SMFM Clinical Guideline ajog - Sieog€¦ · (SMFM) Clinical Guideline #7: nonimmune hydrops fetalis Society for Maternal-Fetal Medicine (SMFM); Mary E. Norton, MD; Suneet P. Chauhan,

SMFM Clinical Guideline ajog.org

Society for Maternal-Fetal Medicine(SMFM) Clinical Guideline #7:nonimmune hydrops fetalis
Society for Maternal-Fetal Medicine (SMFM); Mary E. Norton, MD;Suneet P. Chauhan, MD; and Jodi S. Dashe, MD
ydrops fetalis is a Greek term

OBJECTIVE: Nonimmune hydrops is the presence of �2 abnormal fetal fluid collectionsin the absence of red cell alloimmunization. The most common etiologies include car-diovascular, chromosomal, and hematologic abnormalities, followed by structural fetalanomalies, complications of twinning, infection, and placental abnormalities. We soughtto provide evidence-based guidelines for the evaluation and management of nonimmunehydrops fetalis.METHODS: A systematic literature review was performed using MEDLINE, PubMed,EMBASE, and Cochrane Library. The search was restricted to English-language articlespublished from 1966 through June 2014. Priority was given to articles reporting originalresearch, although review articles and commentaries also were consulted. Abstracts ofresearch presented at symposia and scientific conferences were not consideredadequate for inclusion in this document. Evidence reports and guidelines published byorganizations or institutions such as the National Institutes of Health, Agency for HealthResearch and Quality, American Congress of Obstetricians and Gynecologists, andSociety for Maternal-Fetal Medicine were also reviewed, and additional studies werelocated by reviewing bibliographies of identified articles. Grading of RecommendationsAssessment, Development, and Evaluation methodology was employed for definingstrength of recommendations and rating quality of evidence. Consistent with US Pre-ventive Task Force guidelines, references were evaluated for quality based on the highestlevel of evidence.RESULTS AND RECOMMENDATIONS: Evaluation of hydrops begins with an antibodyscreen (indirect Coombs test) to verify that it is nonimmune, detailed sonography of thefetus(es) and placenta, including echocardiography and assessment for fetal arrhythmia,and middle cerebral artery Doppler evaluation for anemia, as well as fetal karyotype and/or chromosomal microarray analysis, regardless of whether a structural fetal anomaly isidentified. Recommended treatment depends on the underlying etiology and gestational

H that describes pathological fluid(“ὕdur,” Greek for water) accumulationin fetal soft tissues and serous cavities.The features are detected by ultrasound,and are defined as the presence of �2abnormal fluid collections in the fetus.These include ascites, pleural effusions,pericardial effusion, and generalizedskin edema (defined as skin thickness>5 mm).1 Other frequent sonographicfindings include placental thickening(typically defined as a placental thickness�4 cm in the second trimester or �6cm in the third trimester)2,3 and poly-hydramnios (Figure 1). Nonimmunehydrops fetalis (NIHF) refers specificallyto cases not caused by red cell alloim-munization. With the developmentand widespread use of Rh(D) immuneglobulin, the prevalence of Rh(D)alloimmunization and associatedhydrops has dramatically decreased. Asa result, NIHF now accounts for almost90% of cases of hydrops,4 with theprevalence in published series reportedas 1 in 1700-3000 pregnancies.5-7

From the Society for Maternal-Fetal MedicinePublications Committee, Washington, DC; theDivision of Maternal-Fetal Medicine, University ofCalifornia, San Francisco, San Francisco, CA(Dr Norton); the Department of Obstetrics,Gynecology and Reproductive Sciences, TheUniversity of Texas Health Science Center atHouston, Houston, TX (Dr Chauhan); and theDepartment of Obstetrics and Gynecology,University of Texas Southwestern MedicalCenter, Dallas, TX (Dr Dashe).

Received Nov. 5, 2014; accepted Dec. 12,2014.

The authors report no conflict of interest.

Corresponding author: Society for Maternal-Fetal Medicine: Publications Committee. [email protected]

0002-9378/freeª 2015 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.ajog.2014.12.018

age; preterm delivery is recommended only for obstetric indications including devel-opment of mirror syndrome. Candidates for corticosteroids and antepartum surveillanceinclude those with an idiopathic etiology, an etiology amenable to prenatal or postnataltreatment, and those in whom intervention is planned if fetal deterioration occurs. Suchpregnancies should be delivered at a facility with the capability to stabilize and treatcritically ill newborns. The prognosis depends on etiology, response to therapy iftreatable, and the gestational age at detection and delivery. Aneuploidy confers a poorprognosis, and even in the absence of aneuploidy, neonatal survival is often <50%.Mirror syndrome is a form of severe preeclampsia that may develop in association withfetal hydrops and in most cases necessitates delivery.

Key words: fetal complications, hydrops, nonimmune hydrops fetalis

FLA 5.2.0 DTD

However, many of these reports pre-date routine sonography and limitedinformation is available on contempo-rary incidence of NIHF in a prenatalpopulation.

FEBRUARY 2015 Am

� YMOB10177_proof � 30 December 2014 � 3

What is the underlying pathogenesisof NIHF?The common pathophysiology underly-ing the many etiologies of hydrops fetalisis an imbalance in the regulation of

erican Journal of Obstetrics & Gynecology 127:59 pm � ce

mailto:[email protected]

mailto:[email protected]

http://dx.doi.org/10.1016/j.ajog.2014.12.018

http://www.AJOG.org

http://www.AJOG.org

print&web4C=F

PO

FIGURE 1Sonographic features of hydrops fetalis

A, Pericardial effusions. B, Pleural effusions; note midline heart anterior to small lungs with bilateral

effusions. C, Placental thickening, with placenta measuring>4 mm in thickness. D, Skin thickening

at level of fetal skull. E, Ascites, sagittal, with free-floating loops of bowel surrounded by ascites.

F, Ascites in upper abdomen, at level of fetal liver and stomach.SMFM. Nonimmune hydrops fetalis. Am J Obstet Gynecol 2015.


fluid movement between the vascularand interstitial spaces,8 with an increasein interstitial fluid production or adecrease in lymphatic return. Variousmechanisms thought to lead to NIHFinclude increased right heart pressure,resulting in increased central venouspressure (eg, structural heart defects);obstruction of venous or arterialblood flow (eg, pulmonary masses);inadequate diastolic ventricular filling

128 American Journal of Obstetrics & Gynecology

FLA 5.2.0 DTD

(eg, arrhythmias); hepatic venous con-gestion leading to decreased hepaticfunction and hypoalbuminemia; in-creased capillary permeability (eg,congenital infection); anemia leadingto high output cardiac failure andextramedullary hematopoiesis, oftenwith resultant hepatic dysfunction;lymphatic vessel dysplasia and obs-truction (eg, cystic hygroma); andreduced osmotic pressure (eg,

FEBRUARY 2015


congenital nephrosis). The precisepathogenesis of NIHF depends on theunderlying disorder, and in manycases remains unclear. Pathophysio-logic mechanisms that contribute to thedevelopment of hydrops are describedin Table 1 according to underlying eti-ology or category.

What are the causes of NIHF?NIHF can result from a large numberof underlying pathologies (Table 1).The differential diagnosis is extensive,and the success in identifying a causepartially depends on the thoroughnessof efforts to establish a diagnosis.Although older studies considered manycases to be idiopathic,9-11 more recent,larger series and a systematic review re-port that a cause can be found in nearly60% of cases prenatally12 and in 85%when postnatal detection is included.13

A number of series have been pub-lished describing the many disordersassociated with NIHF.8,13-16 Review ofthese indicates that the most commonetiologies ofNIHF include cardiovascularcauses, chromosomal anomalies, andhematologic abnormalities. Other con-ditions associated with NIHF in-clude fetal malformations, particularlythoracic abnormalities, twin-twin tran-sfusion syndrome, congenital infection,placental abnormalities, fetal tumors, andgenetic or metabolic disorders (Table 1).

Overall, cardiovascular abnormalitiesare the most common cause of NIHFin most series, accounting for about20% of cases.13 NIHF can result fromcardiac structural abnormalities, ar-rhythmias, cardiomyopathy, cardiac tu-mors, or vascular abnormalities. Inmost cardiac cases, hydrops is likelycaused by increased central venouspressure due to a structural malforma-tion or from inadequate diastolic ven-tricular filling.8,17 The most commoncongenital heart defects reported in as-sociation with NIHF are right heart de-fects.6,14,18 The prognosis of NIHF dueto cardiac structural abnormalities ispoor, with combined fetal and infantmortality reported as 92%, largely dueto the severity of the heart defects thatcause in utero congestive heart failure.19

:59 pm � ce

http://www.AJOG.org

TABLE 1Etiologies of nonimmune hydrops fetalis6,11,12,14,75

Cause Cases Mechanism

Cardiovascular 17-35% Increased central venous pressure

Chromosomal 7-16% Cardiac anomalies, lymphatic dysplasia, abnormalmyelopoiesis

Hematologic 4-12% Anemia, high output cardiac failure; hypoxia (alphathalassemia)

Infectious 5-7% Anemia, anoxia, endothelial cell damage, andincreased capillary permeability

Thoracic 6% Vena caval obstruction or increased intrathoracicpressure with impaired venous return

Twin-twin transfusion 3-10% Hypervolemia and increased central venous pressure

Urinary tract abnormalities 2-3% Urinary ascites; nephrotic syndrome withhypoproteinemia

Gastrointestinal 0.5-4% Obstruction of venous return; gastrointestinalobstruction and infarction with protein loss anddecreased colloid osmotic pressure

Lymphatic dysplasia 5-6% Impaired venous return

Tumors, includingchorioangiomas

2-3% Anemia, high output cardiac failure, hypoproteinemia

Skeletal dysplasias 3-4% Hepatomegaly, hypoproteinemia, impaired venousreturn

Syndromic 3-4% Various

Inborn errors ofmetabolism

1-2% Visceromegaly and obstruction of venous return,decreased erythropoiesis and anemia, and/orhypoproteinemia

Miscellaneous 3-15%

Unknown 15-25%

SMFM. Nonimmune hydrops fetalis. Am J Obstet Gynecol 2015.

ajog.org SMFM Clinical Guideline

Both tachyarrhythmias and bradyar-rhythmias can lead to NIHF.14,20 Themost common tachyarrhythmias aresupraventricular tachycardia and atrialflutter, and both can often be successfullytreated with transplacental medicaltherapy.14,19,20 We recommend maternaltreatment with antiarrhythmic medica-tions for NIHF secondary to fetaltachyarrhythmia unless the gestationalage is close to term or there is a maternalor obstetrical contraindication. Medica-tion selection and dosing are reviewedelsewhere.21

Fetal bradycardia is most commonlycaused by congenital heart block, whichmay occur secondary to an immuneetiology such as transplacental passageof anti-Sjogren’s-syndrome-related an-tigen A, also called anti-Ro, or thecombination anti SSA/Ro or anti Ro/SSAantibodies associated with maternalautoimmune disease. It may also resultfrom structural abnormalities affectingcardiac conduction, as with endocardialcushion defects in the setting of a het-erotaxy syndrome. Once third-degreeatrioventricular block has developed,treatment with corticosteroid therapyhas not been shown to be beneficial, andin the setting of hydrops the prognosis ispoor.22 For this reason, in-utero therapyfor fetal bradyarrhythmia resulting inhydrops is considered investigationaland is not generally recommendedoutside of a research setting.

Chromosomal abnormalities, particu-larly Turner syndrome (45,X) and Downsyndrome (trisomy 21) are also commoncauses of NIHF, accounting for 13% ina large systematic review.13 In prenatalseries, aneuploidy is the most commoncause of NIHF, particularly when iden-tified early in gestation.4,5,12 Turnersyndrome is associated with 50-80%of cases of cystic hygromas, which resultfrom a lack of communication betweenthe lymphatic system and venousdrainage in the neck.23 Lymphaticdysplasia likely leads to the developmentof NIHF in these cases.

NIHF has been described in associa-tion with other aneuploidies, includingtrisomies 13 and 18, and triploidy.24-27

In some cases, hydrops occurs dueto cardiovascular malformations in

FLA 5.2.0 DTD

aneuploid fetuses. NIHF has also beenreported in trisomy 21 in the absence ofstructural heart defects.24,25,27,28 Somesuch cases occur due to a transientabnormal myelopoiesis, a leukemiccondition that occurs in about 10%of infants with Down syndrome.25,27

Postnatally, transient abnormal myelo-poiesis is often mild and self-limiting;prenatally, it is less common but typi-cally more severe. For these reasons,we recommend that NIHF is an indica-tion to offer prenatal diagnosis withkaryotype, fluorescence in situ hybridi-zation, and/or chromosomal microarrayanalysis, even when severe anemiais present (Figure 2). Screening withnoninvasive prenatal testing may detectsome chromosomal causes but providesmore limited information about

FEBRUARY 2015 Am


possible genetic etiologies, and thereforewe recommend diagnostic testing.

Fetal anemia, which can result inimmune hydrops if caused by bloodgroup alloimmunization, can also leadto NIHF. Etiologies include inheritedconditions such as hemoglobinopathies,as well as acquired conditions, such ashemolysis, fetomaternal hemorrhage,parvovirus infection, or red cell aplasia.

Among the hemoglobinopathies, themost common cause of NIHF is alphathalassemia. This autosomal recessivedisorder is common in Southeast Asianpopulations, where it accounts for 28-55% of NIHF.29,30 The incidence in mostother series of NIHF is about 10%.13

Parents can be screened by evaluationof the mean cell volume, which will be<80 fL in thalassemia carriers. Definitive


http://www.AJOG.org

FIGURE 2Workup of nonimmune hydrops fetalis

AFAFP, amniotic fluid alpha fetoprotein; CBC, complete blood count; CMA, chromosomal microarray; CMV, cytomegalovirus; G6PD,glucose-6-phosphate dehydrogenase deficiency; MCA PSV, middle cerebral artery peak systolic velocity; MCV, mean corpuscularvolumne; MoM, multiple of the median; PCR, polymerase chain reaction; RPR, rapid plasma reagin.

*Assuming negative indirect Coombs test, thereby excluding alloimmunization; **CMV/toxoplasmosis testing if fetal anomalies sug-gestive of infection; ***Either amniocentesis or FBS; ****Available in some laboratories.67,76



diagnosis of an affected fetus can bemade by detection of one of the commonDNA deletions or point mutations thataccount for most cases.31 Conversely, afetal blood sample can be evaluated forthe presence of the abnormal Bart’s he-moglobin seen in this condition. Bart’shemoglobin is an ineffective oxygencarrier, thus the fetus with alpha thalas-semia will suffer severe intrauterinehypoxia from an early gestational age.The resultant NIHFwill typically presentin the late second or early third trimester.

Fetal anemia may also occur due tofetal hemorrhage. NIHF occurs only withsignificant fetomaternal bleeding thatis not enough to lead to fetal hypo-volemia and death. Fetomaternal hem-orrhage leading to hydrops may occur


FLA 5.2.0 DTD

as either an isolated acute event, or as achronic, ongoing hemorrhage.32,33 Witheither, a Kleihauer-Betke smear willshow the presence of fetal cells in thematernal peripheral blood in most cases.Flow cytometry can also be used to es-timate the volume of fetal bleedinginto the mother. This is an importantdiagnosis to make, because even witha massive fetomaternal hemorrhage,intravascular fetal transfusion can belifesaving.34-36 For this reason, werecommend that NIHF due to anemiafrom fetomaternal hemorrhage betreated with transfusion, unless thepregnancy is at an advanced gestationalage and risks associated with deliveryare considered to be less than thoseassociated with the procedure.

FEBRUARY 2015


Other, less common causes of fetalanemia and hydrops include G-6-PDdeficiency, erythrocyte enzymopathiessuch as pyruvate kinase deficiency, andmaternal acquired red cell aplasia.37,38

NIHF has been reported in asso-ciation with a number of viral, bacterial,and parasitic infectious diseases, in-cluding parvovirus, cytomegalovirus,syphilis, and toxoplasmosis.39-42 Inmost series, such infections account for5-10% of NIHF.6,13,14,16 Although theassociations are less clear, NIHF hasalso been reported to occur with Cox-sackie virus, trypanosomiasis, varicella,human herpesvirus 6 and 7, herpessimplex type 1, respiratory syncytial vi-rus, congenital lymphocytic choriome-ningitis virus, and leptospirosis.6,43-47

Fetal infection can cause NIHF due toanemia, anoxia, endothelial cell damage,increased capillary permeability, andmyocarditis.

Parvovirus is the most commonly re-ported infectious cause of NIHF. Inthe fetus, the virus has a predilectionfor erythroid progenitor cells, leadingto inhibition of erythropoiesis and sub-sequent anemia.48,49 The risk of a pooroutcome for the fetus is greatest whenthe congenital infection occurs in theearly second trimester (<20 weeks ofgestation).50-53 The risk of fetal death hasbeen reported to be 15% at 13-20 weeksof gestation, and 6% after 20 weeks.54

In most cases, the anemia is transientand fetal intravascular transfusion cansupport a fetus through this aplasticcrisis.55,56 However, development ofNIHF is associated with high mortality,and outcomes are reported to be signif-icantly improved following fetal intra-uterine transfusion.51,53 For this reason,we recommend fetal intrauterine trans-fusion for NIHF due to parvovirusinfection, unless the pregnancy is at anadvanced gestational age and risks asso-ciated with delivery are considered tobe less than those associated with theprocedure.

Fetal thoracic abnormalities, includingmasses as well as congenital hydrothorax,can also be associated with NIHF. Themost frequent pulmonary lesion associ-ated with NIHF is a congenital pulmo-nary airway malformation (CPAM).

:59 pm � ce

http://www.AJOG.org


With a large lesion or effusion, medias-tinal shift may impair venous returnand cardiac output, and the associatedesophageal compression may result inpolyhydramnios. Hydrops occurs in onlyabout 5% of fetuses with CPAM butconfers a poor prognosis without treat-ment.57 If the lesion is macrocystic,the cyst may be treated with needledrainage or thoracoamniotic shuntplacement.58,59 If predominantly solid(microcystic), both corticosteroid ther-apy and in utero resection have beenadvocated, and corticosteroid treatmentis currently recommended as a first-linetreatment.60 Large bronchopulmonarysequestrations have also been treatedwith a needle procedure involvingneodymium:yttrium-aluminium-garnetlaser of the feeding vessel.57

The most common etiology of anisolated effusion leading to NIHF ischylothorax, caused by lymphaticobstruction. The fluidmay be sampled atthe time of needle drainage or shuntplacement, and the diagnosis is con-firmed by the finding of a fetal pleuralcell count with >80% lymphocytes inthe absence of infection. Reported sur-vival exceeds 50% in hydropic fetusestreated with thoracoamniotic shuntplacement.61

Twin-twin transfusion syndrome resultsfrom an imbalance in blood flow causedby anastomoses in the placentas ofmonochorionic twin pregnancies. In se-vere cases, one or both twinsmay developNIHF, although more commonly therecipient twin is affected, likely due tohypervolemia and increased centralvenous pressure.62 Cases of twin-twintransfusion sequence with hydropshave a very poor prognosis withouttreatment, and laser therapy is consid-ered by most experts to be the bestavailable therapeutic approach toimprove the prognosis.63 Selective ter-mination via umbilical cord coagulationis also an option for pregnancies withtwin-twin transfusion sequence result-ing in NIHF. Another complication ofmonochorionic twinning that mayresult in NIHF is twin-reversed arterialperfusion sequence. Radiofrequencyablation of the acardiac twin has beenadvocated for severe cases, including

FLA 5.2.0 DTD

those with hydrops, with reportedoverall survival of 80%.64

Structural urinary and gastrointestinalabnormalities are less common causes ofNIHF. A ruptured bladder or renal col-lecting system may cause urinary ascitesand mimic NIHF. Congenital nephroticsyndromes have been reported to causeNIHF due to hypoproteinemia.19,65,66

Surviving infants may have massiveproteinuria at birth and develop renalfailure in childhood.Few primary abnormalities of the

gastrointestinal tract have been associ-ated with NIHF. Those that have beenreported include diaphragmatic hernia,midgut volvulus, gastrointestinal ob-struction, jejunal atresia, malrotation ofthe intestines, and meconium perito-nitis.6,19 Intraabdominal masses maycause NIHF due to obstruction ofvenous return, while gastrointestinalobstruction and infarction may lead todecreased colloid osmotic pressure dueto protein loss.19 Hepatic disorderssuch as cirrhosis, hepatic necrosis,cholestasis, polycystic disease of the liver,and biliary atresia have been reported inassociation with NIHF, most likely dueto hypoproteinemia.7 Hemangiomaof the liver has also been reported as acause of NIHF, probably due to arterio-venous shunting resulting in cardiacfailure.Neoplastic diseases or fetal tumors can

occur in utero and have been associatedwith NIHF. Relatively common in thiscategory are lymphangiomas, hemangi-omas, sacrococcygeal, mediastinal, andpharyngeal teratomas, and neuroblas-tomas.19,67,68 Many of these are veryvascular and lead to NIHF due to highoutput cardiac failure. Fetal therapy hasbeen offered for cases of solid sacro-coccygeal teratoma resulting in NIHF,and in a recent systematic review, openfetal surgery resulted in survival in 6 of11 cases (55%), and minimally invasivetherapy was associated with survival in6 of 20 (30%).69 Tuberous sclerosis isan autosomal dominant disorder char-acterized by fibroangiomatous tumorsin multiple organs, most typically thecortex of the brain, the skin, and thekidneys. Cardiac rhabdomyomas andliver fibrosis are also sometimes present.

FEBRUARY 2015 Am


NIHF has been reported in associationwith tuberous sclerosis, probably eitheras a result of cardiac failure due torhabdomyomas (resulting in obstructionto filling or outflow), or hepatic failuredue to fibrosis.70

Placental and cord lesions that havebeen associated with NIHF include cho-rioangiomas, angiomyxoma of the cord,aneurysm of the umbilical artery, cordvein thrombosis, umbilical vein torsion,true knots, and amniotic bands.7,19,68

Placental chorioangiomas are relativelycommon, occurring in about 1% ofpregnancies. While small lesions areusually not clinically significant, thosemeasuring >5 cm can act as highvolume arteriovenous shunts and leadto hydrops due to high output cardiacfailure. Other vascular tumors and arte-riovenous malformations can similarlycause NIHF. Hemangiomas have beenreported to cause NIHF, likely due tosevere anemia, hypoproteinemia, and/orextramedullary erythropoiesis.

A large number of skeletal dysplasiashave been associatedwithNIHF, includingachondroplasia, achondrogenesis, osteo-genesis imperfecta, osteopetrosis, thana-tophoric dysplasia, short-rib polydactylysyndrome, and asphyxiating thoracicdysplasia.14,19,71-73 In all of these, themechanism is unclear, although it hasbeen proposed that hepatic enlargementoccurs secondary to intrahepatic prolifer-ation of blood cell precursors tocompensate for a small bone-marrowvolume. This may cause large vesselcompression and lead to anasarca in thesefetuses.

Inborn errors of metabolism and othergenetic conditions are historically asso-ciated with 1-2% of cases of NIHF, whichmay be transient or manifest as isolatedascites. Inherited metabolic disordersthat have been implicated as a cause ofNIHF are most typically lysosomalstorage diseases such as various muco-polysaccharidoses, Gaucher disease, andNiemann-Pick disease.74,75 In a recentreview of the literature including 678cases of NIHF, lysosomal storage diseasesoccurred in 5.2% of all NIHF cases, andin 29.6% of idiopathic NIHF cases ifa comprehensive workup for these con-ditions is done.76 Proposed mechanisms


http://www.AJOG.org


involve visceromegaly and obstructionof venous return, decreased erythropoi-esis and anemia, and/or hypopro-teinemia. Although such disorders are arelatively uncommon cause of NIHF,they are important because of the highrecurrence risk of these mainly auto-somal recessive disorders. Careful his-tology of the placenta, liver, spleen,and bone marrow will often provide aclue that a metabolic storage disorderwas present. For many such disorders,testing is available to determine a diag-nosis and for prenatal diagnosis in asubsequent pregnancy. Panels of causa-tive storage disorders can be tested forin some laboratories, and this shouldbe considered for cases of NIHF ina structurally normal fetus in whichanother cause has not been identified,or with cases of recurrence within afamily.76,77

A number of other syndromes havebeen associated with NIHF. Many ofthese are disorders associated withlymphatic dysfunction, such as Noonanand multiple pterygium syndrome, bothof which frequently present with cystichygroma; idiopathic chylothorax, inwhich a local pleuromediastinal lymphvessel disturbance occurs as the possiblepathogenic mechanism; yellow nail syn-drome, a dominantly inherited congen-ital lymphedema syndrome; andcongenital pulmonary lymphangiectasia.Familial recurrence in some of thesecases suggests a hereditary maldevelop-ment of lymphatic vessels.6,14,19,78

What is the appropriate evaluationwhen fetal hydrops is detected?Sonographic identification of thehydropic fetus is not difficult. The diag-nostic challenge is to establish the etiol-ogy and determine the appropriatetherapy (if available) and timing of de-livery. It has been reported that the causeof hydrops can be determined in about60-85% of cases, although this includespostnatal evaluation.13

Figure 2 outlines the various steps inthe evaluation of the hydropic fetus. It isespecially important to rule out poten-tially treatable conditions, as well as ge-netic disorders with a risk of recurrencein future pregnancies. Often, the etiology


FLA 5.2.0 DTD

of the hydrops can be determined at thetime of diagnosis, since several etiologiesare confirmed or excluded based uponultrasound findings (eg, twin-to-twintransfusion, cardiac arrhythmias, andstructural anomalies associated withNIHF).Management is guided by the pres-

ence or absence of additional anomalies.Sonographic evaluation should includea detailed survey for anomalies of thefetus, umbilical cord, and placenta, andestimation of amniotic fluid volume. Afetal echocardiogram should be in-cluded, as fetal cardiac anomalies areamong the most common causes ofNIHF.In a structurally normal fetus, the first

step is to rule out alloimmunization as acause. The maternal blood type andRh(D) antigen status are assessed as partof routine prenatal care, along with anindirect Coombs test (an antibodyscreen) to evaluate for circulating redblood cell antibodies. These resultsshould be reviewed, and if the indirectCoombs test was previously normal, itshould be repeated. Maternal bloodstudies should also include a completeblood cell count with differentialand indices, Kleihauer-Betke stain forfetal hemoglobin, and parvovirus B19serology. Serologic test results for syph-ilis should be reviewed or repeated, andconsideration should be given to acutephase titers for cytomegalovirus andtoxoplasmosis.It is particularly important to perform

middle cerebral artery Doppler studiesto assess for the presence of fetal anemia,which may be treatable with intravas-cular transfusion. The fetus with NIHFdue to severe anemia will have increasedvelocity through the middle cerebralartery.79

A fetal karyotype, fluorescence in situhybridization studies, and/or chromo-somal microarray analysis should beoffered with or without identified so-nographic anomalies.80 This can beperformed by amniocentesis or fetalblood sampling; the latter allows directanalysis of fetal hematocrit and hemo-globin if anemia is suspected. Invasivetesting also allows testing for lysosomalstorage disorders, and polymerase chain

FEBRUARY 2015


reaction studies for parvovirus, toxo-plasmosis, and cytomegalovirus infec-tion. Such testing should be performedin a structurally normal fetus in whichno other cause has been identified.

An important step in the evalua-tion of NIHF is to exclude a geneticabnormality. Genetically transmitteddisorders account for about onethird of cases of NIHF, and includechromosomal abnormalities, hemo-globinopathies, skeletal dysplasias,metabolic storage disorders, and eryth-rocyte enzymopathies. A complete fam-ily history is thus imperative to rule out aknown inherited disorder in the familyand to assess for consanguinity, whichwill increase the likelihood of a recessivedisorder. Although idiopathic NIHFhas a low recurrence risk, the risk forsome cases of NIHF may be as highas 25%, making genetic counseling anintegral part of the management of anypatient with NIHF.

What maternal risks are associatedwith NIHF?Women with NIHF may develop mirrorsyndrome, an uncommon complicationin which the mother develops edemathat “mirrors” that of her hydropic fetus.Mirror syndrome may represent a formof preeclampsia, and is characterized byedema in approximately 90%, hyper-tension in 60%, and proteinuria in 40%of cases.81 As it is uncommon and likelyunderdiagnosed, the incidence is un-clear. Additional associated findings withthe syndrome include headache, visualdisturbances, oliguria, elevated uric acid,liver function tests, or creatinine levels,low platelets, anemia, and hemodilu-tion.82 A review of the literature (1956through 2009) by Braun et al81 notedthat among 56 cases ofmirror syndrome,the major maternal morbidity was pul-monary edema, which occurred in 21%.Resolution occurs with either the treat-ment of the hydrops or with delivery.81,82

There have been case reports inwhich pregnancies with mirror syn-drome and various treatable causes ofhydropsesecondary to fetal arrhythmia,hydrothorax, parvovirus, and bladderobstructionehave experienced resolu-tion of both hydrops and mirror

:59 pm � ce

http://www.AJOG.org


syndrome following treatment.83-86 Thesame imbalance of angiogenic and anti-angiogenic factors described with severepreeclampsia has also been observed incases of mirror syndrome, with correc-tion following treatment and resolutionof the NIHF.83,84 However, there areno data regarding the likelihood of res-olution or long-term benefits. Givenrisks of expectant management ofsevere preeclampsia, it is recommendedthat this approach be taken only withcaution, and that delivery not be delayedif the maternal condition deteriorates.Thus for most cases of NIHF, includingall cases without a treatable etiology,development of mirror syndrome ne-cessitates delivery.

What obstetric complications areassociated with NIHF?Polyhydramnios and preterm birthoccur frequently with NIHF, with

TABLE 2Therapy for selected etiologies of noEtiology Thera

Cardiac tachyarrhythmia,supraventricular tachycardia,atrial flutter, or atrial fibrillation

Materof ant

Fetal anemia secondary toparvovirus infection orfetomaternal hemorrhage

Fetalintrau

Fetal hydrothorax, chylothorax,or large pleural effusion associatedwith bronchopulmonary sequestration

Fetalor plashunt;needleselect

Fetal CPAM Macroof effushunt;adminbetamh � 26.25 m

TTTS or TAPS Laseror sel

Twin-reversed arterialperfusion sequence

Percu

For each of these etiologies, it is recommended that treatment

CPAM, congenital pulmonary airway malformation; IM, intramus

SMFM. Nonimmune hydrops fetalis. Am J Obstet Gynecol

FLA 5.2.0 DTD

reported incidences as high as 29%71 and66%,87 respectively. If the poly-hydramnios is associated with maternalrespiratory symptoms, reported man-agement options have included a shortcourse of a prostaglandin inhibitor orserial amnioreduction. Since both treat-ment modalities lack evidence of bene-fit and have potential complications,including in utero constriction of theductus arteriosus, abruption, prematurerupture of the membranes, and neo-natal complications such as necrotizingenterocolitis and patent ductus arterio-sus, they should be used judiciously.88,89

Tocolytic agents are a consideration<24weeks if contractions occur secondary toa known inciting event, such as aninvasive procedure performed for thediagnosis or management of NIHF.88

Although in the past preterm deliveryhas been advocated by some to poten-tially improve the outcome of NIHF,7

nimmune hydrops21,54,58,59,61,63,64

py Recomm

nal transplacental administrationiarrhythmic medication(s)

Treatmengestationaor obstetr

blood sampling followed byterine transfusion

Fetal intraunless preand risksbe less th

needle drainage of effusioncement of thoracoamnioticif gestational age is advanced,drainage prior to delivery in

ed cases

Considerresultingconsidera

cystic type: fetal needle drainagesion or placement of thoracoamnioticmicrocystic type: maternalistration of corticosteroids,ethasone 12.5 mg IM q24doses or dexamethasoneg IM q12 h � 4 doses

Considerresulted iresulted ininclude m

ablation of placental anastomosesective termination

Consideraplacentalresulted i

taneous radiofrequency ablation Referral fradiofrequ

be performed at tertiary care center or center with expertise in releva

cular; NIHF, nonimmune hydrops fetalis; TAPS, twin-anemia polycyth

2015.

FEBRUARY 2015 Am


prematurity is likely to worsen theprognosis. For this reason, we recom-mend that preterm delivery be under-taken only for obstetric indications.

What is the prognosis of NIHF?The prognosis of NIHF depends on theunderlying etiology, gestational age atdetection and delivery, Apgar scores,extent of resuscitation in the deliveryroom, and whether the newborn re-quires transport.90 In one prenatal series,nearly half of those diagnosed<24weekshad aneuploidy, with extremely poorsurvival. However, even in the absence ofa chromosomal abnormality, survivalwas<50%.4 In a recent prenatal series of71 pregnancies that continued >20weeksethereby excluding many withaneuploidyesurvival was approximately50%, and only 25% survived withoutmajor morbidities.12 Among liveborninfants, neonatal mortality with NIHF is

endation

t with antiarrhythmic medication unlessl age is close to term or there is maternalical contraindication to therapy

uterine transfusion if anemia is confirmed,gnancy is at an advanced gestational ageassociated with delivery are considered toan those associated with procedure

drainage of large unilateral pleural effusion(s)in NIHF, or, if gestational age is advanced,tion of needle drainage prior to delivery

drainage of large macrocystic CPAM that hasn NIHF; if large microcystic CPAM hasNIHF, we suggest that management optionsaternal corticosteroid administration

tion of fetoscopic laser photocoagulation ofanastomoses for TTTS or TAPS that hasn NIHF <26 wk

or consideration of percutaneousency ablation that has resulted in NIHF

nt therapy.

emia sequence; TTTS, twin-twin transfusion sequence.


http://www.AJOG.org


reported to be as high as 60%.91 Withchylothorax as the underlying etiology,the mortality may be as low as 6%;however, when the infant has associatedanomalies, almost two-thirds do notsurvive.14 Treatable causes of hydrops,such as fetal arrhythmia or infectionwith parvovirus B19,92 have a betterprognosis. In a large series of newbornsadmitted to the neonatal intensive careunit, the independent risk factors fordeath in logistic regression analyses wereyounger gestational age, low 5-minuteApgar score, and need for high levels ofsupport during the first day after birth(higher levels of inspired oxygen supportand greater need for high-frequencyventilation).14

Temporal trends suggest that amongcases of liveborn infants with NIHF, theassociated mortality has not improvedover 2 decades. Comparing the mortalityamong hydropic newborns delivered in1993 through 2003 vs 2003 through2009, there was no significant differencein mortality in the 2 time periods, 47%vs 67%, respectively.91 In addition to thesmall sample size, an explanation for thelack of improvement in survival overtime may be that the more severe casesare now more frequently diagnosed

TABLE 3Society for Maternal-Fetal MedicineRecommendations

� We recommend that initial evaluation of hydroCoombs test) to verify that it is nonimmune,ography to evaluate for fetal and placental abnanemia, and fetal karyotype or chromosomawhether structural fetal anomalies are identi

� We recommend that fetal therapy decisionsparticular whether there is a treatable causeNIHF develops or is first identified

� As prematurity is likely to worsen prognosis,be undertaken only for obstetric indications

� We recommend that pregnancies with NIHF detiologies be considered candidates for cortisurveillance, and that they be delivered at aand treat critically ill neonates

� We recommend that in most cases, developmfor delivery

MCA, middle cerebral artery; NIHF, nonimmune hydrops fetalis.

SMFM. Nonimmune hydrops fetalis. Am J Obstet Gynecol


FLA 5.2.0 DTD

prenatally and referred to tertiary cen-ters, which are more likely to contributeto large series in the literature.The long-term prognosis for survivors

of NIHF also depends upon the under-lying etiology. After intrauterine trans-fusion for hydrops secondary toinfection with parvovirus B19, there ispotential for delayed psychomotordevelopment and abnormal neurologicaloutcomes.93 It is unclear if this isbecause of the hydrops, a direct conse-quence of the parvovirus infection,from severe anemia, or associatedwith the transfusion. Finally, fetuseswith supraventricular tachycardia maydevelop Wolff-Parkinson-White syn-drome later in life.94

Management of NIHFThe cornerstone of counseling andmanagement for this condition is athorough evaluation for the underlyingetiology of the hydrops (Figure 2).Pregnancy management decisions willdepend on the etiology, in particularwhether there is a treatable cause andthe gestational age that NIHF developsor is first identified. Cases generally fallinto 1 of 3 categories: those amenable tofetal therapyewhich often require

recommendations for nonimmune hydroG

ps include an antibody screen (indirecttargeted sonography with echocardi-ormalities, MCA Doppler evaluation forl microarray analysis, regardless offied (Figure 2)

1CSt

be based on underlying etiology, in(Table 2) and the gestational age that

1CSt

we recommend that preterm delivery 1CSt

ue to nonlethal or potentially treatablecosteroid therapy and antepartumcenter that has capability to stabilize

1CSt

ent of mirror syndrome is an indication 1CSt

2015.

FEBRUARY 2015


urgent treatment or referral to aspecialized center; those with a lethalprognosis, for whom pregnancy termi-nation or comfort care are the only op-tions realistic to offer; and cases in whichthe etiology is idiopathic and the prog-nosis is likely poor but uncertain. Giventhe poor overall prognosis, pregnancytermination should be offered if NIHF isidentified prior to viability. It is impor-tant in counseling that the potential formaternal complications with expectantmanagement be anticipated, includingmirror syndrome. Serial evaluation ofmaternal blood pressure is thereforerecommended.

What are the fetal therapy optionsavailable for NIHF?Selected etiologies of NIHF for whichfetal therapy should be considered arelisted in Table 2. Therapy options mayinclude intrauterine transfusion(s)for fetal anemia, medications such asantiarrhythmic agents, drainage of largepleural effusions, corticosteroids forCPAMs, or specialized procedures suchas laser coagulation of placental anasto-moses for twin-twin transfusion syn-drome. The list is not intended tobe comprehensive but rather to serve as a

psrading of recommendations (Table 4)

rong recommendation, low-quality evidence





:59 pm � ce

http://www.AJOG.org

TABLE 4Grading of Recommendations Assessment, Development, and Evaluation recommendationsGrade of recommendation Clarity of risk/benefit Quality of supporting evidence Implications

1AStrong recommendation,high-quality evidence

Benefits clearly outweigh riskand burdens, or vice versa

Consistent evidence from well-performedrandomized, controlled trials or overwhelmingevidence of some other form; further researchis unlikely to change our confidence in estimateof benefit and risk

Strong recommendations, can apply to mostpatients in most circumstances withoutreservation; clinicians should follow strongrecommendation unless clear and compellingrationale for an alternative approach is present

1BStrong recommendation,moderate-quality evidence

Benefits clearly outweigh riskand burdens, or vice versa

Evidence from randomized, controlled trials withimportant limitations (inconsistent results,methodologic flaws, indirect or imprecise), or verystrong evidence of some other research design;further research (if performed) is likely to have animpact on our confidence in estimate of benefitand risk and may change estimate

Strong recommendation and applies to mostpatients; clinicians should follow strongrecommendation unless clear and compellingrationale for an alternative approach is present

1CStrong recommendation,low-quality evidence

Benefits appear to outweigh riskand burdens, or vice versa

Evidence from observational studies, unsystematicclinical experience, or from randomized, controlledtrials with serious flaws; any estimate of effect isuncertain

Strong recommendation, and applies to mostpatients; some of evidence base supportingrecommendation is, however, of low quality

2AWeak recommendation,high-quality evidence

Benefits closely balanced withrisks and burdens

Consistent evidence from well-performedrandomized, controlled trials or overwhelmingevidence of some other form; further researchis unlikely to change our confidence in estimateof benefit and risk

Weak recommendation, best action maydiffer depending on circumstances orpatients or societal values

2BWeak recommendation,moderate-quality evidence

Benefits closely balanced withrisks and burdens, some uncertainlyin estimates of benefits, risks,and burdens

Evidence from randomized, controlled trials withimportant limitations (inconsistent results,methodologic flaws, indirect or imprecise), orvery strong evidence of some other researchdesign; further research (if performed) is likelyto have an impact on our confidence in estimateof benefit and risk and may change estimate

Weak recommendation, alternative approacheslikely to be better for some patients under somecircumstances

2CWeak recommendation,low-quality evidence

Uncertainty in estimates of benefits,risks, and burdens; benefits may beclosely balanced with risks and burdens

Evidence from observational studies, unsystematicclinical experience, or from randomized, controlledtrials with serious flaws; any estimate of effectis uncertain

Very weak recommendation; other alternativesmay be equally reasonable

Best practice A recommendation in which either:(i) there is an enormous amount of indirectevidence that clearly justifies strongrecommendation; direct evidence would bechallenging, and an inefficient use of timeand resources, to bring together andcarefully summarize; or (ii) recommendationto contrary would be unethical


ajog.o

rgSM

FM

Clin

icalGuidelin

e

FEBRUARY2015

American

JournalofObstetrics

&Gynecology

135FLA

5.2.0

DTD

�YMOB10177_proof�

30Decem

ber

2014�

3:59pm

�ce

http://www.AJOG.org


guideline. With the exception of openfetal surgery, therapy is sometimesoffered to pregnancies identified asbeing at risk for NIHF, with the under-standing that the prognosis worsens ifhydrops develops.

Counseling for pregnancies withNIHF amenable to fetal therapy shouldinclude a discussion of potential risks,benefits, and alternatives that takes intoconsideration the severity of the under-lying condition and the anticipatedresponse to the intervention. If the pa-tient declines therapy or is unable toreceive therapy, the prognosis is poor.Given the specialized nature of fetaltherapy, patients should receive carefrom physicians with expertise providingthe treatment offered, which in somecases may require evaluation at aspecialized center.

When is antepartum fetal surveillanceappropriate in NIHF?Antepartum surveillance is generallyused in the setting of maternal orpregnancy complications associatedwith an increased risk for fetal demise,and when findings from surveillancewill assist with delivery decisions. ForNIHF, antepartum testing has not beendefinitively shown to improve perinataloutcomes, and all indications fortesting are considered relative.95 Thereare no management trials or observa-tional series of the utility of ante-partum surveillance in the setting ofNIHF upon which to base recommen-dations. Whether an individual preg-nancy with NIHF may benefit fromsurveillance depends on the etiology ofthe hydrops, the underlying patho-physiology, and the potential for pre-natal or postnatal treatment.

Fetuses with NIH may be candidatesfor antepartum surveillance if: (1)the underlying etiology of the hydropsis not considered lethal, (2) the preg-nancy has reached a viable gestationalage, and (3) the findings from surveil-lance would be used to assist with timingof delivery. In such cases, deteriorationof testing results or worsening of thesonographic findings of hydrops mightprompt delivery.


FLA 5.2.0 DTD

Most fetuses with NIHF secondaryto an etiology listed in Table 2 are can-didates for antepartum surveillance. Iffetal therapy is attempted but does notameliorate the hydrops, the prognosis issignificantly worse.20,61 If the NIHFis idiopathic, counseling about theguarded prognosis should include limi-tations in available treatment options,but in the absence of a contraindication,antepartum testing may be considered. Ifthere are questions about the postnatalprognosis, consultation with a neonatol-ogist or other pediatric subspecialist maybe helpful.

When is the optimal timing of delivery?There are no management trials ofdelivery timing in the setting of NIHFupon which to base recommendations.Many hydropic fetuses succumb priorto viability. There is no evidence thatelective preterm delivery will improvethe outcome. In one retrospective series,preterm birth <34 weeks was a poorprognostic factor.90 Based on expertopinion, development or worsening ofNIHF in a pregnancy that has reachedabout 34 weeks would seem a reason-able indication for delivery, althoughgiven the wide spectrum of etiologiesand severity of NIHF, care should beindividualized. In the absence of clinicaldeterioration or other indication forearlier intervention, delivery by 37-38weeks should be considered. As dis-cussed previously, we recommend de-livery in most cases if mirror syndromedevelops.

Should corticosteroids be given?There are no studies that specificallyaddress the utility of antepartum corti-costeroid therapy to ameliorate thesequelae of prematurity in the settingof NIHF, and there are similarly no datato suggest that corticosteroid adminis-tration is detrimental in pregnanciescomplicated by hydrops. In 2 retrospec-tive series, neonatal survival was notimproved in those who received corti-costeroids.9,96 This was likely due tothe overall extremely high morbidityand mortality among infants withNIHF in these cohorts, and that

FEBRUARY 2015


hydrops represents an advanced stageof multiple underlying pathophysiol-ogies. Fetuses with NIHF are at riskfor preterm delivery and thus forprematurity-related morbidities thatmay compound their hemodynamiccompromise. Pregnancies with NIHFwould reasonably be candidates forantepartum corticosteroid therapy ifthe gestational age is between 24-34weeks, if the underlying etiologyof the hydrops is not considered lethal,and if intervention is planned on behalfof the fetus should deterioration of thefetal condition occur. If any type of fetaltherapy is planned (Table 2) and thegestational age is between 24-34 weeks,corticosteroid administration should beconsidered.

What is the optimal mode of delivery?If the fetus is potentially treatable orconsidered viable, and if the decisionto proceed with delivery is based onfindings of antepartum surveillance orconcern about deterioration of thefetal condition (eg, based on sono-graphic findings), cesarean delivery maybe indicated. Prior to delivery of thehydropic fetus, consideration should begiven to whether drainage of a largeeffusion may improve the efficacy ofneonatal resuscitative efforts. Rarely, ef-fusions may be so large as to pose a riskfor trauma to the infant during delivery.Depending on the degree of associatedeffusions and anasarca, considerationshould be given to the potential fordystocia at delivery. If a decision hasbeen made not to intervene for fetalindicationseto provide comfort careonly, vaginal delivery is preferred unlessotherwise contraindicated.

Where should delivery occur?If the NIHF is considered to have anetiology that is potentially amenable topostnatal treatment, or if the etiology ofthe hydrops is idiopathic, the pregnancyshould be delivered at a center with alevel-III neonatal intensive-care unit thathas the capability to stabilize and treatcritically ill neonates. This may requiretransfer of the pregnant patient prior todelivery.

:59 pm � ce

http://www.AJOG.org


RECOMMENDATIONS

Recommendations regarding NIHF arepresented in Table 3. The gradingscheme classifies recommendations aseither strong (grade 1) or weak (grade 2),and classifies the quality of evidence ashigh (grade A), moderate (grade B), orlow (grade C). Thus, the recommenda-tions can fall into 1 of the following6 categories: 1A, 1B, 1C, 2A, 2B, 2C(Table 4).

Quality of evidenceThe quality of evidence for each articlewas evaluated according to the methodoutlined by the US Preventative ServicesTask Force:

I Properly powered and conductedrandomized controlled trial (RCT);well-conducted systematic review ormetaanalysis of homogeneous RCTs.

II-1 Well-designed controlled trial withoutrandomization.

II-2 Well-designed cohort or case-controlanalytic study.

II-3 Multiple time series with or withoutthe intervention; dramatic resultsfrom uncontrolled experiment.

III Opinions of respected authorities,based on clinical experience;descriptive studies or case reports;reports of expert committees.

This opinion was developed by thePublications Committee of the Societyfor MaternaleFetal Medicine (SMFM)with the assistance of Mary E. Norton,MD, Suneet P. Chauhan, MD, and Jodi S.Dashe, MD and was approved by theexecutive committee of the society onSept. 29, 2014. Each member of thepublications committee (Sean Blackwell,MD [Chair], Mary Norton, MD [ViceChair], Vincenzo Berghella, MD, JosephBiggio,MD, AaronCaughey,MD, SuneetChauhan, MD, Sabrina Craigo, MD, JodiDashe, MD, Brenna Hughes, MD, JamieLo, MD, Tracy Manuck, MD, BrianMercer, MD, Eva Pressman, MD, An-thony Sciscione, DO, Neil Silverman,MD, Alan Tita, MD, and GeorgeWendel,MD) has submitted a conflict of interest

FLA 5.2.0 DTD

disclosure delineating personal, profes-sional, and/or business interests thatmight be perceived as a real or potentialconflict of interest in relation to thispublication. -

REFERENCES

1. Skoll MA, Sharland GK, Allan LD. Is the ul-trasound definition of fluid collections in non-immune hydrops fetalis helpful in defining theunderlying cause or predicting outcome? Ul-trasound Obstet Gynecol 1991;1:309-12(Level II-2).2. Lee AJ, Bethune M, Hiscock RJ. Placentalthickness in the second trimester: a pilot study todetermine the normal range. J Ultrasound Med2012;31:213-8 (Level II-3).3. Hoddick WK, Mahony BS, Callen PW,Filly RA. Placental thickness. J Ultrasound Med1985;4:479-82 (Level II-3).4. Santolaya J, Alley D, Jaffe R, Warsof SL.Antenatal classification of hydrops fetalis. ObstetGynecol 1992;79:256-9 (Level III).5. Heinonen S, Ryynänen M, Kirkinen P.Etiology and outcome of second trimesternon-immunologic fetal hydrops. ActaObstet Gynecol Scand 2000;79:15-8 (LevelII-2).6. Machin GA. Hydrops revisited: literaturereview of 1,414 cases published in the1980s. Am J Med Genet 1989;34:366-90(Level III).7. Hutchison AA, Drew JH, Yu VY,Williams ML, Fortune DW, Beischer NA.Nonimmunologic hydrops fetalis: a review of61 cases. Obstet Gynecol 1982;59:347-52(Level III).8. Bellini C, Hennekam RC. Non-immunehydrops fetalis: a short review of etiology andpathophysiology. Am J Med Genet A2012;158A:597-605 (Level III).9. Wy CA, Sajous CH, Loberiza F, Weiss MG.Outcome of infants with a diagnosis of hydropsfetalis in the 1990s. Am J Perinatol 1999;16:561-7 (Level III).10. Larroche JC, Aubry MC, Narcy F. In-trauterine brain damage in nonimmunehydrops fetalis. Biol Neonate 1992;61:273-80 (Level III).11. Laneri GG, Classen DL, Scher MS. Brainlesions of fetal onset in encephalopathic infantswith nonimmune hydrops fetalis. Pediatr Neurol1994;11:18-22 (Level III).12. Santo S, Mansour S, Thilaganathan B, et al.Prenatal diagnosis of non-immune hydropsfetalis: what dowe tell the parents? Prenat Diagn2011;31:186-95 (Level II-2).13. Bellini C, Hennekam RC, Fulcheri E, et al. Eti-ology of nonimmune hydrops fetalis: a systematicreview. Am J Med Genet A 2009;149A:844-51(Level I).14. Abrams ME, Meredith KS, Kinnard P,Clark RH. Hydrops fetalis: a retrospective re-view of cases reported to a large nationaldatabase and identification of risk factors

FEBRUARY 2015 Am


associated with death. Pediatrics 2007;120:84-9 (Level II-2).15. Lallemand AV, Doco-Fenzy M, Gaillard DA.Investigation of nonimmune hydrops fetalis:multidisciplinary studies are necessary fordiagnosisereview of 94 cases. Pediatr DevPathol 1999;2:432-9 (Level III).16. Jauniaux E, Van Maldergem L, DeMunter C, Moscoso G, Gillerot Y. Nonimmunehydrops fetalis associated with genetic ab-normalities. Obstet Gynecol 1990;75:568-72(Level II-3).17. Fesslova V, Villa L, Nava S, Boschetto C,Redaelli C, Mannarino S. Spectrum andoutcome of atrioventricular septal defect infetal life. Cardiol Young 2002;12:18-26 (LevelII-2).18. Hofstaetter C, Hansmann M, Eik-Nes SH,Huhta JC, Luther SL. A cardiovascular profilescore in the surveillance of fetal hydrops.J Matern Fetal Neonatal Med 2006;19:407-13(Level II-2).19. Randenberg AL. Nonimmune hydropsfetalis part II: does etiology influencemortality? Neonatal Netw 2010;29:367-80(Level III).20. Moodley S, Sanatani S, Potts JE,Sandor GG. Postnatal outcome in patients withfetal tachycardia. Pediatr Cardiol 2013;34:81-7(Level II-2).21. Donofrio MT, Moon-Grady AJ,Hornberger LK, et al. American Heart Asso-ciation Adults with Congenital Heart DiseaseJoint Committee of the Council on Cardio-vascular Disease in the Young and Council onClinical Cardiology, Council on Cardiovascu-lar Surgery and Anesthesia, and Council onCardiovascular and Stroke Nursing. Diag-nosis and treatment of fetal cardiac disease:a scientific statement from the AmericanHeart Association. Circulation 2014;129:2183-242 (Level III).22. Friedman DM, Kim MY, Copel JA,Hanos C, Davis C, Beyon JP. Prospectiveevaluation of fetuses with autoimmune-associated congenital heart block followed inthe PR Interval and Dexamethasone Evalua-tion (PRIDE) study. Am J Cardiol 2009;103:1102-6 (Level II-1).23. Alpman A, Cogulu O, Akgul M, et al.Prenatally diagnosed Turner syndrome andcystic hygroma: incidence and reasons forreferrals. Fetal Diagn Ther 2009;25:58-61(Level II-2).24. Sükür YE, Gözüküçük M, Bayramov V,Koç A. Fetal hydrops and anemia as signs ofDown syndrome. J Formos Med Assoc2011;110:716-8 (Level III).25. Akdag A, Tunç B, O�guz S, Dilli D, Dilmen U.A newborn with Down syndrome-transientmyeloproliferative disorder. J Perinat Med2010;38:445-7 (Level III).26. Acar A, Balci O, Gezginc K, et al. Evaluationof the results of cordocentesis. Taiwan J ObstetGynecol 2007;46:405-9 (Level II-2).27. Malin GL, Kilby MD, Velangi M. Transientabnormal myelopoiesis associated with Down


http://refhub.elsevier.com/S0002-9378(14)02443-0/sref1






























































































































http://www.AJOG.org


syndrome presenting as severe hydrops fetalis:a case report. Fetal Diagn Ther 2010;27:171-3(Level III).28. Hojo S, Tsukimori K, Kitade S, et al. Prenatalsonographic findings and hematological abnor-malities in fetuses with transient abnormal mye-lopoiesis with Down syndrome. Prenat Diagn2007;27:507-11 (Level II-2).29. Liao C, Wei J, Li Q, Li J, Li L, Li D.Nonimmune hydrops fetalis diagnosed duringthe second half of pregnancy in SouthernChina. Fetal Diagn Ther 2007;22:302-5 (LevelII-2).30. Suwanrath-Kengpol C, Kor-anantakul O,Suntharasaj T, Leetanaporn R. Etiology andoutcome of non-immune hydrops fetalis insouthern Thailand. Gynecol Obstet Invest2005;59:134-7 (Level II-2).31. Origa R, Moi P, Galanello R, Cao A. Alpha-Thalassemia. Nov. 1, 2005 (updated Nov 21,2013). In: Pagon RA, Adam MP, Ardinger H,et al, eds. Seattle,WA: University ofWashington,Seattle; 1993-2014. (Level III).32. Giacoia GP. Severe fetomaternal hemor-rhage: a review. Obstet Gynecol Surv 1997;52:372-80 (Level III).33. Ahmed M, Abdullatif M. Fetomaternaltransfusion as a cause of severe fetal anemiacausing early neonatal death: a case report.Oman Med J 2011;26:444-6 (Level III).34. Wylie BJ, D’Alton ME. Fetomaternal hem-orrhage. Obstet Gynecol 2010;115:1039-51(Level III).35. Rubod C, Houfflin V, Belot F, et al. Suc-cessful in utero treatment of chronic andmassive fetomaternal hemorrhage with fetalhydrops. Fetal Diagn Ther 2006;21:410-3(Level III).36. Tannirandorn Y, Nicolini U, Nicolaidis P,Nasrat H, Letsky EA, Rodeck CH. Intrauterinedeath due to fetomaternal hemorrhage despitesuccessful treatment of fetal anemia. J PerinatMed 1990;18:233-5 (Level III).37. Masson P, Rigot A, Cécile W. Hydropsfetalis and G-6-PD deficiency. Arch Pediatr1995;2:541-4 (Level III).38. Ferreira P, Morais L, Costa R, et al. Hydropsfetalis associated with erythrocyte pyruvate ki-nase deficiency. Eur J Pediatr 2000;159:481-2(Level III).39. Goncé A, Marcos MA, Borrell A, et al.Maternal IgM antibody status in confirmed fetalcytomegalovirus infection detected by sono-graphic signs. Prenat Diagn 2012;32:817-21(Level II-3).40. Syridou G, Spanakis N, Konstantinidou A,et al. Detection of cytomegalovirus, parvovirusB19 and herpes simplex viruses in cases of in-trauterine fetal death: association with patho-logical findings. J Med Virol 2008;80:1776-82(Level II-2).41. Carles G, Lochet S, Youssef M, et al.Syphilis and pregnancy. J Gynecol Obstet BiolReprod (Paris) 2008;37:353-7 (Level II-2).42. Al-Buhtori M, Moore L, Benbow EW,Cooper RJ. Viral detection in hydrops fetalis,spontaneous abortion, and unexplained fetal


FLA 5.2.0 DTD

death in utero. J Med Virol 2011;83:679-84(Level II-2).43. Tanimura K, Kojima N, Yamazaki T,Semba S, Yokozaki H, Yamada H. Secondtrimester fetal death caused by varicella-zostervirus infection. J Med Virol 2013;85:935-8(Level III).44. Dubois-Lebbe C, Houfflin-Debarge V,Dewilde A, Devisme L, Subtil D. Nonimmunehydrops fetalis due to herpes simplex virus type1. Prenat Diagn 2007;27:188-9 (Level III).45. Bachmaier N, Fusch C, Stenger RD,Grabow D, Mentel R, Warzok R. Nonimmunehydrops fetalis due to enterovirus infection. Eur JObstet Gynecol Reprod Biol 2009;142:83-4(Level III).46. Oyer CE, Ongcapin EH, Ni J, Bowles NE,Towbin JA. Fatal intrauterine adenoviral endo-myocarditis with aortic and pulmonary valvestenosis: diagnosis by polymerase chain reac-tion. Hum Pathol 2000;31:1433-5 (Level III).47. Anderson JL, Levy PT, Leonard KB,Smyser CD, Tychsen L, Cole FS. Congenitallymphocytic choriomeningitis virus: when toconsider the diagnosis. J Child Neurol 2013;29:837-42 (Level III).48. Young N, Harrison M, Moore J, Mortimer P,Humphries RK. Direct demonstration of the hu-man parvovirus in erythroid progenitor cellsinfected in vitro. J Clin Invest 1984;74:2024-32(Level III).49. Chisaka H, Morita E, Yaegashi N,Sugamura K. Parvovirus B19 and the patho-genesis of anemia. Rev Med Virol 2003;13:347-59 (Level III).50. Enders M, Klingel K, Weidner A, et al. Riskof fetal hydrops and non-hydropic late intra-uterine fetal death after gestational parvovirusB19 infection. J Clin Virol 2010;49:163-8 (LevelII-2).51. Enders M, Weidner A, Zoellner I, Searle K,Enders G. Fetal morbidity and mortality afteracute human parvovirus B19 infection in preg-nancy: prospective evaluation of 1018 cases.Prenat Diagn 2004;24:513-8 (Level II-2).52. Yaegashi N, Niinuma T, Chisaka H, et al. Theincidence of, and factors leading to, parvovirusB19-related hydrops fetalis following maternalinfection; report of 10 cases and meta-analysis.J Infect 1998;37:28-35 (Meta-analysis).53. Lamont RF, Sobel JD, Vaisbuch E, et al.Parvovirus B19 infection in human pregnancy.BJOG 2011;118:175-86 (Level III).54. Centers for Disease Control (CDC). Risksassociatedwith humanparvovirusB19 infection.MMWR Morb Mortal Wkly Rep 1989;38:81-8,93-7. (Level III).55. Miller E, Fairley CK, Cohen BJ, Seng C.Immediate and long term outcome of hu-man parvovirus B19 infection in pregnancy.Br J Obstet Gynaecol 1998;105:174-8(Level II-2).56. Dijkmans AC, de Jong EP, Dijkmans BA,et al. Parvovirus B19 in pregnancy: prenataldiagnosis and management of fetal complica-tions. Curr Opin Obstet Gynecol 2012;24:95-101 (Level III).

FEBRUARY 2015


57. Cavoretto P, Molina F, Poggi S,Davenport M, Nicolaides KH. Prenatal diagnosisand outcome of echogenic fetal lung lesions.Ultrasound Obstet Gynecol 2008;32:769-83(Level II-2).58. Wilson RD, Baxter JK, Johnson MP, et al.Thoracoamniotic shunts: fetal treatment ofpleural effusions and congenital cystic adeno-matoid malformations. Fetal Diagn Ther2004;19:413-20 (Level II-2).59. Schrey S, Kelly EN, Langer JC, et al. Fetalthoracoamniotic shunting for large macrocysticcongenital cystic adenomatoid malformations ofthe lung. Ultrasound Obstet Gynecol 2012;39:515-20 (Level II-2).60. Loh KC, Jelin E, Hirose S, Feldstein V,Goldstein R, Lee H. Microcystic congenital pul-monary airway malformation with hydrops feta-lis: steroids vs open fetal resection. J PediatrSurg 2012;47:36-9 (Level II-2).61. Yinon Y, Grisaru-Granovsky S, Chaddha V,et al. Perinatal outcome following fetal chestshunt insertion for pleural effusion. UltrasoundObstet Gynecol 2010;36:58-64 (Level II-2).62. Van Den Wijngaard JP, Ross MG, VanGemert MJ. Twin-twin transfusion syndromemodeling. Ann N Y Acad Sci 2007;1101:215-34(Level III).63. Society for Maternal-Fetal Medicine,Simpson LL. Twin-twin transfusion syndrome.Am J Obstet Gynecol 2013;208:3-18 (Level III).64. Lee H, Bebbington M, Crombleholme TM;North American Fetal Therapy Network.The North American Fetal Therapy NetworkRegistry data on outcomes of radiofrequencyablation for twin-reversed arterial perfusionsequence. Fetal Diagn Ther 2013;33:224-9(Level II-2).65. Overstreet K, Benirschke K, Scioscia A,Masliah E. Congenital nephrosis of the Finnishtype: overview of placental pathology and liter-ature review. Pediatr Dev Pathol 2002;5:179-83(Level III).66. Zenker M, Tralau T, Lennert T, et al.Congenital nephrosis, mesangial sclerosis, anddistinct eye abnormalities with microcoria: anautosomal recessive syndrome. Am J MedGenet A 2004;130A:138-45 (Level III).67. Noreen S, Heller DS, Faye-Petersen O.Mediastinal teratoma as a rare cause of hydropsfetalis and death: report of 3 cases. J ReprodMed 2008;53:708-10 (Level III).68. Isaacs H Jr. Fetal hydrops associated withtumors. Am J Perinatol 2008;25:43-68 (Level III).69. Van Mieghem T, Al-Ibrahim A, Deprest J,et al. Minimally invasive therapy for fetal sacro-coccygeal teratoma: case series and systematicreview of the literature. Ultrasound ObstetGynecol 2014;43:611-9 (Level III).70. Isaacs H. Perinatal (fetal and neonatal) tu-berous sclerosis: a review. Am J Perinatol2009;26:755-60 (Level II-2).71. McCoy MC, Katz VL, Gould N, Kuller JA.Non-immune hydrops after 20weeks’ gestation:review of 10 years’ experience with suggestionsfor management. Obstet Gynecol 1995;85:578-82 (Level III).

:59 pm � ce





































































































































































































http://www.AJOG.org

The practice of medicine continues toevolve, and individual circumstanceswill vary. This opinion reflects informa-tion available at the time of its submis-sion for publication and is neitherdesigned nor intended to establish anexclusive standard of perinatal care.This publication is not expected to reflectthe opinions of all members of theSociety for MaternaleFetal Medicine.


72. Vintzileos AM, Campbell WA,Weinbaum PJ, Nochimson DJ. Perinatal man-agement and outcome of fetal ventriculomegaly.Obstet Gynecol 1987;69:5-11 (Level III).73. Mathur BP, Karan S. Non-immune hydropsfetalis due to osteopetrosis congenita. IndianPediatr 1984;21:651-3 (Level III).74. Whybra C, Mengel E, Russo A, et al. Lyso-somal storage disorder in non-immunologicalhydrops fetalis (NIHF): more common thanassumed? Report of four cases with transientNIHF and a review of the literature. Orphanet JRare Dis 2012;7:86 (Level III).75. Staretz-Chacham O, Lang TC,LaMarca ME, Krasnewich D, Sidransky E.Lysosomal storage disorders in the newborn.Pediatrics 2009;123:1191-207 (Level III).76. Gimovsky AC, Luzi P, Berghella V. Lysomalstorage diseases as an etiology of non-immunehydrops: a systematic review. Am J ObstetGynecol 2014. Oct 8 [Epub ahead of print](Level I).77. Gort L, Granell MR, Fernández G, Carreto P,Sanchez A, Coll MJ. Fast protocol for the diag-nosis of lysosomal diseases in nonimmunehydrops fetalis. Prenat Diagn 2012;32:1139-42(Level III).78. Govaert P, Leroy JG, Pauwels R, et al.Perinatal manifestations of maternal yellow nailsyndrome. Pediatrics 1992;89:1016-8 (Level III).79. Mari G, Deter RL, Carpenter RL, et al.Noninvasive diagnosis by Doppler ultrasonog-raphy of fetal anemia due to maternal red-cellalloimmunization; Collaborative Group forDoppler Assessment of the Blood Velocity inAnemic Fetuses. N Engl J Med 2000;342:9-14(Level II-2).80. Désilets V, Audibert F; Society of Obste-trician and Gynecologists of Canada. Investiga-tion and management of non-immune fetalhydrops. J Obstet Gynaecol Can 2013;35:923-38 (Level III).

FLA 5.2.0 DTD

81. Braun T, Brauer M, Fuchs I, et al. Mirrorsyndrome: a systematic review of fetal associ-ated conditions, maternal presentation andperinatal outcome. Fetal Diagn Ther 2010;27:191-203 (Level III).82. Gedikbasi A, Oztarhan K, Gunenc Z, et al.Preeclampsia due to fetal non-immune hydrops:mirror syndrome and review of literature. Hyper-tens Pregnancy 2011;30:322-30 (Level III).83. Goa S, Mimura K, Kakigano A, et al.Normalization of angiogenic imbalance afterintra-uterine transfusion for mirror syndromecaused by parvovirus B19. Fetal Diagn Ther2013;34:176-9 (Level III).84. Llurba E, Marsal G, Sanchez O, et al.Angiogenic and antiangiogenic factors beforeand after resolution of maternal mirror syn-drome. Ultrasound Obstet Gynecol 2012;40:367-9 (Level III).85. Livingston JC, Malik KM,Crombleholme TM, Lim FY, Sibai BM. Mirrorsyndrome: a novel approach to therapywith fetalperitoneal-amniotic shunt. Obstet Gynecol2007;110:540-3 (Level III).86. Midgley DY, Hardrug K. The mirror syn-drome. Eur J Obstet Gynecol Reprod Biol2000;88:201-2 (Level III).87. Mascaretti RS, Falcão MC, Silva AM,Vaz FA, Leone CR. Characterization of new-bornswith non-immune hydrops fetalis admittedto a neonatal intensive care unit. Rev Hosp ClinFac Med Sao Paulo 2003;58:125-32 (Level III).88. American College of Obstetricians and Gy-necologists. Management of preterm labor.ACOGPractice bulletin no. 127. Obstet Gynecol2012;119:1308-17 (Level III).89. Sandlin AT, Chauhan SP, Magann EF.Clinical relevance of sonographically estimatedamniotic fluid volume: polyhydramnios.J Ultrasound Med 2013;32:851-63 (Level III).90. Huang HR, Tsay PK, Chiang MC, Lien R,Chou YH. Prognostic factors and clinical

FEBRUARY 2015 Am


features in liveborn neonates with hydropsfetalis. Am J Perinatol 2007;24:33-8 (Level II-2).91. Czernik C, Proquitté H, Metze B, Bührer C.Hydrops fetalisehas there been a change indiagnostic spectrum and mortality? J MaternFetal Neonatal Med 2011;24:258-63 (Level II-2).92. Bonvicini F, Puccetti C, Salfi NC, et al.Gestational and fetal outcomes in B19 maternalinfection: a problem of diagnosis. J Clin Micro-biol 2011;49:3514-8 (Level II-3).93. Nagel HT, de Haan TR, Vandenbussche FP,Oepkes D,Walther FJ. Long-term outcome afterfetal transfusion for hydrops associated withparvovirus B19 infection. Obstet Gynecol2007;109:42-7 (Level III).94. Hahurij ND, Blom NA, Lopriore E, et al.Perinatal management and long-term cardiacoutcome in fetal arrhythmia. Early Hum Dev2011;87:83-7 (Level III).95. American College of Obstetricians and Gy-necologists. Antepartum fetal surveillance.Practice bulletin no. 145. Obstet Gynecol2014;124:182-92 (Level III).96. Simpson JH, McDevitt H, Young D,Cameron AD. Severity of non-immune hydropsfetalis at birth continues to predict survivaldespite advances in perinatal care. Fetal DiagnTher 2006;21:380-2 (Level III).


















































































































http://www.AJOG.org

Documents

SMFM Clinical Guideline ajog - Sieog€¦ · (SMFM) Clinical Guideline #7: nonimmune hydrops fetalis Society for Maternal-Fetal Medicine (SMFM); Mary E. Norton, MD; Suneet P. Chauhan,