a range of 25-31 mm Hg in one study. In other reports, hydrops induced by sustained fetal tachycardia was unrelated to blood gases, plasma protein, or albumin turnover; however, a 75-100% increase in CVP was observed in the fetuses that developed hydrops.

Excision of major lymphatic ducts produces fetal hydrops in the sheep model. A related study demonstrates an exquisite, linear, inverse relationship between lymphatic outflow pressure and CVP; a rise in CVP of 1 mm Hg reduces lymph flow 13%, and flow stops at a CVP of 12 mm Hg. These results are confirmed by other observations of linear decline in lymph flow when CVP exceeds 5 mm Hg and a cessation of flow at CVPs greater than 18 mm Hg.

Also of note is a computer simulation model in which cardiovascular and fluid electrolyte disturbances (eg, severe anemia, lymphatic obstruction, excess fluid and electrolyte loads, elevation in angiotensin levels) and compensating homeostatic mechanisms have been examined. This model demonstrated that "...fetal cardiac failure constituted the strongest stimulus for the formation of fetal edema..."³ , thus further substantiating the pivotal role of CVP in the development of fetal hydrops.

Many other physiologic disturbances are associated with human fetal hydrops. Elevations in aldosterone, renin, norepinephrine, and angiotensin I levels are likely to be secondary consequences. Although infusion of angiotensin I led to fetal hydrops in nephrectomized sheep, the 4-fold rise in CVP was probably the primary cause of the hydrops. The meaning of increased levels of coenzyme Q10, placental vascular endothelial growth factor, and endothelin and decreased cytokine interleukin-3 levels is unclear at this time.

However, of particular interest is the 3- to 5-fold increase in atrial natriuretic peptide (ANP) that accompanies both human fetal hydrops (with cardiac anomaly or isoimmunization) and ovine hydrops (induced by obstruction of venous return, sustained tachycardia, or induced anemia). A return of ANP levels to normal parallels the resolution of hydrops. These observations and the observations that vascular permeation of albumin is enhanced and cardiovascular and renal homeostatic adaptations are influenced by this peptide suggest an important role for ANP in fetal hydrops. 

Evidence of low fetal plasma levels of cyclic guanosine monophosphate suggests that reduced nitric oxide production due to injury of fetal vascular endothelial cells may be involved in the development of fetal hydrops. This isolated observation requires confirmation and further study.

Frequency

United States

The precise incidence of hydrops fetalis is difficult to elucidate, because many cases are not detected prior to intrauterine fetal death and some cases may resolve spontaneously in utero. The best estimate for how common hydrops fetalis is in the United States is approximately 1 in 600 to 1 in 4000 pregnancies. The incidence of immune hydrops has significantly decreased with the wide use of passive immunization using Rh immunoglobulin for Rh-negative mothers at 28 weeks' gestation (following suspected fetomaternal hemorrhage) and postpartum (following the delivery of an Rh-positive infant). The efficacy of this program has been demonstrated by a decline in the incidence of Rh hemolytic disease of the fetus or newborn, from 65 in 10,000 births in the United States in 1960 to 10.6 in 10,000 births in 1990.

International

Hydrops fetalis is much more common in Southeast Asia. The best figures come from Thailand, where the expected frequency of hydrops, from homozygous alpha-thalassemia or Bart hydrops alone, is 1 in 500 to 1 in 1500 pregnancies.⁴ Accurate figures from the Mediterranean region are not available; however, the commonness of glucose-6-phosphate dehydrogenase (G-6-PD) deficiency and of defects in alpha-chain hemoglobin production in several populations from this region lead to the suspicion that the incidence of hydrops in that region is much higher than it is in the United States.

Mortality/Morbidity

Estimates of mortality vary widely, from nearly zero to virtually 100%. Most case series report 60-90% mortality, although some improvements are notable in more recent reports. Many causes for these variations are recognized, not the least of which include the sophistication of diagnostic methods used and the complexity and costs of treatment. However, the most important single factor is the cause of the hydrops. A significant proportion of these cases are caused or accompanied by multiple and complex congenital malformations of genetic and/or chromosomal origin, which by themselves are fatal at an early age. Many other causes are accompanied by masses or fluid accumulations, which compress the developing fetal lung and preclude its normal development. Thus, the presence or absence and potential prevention of pulmonary hypoplasia are of crucial importance.

Another highly important factor is the very premature delivery of most babies with hydrops consequent to conditions that distend the uterus and provoke early labor or to therapeutic interventions (eg, fetal thoracentesis, paracentesis, complex fetal surgical procedures). 

One study showed that mortality rate was highest among neonates with congenital anomalies and lowest among neonates with congenital chylothorax.⁵ Infants who died were more likely to be more premature, were sicker after birth (with lower 5-minute Apgar scores), and needed higher levels of support during the first day after birth.

Race

Ethnic influences are related almost entirely to cause. Selected examples include the importance of genetic variations in the alpha-chain structure of hemoglobin in Asian and Mediterranean populations in addition to the more serious nature of the hemolytic disease in the African American fetus affected by maternal ABO-factor isoimmunization.

Sex

Sex influences in incidence or outcome of hydrops fetalis are largely related to the cause of the condition. A significant proportion of hydrops is caused by or associated with chromosomal abnormalities or syndromes. Many of these are X-linked disorders.

Because most individuals with hydrops fetalis are delivered quite prematurely, and because fetal pulmonary maturation takes place earlier in female than in male fetuses, male preterm infants are at greater risk for the pulmonary complications of very preterm delivery. They are also at greater risk for infections (nosocomial or otherwise), which are quite common in the very preterm infant. A striking example of greater male risk is the nearly 13-fold increase in the odds ratio for development of hydrops in the male fetus with Rh D hemolytic disease. Although a single precise risk figure is not available for the heterogenous collection of cases that comprise hydrops fetalis, male fetuses appear to have greater risk for occurrence, morbidity, and mortality.

Clinical

History

A history suggesting the presence of any of the following factors should trigger an extensive diagnostic study:

• Maternal history
  • Rh negative (d;d) blood type
  • Known presence of isoimmune blood group antibodies
  • Prior administration of blood products
  • Risks of illicit drug use
  • Collagen-vascular disease
  • Thyroid disease or diabetes
  • Organ transplant (liver, kidney)
  • Blunt abdominal trauma (abuse, auto accident)
  • Coagulopathy
  • Use of indomethacin, sodium diclofenac, or potentially teratogenic drugs during pregnancy
  • Younger (<16 y) or older (>35 y) maternal age
  • Risk factors for sexually transmitted diseases
  • Hemoglobinopathy (especially with Asian or Mediterranean ethnicity)
  • Occupational exposure to infants or young children
  • Pet cat
  • Current or recent community epidemic of viral illness
• Family history
  • Jaundice in other family members or in previous child
  • Family history of twinning (specifically, monozygotic)
  • Family history of genetic disorders, chromosomal abnormalities, or metabolic diseases
  • Congenital malformation in previous child
  • Previous fetal death
  • Hydramnios in earlier pregnancies
  • Prior hydrops fetalis
  • Previous fetomaternal transfusion
  • Congenital heart disease in previous child

Physical

The presence of any of the following maternal or fetal physical findings should prompt further diagnostic evaluation:

• Twinning
• Hydramnios
• Exanthem or other evidence of intercurrent viral illness
• Herpetic lesion or chancre
• Decrease in fetal movements

Causes

Hydrops fetalis is a nonspecific finding that is easily detected using prenatal ultrasonography and may be associated with a wide range of associated abnormalities. However, despite extensive pre- and postnatal investigations, including postmortem pathological examination of the fetus, no definite cause can be found in about 26% of cases of nonimmune hydrops.
 
Hydrops is an end-stage process for numerous fetal diseases. Causes can be mainly divided in 6 broad categories: cardiovascular, genetic abnormalities, intrathoracic malformations, hematological disorders, infectious conditions, and idiopathic forms. The advent of routine Rh antagonism screening drastically reduced the occurrence of immune hydrops fetalis. 

Hematologic causes that lead to profound anemia and have been recognized to trigger hydrops fetalis are as follows: 

• Isoimmunization (hemolytic disease of the newborn, erythroblastosis)
  • Rh (most commonly D; also C, c, E, e)
  • Kell (K, k, Kp, Js[B])
  • ABO
  • MNSs (M, to date)
  • Duffy (Fy^b
• Other hemolytic disorders
  • Glucose phosphate isomerase deficiency (autosomal recessive)
  • Pyruvate kinase deficiency (autosomal recessive)
  • G-6-PD deficiency (X-linked dominant
• Disorders of red cell production
  • Congenital dyserythropoietic anemia types I and II (autosomal dominant)
  • Diamond-Blackfan syndrome (autosomal dominant)
  • Lethal hereditary spherocytosis (spectrin synthesis defects) (autosomal recessive)
  • Congenital erythropoietic porphyria (Günther disease) (autosomal recessive)
  • Leukemia (usually associated with Down or Noonan syndrome)
  • Alpha-thalassemia (Bart hemoglobinopathy)
  • Parvovirus B19 (B19V
• Fetal hemorrhage
  • Intracranial or intraventricular
  • Hepatic laceration or subcapsular
  • Placental subchorial
  • Tumors (especially sacrococcygeal teratoma)
  • Fetomaternal hemorrhage
  • Twin-to-twin transfusion
  • Isoimmune fetal thrombocytopenia

Several years ago, Rh disease was considered the usual cause of fetal hydrops. The use of Ig in the at-risk mother, administered prior to maternal isoimmunization, should have made this an entirely preventable disorder. Sadly, this has not been the case. Although a dramatic reduction in Rh D sensitization has been realized, the disorder has stubbornly persisted in a small group of women, many of whom have become isoimmunized from repeated exposure to foreign RBC antigens that contaminate needles used for illicit drug use. One study noted this cause for 1 in 5 women with Rh sensitization; the prevalence of hydrops in this group was a stunning 80%.⁶

The reduced prevalence of Rh D disease has made fetal hemolytic anemias secondary to maternal isoimmunization with other Rh-group and other blood group antigens more apparent. Many of these result in profound fetal anemia and hydrops. Because many others are likely, maternal antibody screening should at least search for those already demonstrated to lead to fetal hydrops.

Molecular genetic technologies, specifically polymerase chain reaction (PCR) testing, have been particularly demonstrated to provide more precise and complete genotyping. Other heritable fetal hemolytic anemias have been associated with fetal hydrops. Most are uncommon, autosomal recessive genetic diseases (eg, pyruvate kinase deficiency, glucose phophate isomerase deficiency), and their association with fetal hydrops is limited to 1 or 2 reports. G-6-PD deficiency is a more common, X-linked recessive disorder; however, G-6-PD has been in frequently associated with fetal hydrops.

Diagnosis is important in these rare conditions because they are compatible with a relatively normal life, and fetal transfusions should be effective. Fetal RBC hemolysis from placental transfer of maternal IgG antibody against fetal RBC antigens (termed isoimmune disease) continues to account for approximately 15-20% of individuals with hydrops fetalis.

Early and precise diagnosis is of enormous importance because highly effective fetal therapy is now available, and long-term outcome is unimpaired in babies with these causes for hydrops. Although fetal imaging confirms the presence of hydrops, it does so only after the fact. Studies preceding and predicting fetal deterioration include amniotic fluid (AF) bilirubin (delta optical density at 450 µ using Liley extrapolations) and, more recently, measurement of fetal hematocrit and hemoglobin levels by direct sampling using cordocentesis.

Disorders of RBC production, resulting in functional fetal aplastic anemia, are particularly important causes of fetal hydrops. The importance of infection with B19V (parvovirus) is increasingly recognized. Use of a sensitive and precise diagnostic test (ie, PCR) has demonstrated that perhaps 20% of fetal hydrops is associated with this maternal/fetal infection. During seasons of particularly high prevalence, the proportion is much higher. Early diagnosis is of crucial importance because fetal treatment by direct transfusion has been demonstrated to be effective, the virus has not been demonstrated to have teratogenic effects, and growth and development of the survivors appear to be normal.

Heritable disorders of hemoglobin alpha-chain production are important causes of hydrops in Asian populations. These hemoglobinopathies have become increasingly important in the United States because of recent immigration patterns, particularly in the West. A report from Hawaii over a 10-year period identifies alpha-thalassemia as the single most important cause of fetal hydrops.⁷ Homozygous alpha-thalassemia, with deletion of all 4 alpha-globin genes, results in the total absence of alpha-hemoglobin chains in the fetus. This condition, ranging from 1 in 500 to 1 in 1500 in a Thai population, has been considered to be a fatal fetal condition (Bart hydrops).

More recently, a handful of survivors of hydrops fetalis due to alpha-thalassemia have been reported;⁸ however, all required fetal transfusions, all required repeated frequent transfusions after birth, and all surviving males had hypospadias. Thus, some health care professionals have questioned the practical and ethical basis of fetal and neonatal treatment. However, opportunities for treatment, such as stem cell transplantation, bone marrow transplantation, and gene replacement therapy, may hold promise for babies with this condition in the future. Fetal diagnosis of the condition has been confirmed (using PCR) from fetal DNA samples of chorionic villus, fetal fibroblast (AF), and from fetal blood.

Once disorders of hemoglobin alpha-chain production are confirmed, fetal interventions have been based on hematocrit and hemoglobin levels obtained by direct cordocentesis. Ultrasound findings are nonspecific, and they occur late. Several simple maternal screening techniques have been suggested, but DNA-based studies using a testing system that allows unequivocal identification of haplotypes commonly detected in Asian Americans (-SEA in 62%, -alpha 3.7 in 27%, -FIL in 11%) appear to be most promising in this country. Despite the current generally gloomy outlook and uncertain treatment of the baby with fetal hydrops, early diagnosis of the condition is important because maternal morbidity is very high with fetal hydrops due to alpha-thalassemia.

Other heritable disorders of RBC production are listed above, but none is very common. Some are fatal, but most are manageable after birth; some are associated with malformation syndromes. These heritable disorders all lead to hydrops in the same manner, as do the other conditions listed above.

Profound anemia leads to high-output cardiac failure and increased CVP. Early and precise diagnosis is important for fetuses with correctable conditions (eg, need for and timing of fetal transfusions) and for fetuses with conditions that are not correctable (to permit parents to understand options and participate in decisions about pregnancy management). Gene therapy may also hold promise for some of these babies in the future.

Fetal hemorrhage is another important cause of fetal hydrops. Acute bleeding may be local or more generalized. Unless the origin is from a tumor mass, the bleeding may not be recognized early enough to intervene. Thus, fetal imaging is of critical importance, and a careful examination, particularly of those sites where bleeding has been associated with hydrops, is essential for prompt and proper fetal treatment.

Isoimmune fetal thrombocytopenia is probably more common than has been reported, and, because treatment may be effective in this condition, maternal screening for platelet antibodies should be routine in all incidents in which the cause of fetal hydrops remains undetermined.
 
Sacrococcygeal teratoma is relatively common, accounting for a measurable proportion of incidents of fetal hydrops. Controlled trials are needed to be certain that currently proposed interventions are more helpful than harmful, but these interventions hold considerable promise. Effective treatment is especially important for this condition because associated anomalies are rare, and fully normal development is possible. Once again, fetal imaging studies are the cornerstone for diagnosis and management of sacrococcygeal teratoma.

The fetus may bleed into the mother, and this hemorrhage may be severe enough to lead to fetal death or hydrops. Disruptions of the fetomaternal circulation may be placental or related to tumors (choriocarcinoma, chorangioma), trauma, or partial placental abruption.

Early diagnosis of fetomaternal hemorrhage is deceptively simple, requiring a maternal blood smear to assess the proportion of circulating cells with fetal hemoglobin (resistant to acid elution). Unfortunately, recent automated modifications of this test are less specific and sensitive than the original Betke-Kleihauer test, and several newer tests have been proposed. Of these tests, the most promising appear to be either immunofluorescent flow cytometry or DNA analysis using PCR. More difficult than determining which test to use is knowing when to perform the tests because, in most reported cases, the diagnosis is usually too late to allow effective fetal intervention.

The earliest warning of the condition in most recent series has been reduced fetal body movements accompanied by sinusoidal fetal heart rate patterns and altered fetal biophysical profile. Confirmation of fetal anemia by direct cordocentesis is the final step to transfusion. Unfortunately, fetal transfusion has often been ineffective due to continued, repeated, massive fetal hemorrhages.

Placental vascular anastomoses are present in virtually all monochorionic monozygotic pregnancies. Twin-to-twin transfusion is balanced in most circumstances, with no excessive accumulation or loss for either twin. Sizable hemorrhages or unbalanced transfusion occurs in 5-30% of these pregnancies, leaving one twin anemic and the other polycythemic. This may lead to fetal death, impaired fetal growth, high-output cardiac failure from hypovolemic shock, congestive failure from volume overload, or hydrops fetalis, depending on the size of the bleed and whether it is acute or chronic. Extremely early twin-to-twin transfusion may result in fetal acardia; somewhat later, they may be detected as fetus papyraceous or as a stuck twin or vanishing twin.

Although some placental studies suggest fewer (rather than more) vascular anastomoses with resultant trapping of blood in the recipient fetus, other placental studies demonstrate excessive and abnormal placental vascular communications. Velamentous cord insertion is much more common in those fetuses with large shunts. Curiously, the recipient (polycythemic) twin usually develops hydrops, not the (anemic) donor. Even more curiously, death of the hydropic twin (whether untreated and/or spontaneous, following fetal therapy, or after selective feticide) is not uncommonly followed by the development of hydrops in the remaining twin.

Reasons for all these events remain causes for speculation. Definitive diagnosis is also surprisingly difficult because hydrops may occur in either (or both) twin, disparities in fetal size may not be present, and fetal hemoglobin or hematocrit levels may be well outside the reference range (high or low) in the absence of any hydrops.

Ultrasound evidence of same-sex twins, a monochorionic placenta, with hydramnios in one sac and oligohydramnios in the other sac, is often used to make the diagnosis. These findings and disparities in fetal sizes (15-25%) are useful, but unfortunately they are not definitive. Determination of fetal hemoglobins by cordocentesis is used; however, differences in fetal hemoglobin concentration exceeding 5 g/dL are common in the absence of hydrops, and, conversely, differences less than this may be found in individuals with hydrops.

Significant differences in serum protein levels may also be observed in twins with hydrops fetalis, and atrial natriuretic factor concentrations are usually high. Unfortunately, none of these findings are diagnostic. Clearly, earlier and more precise fetal diagnostic methods, which measure degree of functional dysfunction, are needed.

Most promising in this regard are pulsed Doppler ultrasound measurements of umbilical vessel blood velocity. Such studies hold promise of providing an earlier window of opportunity for fetal diagnosis and treatment. Outcome is surprisingly poor in this condition. Most twins with hydrops die before birth (42-86%), and a shocking proportion of survivors of the condition have cardiovascular and neurologic damage. Ultrasound studies demonstrate cerebral white matter damage, suggesting antenatal necrosis in approximately one third. Follow-up studies of neurodevelopment suggest serious impairment in approximately one quarter of surviving twins.

Many (if not most) surviving twins have significant cardiomyopathy (predominantly right-sided), usually associated with pulmonary outflow obstruction; pulmonary artery calcification and endocardial fibroelastosis also are common. Neutropenia, impaired fetal growth, reduced bone density, and mineralization have been observed in the surviving donors. Optic nerve hypoplasia has been reported, and peripheral vascular ischemic necrosis with gangrene of distal extremities has been observed in several individuals with the condition. Coagulopathy and embolic phenomena were speculated in many early studies; however, scant evidence for them is present in recent reports. Very premature delivery is common and contributes undoubtedly to the morbidity and mortality.

Treatment successes have been reported with transfusion of the anemic fetus, plasmapheresis of the polycythemic twin, laser ablation of placental vascular anastomoses, and amnioreduction; however, failures and serious complications have also been reported with each of these. See Twin to Twin Transfusion Syndrome.

Cardiovascular problems that cause or associated with hydrops are summarized below. Although extensive, the list is inevitably incomplete because new associations are reported each year.Cardiac causes are as follows:

• Structural anomalies
  • Abnormalities of left ventricular outflow
    • Aortic valvular stenosis
    • Aortic valvular atresia
    • Coarctation of the aorta
    • Aortico-left ventricular tunnel
    • Atrioventricular canal
    • Left ventricular aneurysm
    • Truncus arteriosus
    • Hypoplastic left heart
    • Spongiosum heart
    • Endocardial fibroelastosis
  • Abnormalities of right ventricular outflow
    • Pulmonary valvular atresia or insufficiency
    • Ebstein anomaly
  • Other vascular malformations
    • Arteriovenous malformations
    • Diffuse hemangiomatosis
    • Placental hemangioma
    • Umbilical cord hemangioma
    • Hepatic hemangioendothelioma
    • Abdominal hemangioma
    • Pulmonary arteriovenous fistula
    • Cervical hemangioendothelioma
    • Paratracheal hemangioma
    • Cutaneous cavernous hemangioma
    • Arteriovenous malformations of the brain
• Nonstructural anomalies
  • Obstruction of venous return
    • Superior or inferior vena cava occlusion
    • Absent ductus venosus
    • Umbilical cord torsion or varix
    • Intrathoracic or abdominal tumors or masses
    • Disorders of lymphatic drainage
  • Supraventricular tachycardia
  • Congenital heart block
  • Prenatal closure of the foramen ovale or ductus arteriosus
  • Myocarditis
  • Idiopathic arterial calcification or hypercalcemia

Congenital structural anomalies of the heart may be present in as many as 1 in 4 babies with hydrops; both right-heart and left-heart anomalies, systolic-overload and diastolic-overload conditions, high-output, and congestive situations are represented.

Structural cardiac defects are commonly accompanied by other anomalies and are often associated with cytogenic abnormalities. Examples include the association between coarctation of the aorta and Turner syndrome, the relation between atrioventricular (AV) canal and/or endocardial cushion defects and Down syndrome, and the common association of Turner syndrome with cystic hygroma, left-sided lymphatic flow defects, and left-heart outflow defects.

Fibroelastosis may be an isolated abnormality; however, fibroelastosis more commonly represents an endocardial response to chronic fetal myocardial stress. Prenatal detection of a cardiac defect should always trigger a careful search for other malformations, and karyotyping should be performed in all such fetuses. Arteriovenous malformations (AVMs) are often cited causes of hydrops; they are listed individually above.

Impaired right-heart filling is also an important cause of hydrops. Although uncommon, umbilical or vena caval thromboses have been noted. Theoretically, they may be correctable if diagnosed early enough. Conversely, tumor compression is a frequently reported cause of hydrops. Several of these masses involve lymphatic malformation and/or obstruction; cystic hygroma is a particularly important example.

Prenatal closure of the foramen ovale or ductus arteriosus prematurely converts the (parallel) fetal circulation to a (serial) postnatal circulation; associated problems are obvious. Most recorded instances of premature ductal closure are iatrogenic, related to maternal administration of indomethacin or sodium diclofenac.

Several instances of idiopathic arterial calcification with hydrops have been reported. In one such incident, fetal serum calcium levels were elevated, and a possible association with Williams syndrome was suggested. In 3 other cases, lysosomal storage diseases were present (Gaucher, sialidosis, galactosialidosis). No associations were noted in 4 cases. Hydropic recipients of twin-to-twin transfusion who survive also usually have pulmonary artery calcification.

Fetal supraventricular tachycardias are important causes of hydrops because they can be diagnosed accurately by cardiac imaging in early pregnancy, they may be treated effectively before hydrops develops, and, since associated malformations or syndromes are rare, they have anticipated good outcomes. Whether an AV block is present (atrial flutter) or not (tachyarrhythmia), survival rates of 85-95% are typical, and neurodevelopmental outcome is usually normal. The condition is more common in males than in females (2:1). Clinical experience and animal model studies indicate that hydrops can occur with sustained cardiac rates of less than 220-230 beats per minute (bpm) and that the risk is related directly to the degree of prematurity.

Congenital heart block is also often associated with hydrops. Diagnosis is made using cardiac imaging or with an ECG in the newborn; rates are always less than 90 beats per minute (bpm) and usually less than 65 bpm. Approximately two thirds to three fourths occur in pregnancies complicated by maternal collagen disease. Maternal IgG antinuclear antibodies cross the placenta and attack fetal collagen in the conduction bundle. Why some fetuses develop congenital heart block and some do not is unclear; however, an association with human leukocyte antigen (HLA) types (HLA-DR3, among others) has been suggested.

Treatment with various drugs has generally been unsuccessful, as has fetal surgery for pacing. Recent evidence suggests corticosteroid therapy may be of benefit.

Virtually all of the remaining babies, whose mothers have no collagen disorder, have serious, complicated, cardiac structural defects. The most common lesions are AV canal and/or endocardial cushion defects, transposition of the great vessels, and other isomerisms. Outcomes for these babies are grim. Mortality is 25-35% if cardiac structure is normal; many survivors require neonatal surgery for pacing, and no information is available on long-term outlooks. Because the cardiac structural abnormalities are so serious and complex, mortality and morbidity are much higher if cardiac anomalies are present.

The immature fetus is particularly susceptible to overwhelming viral and bacterial infection. Those agents, which do not kill quickly, may cause smoldering generalized infections with myocarditis, suppressed erythropoiesis and myelopoiesis, hemolysis, and hepatitis. Such infections may lead to hydrops fetalis. Those agents reported to be causative, to date, are listed below. This list will change over time as other agents, yet to be identified, are demonstrated to cause hydrops.

• B19V
• Cytomegalovirus (CMV)
• Syphilis
• Herpes simplex
• Toxoplasmosis
• Hepatitis B
• Adenovirus
• Ureaplasma urealyticum
• Coxsackievirus type B
• Listeria monocytogenes

The association of congenital syphilis with hydrops is classic. Fetal and placental edema accompanied by serous effusions first was described generations ago. However, the surprising frequency with which maternal serologic tests for syphilis may appear negative in this condition is less well known.

The prozone phenomenon, observed during primary and secondary maternal syphilis, occurs when a higher-than-optimal amount of antisyphilis antibody in the tested maternal sera prevents the flocculation reaction typifying a positive result in reagin tests. In these circumstances, dilution of the tested serum is necessary to make the correct diagnosis. Thus, serum dilution (to as much as 1:1024 or greater) should be routine in high-risk situations and should certainly be used in any individual in whom fetal hydrops of unknown etiology is present.

Early, accurate diagnosis of this infection is critical because fetal treatment is available and effective. Several viral infections have been associated with fetal hydrops. The number of viruses implicated and the frequency of these cases have paralleled the increased recognition of this association and the improved simplicity and sensitivity of diagnostic methods. Hydrops in these conditions appears to be the cumulative result of viral effects on marrow, myocardium, and vascular endothelium. Currently, reports of effective fetal treatment are rare.

Of particular interest is recent evidence of how commonly acute B19V infection is a cause of fetal hydrops.⁹ The virus was first identified in 1974 and first linked with fetal hydrops 10 years later. Evidence published since then suggests this virus may be the single most important currently recognized cause of fetal hydrops. Parvovirus may be the cause of as much as one third of all incidents of hydrops fetalis. The virus directly attacks red cell precursors and is visible as intranuclear inclusions in stained RBC preparations. Thrombocytopenia is also usually present.

Outcome in such fetuses is surprisingly good; spontaneous resolution occurs in approximately one third of such incidents, and approximately 85% of those who receive fetal transfusions survive. The virus is not teratogenic and, despite reports of viral persistence in myocardial and brain tissues, neurodevelopmental outcome in survivors appears to be normal. Early, accurate diagnosis, using maternal serologic and/or molecular biologic PCR techniques, is essential. Positive results are usually confirmed by direct fetal PCR, hemoglobin, hematocrit, and platelet studies to plot a proper treatment plan.

An interesting association between hydrops and fetal meconium peritonitis is recognized. At least 16 such cases are found in the literature. No baby reported before 1991 had evidence of infection; however, CMV (1), hepatitis B (1), and B19V (5) were found in 7 of 8 cases reported since 1991. The only instance of meconium peritonitis and hydrops without confirmed infection in these later reports was probably iatrogenic because it followed paracentesis with subsequent placement of a peritoneoamniotic shunt. These observations suggest that the coexistence of hydrops and meconium peritonitis should be assumed to be related to fetal infection until proven otherwise.

Hydrops fetalis has been associated with more than 75 inborn errors of metabolism, chromosomal aberrations, and genetic syndromes. Approximately 50 of the more common errors are listed below. An additional 20 or more reports of imprecisely defined chromosomal or genetic syndromes identify hydrops as an incidental finding.

• Inborn errors of metabolism
  • Glycogen-storage disease, type IV
  • Lysosomal storage diseases
    • Gaucher disease, type II (glucocerebroside deficiency)
    • Morquio disease (mucopolysaccharidosis, type IV-A)
    • Hurler syndrome (mucopolysaccharidosis, type 1H; alpha1–iduronidase deficiency)
    • Sly syndrome (mucopolysaccharidosis, type VII; beta-glucuronidase deficiency
    • Farber disease (disseminated lipogranulomatosis)
    • G_M1 gangliosidosis, type I (beta-galactosidase deficiency)
    • Mucolipidosis I
    • I-cell disease (mucolipidosis II)
    • Niemann-Pick disease, type C
  • Salla disease (infantile sialic acid storage disorder [ISSD] or sialic acid storage disease, neuroaminidase deficiency)
  • Hypothyroidism and hyperthyroidism
  • Carnitine deficiency
• Genetic syndromes (autosomal recessive, unless otherwise noted)
  • Achondrogenesis, type IB (Parenti-Fraccaro syndrome)
  • Achondrogenesis, type II (Langer-Saldino syndrome)
  • Arthrogryposis multiplex congenita, Toriello-Bauserman type
  • Arthrogryposis multiplex congenita, with congenital muscular dystrophy
  • Beemer-Langer (familial short-rib syndrome)
  • Blomstrand chondrodysplasia
  • Caffey disease (infantile cortical hyperostosis; uncertain inheritance)
  • Coffin-Lowry syndrome (X-linked dominant)
  • Cumming syndrome
  • Eagle-Barrett syndrome (prune-belly syndrome; since 97% males, probably X-linked)
  • Familial perinatal hemochromatosis
  • Fraser syndrome
  • Fryns syndrome
  • Greenberg dysplasia
  • Lethal congenital contracture syndrome
  • Lethal multiple pterygium syndrome (excess of males, so probably X-linked)
  • Lethal short-limbed dwarfism
  • McKusick-Kaufman syndrome
  • Myotonic dystrophy (autosomal dominant)
  • Nemaline myopathy with fetal akinesia sequence
  • Noonan syndrome (autosomal dominant with variable penetrance)
  • Perlman/familial nephroblastomatosis syndrome (inheritance uncertain)
  • Simpson-Golabi-Behmel syndrome (X-linked [Xp22 or Xp26])
  • Sjögren syndrome A (uncertain inheritance)
  • Smith-Lemli-Opitz syndrome
  • Tuberous sclerosis (autosomal dominant)
  • Yellow nail dystrophy with lymphedema syndrome (autosomal dominant
• Chromosomal syndromes
  • Beckwith-Wiedemann syndrome (trisomy 11p15)
  • Cri-du-chat syndrome (chromosomes 4 and 5)
  • Dehydrated hereditary stomatocytosis (16q23-qter)
  • Opitz G syndrome (5p duplication)
  • Pallister-Killian syndrome (isochrome 12p mosaicism)
  • Trisomy 10, mosaic
  • Trisomy 13
  • Trisomy 15
  • Trisomy 18
  • Trisomy 21 (Down syndrome)
  • Turner syndrome (45, X)

The heterogeneity of this collection of associations is bewildering at first glance. However, the common thread that runs through is useful for the clinician to understand. Most of the babies with hydrops associated with the conditions listed above have severe complex cardiac defects, disorders of lymphatic drainage, arteriovenous malformations, impaired production of properly functioning red cells, and/or thoracoabdominal masses that impair venous return to the heart. Thus, the same disturbed pathophysiology identified as causing hydrops in the animal studies is reflected in these conditions.

Inheritance for most of these conditions (when known) is autosomal, most commonly recessive. Because a few of these conditions are X-linked recessive, slightly more males are affected among this particular set of causes. Gene therapy may hold therapeutic promise for the future; however, outcomes are generally grim for babies with hydrops related to these causes. Accurate diagnosis is particularly important in these babies, despite their poor prognosis because parental counseling is of critical importance in the management of current and future pregnancies for these families.

Fetal hydrops has been associated with approximately 10 of the approximately 50 lysosomal storage disorders. Little doubt appears to exist that hydrops will be linked with most such inborn errors of metabolism in the near future.

Cystic hygroma is often found in several of these conditions. These vascular tumors are associated commonly with complex profound aberrations of lymphatic drainage. They are usually found in the neck but may also be present in the abdomen or thoracic cavity. Two thirds to three fourths of fetuses with this tumor have chromosomal abnormalities (most commonly 45,XO), and those fetuses with normal chromosomes often have major malformations. This association with Turner, Noonan, and lethal multiple pterygium syndromes is particularly notable. Mortality is extremely high (85-96%), but early precise diagnosis is important for purposes of genetic counseling and pregnancy management. The reports of spontaneous remissions with healthy long-term survival are important to note.

Thoracic and abdominal tumors are common causes of fetal hydrops. This association makes physiologic sense because the location and size of these masses are likely to obstruct the return of venous or lymphatic fluids to the heart. Some are commonly associated with major malformations and/or chromosomal abnormalities and, consequently, have a poor long-term prognosis. For example, upper airway obstructions are associated with other major malformations in more than one half of the cases reported, and the association of fetal rhabdomyomas with tuberous sclerosis and complex cardiac malformations is well recognized.

Tumor or mass causes of hydrops fetalis are as follows:

• Intrathoracic tumors or masses
  • Pericardial teratoma
  • Rhabdomyoma
  • Mediastinal teratoma
  • Cervical vascular hamartoma
  • Pulmonary fibrosarcoma
  • Leiomyosarcoma
  • Pulmonary mesenchymal malformation
  • Lymphangiectasia
  • Bronchopulmonary sequestration
  • Cystic adenomatoid malformation of the lung
  • Upper airway atresia or obstruction (laryngeal or tracheal)
  • Diaphragmatic hernia
  • Eventration of the diaphrag
• Abdominal tumors or masses
  • Metabolic nephroma
  • Polycystic kidneys
  • Neuroblastoma
  • Hepatic mesenchymal hamartoma
  • Hepatoblastoma
  • Ovarian cyst
• Other conditions
  • Placental choriocarcinoma
  • Placental chorangioma
  • Cystic hygroma
  • Intussusception
  • Meconium peritonitis
  • Intracranial teratoma
  • Sacrococcygeal teratoma

Venous return is directly impaired by such conditions as pericardial teratomas and cardiac rhabdomyosarcomas. Upper airway (laryngeal, tracheal) atresia or obstruction leads to massive pulmonary overdistention and, thus, to impaired cardiac filling. Cystic hygromas are mentioned again because they comprise an important and common example of mass compression with obstruction of venous-lymphatic return. Meconium peritonitis is noted in tumor or mass causes and in the discussion of infectious causes. This redundancy is due to the fact that some observers have postulated an association with hydrops on the basis of mass effects on venous return; as noted earlier, the association is almost certainly one with fetal infection and consequent red cell aplasia.

Some of these conditions may lead to fetal hydrops not because of mass compression effects but because their intense vascularization may lead to arteriovenous shunting and/or to massive fetal hemorrhage. Such consequences are especially common with sacrococcygeal teratomas and with placental chorioangiomas. In both instances, fetal high-output cardiac failure may ultimately lead to fetal hydrops and/or death. Sacrococcygeal teratoma is associated with hydrops in one fifth to one third of cases in several case series; fetal coagulopathy, most commonly thrombocytopenia, is found in approximately the same proportion of cases. Tumor size as assessed using ultrasonography has not been demonstrated to be an independent prognostic factor; however, solid, highly vascular tumors lead to hydrops more often than those with a more cystic, less vascular structure.

Because chromosomal abnormalities and life-threatening anomalies are rare with sacrococcygeal sequestration, early diagnosis and aggressive fetal treatment are particularly important with this condition. Although bloody AF secondary to rupture of the highly vascular teratoma is not uncommon, diagnosis in most cases has been made only after hydrops has developed.

Early routine fetal imaging may be the only way in which early diagnosis can be made in this condition; however, the low incidence of sacrococcygeal teratoma may preclude cost-effective screening for this condition. Elevated concentrations of alpha-fetoprotein (AFP) and/or acetylcholinesterase in AF have been found to accompany fetal sacrococcygeal teratoma, but the invasive sampling and low specificity appears to preclude these tests as routine screening procedures. While placental chorioangiomas are common (present in approximately 1% of pregnancies), large vascular tumors with cardiovascular and hematologic consequences are very uncommon. When present, the pathophysiology is remarkably similar to that found with fetal sacrococcygeal teratomas. Diagnosis and techniques for early intervention are also similar.

Bronchopulmonary sequestration is a condition in which abnormal vascular supply and misplacement of a portion of the lung may lead to torsion of the affected lobes, profound obstruction of lymphatic and venous return, and tension hydrothorax. This sequence of events leads to fetal hydrops in perhaps one third of such cases. Although drainage of the hydrothorax, definitive diagnosis using color Doppler imaging, and fetal angiography have been described, and though fetal surgical excision of the affected portion of the lung may improve survival in this condition, nearly two thirds of these cases fail to be diagnosed before fetal death or birth occurs.

Cystic adenomatoid malformation of the lung may also lead to hydrops by mass compression of venous return. Because this condition is seldom associated with other malformations or with chromosomal abnormalities and because fetal surgical maneuvers have demonstrated considerable promise with some forms of the disorder, early and precise diagnosis using fetal imaging techniques is of critical importance.

Pulmonary capillary-alveolar development is abnormal in this condition, and 3 degrees of severity, described initially by Stocker, have been used to predict prognosis.¹⁰ The 3 degrees of severity are as follows:

• Type I: The fetus with large (>2 mm), isolated cysts seldom develops hydrops, and spontaneous remissions have been reported. Drainage or excision of individual cysts has also been reported with generally favorable outcomes.
• Type II: Poorer prognosis is associated in the fetus with smaller (<2 mm) diffuse macrocysts, and isolated fetal pulmonary excisions have been proposed in those who develop hydrops.
• Type III: In the fetus with microcystic disease, the affected lung appears solid, hydrops is common, and outcome is generally unfavorable.

Compression of fetal lung, common in so many of the tumor and mass causes, not only impairs cardiac return but also has an additional particularly serious consequence. External compression of developing fetal lung is known to impair both anatomic and biochemical maturation. Pulmonary hypoplasia, with a profound reduction in the number of functional alveolar units, is a common finding when fetal hydrops accompanies these conditions. Delayed or impaired maturation of pulmonary surfactant production is another consequence of impaired expansion of the fetal lung, thus worsening the already serious compromise of extreme prematurity in these babies.

• Search for CME/CE on This Topic »

• Hydrops Fetalis (Radiology)
• Endocardial Fibroelastosis (Pediatrics: Cardiac Disease and Critical Care Medicine)
• Thalassemia (Pediatrics: General Medicine)
• Hemolytic Disease of Newborn (Pediatrics: Cardiac Disease and Critical Care Medicine)

• Heart Center
• Supraventricular Tachycardia Overview

• Which Diagnosis Best Fits These Symptoms?
• Immunological Complications of Blood Transfusion
• How Should an Infant With Abnormal P Wave Axis Be Managed?