I write frequently on this blog about how prejudice regarding developmental and neurologic problems, and prejudice about preterm infants, combine to over-emphasize the difficulties that former preterm babies have, to the extent that professional societies develop guidelines for withholding care to at-risk babies; based on outcomes that would never be considered grounds for withholding or withdrawing care in other areas of medicine.

Nevertheless, I am not blind to the increased problems experienced by preterm infants, from the mildly preterm to the most immature, and the need to do everything we can to mitigate those impacts.

One thing I often explain to my trainees is that during the 3rd trimester, the developing brain produces 250,000 neurones every minute. In addition to the development of the neuroglial substrate, the production and pruning of synapses and the on-going exquisitely fine structural development of the cortex, grey nuclei, and the connections between them that we call white matter.

In order to do all that you need a constant flow of nutrients, enough calories, enough amino acids in a reasonable proportion, fatty acids of the right type in reasonable proportions, minerals and trace elements, growth factors, and probably all sorts of other things that we know little about.

Good neonatal nutrition is not just a matter of calories, or total protein intake, but the best components also, for which we need much more research. At present we try to mimic intra-uterine weight gain (and often fail badly) but have little knowledge of the impact of our practices on the quality of growth, especially cerebral growth. And to be honest, we don’t know if human breast milk as a nutritional source, is the best way to try to mimic in-utero brain development, which normally works with a substrate of trans-placental supply of nutrients. That is probably quite different to the trans-intestinal supply of nutrients derived from fortified breast milk. On the other hand, I really don’t think that bovine milk derived products could possibly be better!

With this long preamble in mind, I wanted to introduce some recent articles that address these issues. The first is an abstract from the 2017 PAS-meeting which compared early neonatal nutrition among extremely preterm babies, and development of cerebral connections on term equivalent MRI. The changed abstract system for the latest PAS-meeting was a disaster, and it seems that the abstracts for all the recent years, since there was no longer a paper issue of Pediatric research with the abstracts in, are no longer available. Abstracts2view/PASALL and /PAS no longer function, which is a disastrous situation, if you ask me. So when I was trying to find this abstract and remembering that Steven Miller was one of the authors, I eventually found an enormous web page https://registration.pas-meeting.org/2017/reports/rptPAS17_Abstracts.asp that exists from the latest PAS-meeting (but not from the previous ones) and using a web page search (ctl-F) I finally found this badly formatted abstract, I’ll put the whole thing here so you don’t have to go searching for it:

Background: Background: Optimizing early nutritional intake in preterm neonates might enhance brain maturation and improve neurodevelopment. The relationship of energy and nutrient intake in the first weeks of life with brain growth during neonatal intensive care needs to be determined. Objective: Objective: To determine the association of early energy and nutrient intake with brain regional and total growth, and white matter maturation assessed by serial magnetic resonance imaging in very preterm (VPT) neonates. Design/Methods: Methods: 49 VPT (21 males, median[IQR] gestational age (GA): 27.6[2.3] weeks) were scanned serially at median postmenstrual weeks (PMA): 29.4, 31.7 and 41. Thalamus, basal ganglia, cerebellum and total brain were semi-automatically segmented in the T1-weighted images. Fractional anisotropy (FA) was extracted from the diffusion-tensor imaging (DTI) data using tract-based spatial statistics (TBSS). Nutritional intake from days of life 1 to 14 was collected. Multivariate linear regression and generalized estimating equations (GEE) for repeated measures were used to assess the association between nutrient intake and volumes, and FA values in separate models.Results: Results: In GEE models, greater energy [kcal/kg/d] and lipids [g/kg/d] intake predicted increased basal ganglia (?=29.7, p=0.002; ?=28.9, p=0.005, respectively) and total brain (?=776.5, p=0.036; ?=12824.9, p=0.019, respectively) volumes [mm3] over the course of neonatal intensive care to term age, adjusting for PMA, birth GA and sex. A similar association was found with carbohydrates intake and basal ganglia volume (?=12.1, p=0.043). Examining volumes at each scan, adjusting for PMA at MRI, the associations of energy and lipid intake with thalamic, basal ganglia, cerebellar and total brain volumes became increasingly robust on the second and third scans. Each 10-kcal/kg/d-energy intake increase in early life predicted a 2% increase in brain volume at term. Similarly, FA values in the posterior corona radiata and posterior thalamic radiations were significantly associated with early calories and lipid intake, both in linear regression and GEE models (all p<0.05).Conclusion(s): Conclusion: In VPT neonates, greater energy and lipid intake during the first two weeks of life predicted more robust brain growth particularly in subcortical structures and cerebellum, and accelerated white matter maturation. Optimizing early nutrition in VPT neonates warrants further attention as a potential avenue to improve brain health outcomes

There are clearly multiple limitations for a small observational study of this type, but it is nevertheless suggestive and consistent with other data. we shouldn’t be complacent about the period of poor nutrient intake, poor growth, and poor head growth that accompany that poor intake. The millions of lost neurones and disturbed cerebral maturation that occur during the first 14 days of life may have permanent impacts.

The second is a study from the CNN comparing head circumference growth (a reasonable indicator of brain growth, at least in terms of size) between birth and discharge and longer term follow-up, with neurodevelopmental outcomes. (Raghuram K, et al. Head Growth Trajectory and Neurodevelopmental Outcomes in Preterm Neonates. Pediatrics. 2017) They analyzed data from nearly 2000 babies of less than 29 weeks gestation, they divided the babies up into groups depending on changes of head circumference z-scores between birth and discharge from NICU, and between birth and follow-up at 16 to 36 months. Babies were born between april 2009 and september 2011. Sadly, 25% of babies dropped their head circumference z-scores by between 1 and 2 between birth and discharge and another 25% by more than 2. I say sadly, because this is avoidable, in our study of enhanced a nutritional protocol the mean head circumference z-score change from birth to discharge was about 0 (-0.1 to be exact). In this new CNN publication there is a clear association between loss of head circumference growth, the infants in the worst group of head growth between birth and discharge had double the Odds of having significant  developmental delay or neurological impairment (Bayley 3 motor, language or composite scores less than 70, or CP with a GMFCS <2).  Catch up growth of the head after discharge did improve things a little, but not back to the original potential (if I over-interpret the results correctly).

The third is a study just examining a change in nutritional practice and growth outcomes. Not dissimilar to our publication of 4 years ago, then demonstrated that post-natal growth can be improved, to be similar to intra-uterine standards with little postnatal growth delay. Genoni G, et al. Non-randomised interventional study showed that early aggressive nutrition was effective in reducing postnatal growth restriction in preterm infants. Acta Paediatr. 2017  However they still overall had a loss of head circumference from a mean of 0.14 at birth to a mean of -0.78 in the group with the improved nutrition. In their ELBW population there were still very many infants who started with a birth weight above the 10th %le, but were discharged below the 10% percentile 70% with their new protocol, compared to 90% with their old protocol.  This is better than some results, but could be substantially better. Early nutritional intakes can be pushed up faster, than this group do, with a quicker increase in glucose, and perhaps starting lipids at much higher doses immediately after birth. To avoid catabolism we need to achieve over 70 kcal/kg/d as soon as possible after birth, to achieve reasonable growth 100 kcal/kg/d intravenously, and 3.5 g/kg/d of good quality protein are needed as quickly as that can be achieved. Enteral needs are of course higher, at 120 kcal/kg/d and 4 g of protein (maybe 4.5). Focusing on nutrition every day, even when the babies are critically ill can avoid much post-natal growth restriction, and it leads to bigger babies with bigger cerebellums.

Paviotti G, et al. Higher growth, fat and fat-free masses correlate with larger cerebellar volumes in preterm infants at term. Acta Paediatr. 2017;106(6):918-25.  42 VLBW babies were assessed with body composition measures and MRI at 40 weeks, and basically they showed what it says in the title, bigger babies had bigger ‘little brains’. As far as I can see from the limited data presented about total cerebral volume, daily weight gain from birth to term age also was correlated with the size of the entire brain.

Adequate quantities of nutrition and appropriate quality can dramatically reduce postnatal growth restriction, and if we pay enough attention to it, can probably eliminate it. The first couple of weeks are vitally important, and I don’t think we can overcome several days of undernutrition by trying to catch up later. Finding ways to improve fat-free growth and brain growth should be priorities for future.