[Evaluation of the concordance between guidelines and transfusion practices in neonatal intensive care units]. / Évaluation de la concordance entre recommandations et pratiques transfusionnelles en unités de soins intensifs néonataux.

OBJECTIVE: To assess the concordance rate between the prescriptions of blood products and Afssaps guidelines for transfusion practices in neonatology. STUDY DESIGN: Retrospective monocentric study. PATIENTS AND METHODS: Children transfused in the neonatology intensive care units in the Universitary Hospital of Rouen were included. Concordance rates between transfusion prescriptions, delivered and transfused products and the Afssaps guidelines were assessed. Any additional costs resulting from a theoretical discordance was also assessed. RESULTS: In 2006, 380 PSL were administered to 168 newborn children (NBC). Packed red blood cells (PRBC) represented the most often transfused products (n=290, 76%). Fifty packed platelets (PP) were transfused (13% of the blood products) and 41 fresh frozen plasmas (11%). Overall concordance rate between the Afssaps guidelines and the prescriptions was 64.9%. In 35.1% of cases, the prescription was excessive, compared to the recommendations. Excessive transfusion represented a total of 27,307 euros. CONCLUSION: Global concordance's rate between the guidelines and the prescriptions is fairly well. PRBC are the most blood product transfused. Excessive transfusions related to this concordance rate are leading to important theoretical costs. New informations to the guidelines are warranted to improve transfusional practices in this institution.

[Perinatal management and neurological outcome of newborns hospitalized with Rhesus hemolytic disease]. / Prise en charge périnatale et devenir neurologique à moyen terme des nouveau-nés hospitalisés pour maladie hémolytique par immunisation anti-D.

OBJECTIVES: To evaluate perinatal management and neurological outcome in a group of infants born with Rhesus fetomaternal allo-immunization. PATIENTS AND METHODS: Between 1 January and 31 December 2005, all newborns admitted to neonatal unit of Rouen tertiary centre for Rhesus hemolytic disease were included in a retrospective study and divided in two groups. The newborns who were treated with intrauterine transfusion are in the group 1 and those who needed only postnatal treatment in the group 2. In each case, were considered antenatal management (ultrasonographic data, middle cerebral artery peak systolic velocity, intrauterine transfusion), postnatal treatment (phototherapy, exchange transfusion, transfusion requirements) and neurological outcome. RESULTS: Among 42 cases of Rhesus allo-immunization observed in six years, 28 newborns (67%) were admitted for neonatal cares. No case of fetal hydrops was noted. But 16/28 (57%) were preterm with a median term of 35 weeks gestation (32-36 weeks). In group 1 of six infants who had received intrauterine transfusion (IUT), only one (17%) needed postnatal exchange transfusion, and all six received one to three blood transfusions after their birth. In group 2 of 22 infants who did not receive IUT, 6/22 (27%) needed postnatal exchange and 18/22 (82%) of them received one to four blood transfusions. Phototherapy duration and albumin requirements were similar in both groups. Three deaths occurred, one due to necrotizing enterocolitis and the other two later on due to sudden infant death and fulminant meningococcemia. Neurological outcome of the remaining 25 children was normal. DISCUSSION AND CONCLUSION: Rhesus alloimmunization remain a situation at risk. Neonatal clinical presentation is less severe than previously described due to improvement in antenatal management. Infants required less postnatal exchange transfusion when they received intrauterine transfusion but more frequent blood transfusions.

[Breastfeeding and Cytomegalovirus infection in preterm infants. How to reduce the risks?]. / Cytomégalovirus néonatal et allaitement maternel chez le nouveau-né prématuré. Quelles propositions?

In France, screening for cytomegalovirus infection (CMV) during pregnancy is not recommended in routine. The transmission of CMV through breastmilk from mothers to preterm infants is frequent (15-20%). The frequency of neuro-sensorial handicap related to congenital CMV infection in very preterm infants is not well documented. We report the case of a female infant born at 30 weeks of gestation. At 15 days, she developed cholestatic jaundice. Urine cultures were positive for CMV. Diagnostic procedure showed no other cause for jaundice. At 40 days, the infant presented with hepato-splenomegaly, purpura and abnormal skin color related to a symptomatic, secondary CMV infection, probably transmitted through breastmilk. Ganciclovir was begun for 21 days. At 12 months, she presents with normal development. This observation raises questions about breastfeeding in very preterm infants. Unexplained prematurity could reflect recent infection or reactivation in the mother. Thus, because of the well-known risks of prematurity on one hand, and CMV infection on the other, we suggest that detection of CMV seropositive mothers should be considered before allowing breastfeeding. If the mother has serologic evidence of recent infection or reactivation, freezing breastmilk at -20 degrees C for 3 days may be an option in order to reduce virolactia, especially during early lactation. This may reduce the risk of postnatal vertical virus transmission with minimal logistical difficulties and without interrupting breastfeeding.

Re-hospitalization in infants younger than 29 weeks' gestation in the EPIPAGE cohort.

AIM: To estimate the re-hospitalization rate of extremely preterm children during infancy and associated factors after the recent improvement in survival rates. METHOD: The cohort included all children born before 29 wk of gestation in nine French regions in 1997. All admissions between discharge from initial hospitalization and 9 mo after birth were considered. Factors studied included the child's characteristics at birth and during neonatal hospitalization, risk factors for infection after discharge and parents' socio-demographic characteristics. Adjusted odds ratios (aOR) for re-hospitalization for all reasons and for respiratory disorders were obtained from logistic regression models. RESULTS: Of the 376 children, 178 were re-admitted at least once (47.3%; 95% CI: 42.3-52.4). Fifty-five percent of the hospitalized children were admitted at least once for respiratory disorders. The re-hospitalization rate was higher for children who had had chronic lung disease (aOR: 2.2; 95% CI: 1.3-3.7), those initially discharged between August and October (aOR: 2.5; 95% CI: 1.2-5.1) or between November and January (aOR: 3.2; 95% CI: 1.5-6.8), and children living with other children under six (aOR: 3.4; 95 %CI: 1.6-7.5). Re-hospitalizations were associated with neither gestational age nor the duration of neonatal hospitalization. Adjusted odds ratios for re-hospitalization for respiratory tract disorders were very similar to those for the overall hospitalizations. CONCLUSION: Infants born before 29 wk have a very high risk of re-hospitalization. The associated factors can help define high-risk groups at discharge from the neonatal unit who need special surveillance.

Fetal and neonatal cerebral infarcts.

Focal arterial infarction in the full-term newborn is an important cause of acquired cerebral lesions in the perinatal period. Clinical motor seizures, most often unifocal, are the nearly constant disclosing symptom confirmed by focal EEG abnormalities. A multifactorial physiopathology is usual, including genetic and perinatal environmental factors. In the past decade, various acquired or genetic thrombophilias have been discussed as risk factors. For several of the involved mechanisms, the excitotoxic cascade could represent a common final pathway leading to neuronal cell death. Early magnetic resonance imaging studies and EEG help to identify the newborns with strokes who are likely to develop hemiplegia and disabilities at school. Protection of the human fetal brain remains difficult, since the triggering factor initiating the excitotoxic cascade is rarely observed. Treatment of seizures is nevertheless necessary, because it seems that they accelerate anoxia-induced neuronal death in animal models of focal hypoxic ischemia.