AimLong‐term electroencephalogram (EEG) recording is increasingly being used in the neonatal period, but application and maintenance of the EEG electrodes is challenging, especially in preterm infants. This study proposes a practical method of electrode application that can be used in the neonatal intensive care unit (NICU).Methods EEG recording in preterm infants of <32 weeks of gestational age is often challenging and requires careful preparation and strict adherence to NICU protocols. An effective technique for EEG application in preterm infants is to use prepackaged, sterile, disposable, flat‐surfaced EEG electrodes. The use of these electrodes in combination with a continuous positive airway pressure hat provides good security for electrodes and good quality EEG recordings. It also limits the handling of the infant, while strictly adhering to infection control policies.ResultsLong‐term monitoring for >72 h has been achieved using this technique. Important steps to consider are efficient preparation of the recording machine and materials, careful electrode application and infection control.ConclusionA fast and effective method of EEG electrode placement is required for neonatal EEG monitoring. The practical techniques described in this article outline a reliable method of EEG electrode placement, suitable for even extremely preterm infants.

FFM and FM can define nutritional depletion or obesity. We present longitudinal data from a large dataset of stable growing infants. Methods BC data (n = 857) from 574 infants (22-42 wks) enrolled in 4 longitudinal studies, 2/3rd were preterm, was considered as independent data points. Preterms on fortified breast milk or formula (80 kcal/dL), term infants on breast milk or formula (67 kcal/dL). Time points of measurement : after reaching full enteral feeding, at term and two further time points until a maximum of 6 months of corrected age. BC was measured by Dual energy X-ray absorptiometry (QDR 1500; Hologic). R software (GAMLSS) used for construction of growth curves. Results Length of preterms remain shorter than terms, both increases linearly at 0.7 cm/wk. Median FM/FFM in preterm is 500 g/2450 g (40 wks) and 1700 g/4500 g (60 wks) compared to 430 g/2790 g (40 wks) and 2400 g/4700 g (60 wks) for term. Preterm FMI centiles are higher than terms till 45-50 wks corrected. Preterm FFMI increases progressively till 40 wks, then remains constant over time like terms (Figure 1). Conclusion Growth pattern for preterm vary from term, justifying their higher nutritional requirement to support rapid FFM growth initially. FM being inverse of FFM, length normalised indices rather than percentages allow independent assessment of growth in each body compartment, while compensating for difference in body size of term and preterm infants. Background Growth of preterm infants should follow intrauterine rates. Postnatal loss of extracellular fluid shifts growth trajectories to a percentile below that in-utero. Which 'new' trajectory a preterm infant should adjust to after completed postnatal adaptation is unknown. Results 890 infants were eligible of 6915 meeting inclusion criteria. Infants had maximum weight loss by DoL 5, regained birth weight by DoL 11 and showed stable growth parallel to intrauterine percentiles during DoL 7-21. Surprisingly the new trajectory was independent from GA with a z-score difference from birth of (A) -0.96 ± 0.75 and (B) -0.88 ± 0.67 at DoL14. Linear regression models predicted weight at DoL 14 (R2=0.88) and 21 (R2=0.82). Conclusions 1) The study provides robust estimates of ideal postnatal growth trajectories for preterm infants. 2) The impact on long-term outcome using these trajectories for nutritional adjustment needs to be assessed, ideally in an RCT. PO-0460 IDENTIFYING TRAJECTORIES FOR HEALTHY POSTNATAL GROWTH OF PRETERM INFANTS PO-0461 HEART RATE VARIABILITY IN FULL-TERM NEONATES WITH HYPOXIC ISCHAEMIC ENCEPHALOPATHY

Hypernatremia causes brain shrinkage and resultant vascular rupture with cerebral and IVH. However, it is not known if rapid fluctuation in serum sodium in hypernatremic preterm infants results in IVH or death. Objective To determine if the rapid rise in serum sodium or rapid correction of hypernatremia predict the composite outcome of severe IVH (grade 3 and 4) or death during the first 10 days of life. Methods Single center retrospective review of 167 preterm infants with GA ≤26 weeks who had serum sodium monitored at least every 12-24 hours and more frequently, if indicated. Logistic regression analysis identified which of the commonly cited risk factors of IVH, including rapid (>10 and >15 mmol/l/day) rise or fall in serum sodium could predict composite outcome in hypernatremic infants. Results 98 (59%) of 167 infants studied developed hypernatremia (serum sodium>150 mmol/L), with a maximum median serum sodium of 154 mmol/l (range 150-181, IQR 152-157), occurring on median postnatal age of 4 days (IQR 3-5). Grade 4 IVH was more frequent in hypernatremic compared to normonatremic infants (p=0.032, OR 3.4, 95% CI 1.1-10.6). Among 98 infants with hypernatremia, severe IVH or death occurred in 33 and 21 infants with rapid (>10 mmol/l/day) rise and drop in serum sodium, respectively. However, rapid (>10 and >15 mmol/l/day) rise or fall in serum sodium was not associated with composite outcome on multivariate analysis. Conclusion Correction of hypernatremia not exceeding 10 to 15 mmol/l/day in hypernatremic preterm infants was not associated with severe IVH or death. Background EEG monitoring provides important information about the neurological status of the preterm infant but is difficult to interpret for most. We aim to automatically detect the typical bursting pattern (trace discontinu) of the preterm EEG and compare the detections with expert manual annotations. Methods The method was based on the single channel EEG method of Palmu et al. but extended to 8-channel recordings for the first time. The EEG signal was first filtered with a Kaiser-window filter and the output of a non-linear energy operator (NLEO) was calculated. The NLEO signal was smoothed and corrected for baseline artefacts. A burst was identified if the resulting signal remained AUTOMATIC IDENTIFICATION OF ACTIVITY BURSTS IN EEG OF PRETERM INFANTS