ABSTRACT
BACKGROUND: Congenital diaphragmatic hernia (CDH) is a rare disease associated with major nutritional and digestive morbidities. Oral feeding autonomy remains a major issue for the care and management of these patients. The aim of this study was to specify the perinatal risk factors of delayed oral feeding autonomy in patients treated for CDH. METHODS: This monocentric cohort study included 138 patients with CDH. Eighty-four patients were analyzed after the exclusion of 54 patients (11 with delayed postnatal diagnosis, 5 with chromosomal anomaly, 9 with genetic syndrom, 13 with right-sided CDH, and 16 who died before discharge and before oral feeding autonomy was acquired). They were divided into two groups: oral feeding autonomy at initial hospital discharge (group 1, n = 51) and nutritional support at discharge (group 2, n = 33). Antenatal, postnatal, and perisurgical data were analyzed from birth until first hospital discharge. To remove biased or redundant factors related to CDH severity, statistical analysis was adjusted according to the need for a patch repair. RESULTS: After analysis and adjustment, delayed oral feeding autonomy was not related to observed/expected lung-to-head ratio (LHR o/e), intrathoracic liver and/or stomach position, or operative duration. After adjustment, prophylactic gastrostomy (OR adjusted: 16.3, IC 95%: 3.6-74.4) and surgical reoperation (OR adjusted: 5.1, IC 95% 1.1-23.7) remained significantly associated with delayed oral feeding autonomy. CONCLUSIONS: Delayed oral feeding autonomy occurred in more than one third of patients with CDH. Both prophylactic gastrostomy and surgical reoperation represent significant risk factors. Bowel obstruction might also impact oral feeding autonomy. Prophylactic gastrostomy seems to be a false "good idea" to prevent failure to thrive. This procedure should be indicated case per case. Bowel obstruction and all surgical reoperations represent decisive events that could impact oral feeding autonomy.
ABSTRACT
Congenital diaphragmatic hernia (CDH) is associated with pulmonary hypoplasia and respiratory morbidity. To assess whether respiratory morbidity during the first 2 years of life in infants with left-sided CDH is associated with fetal lung volume (FLV) evaluated by the observed-to-expected FLV ratio (o/e FLV) on antenatal magnetic resonance imaging (MRI). In this retrospective study, o/e FLV measures were collected. Respiratory morbidity in the first 2 years of life was studied according to two endpoints: treatment with inhaled corticosteroids for >3 consecutive months and hospitalization for any acute respiratory disease. The primary outcome was a favorable progression defined by the absence of either endpoint. Forty-seven patients were included. The median o/e FLV was 39% (interquartile range, 33-49). Sixteen (34%) infants were treated with inhaled corticosteroids and 13 (28%) were hospitalized. The most efficient threshold for a favorable outcome was an o/e FLV ≥ 44% with a sensitivity of 57%, specificity of 79%, negative predictive value of 56%, and positive predictive value of 80%. An o/e FLV ≥ 44% was associated with a favorable outcome in 80% of cases. These data suggest that lung volume measurement on fetal MRI may help to identify children at lower respiratory risk and improve information during pregnancy, patient characterization, decisions about treatment strategy and research, and personalized follow-up.
ABSTRACT
BACKGROUND: To define a threshold of maternal antibodies at risk of severe fetal anemia in patients followed for anti-RH1 alloimmunization (AI). STUDY, DESIGN, AND METHODS: We conducted a retrospective study of patients followed for anti-RH1 AI at the Lille University Hospital. The first group, severe anemia, included patients who received one or more in utero transfusions (IUT) or who were induced before 37 weeks of pregnancy for suspected severe fetal anemia. The second group, absence of severe anemia, corresponded to patients without intervention during pregnancy related to AI. Sensitivities, specificities, and positive and negative predictive values for screening for severe fetal anemia were calculated for the antibody thresholds of 3.5 and 5 IU/ml for the quantification. RESULTS: Between 2000 and 2018, 207 patients were included 135 in the severe anemia group and 72 in the no severe anemia group. No severe anemia was observed for an antibody titer below 16. For an antibody threshold of 3.5 IU/ml, the sensitivity was 98.2%, with 30.2% false positives. All severe anemias were detected in the second trimester; two cases of severe anemia were not detected in the third trimester. For an antibody threshold of 5 IU/ml, the sensitivity was lower at 95.6%, with five cases of severe anemia not detected. CONCLUSION: The antibody threshold of 3.5 IU/ml for the quantification and 16 for the titration allow targeting patients requiring close monitoring by an experienced team in case of anti-RH1 AI.
Subject(s)
Anemia, Hemolytic, Autoimmune , Fetal Diseases , Rh Isoimmunization , Pregnancy , Female , Humans , Retrospective Studies , Prenatal Care , Isoantibodies , Blood Transfusion, IntrauterineABSTRACT
Our objective is to determine perinatal factors contributing to the development of pulmonary hypertension (PH) in patients with isolated giant omphaloceles (GO). All cases of omphaloceles that underwent prenatal and postnatal care at the University Hospital of Lille between 1996 and 2021 were reviewed. We included all infants with isolated GO, including at least a part of the liver, who were treated by delayed surgical closure. Prenatal and postnatal data were recorded and correlated with postnatal morbidities. We compared outcomes between a group of infants with GO who developed PH and infants with GO with no PH. We identified 120 infants with omphalocele. Fifty isolated GO cases fulfilled the inclusion criteria of our study. The incidence of PH was 30%. We highlighted a prolonged inflammatory state, defined as a CRP superior to 15 mg/L, platelets higher than 500 G/L, and white blood cells higher than 15 G/l for more than 14 days in patients who developed PH. This event occurred in 73% of patients with PH versus 21% of patients without PH (p < 0.05). Late-onset infection was not different between the two groups. We speculate that prolonged inflammatory syndrome promotes PH in infants with GO treated with delayed surgical closure.
ABSTRACT
Congenital diaphragmatic hernia (CDH) is associated with abnormal pulmonary development, which is responsible for pulmonary hypoplasia with structural and functional abnormalities in pulmonary circulation, leading to the failure of the cardiorespiratory adaptation at birth. Despite improvement in treatment options and advances in neonatal care, mortality remains high, at close to 15 to 30%. Several risk factors of mortality and morbidities have been validated in fetuses with CDH. Antenatal assessment of lung volume is a reliable way to predict the severity of CDH. The two most commonly used measurements are the observed/expected lung to head ratio (LHRo/e) and the total pulmonary volume (TPV) on MRI. The estimation of total pulmonary volume (TPVo/e) by means of prenatal MRI remains the gold standard. In addition to LHR and TPV measurements, the position of the liver (up, in the thorax or down, in the abdomen) also plays a role in the prognostic evaluation. This prenatal prognostic evaluation can be used to select fetuses for antenatal surgery, consisting of fetoscopic endoluminal tracheal occlusion (FETO). The antenatal criteria of severe CDH with an ascended liver (LHRo/e or TPVo/e < 25%) are undoubtedly associated with a high risk of death or significant morbidity. However, despite the possibility of estimating the risk in antenatal care, it is difficult to determine what is in the child's best interest, as there still are many uncertainties: (1) uncertainty about individual short-term prognosis; (2) uncertainty about long-term prognosis; and (3) uncertainty about the subsequent quality of life, especially when it is known that, with a similar degree of disability, a child's quality of life varies from poor to good depending on multiple factors, including family support. Nevertheless, as the LHR decreases, the foreseeable "burden" becomes increasingly significant, and the expected benefit is increasingly unlikely. The legal and moral principle of the proportionality of medical procedures, as well as the prohibition of "unreasonable obstinacy" in all investigations or treatments undertaken, is necessary in these situations. However, the scientific and rational basis for assessing the long-term individual prognosis is limited to statistical data that do not adequately reflect individual risk. The risk of self-fulfilling prophecies should be kept in mind. The information given to parents must take this uncertainty into account when deciding on the treatment plan after birth.
ABSTRACT
Hydrolethalus syndrome (HLS) is a rare lethal fetal malformation disorder related to ciliogenesis disruption. This condition is more frequent in Finland where a founder missense variant in the HYLS1 gene was identified. No other HYLS1 variant has hitherto been implicated in HLS. We report two unrelated French fetuses presenting with a phenotype of HLS with brain abnormalities, limbs malformations with pre and postaxial hexadactyly and abnormal genitalia. These two fetuses have compound heterozygous variants in HYLS1. The first allele carries the same Finnish missense variant (NM_145014.2: c.632A > G, p.[Asp211Gly]) in both fetuses and the second allele carries a new missense variant (c.662G > C, p.[Arg221Pro]) in the first fetus, and a new nonsense variant (c.613C > T, p.[Arg205*]) in the second fetus. This is the first report of HYLS1 mutated cases outside Finland. Both cases presented here are consistent with HLS with additional malformations, allowing expansion of the phenotypic presentation previously described.
Subject(s)
Genetic Predisposition to Disease , Genetic Variation , Hand Deformities, Congenital/diagnosis , Hand Deformities, Congenital/genetics , Heart Defects, Congenital/diagnosis , Heart Defects, Congenital/genetics , Hydrocephalus/diagnosis , Hydrocephalus/genetics , Phenotype , Proteins/genetics , Alleles , Amino Acid Substitution , Autopsy , Comparative Genomic Hybridization , Female , Fetus , Genetic Association Studies , Genotype , Humans , Immunohistochemistry , Pedigree , Pregnancy , Ultrasonography, PrenatalABSTRACT
Resuscitation at birth of infants with Congenital Diaphragmatic Hernia (CDH) remains highly challenging because of severe failure of cardiorespiratory adaptation at birth. Usually, the umbilical cord is clamped immediately after birth. Delaying cord clamping while the resuscitation maneuvers are started may: (1) facilitate blood transfer from placenta to baby to augment circulatory blood volume; (2) avoid loss of venous return and decrease in left ventricle filling caused by immediate cord clamping; (3) prevent initial hypoxemia because of sustained uteroplacental gas exchange after birth when the cord is intact. The aim of this trial is to evaluate the efficacy of intact cord resuscitation compared to immediate cord clamping on cardiorespiratory adaptation at birth in infants with isolated CDH. The Congenital Hernia Intact Cord (CHIC) trial is a prospective multicenter open-label randomized controlled trial in two balanced parallel groups. Participants are randomized either immediate cord clamping (the cord will be clamped within the first 15 s after birth) or to intact cord resuscitation group (umbilical cord will be kept intact during the first part of the resuscitation). The primary end-point is the number of infants with APGAR score <4 at 1 min or <7 at 5 min. One hundred eighty participants are expected for this trial. To our knowledge, CHIC is the first study randomized controlled trial evaluating intact cord resuscitation on newborn infant with congenital diaphragmatic hernia. Better cardiorespiratory adaptation is expected when the resuscitation maneuvers are started while the cord is still connected to the placenta.
ABSTRACT
OBJECTIVES: The objective of this study was to assess whether the laterality of congenital diaphragmatic hernia (CDH) was a prognostic factor for neonatal survival. METHODS: This was a cohort study using the French national database of the Reference Center for Diaphragmatic Hernias. The principal endpoint was survival after hospitalization in intensive care. We made a comparative study between right CDH and left CDH by univariate and multivariate analysis. Terminations and stillbirths were excluded from analyses of neonatal outcomes. RESULTS: A total of 506 CDH were included with 67 (13%) right CDH and 439 left CDH (87%). Rate of survival was 49% for right CDH and 74% for left CDH (P < .01). Multivariate analysis showed two factors significantly associated with mortality: thoracic herniation of liver (OR 2.27; IC 95% [1.07-4.76]; P = .03) and lung-to-head-ratio over under expected (OR 2.99; IC 95% [1.41-6.36]; P < .01). Side of CDH was not significantly associated with mortality (OR 1.87; IC 95% [0.61-5.51], P = .26). CONCLUSION: Rate of right CDH mortality is more important than left CDH. Nevertheless after adjusting for lung-to-head-ratio and thoracic herniation of liver, right CDH does not have a higher risk of mortality than left CDH.
Subject(s)
Hernias, Diaphragmatic, Congenital/diagnosis , Hernias, Diaphragmatic, Congenital/pathology , Lung/pathology , Adult , Cohort Studies , Female , France/epidemiology , Hernias, Diaphragmatic, Congenital/mortality , Humans , Infant , Infant Mortality , Infant, Newborn , Lung/diagnostic imaging , Male , Pregnancy , Prenatal Diagnosis , Prognosis , Reproducibility of Results , Retrospective StudiesSubject(s)
Amniotic Band Syndrome/etiology , Polymicrogyria/etiology , Humans , Infant, Newborn , MaleABSTRACT
BACKGROUND: The antibody primarily responsible for fetal anemia may influence treatment and prognosis. The primary objective was to compare ante- and postnatal management and the outcomes of maternal red blood cell (RBC) alloimmunizations according to the antibody involved. The secondary objective was to compare anti-D alloimmunizations according to associated number of antibodies. STUDY DESIGN AND METHODS: A single-center study from 1999 to 2015 including maternal RBC alloimmunizations requiring intrauterine transfusion (IUT) was conducted. Patients were classified according to the antibody involved: anti-D, other Rh (anti-c and anti-E), and anti-K1. Obstetric data, IUT characteristics, and neonatal outcome were compared. A specific study on the anti-D, when isolated or associated, was then conducted. RESULTS: There were 106 pregnancies included, with 77.4% having anti-D, 9.4% having another anti-Rh (Rh group), and 13.2% having anti-K1. No significant difference between the anti-D and Rh groups was found for management and prognosis. The hemoglobin level in the first IUT was higher in the anti-D group than in the Kell group (6.8 vs. 4.7 g/dL, p = 0.008). Newborns in the anti-D group had significantly higher bilirubin levels and phototherapy duration than those in the Kell group. The mean estimated daily decrease in hemoglobin and that between the first two IUTs were lower with an isolated anti-D, compared with anti-D associated with two antibodies (p = 0.04). CONCLUSION: Anti-K1 alloimmunizations seem to cause more severe fetal anemia than anti-D alloimmunizations. Moreover, a decrease in hemoglobin appears to be more rapid when anti-D is associated with other antibodies.
Subject(s)
Blood Transfusion, Intrauterine , Erythrocytes/immunology , Kell Blood-Group System/immunology , Rh Isoimmunization , Rh-Hr Blood-Group System/immunology , Adult , Disease Management , Erythroblastosis, Fetal , Female , Humans , Pregnancy , Rho(D) Immune Globulin , Treatment Outcome , Young AdultABSTRACT
BACKGROUND: Starting resuscitation before clamping the umbilical cord at birth may progressively increase pulmonary blood flow while umbilical venous blood flow is still contributing to maintenance of oxygenation and left ventricle preload. OBJECTIVE: To evaluate the feasibility, safety, and effects of intact cord resuscitation (ICR) on cardiorespiratory adaptation at birth in newborn infants with CDH. STUDY DESIGN: Prospective, observational, single-center pilot study. METHODS: Physiologic variables and outcomes were collected prospectively in 40 consecutive newborn infants with an antenatal diagnosis of isolated CDH. RESULTS: Infants were managed with immediate cord clamping (ICC group) from 1/2012 to 5/2014 or the cord was clamped after initiation of resuscitation maneuvers (ICR group) from 6/2014 to 4/2016 (20 in each group). Ante- and postnatal markers of CDH severity were similar between groups. Resuscitation before cord clamping was possible for all infants in the ICR group. No increase in maternal or neonatal adverse events was observed during the period of ICR. The pH was higher and the plasma lactate concentration was significantly lower at one hour after birth in the ICR than in the ICC group (pH=7.17±0.1 vs 7.08±0.2; lactate=3.6±2.3 vs 6.6±4.3mmol/l, p<0.05). Mean blood pressure was significantly higher in the ICR than in the ICC group at H1 (52±7.7 vs 42±7.5mmHg), H6 (47±3.9 vs 40±5.6mmHg) and H12 (44±2.9 vs 39±3.3mmHg) (p<0.05). CONCLUSION: Commencing resuscitation and initiating ventilation while the infant is still attached to the placenta is feasible in infants with CDH. The procedure may support the cardiorespiratory transition at birth in infants with CDH.
Subject(s)
Delivery, Obstetric/methods , Hernias, Diaphragmatic, Congenital , Resuscitation/methods , Umbilical Cord , Adult , Apgar Score , Feasibility Studies , Female , Gestational Age , Humans , Infant, Newborn , Intubation, Intratracheal/methods , Male , Pilot Projects , Pregnancy , Prospective Studies , Safety , Time Factors , Young AdultABSTRACT
BACKGROUND: To best predict the recurrence of fetal anemia after intrauterine transfusion (IUT), the measurement of middle cerebral artery peak systolic velocity (PSV) and the estimation of hemoglobin (Hb) daily decrease are compared. STUDY DESIGN AND METHODS: A retrospective study including 38 patients who had at least two IUTs in a context of red blood cell alloimmunization was conducted. PSV values before first, second, and third IUTs were collected and expected Hb level was calculated according to various Hb daily decrease formulas as proposed in the literature. RESULTS: Comparison of PSV receiver operating characteristic curves with the various Hb levels did not find any significant difference between first and second IUTs. On the other hand, we found a significant difference between the second and third IUTs, with better prediction of fetal anemia through Hb decrease calculation, whatever the formula. Between the second and third IUTs, no formula was significantly better than the others. CONCLUSION: The timing of a second transfusion can be difficult to determine with certainty, but PSV can give an accurate assessment of when to resample the fetus with probably a higher recommended threshold for the diagnosis of fetal anemia. Subsequent to a second transfusion, the intertransfusion interval should be based on estimated Hb decrease rather than PSV thresholds, whatever the chosen formula proposed in the literature. Larger numbers are needed to definitely make this recommendation and it will be interesting to evaluate correlation between different antibodies.
Subject(s)
Fetal Diseases , Fetomaternal Transfusion , Middle Cerebral Artery/physiopathology , Adult , Blood Flow Velocity , Female , Fetal Diseases/blood , Fetal Diseases/diagnosis , Fetal Diseases/diagnostic imaging , Fetal Diseases/physiopathology , Fetomaternal Transfusion/diagnosis , Fetomaternal Transfusion/diagnostic imaging , Fetomaternal Transfusion/physiopathology , Humans , PregnancyABSTRACT
BACKGROUND AND OBJECTIVE: Several studies have shown the benefits of delayed cord clamping (DCC) in preterm and in healthy newborns at short and long term. Our objective was to evaluate the potentials benefits and risks of DCC in red cell alloimmunization. METHODS: This was a comparative before/after study of all living born neonates followed after fetal anemia requiring in utero transfusion. DCC was defined as cord clamping 30 seconds after birth. RESULTS: We included a continuous series of 72 neonates: 36 without DDC (group 1) and 36 with DDC (group 2). Hemoglobin at birth was lower in group 1 (10.2 vs 13.4 g/dL, P = .0003); 7 (25%) neonates in group 1 vs 24 (70.6%) in group 2 had no anemia at birth (P = .004). The rate of transfusion was similar between the 2 groups. Postnatal exchange transfusions were more likely performed in the group without DCC than in the group with DCC (47.2% vs 19.4%, P = .0124). Delay between birth and first transfusion was higher in group 2 (0 [0-13] vs 1 [0-21], P = .0274). The maximum level of bilirubin, the rate of intensive phototherapy, and the total duration of phototherapy were similar in the 2 groups. CONCLUSIONS: This study highlights a significant benefit of DCC in anemia secondary to red blood cell alloimmunization with a resulting decreased postnatal exchange transfusion needs, an improvement in the hemoglobin level at birth and longer delay between birth and first transfusion with no severe hyperbilirubinemia.
Subject(s)
Anemia, Hemolytic, Autoimmune/immunology , Anemia, Neonatal/immunology , Autoimmunity , Blood Transfusion/statistics & numerical data , Delivery, Obstetric/methods , Erythrocytes/immunology , Umbilical Cord , Anemia, Hemolytic, Autoimmune/blood , Anemia, Hemolytic, Autoimmune/therapy , Anemia, Neonatal/blood , Anemia, Neonatal/therapy , Constriction , Exchange Transfusion, Whole Blood/statistics & numerical data , Female , Follow-Up Studies , Humans , Infant, Newborn , Infant, Premature , Male , Prognosis , Time FactorsABSTRACT
Tianeptine, an atypical antidepressant, has been found to exhibit a potential for abuse. The use of therapeutic doses of tianeptine during pregnancy has never raised safety concerns. However, the impact of tianeptine abuse on the mother-child dyad has never been assessed. We report herein the case of a female patient who presented with dependence on tianeptine, with the use of >650 mg of the drug per day. She had 2 successive pregnancies with similar doses. The state of dependence remained unidentified throughout the first pregnancy, but just after delivery, her full-term newborn exhibited unexpected neonatal abstinence syndrome (NAS). The NAS was successfully treated with morphine, although both the mother's and newborn's urine drug screen was negative. The causality of tianeptine in inducing NAS was retrospectively assessed as "probable" by using a validated causality algorithm. During the second pregnancy, this patient sought addiction treatment and was admitted for residential detoxification treatment in her seventh month of pregnancy. Delivery occurred at full term with a low birth weight neonate. No further developmental insults or medical problems were subsequently identified in the 2 children. Maternal tianeptine dependence during pregnancy may induce a type of NAS that mimics opiate NAS. This finding appears to be consistent with a recent finding of the agonist action of tianeptine on the opiate µ-receptor.
Subject(s)
Analgesics, Opioid/therapeutic use , Antidepressive Agents, Tricyclic/adverse effects , Morphine/therapeutic use , Neonatal Abstinence Syndrome/drug therapy , Thiazepines/adverse effects , Adult , Female , Humans , Infant, Newborn , Male , Neonatal Abstinence Syndrome/etiology , Pregnancy , Pregnancy Complications/drug therapySubject(s)
Gaucher Disease/diagnostic imaging , Phenotype , Ultrasonography, Prenatal , Abortion, Eugenic , Adult , Consanguinity , Female , Fetal Death , Humans , Pregnancy , Young AdultABSTRACT
OBJECTIVE: To explore anxiety and pain felt by women undergoing chorionic villus sampling (CVS) and amniocentesis (AC). METHOD: We prospectively questioned 254 women (67 undergoing CVS, 187 AC) before the procedure on their anxiety, after the procedure on their pain felt, and the support they received or desired. The medical team collected technical information concerning each procedure. RESULTS: The level of anxiety was significantly higher in women undergoing CVS than AC, in those who had received complete information before the procedure, and when indication for the procedure was fetal structural abnormalities. The level of pain was significantly higher in cases of anxious women, those undergoing a CVS rather than AC, those who had undergone invasive prenatal diagnostic procedures in previous pregnancies, in procedures deemed difficult by the operator, and with needle insertion in the lateral part of the uterus. About 30% of women undergoing CVS and 8% of those undergoing AC would have desired some form of pain prevention, mostly with nonpharmacologic therapy. CONCLUSION: Prenatal diagnosis is frequently associated with anxiety and pain. We identified factors that could exacerbate either one. When questioned, patients would desire a nonpharmacologic means for pain prevention.
Subject(s)
Anxiety , Pain , Prenatal Diagnosis/psychology , Adult , Amniocentesis/psychology , Analgesia , Anxiety/epidemiology , Anxiety/prevention & control , Chorionic Villi Sampling/psychology , Female , Gestational Age , Humans , Middle Aged , Pain/epidemiology , Pain/prevention & control , Pregnancy , Prospective Studies , Surveys and QuestionnairesSubject(s)
Pregnancy Complications, Hematologic/physiopathology , Thrombocytopenia/complications , Upper Extremity Deformities, Congenital/complications , Adult , Congenital Bone Marrow Failure Syndromes , Female , Glucocorticoids/therapeutic use , Humans , Platelet Count , Pregnancy , Pregnancy Outcome , Radius/abnormalities , Radius/physiopathology , Thrombocytopenia/diagnosis , Thrombocytopenia/physiopathology , Upper Extremity Deformities, Congenital/diagnosis , Upper Extremity Deformities, Congenital/physiopathologyABSTRACT
INTRODUCTION: Fetal hydrops caused by anemia from parvovirus B19 infection (FH-B19) is rare. Doppler measurement of the middle cerebral artery peak systolic velocity (PSV-MCA) improves its prenatal diagnosis, but its frequency and prognosis are still poorly known. Despite improved survival due to in utero transfusions, the possibility of late neurological sequelae makes prognosis uncertain. OBJECTIVES: To assess the frequency, management and prognosis of a consecutive series of FH-B19 observed over a 15-year period. METHODS: Retrospective study of 27 cases of FH-B19, that is, 3/100,000 births, 24 of them discovered during routine second-trimester ultrasound. All but 1 case (96.2%) had at least four of the six ultrasound signs that Saltzman et al. [Obstet Gynecol 1989;74:106-111] suggested as indicators of anemia. Of the fetuses tested, 80% had a PSV-MCA >1.5 MoM, also indicative of anemia. Of the 19 fetuses treated by exchange transfusions, 11 were liveborn compared with 2 of the 6 not so treated (57.8 vs. 33.3%, NS). The survival rate was higher during the second half of the study period (23.1 vs. 71.4%, p < 0.02) for less severe anemia (p < 0.03) and for repeated transfusions (p = 0.03). In our series, 1 case of prenatal cerebral atrophy was identified on screening. All 13 liveborn children appeared healthy at the age of 1 year. CONCLUSION: In cases of fetal hydrops, Saltzman et al.'s ultrasound criteria and PSV-MCA measurement made it possible to determine the likelihood that anemia is the cause of the hydrops and to measure its intensity. Use of these techniques allowed us to choose the most appropriate treatment (transfusion or not, depending on the degree of anemia), and survival improved notably in our series.