Nutritional Interventions to Improve Brain Outcomes in Preterm Infants

Abstract

The last 20 years have seen dramatic improvements in survival for preterm infants in both high- and low-income settings. Survival rates of over 50% in infants born 16 weeks early (24 weeks' gestation) are now commonplace in well-resourced neonatal intensive care units. However, ensuring adequate nutrient intakes especially in the first few days and weeks is challenging, and many infants show poor growth and nutritional status. Good nutritional management should be seen as the cornerstone of good neonatal care and is key to improving a range of important outcomes including reduced rates of retinopathy of prematurity, chronic lung disease, necrotizing enterocolitis (NEC), and sepsis. Equally importantly, is that good nutritional status is essential to optimize brain growth and differentiation. There are multiple potential mechanisms that link nutrition to brain outcomes in preterm infants including needs for tissue accretion, energy supply, signaling roles, functional components in human milk, epigenetic regulation, prevention of NEC and disease, and impacts on the gut brain axes. This article will review data in support of different mechanistic links for the impact of nutrition on brain outcomes in preterm infants.

Introduction

Neonatal care has progressed rapidly in the last 20-30 years, and survival at 24 weeks gestation is now common in well-resourced settings. Neonatal care ex-panded rapidly in the 1960s and 1970s, yet in the early days practice on neonatal intensive care units (NICUs) was focused on cardiorespiratory care. The next few decades showed dramatic decreases in death from respiratory distress syndrome and other complications; however, there was a general lack of focus on nutrition. Whilst parenteral nutrition (PN) has been available for neonates since the last 1960s, it was rarely used routinely for preterm infants until the 1990s. Similarly, whilst the importance of human milk (HM) for promoting lifelong health has always been recognized [1], many NICUs start enteral feeds quite slowly meaning exposure to the benefits of HM are limited in the first few days. Preterm infants present many challenges to achieving adequate nutrition. This includes metabolic intolerance, for example hyperglycemia, hypophosphatemia, etc. during PN, and enteral tolerance, which results in many preterm infants not achieving full enteral milk volumes until around day 10-14. Even at this stage, where “full” milk feed volumes of around 150-180 mL/ kg/day are achieved, the relatively low macronutrient density of HM means full milk feeds do not equate to achieving adequate macronutrient intakes [2]. A lack of focus on nutritional management in the first few days and weeks, combined with challenges in providing both PN and enteral nutrition mean that postnatal malnutrition remains universal in many NICUs [3]. Unfortunately, signs of malnutrition are not easy to see in the first few days. Thereafter, growth (most often assessed simply by weight gain) is often slower than fetal references. Whilst there is a robust and appropriate debate about optimal weight gain in preterm infants, it is clear to most healthcare practitioners (HCPs) that suboptimal nutritional management is common. Nutrient intakes rarely meet needs in the first few days and weeks resulting in poor growth that is associated with worse longer-term outcomes, especially suboptimal neuro- cognitive function.

Optimizing Nutritional Status

Whilst all HCPs recognize the importance of good nutrition, detailed knowledge of mechanisms and requirements tends to be restricted to those with specialized knowledge, and many clinicians lack a conceptual framework for assessing nu-trition. Indeed, many HCPs tend to focus primarily on nutrient intakes without always appreciating the wider aspects that are needed to form a holistic assessment of nutrition for that infant. Nutritional practice includes the following areas:

1.    Macronutrient and micronutrient intakes: whilst there is a lack of randomized controlled trials (RCTs) to determine optimal intakes for most nutrients, expert consensus guidelines provide details of recommended intakes that are widely accepted [4].
2.    Functional components: most nutrients provided in PN and formula milk simply provide nutrients for tissue accretion or act as an energy source al-though even in these relatively “simple” fluids there are components that have wider functional activities such as prebiotics and certain amino acids (e.g., taurine and glutamine). However, these are dwarfed by the huge multitude of functional components provided in HM such as HM oligosaccharides (HMOs), growth factors such as IGF-1, and hormones such as insulin, none of which are in formula milk [5].
3.    Microbiomic aspects: the last 10 years have seen a huge increase in research on gut microbiota that demonstrate the dramatic impact gut microbes have on mortality and morbidity (such as necrotizing enterocolitis, NEC) in preterm infants. Newborn infants are born relatively free of microbes but must rapidly develop immune tolerance against hundreds of microbial species. Gut micro-bial community composition is affected by the quantity and quality of enteral intakes, i.e. human versus cow milk, as well as the specific composition of HM [6], and supplements such as breast milk fortifiers and iron. In turn, gut mi-crobes produce compounds essential for health such as vitamins, short-chain fatty acids and amino acids, as well as being involved in bile acid metabolism. Gut microbial composition is therefore integral to nutritional health.
4.    Socio-behavioral and technical aspects: attitudes and behaviors of HCPs and use of donor HM (DHM) affect whether mothers continue to produce mother's own milk (MOM) [7], use of nasogastric tubes impacts on upper gastro-intestinal microbial colonization, and delivery of milk using continuous, or bolus feeds impacts on gastric emptying, gall bladder emptying, and nutrient absorption.

Considering these different elements, the goal of nutritional management in preterm infants is to optimize nutritional status rather than simply maximize weight gain. Nutritional status in turn, is determined by:
1.    What happened in the past, i.e. fetal growth restriction, maternal nutritional status such as vitamin D concentration, or previous disease, for example NEC.
2.    Present body composition and metabolic tolerance, i.e. an infant's ability to tolerate a carbohydrate load, or dispose of excess amino acids (autophagy), as well as enteral/gastrointestinal tolerance, e.g. gastric emptying, intestinal motility, etc.
3.    Desired future outcomes: whilst growth on the NICU is the most tangible evidence of nutritional intakes, later life outcomes such as metabolic disease and neurocognitive outcome are far more important and relevant to life-long health.

These three components combine to affect the ideal nutritional management for an individual baby that will optimize overall nutritional status. An infant with good nutritional status will demonstrate appropriate weight gain, with acceptable body composition, be able to tolerate nutrient intakes to meet needs and have high intakes of functional components from HM that enable the infant to optimize long-term brain outcomes.

Brain Growth and Nutrition in Early Life

Humans differ from other mammals by having a very large brain which is re-sponsible for around 60-70% of all energy expenditure in early infant life. Between 24 weeks gestation and 2 years of age, humans acquire about 85-90% of the final adult volume. More importantly however, there are multiple overlapping neurological processes that take place in a time-coordinated fashion in early life (Fig. 1) [8, 9]. Throughout all three trimesters of pregnancy, neuronal migration is critical for development of all brain regions. During the 2nd and 3rd trimesters, apoptosis (programmed cell death) is particularly active but continues post-term into early infancy. Synaptogenesis is especially active during the 3rd trimester, throughout infancy, and into early childhood, and myelination that starts in the 3rd trimester does not complete until the 2nd decade of life. The impact of nutrient deficiency depends on which nutrient is involved (depicted in Figure 1 by theoretical nutrients A, B, C), the amount of the deficit (represented by the height of the box in Figure 1), and the timing and duration. A large but short-lived deficit of nutrient A (for example protein in first few days) will result in a different long-term neurocognitive phenotype to nutrient C where the deficit occurs later, is less severe, but lasts longer (for example choline). There are strong data to show that IGF-1 and other growth factors are responsible for brain growth and differentiation in early life. Recent data also show that IGF-1 concentrations are linked to brain growth in preterm infants [10]. Of major concern is that IGF-1 concentrations are frequently lower in preterm infants than those observed in utero or in newborn infants born full term. Further studies are needed, but there are some data to show that IGF-1 concentrations may be related to macronutrient intakes, emphasizing the importance of meeting nutrient needs in the NICU.



Placental nutrient transfer, fetal growth, duration of pregnancy, and com-position of breastmilk evolved over millennia to optimize brain growth at different stages in the first 1,000 days of life. Human physiology aims to maximize survival with good brain outcomes, even if this is at the expense of later metabolic risk. In full-term infants, there is strong evidence of a dose-response effect of HM intakes on longer-term brain outcomes. Except at the extremes, however, there is relatively little evidence that growth or weight gain per se in healthy infants impacts on brain outcomes, unless slow growth is accompanied by specific nutrient deficiencies such as iron or iodine. Healthy full-term babies optimize long-term neurocognitive outcomes by receiving HM for as long as possible, largely independent of how quickly they gain weight. In contrast, there are good observational data linking growth to later brain outcomes, that are supported by a small number of RCTs in preterm infants. However, it is critical to appreciate that “faster” growth itself does not cause better brain outcomes. In fact, growth is simply a marker of good nutritional status rather than weight gain being on the causative pathway to brain growth. Weight gain is a reliable indicator of nutrient intakes in preterm infants, but there will always be individual variation, and it is impossible to know the optimal rate of weight gain for an individual. HM-fed infants on the NICU tend to gain weight more slowly than those receiving artificial formula; however, HM-fed infants will have better brain and metabolic outcomes. This has been termed the so-called “breastfeeding paradox,” although many would consider there is nothing very paradoxical about nutrients of higher quality resulting in better outcomes [11].

One Brain for Life

Preterm infants have “one brain for life,” and every aspect of clinical practice on the NICU should aim to protect and promote brain development. Preterm infants have very high nutrient requirements per kg bodyweight and are prone to macro- and micronutrient deficiencies due to low stores. Furthermore, they frequently lack the metabolic capacity to convert essential nutrients into functional components, meaning some nutrients that are not typically considered to be essential, become essential or conditionally essential such as certain fatty and amino acids. Aside from “nutritional” aspects of brain vulnerability, direct damage to the brain as evidenced by intraventricular hemorrhage or cystic periventricular leukomalacia is common in preterm infants. Whilst there are no data to suggest that typical nutritional management can directly prevent these complications, there are good reasons to think that nutritional management can impact on long-term outcome if these conditions have occurred. Very few RCTs have explored the impact of nutritional interventions specifically in high-risk infants with evidence of “brain damage,” but there are some data to show that higher intakes of macronutrients post-discharge may improve outcome [12]. There are also limited data to suggest that supplementation with docosahexaenoic acid (DHA) and choline may improve neurocognitive outcomes, but large confirmatory trials are needed [13].

The poor neurocognitive outcome of babies who develop NEC has been well documented, but the precise mechanisms are likely to be multifactorial [14]. Most importantly, the associated “cytokine storm” activates cell surface receptors in the brain, especially Toll-like receptors on the surface of microglial cells, that trigger downstream release of inflammatory and pro-oxidant compounds that damage the cell [15]. Whilst there is no data to suggest that specific nutrients could prevent or modulate this process per se, there is (1) strong evidence that NEC is related to gut microbial patterns which are themselves modulated by dietary exposures [6] and (2) strong evidence that NEC is less common in infants

receiving MOM. Therefore, there are mechanisms that link nutrition to disease prevention and thereby better brain outcomes. Providing HM nutrition therefore improves brain outcomes by preventing disease.

Macronutrient and Human Milk Intakes and Later Brain Outcome

Whilst there are few RCTs specifically focused on infants with radiological evi-dence of brain damage, there are several RCTs of “enhanced” nutrition in more typical populations of preterm infants that show better short- and longer-term brain outcomes. Studies show that greater macronutrient supply in just the first week of life is associated with better infant neurodevelopment after correcting for likely confounders; an extra 10 kcal/kg/day energy or an extra 1 g/kg/day protein was associated with a staggering 4.6- or 8.2-point higher developmental score using Bayley Scale of Infant Development (BSID) at 18 months age [16]. A large observational study in extremely preterm infants from Sweden (Express study) showed that energy intakes over the first 4 weeks of life were associated with retinopathy of prematurity (ROP): for every 10 kcal/kg/day increase in energy intakes, there was a 24% reduction in ROP after controlling for known con- founders [17]. Whilst the risks for ROP are multifactorial, it is likely that the reduced risk of ROP was modulated in part via effects on IGF-1. An RCT pro-viding higher nutrient intakes using both enteral nutrition and PN over the first few weeks of life resulted in greater weight gain, but most impressively resulted in measurable differences using MRI at term-corrected age in specific brain regions [18, 19]. This study showed that there was a larger Superior Longitudinal Fasciculus in those receiving enhanced intakes which is known to be involved in motor function, perception, and language. Finally, in long-term follow-up studies of studies initiated by Lucas' group [20], there were significant differences detected at 16 years of age. In this study, adolescents who had received higher nutrient intakes (compared to control) for only 4 weeks on average on the NICU, had verbal IQ that was 8 points higher, matched by MRI evidence of a larger- sized caudate nucleus. Although there are associations between weight gain in the postdischarge period and subsequent brain development, RCTs do not show a clear role for macronutrient enrichment of formula to improve brain outcomes at this stage, although further studies are needed [21, 22].

Numerous observational studies show strong associations between the amount of HM received as a preterm infant and later developmental and cognitive outcome, although there are, of course, no RCTs. Again, the studies of Lucas [23] provide the strongest evidence to show that the amount of HM received as a preterm infant affects verbal and full-scale IQ and white matter volume on MRI conducted 15 years later. However, RCTs exploring the impact of DHM on later brain development do not show any consistent developmental advantage on BSID scores in infancy [24]. This might be because DHM contains relatively less macronutrients (and therefore needs greater fortification), or because the processing of DHM (collection, pasteurization, storage, transport etc.) affects key functional components that result in better brain outcomes. DHM might still be beneficial if it reduces the incidence of NEC [25] that damages the brain but will not result in the long-term brain advantages seen with MOM.

Micronutrients, Trace Elements, and the Brain

There are several micronutrients that are critical for brain development; however, there is relatively little evidence that providing intakes greater than currently recommended is beneficial. Iron is a critical nutrient for the developing brain, and suboptimal supply in 1-2 year-old infants results in worse BSID scores even in the absence of anemia [26]. However, most preterm infants will not become iron deficient if they receive recommended intakes (~2-3 mg/kg/ day starting at around 3 weeks of age) whilst on the NICU and in the postdischarge period. Other metals such as zinc and selenium will also be essential, but again there is limited evidence of widespread deficiency when infants are fed recommended dietary intakes on the NICU. Iodine deficiency is a common cause of impaired brain development worldwide, although the incidence is much lower now that there is widespread use of iodinated salt. Many PN solutions appear to contain little or no iodine, and there was concern that this might be the cause of the commonly observed suboptimal thyroid function in preterm infants. However, a large RCT failed to show consistent benefits of iodine supplementation suggesting most infants somehow receive sufficient intakes, perhaps due to even very small amounts of contamination from iodine antisepsis in the mother or neonate, or some other route [27]. Whilst there is an understandable focus on nutrients that are well known to impact on brain development such as protein, energy, iron, fatty acids, etc., it is important to remember that no nutrients function in isolation, and humans require a diet that provides every essential nutrient for brain development to progress normally. Dietary nutrients may affect brain outcomes either by acting as substrate for new brain tissue, by providing energy to power the system, or by acting as cofactors, enzymes, or signaling compounds. Furthermore, there is strong evidence that certain nutrients such as DHA, folate, B12, iron etc. are involved in several epigenetic processes that may also impact on brain development.

Supplemental Nutrients

There have been some interesting RCTs in recent years suggesting that specific components present in HM might improve brain outcomes. In term infants, milk fat globule membrane supplementation appears to improve brain outcomes in formula-fed infants [28], but no data exist in preterm infants. Sphingomyelin supplementation has been shown to possibly improve neurological outcomes in one very small study, but no confirmatory data exist [29]. The normal transplacental transfer of DHA to the developing fetus does not occur in preterm infants who might also have higher demands. Breastmilk concentrations of DHA also appear to be lower in mother's providing HM compared to several years ago, potentially reflecting differences in maternal dietary intakes. However, supplementing infants or their mothers with DHA is not straightforward, and RCTs have shown slightly higher rates of chronic lung disease in infants receiving higher intakes. This might be due to an imbalance that could be corrected by combined supplementation with arachidonic acid (AA). Indeed, a recent RCT in over 200 Swedish infants showed a 50% reduction in ROP in infants receiving both DHA and AA with no adverse impact on BPD rates [30]. Finally, there are basic scientific data to suggest that inadequate choline intakes might be more common than expected, and that these may adversely impact on brain development [31]. A small RCT in infants at high risk of cerebral palsy showed better language development using the BSID at 2 years of age when infants received a nutrient supplement containing DHA, choline, and a nucleotide [13]. Further large RCTs are being planned.

Conclusions

Preterm infants are at high risk of adverse brain outcomes that will have lifelong consequences on neurocognitive function. Damage to the brain may be a direct consequence of intraventricular hemorrhage or periventricular leukomalacia but might also be secondary to the cytokine storm seen in association with NEC or sepsis which damages white matter. Even in the absence of direct damage, the very high nutrient demands, and limited stores of preterm infants mean they are at high risk of malnutrition. This may limit brain growth as well as adversely affecting the time-dependent coordinated development of complex brain pathways that is especially active in early life. Careful attention to nutritional management in preterm infants has the potential to prevent damage by reducing the risk of NEC, and by ensuring adequate nutrients for brain tissue accretion and sufficient energy to “drive” the process. Whilst further studies are needed, there is also reason to hope that in the future, supplemental “functional” nutrients such as HMOs, DHA, choline, or other fatty acids and phospholipids may improve outcomes.

Conflict of Interest Statement

Dr. Embleton declares research funding to his organization from Danone Early Life Nutrition and Prolacta Bioscience US; lecture honoraria from Nestlé Nutrition Institute and Baxter. Dr. Granger and Dr. Chmelova declare no conflicts.

References
1    Embleton ND. Early nutrition and later outcomes in preterm infants. World Rev Nutr Diet. 2013;106:26-32.
2    Cleminson JS, Zalewski SP, Embleton ND. Nutrition in the preterm infant: what's new? Curr Opin Clin Nutr Metab Care. 2016;19(3):220-5.
3    Embleton NDE, Pang N, Cooke RJ, et al. Postnatal malnutrition and growth retardation: an inevitable consequence of current recommendations in preterm infants? Pediatrics. 2001;107:270-3.
4    Agostoni C, Buonocore G, Carnielli VP, et al. En-teral nutrient supply for preterm infants: com-mentary from the European Society of Paediatric Gastroenterology, Hepatology and Nutrition Committee on Nutrition. J Pediatr Gastroenterol. Nutr. 2010;50:85-91.
5    Granger CL, Embleton ND, Palmer JM, et al. Ma-ternal Breast milk, infant gut microbiome, and the impact on preterm infant health. Acta Paedi- atr. 2021 Feb;110(2):450-457.
6    Masi A, Embleton N, Lamb C, et al. Human milk oligosaccharide DSLNT and gut microbiome in preterm infants predicts necrotising enterocolitis. Gut. 2020 Dec 16;gutjnl-2020-322771.
7    Williams T, Nair H, Simpson J, et al. Use of donor human milk and maternal breastfeeding rates: a systematic review. J Hum Lact. 2016;32:212-20.
8    Georgieff MK, Brunette KE, Tran PV. Early life nutrition and neural plasticity. Dev Psychopathol. 2015;27(2):411-23.
9    Wachs TD, Georgieff M, Cusick S, et al. Issues in the timing of integrated early interventions: con-tributions from nutrition, neuroscience, and psy-chological research. Ann NY Acad Sci. 2014;1308:89-106.
10    Hansen-Pupp I, Lofqvist C, Polberger S, et al. Influence of insulin-like growth factor I and nu-trition during phases of postnatal growth in very preterm infants. Pediatr Res. 2011 May;69(5 Pt 1):448-53.
11    Roze JC, Darmaun D, Boquien CY, et al. The ap-parent breastfeeding paradox in very preterm infants: relationship between breast feeding, early weight gain and neurodevelopment based on results from two cohorts, EPIPAGE and LIFT. BMJ Open. 2012 Apr;2(2):e000834.
12    Dabydeen L, Thomas JE, Aston TJ, et al. High- energy and -protein diet increases brain and cor-ticospinal tract growth in term and preterm infants after perinatal brain injury. Pediatrics. 2008;121:148-56.
13    Andrew MJ, Parr JR, Montague-Johnson C, et al. Neurodevelopmental outcome of nutritional in-tervention in newborn infants at risk of neurode- velopmental impairment: the Dolphin neonatal double-blind randomized controlled trial. Dev Med Child Neurol. 2018;60:906-13.
14    Roze E, Ta BDP, Van Der Ree MH, et al. Functional impairments at school age of children with necrotizing enterocolitis or spontaneous intestinal perforation. Pediatr Res. 2011;70:619-25.
15    Hackam DJ, Sodhi CP, Good M. New insights into necrotizing enterocolitis: from laboratory observation to personalized prevention and treat-ment. J Pediatr Surg. 2019 Mar;54(3):398-404.
16    Stephens BE, Walden RV, Gargus RA, et al. First- week protein and energy intakes are associated with 18-month developmental outcomes in extremely low birth weight infants. Paediatrics. 2009 May;123(5):1337-43.
17    Stoltz Sjostrom E, Lundgren P, Ohlund I, et al. Low energy intake during the first 4 weeks of life increases the risk for severe retinopathy of pre-maturity in extremely preterm infants. Arch Dis Child Fetal Neonatal Ed. 2016 Mar;101(2): F108-13.
18    Strommen K, Blakstad EW, Moltu SJ, et al. En-hanced nutrient supply to very low birth weight infants is associated with improved white matter maturation and head growth. Neonatology. 2015;107:68-75.
19    Strommen K, Haag A, Moltu SJ, et al. Enhanced nutrient supply to very low birth weight infants is associated with higher blood amino acid concen-trations and improved growth. Clin Nutr ESPEN. 2017;18:16-22.
20    Isaacs EB, Morley R, Lucas A. Early diet and gen-eral cognitive outcome at adolescence in children born at or below 30 weeks gestation. J Pediatr. 2009;155:229-34.
21    Embleton ND, Wood CL, Pearce MS, et al. Early diet in preterm infants and later cognition: 10-year follow-up of a randomized controlled trial. Pediatr Res. 2021 May;89(6):1442-1446.
22    Teller IC, Embleton ND, Griffin IJ, et al. Post-dis-charge formula feeding in preterm infants: a sys-tematic review mapping evidence about the role of macronutrient enrichment. Clin Nutr. 2016 Aug;35(4):791-801.
23    Lucas A. Post-discharge nutrition and growth: relationship to later cognition. Pediatr Res. 2021;i:1-2.
24    O'Connor DL, Gibbins S, Kiss A, et al. Effect of supplemental donor human milk compared with preterm formula on neurodevelopment of very low-birth-weight infants at 18 months. JAMA. 2016;316:1897.
25    Quigley M, Embleton ND, McGuire W. Formula versus donor breast milk for feeding preterm or low birth weight infants. Cochrane Database Syst Rev. 2019 Jul 19;7(7):CD002971.
26    Lozoff B, Beard J, Connor J, et al. Long-lasting neural and behavioral effects of iron deficiency in infancy. Nutr Rev. 2006 May;64(5 Pt 2):S34-43; discussion S72-91.
27    Williams FLR, Ogston S, Hume R, et al. Supple-mental iodide for preterm infants and develop-mental outcomes at 2 years: an RCT. Pediatrics. 2017 May;139(5):e20163703.
28    Timby N, Domellof E, Hernell O, et al. Neurode-velopment, nutrition, and growth until 12 mo of age in infants fed a low-energy, low-protein formula supplemented with bovine milk fat globule membranes: a randomized controlled trial. Am J Clin Nutr. 2014;99:860-8.
29    Tanaka K, Hosozawa M, Kudo N, et al. The pilot study: Sphingomyelin-fortified milk has a positive association with the neurobehavioural devel-opment of very low birth weight infants during infancy, randomized control trial. Brain Dev. 2013;35:45-62.
30    Hellstrom A. Effect of enteral lipid supplement on severe retinopathy of prematurity: a randomized clinical trial. JAMA Pediatr. 2021 Apr;175(4):359-367.
31    Caudill MA, Strupp BJ, Muscalu L, et al. Maternal choline supplementation during the third trimester of pregnancy improves infant information processing speed: a randomized, double-blind, controlled feeding study. FASEB J. 2018;32:2172- 80.