[Intrahepatic cholestasis of pregnancy: French College of Obstetricians and Gynecologists guidelines for clinical practice]. / La cholestase gravidique : recommandations pour la pratique clinique du Collège national des gynécologues obstétriciens français.

OBJECTIVE: To identify strategies for reducing neonatal and maternal morbidity associated with intrahepatic cholestasis pregnancy (ICP). MATERIAL AND METHODS: The quality of evidence of the literature was assessed following the GRADE methodology with questions formulated in the PICO format (Patients, Intervention, Comparison, Outcome) and outcomes defined a priori and classified according to their importance. An extensive bibliographic search was performed on PubMed, Cochrane, EMBASE and Google Scholar databases. The quality of the evidence was assessed (high, moderate, low, very low) and a (i) strong or (ii) weak recommendations or (iii) no recommendation were formulated. The recommendations were reviewed in two rounds with external reviewers (Delphi survey) to select the consensus recommendations. RESULTS: Of the 14 questions (from 12 PICO questions and one definition question outside the PICO format), there was agreement between the working group and the external reviewers on 14 (100%). The level of evidence of the literature was insufficient to provide a recommendation on two questions. ICP is defined by the occurrence of suggestive pruritus (palmoplantar, nocturnal) associated with a total bile acid level>10µmol/L or an alanine transaminase level above 2N after ruling out differential diagnoses. In the absence of suggestive symptoms of a differential diagnosis, it is recommended not to carry out additional biological or ultrasound tests. In women with CIP, ursodeoxycholic acid is recommended to reduce the intensity of maternal pruritus (Strong recommendation. Quality of the evidence moderate) and to decrease the level of total bile acids and alanine transaminases. (Strong recommendation. Quality of the evidence moderate). S-adenosyl-methionine, dexamethasone, guar gum or activated charcoal should not be used to reduce the intensity of maternal pruritus (Strong recommendation. Quality of evidence low), and there is insufficient data to recommend the use of antihistamines (No recommendation. Quality of evidence low). Rifampicin (Weak recommendation. Very low quality of evidence) or plasma exchange (Strong recommendation. Very low quality of evidence) should not be used to reduce maternal pruritus and perinatal morbidity. Serum monitoring of bile acids is recommended to reduce perinatal morbidity and mortality (stillbirth, prematurity) (Low recommendation. Quality of the evidence low). The level of evidence is insufficient to determine whether fetal heart rate or fetal ultrasound monitoring are useful to reduce perinatal morbidity (No recommendation). Birth is recommended when bile acid level is above 99µmol/L from 36 weeks gestation to reduce perinatal morbidity, in particular stillbirth. When bile acid level is above 99µmol/L is below 100µmol/L, women should be informed that induction of labor could be considered 37 and 39 weeks gestation to reduce perinatal morbidity. (Strong recommendation. Quality of evidence low). In postpartum, total bile acids and alanine transaminases level should be checked and normalized before prescribing estrogen-progestin contraception, ideally with a low estrogen dose (risk of recurrence of pruritus and cytolysis) (Low recommendation. Quality of evidence very low). CONCLUSION: Although the quality of evidence regarding ICP gestational cholestasis remains low, there is a strong consensus in France, as shown by our Delphi study, on how to manage women with ICP. The reference first-line treatment is ursodeoxycholic acid.

Diagnosis of medulloblastoma in pregnant women: To continue with the pregnancy or not? A case report and literature review.

BACKGROUND: Medulloblastoma (MB) is an uncommon and challenging diagnosis in pregnant women, and especially in pregnancy after in-vitro fertilization (IVF). Clinical features are easily misinterpreted and mistaken for other more common gestation-related pathologies. We report the case of a 34-year-old patient with clinical symptoms of intracranial hypertension. MB was diagnosed and operated on during the pregnancy. OBJECTIVE: To conduct a systematic literature review of other cases of MB operated on during pregnancy, and discuss the clinical and surgical management of MB in pregnancy. METHOD: We conducted a systematic literature review according to PRISMA guidelines. RESULTS: In addition to the present case, 9 cases of MB were reported as operated on during viable pregnancy. In one case, medical abortion was decided on before surgical debulking. Pregnancy term was between 8 and 30 weeks. The most common symptoms were headache, nausea and vomiting followed by dizziness. Tumor prognosis after treatment was favorable in 6 cases out of 10 and unfavorable in 4, with 3 cases of recurrence and 3 of death. CONCLUSION: We report the first case of long-term survival after MB in a woman pregnant via IVF. In standard-risk MB, it is possible to carry the pregnancy to term. Vaginal delivery is not contraindicated a priori. Early diagnosis, close clinical and radiological surveillance and surgery are the key factors for better prognosis. Multidisciplinary collaboration is crucial to determine the best timing and treatment.

Validation of the 'EPICES' social deprivation score in a population of women who have just given birth: a French cross-sectional study.

OBJECTIVES: To assess the diagnostic performance of the EPICES score for identifying social deprivation during pregnancy in a population of women in the immediate postpartum period. STUDY DESIGN: This cross-sectional survey took place between 5th June and 5th August 2017, among women who had just given birth in either of the maternity units in Clermont-Ferrand, France. METHODS: A self-administered questionnaire was completed by women. The questionnaire came in two parts: the EPICES index and the criteria for social deprivation defined by French law. These criteria were chosen to define the reference standard. The women were classified into two groups, living in precarious circumstances or not, according to the criteria defined by the French law (reference standard). To determine the most relevant threshold of the EPICES score, the precision associated with the threshold (the fraction of those predicted positive who are true positives: positive predictive value) was balanced with its sensitivity. EPICES scores above the threshold were classified as deprived, those below as non-deprived. RESULTS: Of the 947 women who gave birth during the study period, 700 (73.9%) completed the self-administered questionnaire. The best trade-off between precision and sensitivity was obtained with a threshold of 22. For this threshold value, the positive predictive value was 42.3% and the sensitivity 70.3%. CONCLUSIONS: The EPICES score with a threshold validated in the population of pregnant women is a useful, rapid, and easy-to-use tool that makes it possible to identify maternal deprivation at an individual level.

[Breech Presentation: CNGOF Guidelines for Clinical Practice - Short Text]. / Présentation du siège. Recommandations pour la pratique clinique du CNGOF ­ Texte court.

OBJECTIVES: To determine the optimal management of singleton breech presentation. MATERIALS AND METHODS: The PubMed database, the Cochrane Library and the recommendations from the French and foreign obstetrical societies or colleges have been consulted. RESULTS: In France, 5% of women have breech deliveries (Level of Evidence [LE3]). One third of them have a planned vaginal delivery (LE3) of whom 70% deliver vaginally (LE3). External cephalic version (ECV) is associated with a reduced rate of breech presentation at birth (LE2), and with a lower rate of cesarean section (LE3) without increases in severe maternal (LE3) and perinatal morbidity (LE3). It is therefore recommended to inform women with a breech presentation at term that ECV could be attempted from 36 weeks of gestation (Professional consensus). In case of breech presentation, planned vaginal compared with planned cesarean delivery might be associated with an increased risk of composite perinatal mortality or serious neonatal morbidity (LE2). No difference has been found between planned vaginal and planned cesarean delivery for neurodevelopmental outcomes at two years (LE2), cognitive and psychomotor outcomes between 5 and 8 years (LE3), and adult intellectual performances (LE4). Short and long term maternal complications appear similar in case of planned vaginal compared with planned cesarean delivery in the absence of subsequent pregnancies. A previous cesarean delivery results for subsequent pregnancies in higher risks of uterine rupture, placenta accreta spectrum and hysterectomy (LE2). It is recommended to offer women who wish a planned vaginal delivery a pelvimetry at term (Grade C) and to check the absence of hyperextension of the fetal head by ultrasonography (Professional consensus) to plan their mode of delivery. Complete breech presentation, previous cesarean, nulliparity, term prelabor rupture of membranes do not contraindicate planned vaginal delivery (Professionnal consensus). Term breech presentation is not a contraindication to labor induction when the criteria for acceptance of vaginal delivery are met (Grade C). CONCLUSION: In case of breech presentation at term, the risks of severe morbidity for the child and the mother are low after both planned vaginal and planned cesarean delivery. For the French College of Obstetricians and Gynecologists (CNGOF), planned vaginal delivery is a reasonable option in most cases (Professional consensus). The choice of the planned route of delivery should be shared by the woman and her caregiver, respecting the right to woman's autonomy.

[Breech Presentation: CNGOF Guidelines for Clinical Practice - Information and Management]. / Présentation du siège. Recommandations pour la pratique clinique du CNGOF ­ Information et organisation.

OBJECTIVE: To provide guidelines regarding parents' information in case of breech presentation and labour ward organisation for trial of labour in breech presentation. METHODS: Medline and Cochrane Library databases search and review of the main foreign guidelines. RESULTS: Information should be in favour of external cephalic version and describe benefits and risks for planned vaginal delivery vs planned caesarean delivery (Professional consensus). Patient should be aware of choice change according to obstetrical context (Professional consensus). Presence of an obstetrician is required at birth as well as immediate assistance of an anaesthesiologist and paediatrician if needed (Professional consensus). CONCLUSION: Information should lead to concerted choice concerning mode of delivery. Labour ward organisation requires presence of an obstetrician at birth and immediate availability of anaesthesiologist and paediatrician.

A randomised controlled trial of amnioexchange for fetal gastroschisis.

OBJECTIVE: Morbidity in fetuses affected by gastroschisis is mainly the result of bowel ischaemic and inflammatory processes. Experimental studies on animal models show that clearing amniotic fluid from the digestive secretions by amnioexchange procedures reduces the inflammatory process. We evaluated the benefit of the amnioexchange procedure for fetal gastroschisis in humans. DESIGN: Prospective, interventional, randomised study. SETTING: Eight referral centres for fetal medicine. POPULATION: Pregnant women carrying a fetus with gastroschisis. METHODS: We compared, in utero, amnioexchange with a sham procedure. The protocol included, in both arms, steroid injections at 30 weeks of gestation and the use of postnatal minimal enteral feeding. MAIN OUTCOME MEASURES: The primary outcome was a composite variable based on the duration of ventilation and parenteral nutrition. Secondary outcomes were the effectiveness and safety of the amnioexchange procedure, including the rate of perinatal death, time to full enteral feeding, primary closure, and late feeding disorders. RESULTS: Sixty-four patients were randomised. There was no difference in the composite criteria between the amnioexchange and control groups. Based on an intention-to-treat analysis, there were no significant between-group differences in pregnancy outcome or complications. When studying the relationship between digestive compounds and amniotic fluid inflammatory markers, a clear correlation was found between bile acid and both ferritin and interleukin 1ß (IL1ß). CONCLUSIONS: In humans, amnioexchange, as described in our protocol, is not an option for fetal care; however, we provide supplementary proof of the involvement of inflammation in the pathogenicity of gastroschisis and suggest that future research should aim at reducing inflammation. ClinicalTrials.gov: NCT00127946. TWEETABLE ABSTRACT: A prospective, interventional, randomised study shows no benefit of amnioexchange for fetal gastroschisis in humans.

[Correlation between automated optical reading and double human reading for two tests of fetal membranes rupture]. / Concordance entre la lecture optique automatisée et la lecture visuelle double pour deux tests de rupture des membranes.

OBJECTIVES: To assess concordance between double human and automated optical reading (AOR) concerning two biological tests for rupture of membranes (ROM) METHODS: We conducted a monocentric, prospective, observational study comparing Actim Prom® (Alere SAS, Jouy-en Josas, France) and Hiprom Duo® (Fumouze, Levallois-Perret, France). Each test was performed simultaneously in patients with suspected ROM and read independently by 2 biologists and AOR device. ROM was clinically confirmed in case of recurrent leakage or spontaneous labour with no perceived membranes within 48hours. RESULTS: Concerning Actim Prom®, concordance was 100 %, 92.5 % and 91.6 % between biologists, biologists-AOR device and biologists or AOR vs. clinical presentation respectively. Concerning Hiprom Duo®, concordance was 97.2 % between biologists, 97.2 % between biologist 1 and AOR, 95.3 % between biologist 2 and AOR, 63.5 % between clinical presentation and human reading, 62.3 % between clinical presentation and AOR. False positive cases were significantly associated with modified cervix (21 % vs. 46 %, P=0.006). CONCLUSION: We demonstrated excellent correlation between biologists and good or excellent correlation between AOR and human reading supporting the use of AOR in clinical practice.

[Preterm premature rupture of membranes: CNGOF Guidelines for clinical practice - Short version]. / Rupture prématurée des membranes avant terme : recommandations pour la pratique clinique du CNGOF ­ Texte court.

OBJECTIVE: To determine management of women with preterm premature rupture of membranes (PPROM). METHODS: Bibliographic search from the Medline and Cochrane Library databases and review of international clinical practice guidelines. RESULTS: In France, PPROM rate is 2 to 3% before 37 weeks of gestation (level of evidence [LE] 2) and less than 1% before 34 weeks of gestation (LE2). Prematurity and intra-uterine infection are the two major complications of PPROM (LE2). Compared to other causes of prematurity, PPROM is not associated with an increased risk of neonatal mortality and morbidity, except in case of intra-uterine infection, which is associated with an augmentation of early-onset neonatal sepsis (LE2) and of necrotizing enterocolitis (LE2). PPROM diagnosis is mainly clinical (professional consensus). In doubtful cases, detection of IGFBP-1 or PAMG-1 is recommended (professional consensus). Hospitalization of women with PPROM is recommended (professional consensus). There is no sufficient evidence to recommend or not recommend tocolysis (grade C). If a tocolysis should be prescribed, it should not last more than 48hours (grade C). Antenatal corticosteroids before 34 weeks of gestation (grade A) and magnesium sulfate before 32 weeks of gestation (grade A) are recommended. Antibiotic prophylaxis is recommended (grade A) because it is associated with a reduction of neonatal mortality and morbidity (LE1). Amoxicillin, 3rd generation cephalosporins, and erythromycin in monotherapy or the association erythromycin-amoxicillin can be used (professional consensus), for 7 days (grade C). However, in case of negative vaginal culture, early cessation of antibiotic prophylaxis might be acceptable (professional consensus). Co-amoxiclav, aminosides, glycopetides, first and second generation cephalosporins, clindamycin, and metronidazole are not recommended for antibiotic prophylaxis (professional consensus). Outpatient management of women with clinically stable PPROM after 48hours of hospitalization is a possible (professional consensus). During monitoring, it is recommended to identify the clinical and biological elements suggesting intra-uterine infection (professional consensus). However, it not possible to make recommendation regarding the frequency of this monitoring. In case of isolated elevated C-reactive protein, leukocytosis, or positive vaginal culture in an asymptomatic patient, it is not recommended to systematically prescribe antibiotics (professional consensus). In case of intra-uterine infection, it is recommended to immediately administer an antibiotic therapy associating beta-lactamine and aminoside (grade B), intravenously (grade B), and to deliver the baby (grade A). Cesarean delivery should be performed according to the usual obstetrical indications (professional consensus). Expectative management is recommended before 37 weeks of gestation in case of uncomplicated PPROM (grade A), even in case of positive vaginal culture for B Streptococcus, provided that an antibiotic prophylaxis has been prescribed (professional consensus). Oxytocin and prostaglandins are two possible options to induce labor in case of PPROM (professional consensus). CONCLUSION: Expectative management is recommended before 37 weeks of gestation in case of uncomplicated PPROM (grade A).

[Diagnosis of rupture of fetal membranes: CNGOF Preterm Premature Rupture of Membranes Guidelines]. / Diagnostic de la rupture des membranes. RPC Rupture prématurée des membranes avant terme CNGOF.

OBJECTIVE: To describe clinical and paraclinical tests diagnosing rupture of fetal membranes (ROM). METHODS: Bibliographic search over the period 1980-2017 considering articles in French and English as well as guidelines from national obstetrical societies. RESULTS: Typical amniotic fluid leakage occurs in ¾ of cases. In this situation, no additional test is required (Professional consensus). For ambiguous cases, a speculum examination can demonstrate pooling of amniotic fluid but suspicion can persist in 50% of cases (evidence level IV). In this context, we recommend to consider performing an IGFBP-1 or PAMG-1 test of vaginal fluid (evidence level III). Ability of these tests to reduce maternal or neonatal morbidity has never been demonstrated (Professional consensus). An isolated positive test should be considered cautiously as false positive does exist (Professional consensus). CONCLUSION: Symptoms suggestive of ROM and speculum examination demonstrating pooling of amniotic fluid are sufficient to diagnose ROM. If pooling is not observed, we recommend to consider performing an IGFBP-1 or PAMG-1 test of vaginal fluid.

Are there risk factors for false-positive malformation diagnoses on obstetric ultrasound? A nested case-control study.

INTRODUCTION: In a population-based study, we found an overall false-positive rate of 8.8% for the second and third trimester ultrasounds. Although numerous studies have been performed to examine factors which lead to false negatives, the same is not true for the factors associated with false positives. The principal objective of this study was to look for risk factors for false-positive diagnoses of fetal malformations on obstetric ultrasound scans. MATERIAL AND METHODS: In this nested case-control study, the case infants were those whose mother had a false-positive antenatal ultrasound diagnosis of a malformation during the second or third trimester (ultrasound false-positives) and who were live - or stillborn in Auvergne in 2006-2010. The control group comprised all children who were ultrasound true-negatives in 2005 and 2007. The study included 46 cases and 184 controls, matched according to the level of the maternity unit in which they were born. RESULTS: Most false-positive diagnoses were minor malformations. The mean term at this false-positive diagnosis was 27.7±5.4 weeks; in 46.8% of cases, the diagnosis was made during the second-trimester ultrasound. A single malformation was suspected in 95.7% of the cases. In 97.9% of cases, only one anatomical system was affected. In all, 49 malformations were identified among the 46 cases and their distribution differed according to anatomical system. The only risk factor identified was a body mass index (BMI)<25 (ORa=1.7; 95%CI: 1.2-2.4). DISCUSSION: A maternal BMI<25 was the only risk factor identified for a false-positive ultrasound diagnosis of a fetal malformation.

Hemoperitoneum in peripartum: A case-series.

Hemorrhages are the first cause of perinatal deaths in French women. Thirteen percent of these deaths are not linked to obstetrical problems but rather to hemoperitoneum. These incidents are under-diagnosed and as a result, treatment is delayed and fetal and maternal mortality increases. We report three cases of patients, all White female in their last trimester of a non-problematic pregnancy presenting with hemoperitoneum and resulting in different outcomes. The analysis of published materials and of our cases leads us to infer that a diagnosis of hemoperitoneum must be considered in pregnant women when abdominal pain, symptoms of shock and a decrease in hemoglobin are associated. An immediate response and intensive care followed by hemostatic surgery give these patients the best chance to survive.

Accuracy of a rapid intrapartum group B Streptococcus test: A new immunochromatographic assay.

OBJECTIVES: To decrease the incidence of early-onset group B streptococcal (GBS) disease, a culture-based screening of all pregnant women at 35-37 weeks is recommended. This gold standard test requires 24-72hours culture. This delay precludes its use for intrapartum screening. This study assesses a new immunoassay, the DIMA test, for identifying GBS-positive patients in the labor ward. MATERIALS AND METHODS: This was a prospective observational study of 195 pregnant women presenting with full-term labor at a single site in France between June and August 2012. We assessed the diagnostic accuracy of intrapartum DIMA testing as compared to intrapartum GBS culture and prenatal screening at 35-38 weeks. RESULTS: The DIMA test sensitivity and specificity were 57.1% and 83.2%, respectively, as compared to 42.9% and 97% for prenatal culture screening. CONCLUSION: The DIMA test assay is a rapid and inexpensive test for the detection of maternal GBS colonization in the labor ward. Its sensitivity is higher than antepartum culture but its specificity is lower. Its performance was inferior to that reported for rapid polymerase chain reaction assays.

Can we perform a prenatal diagnosis of vasa previa to improve its obstetrical and neonatal outcomes?

INTRODUCTION: Vasa previa (VP) is defined as a condition in which the fetal blood vessels, unsupported by the placenta or the umbilical cord, run through the membranes of the lower uterine segment. It is associated with a high risk of stillbirth by exsanguination. This study aimed to assess the clinical context of diagnosis of VP in order to elaborate a strategy for its prenatal diagnosis and to improve its obstetrical and neonatal outcomes. MATERIAL AND METHODS: This historical cohort study covered the period from January 1, 2011 to December 31, 2015. All women who gave birth at our obstetrics and gynecology department (level 3 university hospital) and who had a VP were included. RESULTS: Eight cases of VP among 18,152 deliveries were observed (0.04%). Transvaginal sonography (TVS) with color Doppler allowed a prenatal diagnosis of VP in all cases. The mean gestational age at diagnosis was 26 weeks. Placental abnormalities were noted in 7 cases (87.5%) as bipartita or low-lying placenta. In one case (12.5%), the placenta appeared normal while umbilical cord insertion was velamentous. In 2 cases (25%), concomitant placental and cord abnormalities were objectified. The mean gestational age at delivery was 37±2.1 weeks. Seven deliveries (87.5%) had been by caesarean sections, except one, which occurred by vaginal route at 33 weeks of gestation (twin pregnancy). No case of perinatal death was observed. DISCUSSION: Prenatal diagnosis of VP during screening ultrasounds appears easy to perform and can improve obstetrical and neonatal outcomes. For this purpose, TVS with color and pulsed Doppler remains essential, particularly when an anomaly of the umbilical cord insertion and/or placental location is diagnosed.

[Premature rupture of membranes≥37 weeks of gestation: Predictive factors for labour onset within 24hours]. / Prédiction de l'entrée en travail sous 24heures en cas de rupture des membranes à terme.

OBJECTIVES: To identify predictive criteria for a positive expectation in the context of rupture of membranes after 37 WG. METHODS: Single-center retrospective study including ROM≥37 WG. The primary outcome was labour onset within 24hours. We compared predictive factors for occurrence of spontaneous labour and described obstetrical and neonatal outcomes according to initial Bishop score<6 or ≥6. RESULTS: From January 2013 to December 2014, 520 patients were included. The predictive factors in case of unfavorable cervix were clinical leakage (P<0.001) and a cervical dilatation≥2cm (P<0.001) according to multivariate analysis. When the expectancy failed, there was a higher rate of cesarean section (24.3% vs. 9.6% P<0.001) but no more proven maternal-fetal infection. In case of Bishop≥6, we identified no predictive factor for labour onset but Apgar<7 at 5minutes (18.7% vs. 3.2% P=0.01) and admission to neonatal unit (18.8% vs. 3.2% P=0.04) were more frequent without majoration of maternal-fetal infection. CONCLUSION: The favorable expectation was the outcome for 70.8% of ROM at term. Clinical leakage and dilated cervix appeared as the main predictors in case of Bishop<6. Majoration of low Apgar score and admission to neonatal unit could be increased when no labour onset occurred despite Bishop≥6.

Pregnancy and primary Sjögren's syndrome: management and outcomes in a multicentre retrospective study of 54 pregnancies.

OBJECTIVES: Primary Sjögren's syndrome (pSS) is one of the most common autoimmune diseases, mainly affecting women during the fourth decade of life. During pregnancy, the presence of anti-Ro/SSa and anti-La/SSb antibodies increases the risk of congenital heart block (CHB). Foetal and pregnancy outcomes in pregnant women with pSS compared with the general population are difficult to evaluate because of confounding factors including age and body mass index (BMI). METHOD: The aim of this case-control study was to analyse the impact of pSS in pregnant women on foetal and pregnancy outcomes. RESULTS: We enrolled 19 women with pSS (54 pregnancies) matched by age and BMI to 216 controls. Patients with pSS delivered significantly earlier (38 weeks + 3 days vs. 39 weeks + 2 days) and experienced more spontaneous abortions [< 22 weeks of gestation (WG)] than the controls [n = 16/54 (30.0%) vs. n = 1/216 (0.4%); p < 0.00001]. Preterm delivery (≤ 37+6 WG) was significantly higher in the pSS group than in the control group (29% vs. 12%, p = 0.04). pSS activity significantly affected the birthweight percentile, which was lower in pregnancies occurring after the diagnosis of pSS than in those occurring before (32.43 ± 21.57 vs. 60.46 ± 27.37; p = 0.008). No case of CHB was observed. CONCLUSIONS: pSS is responsible for an increased risk of spontaneous abortion. The duration of pregnancy is lower in patients with than without pSS, with more premature deliveries. Pregnancies that occur after the onset of the disease result in lower birthweight percentile children than when pSS is not clinically overt.

[Iodinated contrast in pregnant women and neonatal thyroid function]. / Injection de produit de contraste iodé en cours de grossesse et fonction thyroïdienne néonatale.

OBJECTIVES: There is a theoretical risk for neonatal hypothyroidism after prenatal exposure to iodinated contrast media. Current recommendations are in favour of neonatal thyroid function assessment. Our aim was to check if recommendations were observed, and if neonatal evaluation demonstrated anomalies. METHODS: Over the period from 01/01/2010 to 01/08/2015, maternal and newborn records were retrospectively reviewed. All pregnant women who underwent a computed tomography and their newborns were included. We collected thyroid-stimulating hormone (TSH), thyroxine (T4) and tri-iodothyronine (T3) levels. RESULTS: A total of 101 maternal and newborn records were reviewed. Mean gestational age at CT scan was 29.3±7.2 weeks. The mean dose of total iodine administered was 82.6±19.1mL. Only 21 newborns had a biological analysis (20.8%). All newborns had normal TSH and T4 levels at birth. Only 7 newborns had a T3 level above the upper threshold value, but according to expert opinion none have been considered pathological. CONCLUSION: Our study revealed that recommendations for neonatal thyroid function assessment after prenatal exposure to iodinated contrast media were not observed. This exposure seemed unlikely to have an important effect on thyroid function at birth.