Article Text



Statistics from


D. M. Murray1, G. B. Boylan1, S. Connolly2, B. P. Murphy1, C. A. Ryan1.1United Maternity Services, Cork, Ireland; 2St Vincent’s University Hospital, Dublin, Ireland

Background: EEG monitoring is useful in the early diagnosis and prognosis of hypoxic ischaemic encephalopathy (HIE) but is not always available in the neonatal intensive care unit. Arterial lactate measurements at 30 min of age have shown promise in predicting which infants are at increased risk of HIE, and proton magnetic resonance spectroscopy has shown persistently raised cerebral lactate up to 1 month post-delivery in infants with HIE.

Aim: To establish the relationship between initial and serial lactate measurements and the severity of EEG changes in full term babies with hypoxic ischaemic encephalopathy.

Method: A prospective study from a large maternity service (6000 deliveries per annum). Continuous video EEG monitoring for 24–72 hours was performed within 6 hours of birth. Encephalopathy was graded as mild, moderate, or severe from analysis of the EEG data, or if unavailable, by Sarnat scoring. EEG seizures if present were measured and the total seizure burden was calculated for each baby. Initial blood gas measurements of pH, base deficit (BD), and lactate were taken within 1 hour of delivery. Subsequent blood gas measurements were taken as clinically required.

Statistical Analysis: Correlation between independent variables was calculated using Spearman’s bivariate analysis. Means were compared using students t test. The threshold for statistical significance was p<0.05.

Results: Twenty four full term babies with HIE were enrolled over an 18 month period; 20 had continuous EEG monitoring. Initial blood gas parameters, pH, Base deficit (BD), and initial lactate measurements did not differ between grades of encephalopathy. Initial lactate was raised in all cases (mean 12.9 (SD 3.98) mmol/l). Lactate clearance rate was increased significantly in severe HIE (mild; 9.1 (6.95) hours, moderate; 11.4 (7.8) hours, severe; 59 (9.9) hours). This delay in lactate clearance correlated with the presence of seizures (r 0.421; p 0.04) and increased with seizure burden (no seizures; 9.7 (6.5) hours, any seizures; 13.4 (11.4) hours, >30 mins seizures; 36.6 (27.5) hours).

Conclusions: The degree of initial acid base disturbance is not helpful in predicting the grade of HIE. Time to lactate clearance is significantly delayed in cases of severe encephalopathy and may be related to the presence of seizures and in particular to seizure burden. Seizure burden has been shown to correlate with neurological outcome.


S. T. Dharmaraj1, M. Henderson1, N. D. Embleton1, A. C. Fenton1, R. J. Cooke2.1Royal Victoria Infirmary and University of Newcastle Upon Tyne, Newcastle Upon Tyne, UK; 2Newborn Centre, University of Tennessee, Center for Health Sciences, Memphis, US

Background: Postnatal growth retardation (PGR) is well recognised in preterm infants, yet limited prospective data exist on the relationship between early PGR, catch up growth (CUG), and later developmental outcome.

Aims: To determine the relationship between PGR, CUG, and developmental outcome at 18 months corrected age in preterm infants born at <32 weeks gestation.

Methods: Infants were recruited from a tertiary NICU over a 12 month period and prospectively followed to 18 months corrected age. Body weight was converted to standard deviation (z score) and stratified into two groups at 28 days of age—those in whom z scores fell by ⩾1.0 between birth—28 days of age (PGR) and those who did not (NPGR). At 18 months corrected age, infants were grouped using the same criteria. Developmental assessment was performed at 18 months using the Bayley Scales of Infant Development. Differences in outcomes were analysed using ANOVA.

Results: 119 infants were recruited, 1 infant died and 108 completed the study (92%). At 28 days, 40 infants were growth retarded (PGR), 68 were not (NPGR). At 18 months, 16 infants remained growth retarded (PGR→PGR) but 24 showed significant CUG (PGR→NPGR). In the NPGR group, 50 infants maintained their growth (NPGR→NPGR) but 18 did not (NPGR→PGR). Mental (MDI) and psychomotor (PDI) developmental index scores for the groups are presented in the table. The PGR→PGR group had poorer MDI (p<0.003) and PDI (p<0.01) scores compared with the NPGR→NPGR group and also a poorer MDI (p<0.003) score compared with the PGR→NPGR group.

Conclusions: Early growth has a significant effect on later development. Factors influencing “catch up” growth may be important in supporting optimal developmental outcome in these high risk infants.


M. S. Fewtrell1, K. L. Loh1, J. Meek2, A. Blake2, D. Ridout1, J. M. Hawdon2.1Institute of Child Health, London, UK; 2University College Hospital, London, UK

Background: Human milk has significant short and long term benefits for preterm infants, but mothers may experience difficulties in expressing breast milk for infants too immature or sick to breastfeed. Oxytocin has been used to assist breastfeeding and milk expression, but few data are available to support this intervention in the neonatal unit setting.

Aims: To test the hypothesis that oxytocin nasal spray increases early milk output in mothers expressing milk for preterm infants.

Methods: A randomised double blind trial of oxytocin nasal spray (100 µl per dose v placebo) was conducted in mothers delivering infants <35 weeks’ gestation. Sprays were used before expressing milk using an electric pump up to day 5.

Main Outcome: Total weight of milk expressed while using spray (study powered to detect >1SD difference between groups). Secondary outcomes: pattern of milk production; number of pumping sessions; weight/fat content of milk expressed during a fixed 20 minute period on day 5 (physiological study); mother’s opinion of expressing; and spray assessed by questionnaire.

Results: 51 mothers were randomised (27 oxytocin, 24 placebo). Total milk production did not differ between groups. Repeated measures ANOVA suggested significantly different patterns of milk production (p 0.001) with initial faster production in group oxytocin then convergence between groups. Parity did not influence the response to the intervention. No significant differences were seen in milk weight or fat content in the physiological study; nor in mothers’ opinions of milk expression and treatment. Milk production in both groups exceeded that from a similar population studied previously using the same methodology.

Conclusions: Despite marginal differences in the pattern of early milk production, use of oxytocin nasal spray did not significantly improve outcome. Most mothers believed they were receiving the active spray, suggesting a significant placebo effect (supported by data from historical controls) and benefits from the extra breastfeeding support available during the study.


C. M. Snowdon, D. R. Elbourne, J. Garcia.London School of Hygiene and Tropical Medicine, London, UK

Background: For some perinatal trials, important decisions about trial participation can be required within very limited timeframes. Parents fearing for the wellbeing of their unborn or newborn baby can be asked to make choices in highly stressful circumstances. The need for swift decisions may compound stress, impede understanding, and compromise parental competency. This qualitative study explores these issues for 78 parents and 42 professionals associated with one or more of four such trials (INNOVO, CANDA, TEAMS, and ORACLE). Professionals were concerned about the speed and quality of parental decisions. Most parents did make very rapid decisions.

Results: Factors associated with rapid decisions were concern for their baby, reactions to staff, and lack of other options. These broad areas are described through the dominant emotions expressed by parents: fear, hope, trust, alienation, acquiescence, and decisiveness. Those who took longer to decide about participation often described similar emotions to those who made rapid decisions. Slower decisions were because more time was available, they wanted further discussion, or they found the decision particularly difficult. Most of those who made rapid decisions felt there were no risks associated with the trial, whereas most of those who made slower decisions felt there were risks. Parents did not appear to share professionals’ perceptions that rapid decisions are problematic. No parent said they were unhappy with the choice they had made. Although there was evidence of parental vulnerability in each trial, parents largely felt that they acted swiftly and responsibly in the best interests of their child.

Conclusion: Parents need to be given every opportunity to access understandable information, even in emergency situations, and then to be trusted to do what they feel is right for their babies and themselves.


M. G. Gnanalingham1, A. Mostyn1, M. E. Symonds1, D. A. Giussani2, T. Stephenson1, D. Gardner1.1Centre for Reproduction and Early Life, University of Nottingham, Nottingham, UK; 2Physiological Laboratory, Cambridge, UK

Background: Chronic umbilical cord compression (UC) in fetal sheep sufficient to restrict fetal blood supply by 30% mimics the degrees of cord compression noted in about 40% of human pregnancies, and results in a range of fetal endocrine adaptations, including increased fetal plasma cortisol. The present study aimed to determine if UC results in up regulation of uncoupling protein (UCP)-2, glucocorticoid receptor (GCR), 11B-hydroxysteroid dehydrogenase type 1 (11BHSD1), voltage dependent anion channel (VDAC), and cytochrome c, involved in energy regulation and cortisol sensitivity within the fetal lung.

Methods: Nine ewes were chronically instrumented with fetal vascular catheters. Five fetuses were then subjected to 3 days UC beginning at 125 days gestation by (term∼147 days) automated compression of the umbilical cord, while the remaining 4 acting as controls. At 137 (SD 2) days gestation all fetuses were humanely euthanased. UCP2, GCR, and 11BHSD1 mRNA abundance was measured by RT-PCR and VDAC and cytochrome c mitochondrial protein abundance by immunoblotting. Results are given as means (SD) (see table). There were no significant differences in body or lung weights between groups. UC resulted in enhanced UCP2, GCR, and 11BHSD1 mRNA, and VDAC and cytochrome c mitochondrial protein abundance in the fetal lung.

Abstract G9

Abstract G12

Conclusions: Chronic UC results in precocious maturation of lung mitochondria. These changes may be important in promoting gaseous exchange within the growth restricted fetal lung.


K. M. Penge, M. Morton, H. Vandecruys, K. Nicolaides, A. Greenough.King’s College Hospital, London, UK

Background: Twin pregnancies are high risk pregnancies accounting for 10% of all perinatal mortality. There are differences in the perinatal mortality rates in relation to chorionicity with the mortality being 3–4 times higher in monochorionic compared with dichorionic twins. Preterm delivery is a major complication in twin pregnancies and is twice as likely to occur in monochorionic than in dichorionic twins. In untreated severe twin to twin transfusion syndrome (TTTS), the mortality rate is over 80% and accounts for 15–20% perinatal deaths in twins. Endoscopic laser coagulation of anastomoses has been shown to be more effective first line treatment than serial amnioreduction for severe twin to twin transfusion syndrome—being associated with a higher survival and a lower incidence of neurological complications at 6 months.1

Aim: To test the hypothesis that neurodevelopmental outcome will be similar in monochronic twins who have received laser surgery (MC laser) for severe twin–twin transfusion to the neurodevelopmental outcome of dichorionic twins and monochronic twins not requiring intervention.

Methods: Women with twin pregnancies cared for in a single centre were enrolled into the study. Perinatal and neonatal data were collected and at 2 years the surviving twins underwent physical examination and assessment using the Griffiths scores corrected for postnatal age. Children were described as having neuromorbidity if they scored less than 78 on any of the Griffiths subscales (that is, significant delay).

Results: The 24 MC Laser and 30 monochronic twins delivered at an earlier gestation (mean 33.6 and 34.7 weeks, respectively) than the 24 dichorionic twins (mean 37.3 weeks) (p<0.001). Although only two MC laser, one monochronic twins, and none of the dichorionic twins had global delay, neuromorbidity was commoner in the MC laser group (p = 0.027), affecting 10 of 24 MC laser, 4 of 30 monochronic twins, and 4 of 24 dichorionic twins. The MC laser twins, despite delivering at a similar gestational age to the monochronic twins, had more neuromorbidity at 2 years of age (p 0.0405).

Conclusion: Neuromorbidity is increased in monochronic twins with severe twin–twin transfusion syndrome, despite intervention with laser coagulation.



D. J. Manning1, P. J. Todd1, M. J. Maxwell1, M. J. Platt2.1Wirral Hospital NHS Trust, Wirral, UK; 2University of Liverpool, Liverpool, UK

Background: Severe neonatal jaundice and bilirubin encephalopathy have been reported in North America and Europe in the past decade. The primary objective of this study was to determine the incidence of severe neonatal hyperbilirubinaemia (unconjugated serum bilirubin ⩾510 μmol/l in the first month of life) in Britain and Ireland. Secondary objectives were to document associated clinical variables and short term outcomes.

Method: Paediatricians reported cases meeting the case definition by means of the British Paediatric Surveillance Unit “orange card” scheme.

Results: Between 1 May 2003 and 30 April 2004, 37 infants met the case definition, 25 of whom were male. None was reported from Ireland. During this period, there were approximately 702 200 live births in Britain, so the incidence of severe neonatal hyperbilirubinaemia was 0.05/1000 live births. Median (range) birth weight was 3.2 (1.8–4.2) kilograms and median (range) gestation was 38 (35–42) weeks. Sixteen infants were of white English origin; 21 were of ethnic minority origin. Twenty eight infants were breast fed, four had mixed feeding, and five were exclusively artificially fed. Readmission was needed for 29 infants, 16 of whom had initially been discharged before 48 hours of life. Median (range) age of maximum hyperbilirubinaemia was 4 (2–9) days. Comorbidity included dehydration (9 infants), haemolysis (11 infants), and sepsis (2 infants, both of whom died). Exchange transfusions (total 28) were done in 24 infants. Bilirubin encephalopathy occurred in six infants.

Conclusion: Severe neonatal jaundice and associated encephalopathy, while rare, are being encountered in Britain. Associations include male sex, relative immaturity, ethnic minority origin, early discharge from the maternity unit, breastfeeding, and comorbidity.


S. J. Broughton, A. Roberts, M. Zuckerman, A. Greenough.1Guy’s, King’s and St Thomas’ Medical School, Division of Asthma, Allergy and Lung Biology, London, UK

Background: Prematurely born infants, particularly if they have had an (respiratory syncytial virus; RSV) infection, frequently suffer troublesome cough and wheeze at follow up.

Objective: To determine if premorbid lung function abnormalities were a risk factor for symptomatic RSV infection and chronic respiratory morbidity at follow up in prematurely born infants.

Methods: Lung function (compliance and resistance of the respiratory system (CRS and RRS) assessed by single breath mechanics and lung volume assessed by measurement of functional residual capacity (FRC) using a helium gas dilution technique) was measured at 36 weeks post-menstrual age. Following NICU discharge, patients were asked to contact the research team if their infant had symptoms consistent with a lower respiratory tract infection (LRTI) and the parents were contacted every 2 weeks to determine if their infant was or had been symptomatic. A nasopharyngeal aspirate (NPA) was obtained on every occasion an infant had a LRTI, whether in hospital or in the community. RSV infection was identified by immunofluorescence and/or culture. When the infants were 11 months corrected for gestational age, their parents completed a diary card for 1 month on a daily basis, recording how many days their infants coughed or wheezed in a month.

Patients: Thirty infants, median gestational age 29 (range 23–31) weeks, 18 (60%) had bronchopulmonary dysplasia.

Results: The 11 infants who suffered an RSV positive LRTI had significantly higher RRS (mean 125.8 (SD 14.1)) than the non-RSV group (mean 82.1 (12.5)) (p 0.02) but the CRS (p 0.69) and FRC (0.80) were similar in the two groups. Regression analysis demonstrated risk factors for cough were RSV LRTI (p<0.001) and a high RRS (p 0.002) and for wheeze were RSV LRTI (p = 0.026) and a high RRS (p = 0.001).

Conclusion: In prematurely born infants, premorbid lung function abnormalities are a risk factor for symptomatic RSV infection and chronic respiratory morbidity.


P. A. Alexander1, D. Sharkey1, P. Dey1, J. Whittaker1, L. M. Williamson2, H. Firth1, A. L. Ogilvy-Stuart1, J. S. Ahluwalia1.1Addenbrooke’s NHS Trust, Cambridge, UK; 2National Blood Service, Cambridge, UK

Aims: In many neonatal units it is usual practice to delay genetic analysis of peripheral blood from newborn babies following blood product transfusion. This practice has not been considered for revision since the introduction of universal leukodepleted blood products in November 1999. According to current National Blood Service quality monitoring data, leukodepleted blood products contain not more than 1×105 white blood cells (WBC) in 250 ml of packed red blood cells (RBC). This study was designed to assess whether donor DNA is detectable in RBC to be transfused into neonates.

Methods: Leukodepleted RBC were transfused into neonatal patients following preparation according to standard neonatal protocols. A sample of 1 ml of RBC from the giving set was saved for genetic analysis. The samples were analysed using a routine DNA extraction procedure (Nucleon extraction kit) for small volume samples and PCR analysis for amelogenin loci (present on X and Y chromosomes). Consent for the use of donated blood for research is given at the time of donation and the relevant local research ethics committee granted this study ethics approval.

Results: Sample collection was conducted in the Neonatal Intensive Care Unit between February 2004 and November 2004. Samples were collected from 63 blood transfusions, from a total of 51 different donors. Donor RBC at time of transfusion had a median age of 16 days post-donation, with a range of 7–35 days. DNA was not detected in any of the samples by routine, non-fluorescent, analysis for the amelogenin loci.

Conclusion: DNA was not detected in samples of leukodepleted RBC for transfusion by routine laboratory PCR testing after routine DNA preparation. It follows that any donor leukocytes that may persist in the peripheral circulation of a neonatal patient following transfusion would be below the level of detection of routine DNA analysis. Thus, transfusion of leukodepleted packed RBC should not contaminate DNA analysis of peripheral blood from a neonatal patient. Our results suggest that the practice of delaying DNA analysis on peripheral blood from newborn babies post-blood transfusion should be reconsidered.


C. Rose, R. Robinson, J. R. Tooley.1NICU, St Michael’s Hospital, Bristol, UK

Background: Information from a high quality postmortem examination is the right of every parent following the death of their child. Postmortem rates worldwide are declining and the perinatal postmortem is no exception. No clear factor has been isolated. It is thought that a change in the public’s willingness to grant permission plus recent media coverage of issues such as retained organs have all contributed to this fall. A new process for obtaining consent to postmortem has been in place for over 6 months. The impact of these issues on the clinician’s willingness to ask for consent to postmortem have not been studied until now.

Aim: To collect consultant neonatologists’ views on the importance of the perinatal postmortem, its role in modern day practice, and views on the new consent process.

Methods: A survey of consultant neonatologists and paediatricians (with an interest in neonatology) from all neonatal intensive care units in England and Wales with ⩾5 ITU cots by telephone questionnaire.

Results: 60 clinicians completed the questionnaire. Regarding the value of a postmortem, one third felt a postmortem offered new information in less than 25% of cases, one third between 25–50%, and one third felt new information would become available >50% of the time. Only 60% routinely offered all parents a postmortem. 53% of all clinicians had specific reasons why they may not offer a postmortem. These reasons for not offering a postmortem included: lack of availability of a perinatal pathologist (35% of clinicians), organ retention issues (9%), and the new consent process (8%). Regarding the new consent process, 66% felt taking consent was now much harder v 15% that felt it was now easier. 20% of respondents had received training for taking consent and 13% had been appraised at taking consent. Of note was that 10% of clinicians had experience of using the coroner to get a postmortem rather than asking parents for consent the standard way.

Conclusions: Although it is every parent’s right, our data suggest that only 60% of clinicians always offer a postmortem to parents. Although more involved, the new consent process does not seem to have reduced the number of postmortems being offered. The single biggest reason given for not offering a postmortem is the availability of a perinatal pathologist.

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.