G Model

ARCPED-4621; No. of Pages 9

Archives de Pédiatrie xxx (2018) xxx–xxx

Available online at

ScienceDirect
www.sciencedirect.com

Review article

Parenteral nutrition for preterm infants: Issues and strategy

D. Darmaun a,*, A. Lapillonne b, U. Simeoni c, J.-C. Picaud d, J.-C. Rozé a, E. Saliba e,
A. Bocquet f, J.-P. Chouraqui g, C. Dupont b, F. Feillet h, M.-L. Frelut i, J.-P. Girardet j, D. Turck k,
A. Briend l, Committee on Nutrition of the French Society of Pediatrics (CNSFP), and
French Society of Neonatology (SFN)
a
Université Nantes-Atlantique, 44300 Nantes, France
b
Université Paris Descartes, 75006 Paris, France
c
Université de Lausanne, CHUV, 1011 Lausanne, Suisse
d
Université Claude-Bernard-Lyon 1, 69008 Lyon, France
e
Université François-Rabelais, 37000 Tours, France
f
Université de Franche-Comté, 25000 Besançon, France
g
Université Joseph-Fourier, 38000 Grenoble, France
h
Université de Lorraine, 54000 Nancy, France
i
Université Paris-Sud, CHU de Bicêtre, 94270 Le Kremlin-Bicêtre, France
j
Université Pierre et Marie Curie-Paris 6, 75005 Paris, France
k
Université Lille 2, LIRIC-Inserm U995, 59037 Lille, France
l
Institut de recherche pour le développement, 13572 Marseille, France

A R T I C L E I N F O A B S T R A C T

Article history: Due to transient gut immaturity, most very preterm infants receive parenteral nutrition (PN) in the first
Received 18 November 2017 few weeks of life. Yet providing enough protein and energy to sustain optimal growth in such infants
Accepted 18 February 2018 remains a challenge. Extrauterine growth restriction is frequently observed in very preterm infants at the
Available online xxx
time of discharge from hospital, and has been found to be associated with later impaired
neurodevelopment. A few recent randomized trials suggest that intensified PN can improve early
Keywords: growth; whether or not such early PN improves long-term neurological outcome is still unclear. Several
Neonatology
other questions regarding what is optimal PN for very preterm infants remain unanswered. Amino acid
Extrauterine growth restriction
Intravenous nutrition
mixtures designed for infants contain large amounts of branched-chain amino acids and taurine, but
Nutritional imprinting there is no consensus on the need for some nonessential amino acids such as glutamine, arginine, and
cysteine. Whether excess growth in the first few weeks of life, at a time when very preterm infants
receive PN, has an imprinting effect, increasing the risk of metabolic or vascular disease at adulthood
continues to be debated. Even though uncertainty remains regarding the long-term effect of early PN, it
appears reasonable to propose intensified initial PN. The aim of the current position paper is to review
the evidence supporting such a strategy with regards to the early phase of nutrition, which is mainly
covered by parenteral nutrition. More randomized trials are, however, needed to further support this
type of approach and to demonstrate that this strategy improves short- and long-term outcome.
C 2018 Elsevier Masson SAS. All rights reserved.

1. What is at stake? Using birth weight instead of prematurity for defining nutritional
strategies is likely inadequate since three different trajectories can
The incidence of preterm birth is on the rise: 7% of live births in result in low birth weight (LBW): (a) growth deceleration in utero
France [1] and 11% worldwide [2] occur before the 37th week of (intrauterine growth restriction, IUGR), (b) steady growth in utero
gestation. Very preterm infants are defined by birth before the 32nd below normal range, or (c) preterm birth with a weight appropriate
week, extremely preterm infants by birth before the 28th week. for gestational age (AGA). The term ‘‘small for gestational age’’
(SGA) refers to the first two settings [3], and it should be borne in
mind that preterm infants can be born SGA as well. This paper only
* Corresponding author. addresses parenteral nutrition (PN) in preterm infants stricto sensu
E-mail address: [email protected] (D. Darmaun).

https://doi.org/10.1016/j.arcped.2018.02.005
0929-693X/ C 2018 Elsevier Masson SAS. All rights reserved.

Please cite this article in press as: Darmaun D, et al. Parenteral nutrition for preterm infants: Issues and strategy. Archives de Pédiatrie
(2018), https://doi.org/10.1016/j.arcped.2018.02.005
G Model
ARCPED-4621; No. of Pages 9

2 D. Darmaun et al. / Archives de Pédiatrie xxx (2018) xxx–xxx

since PN is rarely indicated in low-birth-weight infants born full 2.26 [95% CI, 1.37–3.72]) [16]. Interestingly, regardless of the
term and since nutritional requirements and metabolism differ postnatal growth trajectory, SGA infants showed more frequent
between preterm and LBW term infants. behavioral problems and cognitive deficit at 5 years of age. The
association of slow growth in the first few weeks of life with poor
1.1. Preterm nutrition, a challenge for neonatologists neurodevelopment is more pronounced in boys. In a cohort of
1221 boys and 1056 girls born preterm, the odds ratio for
Thanks to progress in neonatal intensive care, more than 90% of suboptimal neurodevelopment at 2 years of age was 3.2 for boys
very preterm infants survive the neonatal period [4], so that who experienced EUGR vs. those who did not, and the risk was 1.8
nutrition has become a key determinant for their future health (95% CI, 0.7–4.2) and 0.95 (95% CI, 0.4–1.9), respectively, in girls
outcome. In utero, bathed in a nearly sterile, isothermal, and [17]. Most studies in the field, however, are observational. As such,
oxygen-poor milieu, the fetus receives continuous intravenous they cannot prove a causal relationship: premature infants who
feeding through the umbilical vein. Expelled at birth from this grew slowly most likely were frailer or more severely ill, which
cocoon, the premature infant confronts a cold environment that is could have led to the prescription of less optimal nutrition during
high in bacteria and oxygen. Its nutritional supply is cut off, its their hospital stay.
nutrient stores scarce, and its needs are tremendous if this infant is
to triple its weight in 3 months as a normal fetus does over the 1.3. Nutritional imprinting
third trimester of gestation.
Due to gut immaturity during the first postnatal week(s), The concept of nutritional imprinting emerged in the 1990s
feeding through the gastrointestinal route is at first insufficient to when David Barker showed that infants born with a low birth
cover such needs: intravenous nutrition (PN) is required during the weight are exposed to a higher risk of developing obesity, type
first few weeks. Despite a more rapid daily increase of feeds, full 2 diabetes, hypertension, high cholesterol, and coronary disease in
enteral feeding was only achieved around the 22nd day of life in a adulthood [18,19]. The first cohorts included children born SGA in
cohort of preterm infants weighing less than 1000 g [5]. PN takes the 1900s, who were mostly term infants. More recent work shows
over from the maternal–fetal nutrient supply delivered via the that adults born preterm are exposed to increased risk of
placenta, but is far from approaching the complex composition of hypertension and insulin resistance in the long run [20–22]. In a
umbilical blood (e.g., regarding lactate, growth factors, etc.) and meta-analysis of 27 studies, collecting more than 17,000 adults born
should not necessarily mimic the fetal supply of nutrients in view preterm, LDL cholesterol and blood pressure were 0.14 mmol/L (95%
of the dramatic changes in metabolism occurring after birth. PN CI, 0.05–0.21) and 4.2 mmHg (95% CI, 2.8–5.7) higher for LDL and
may result in severe complications (e.g., sepsis, cholestasis, blood pressure, respectively, in adults born preterm than in those
thrombosis) and often falls short of covering needs. Lack of fine born full term [23]. Such differences, albeit small, are associated
tuning of the parenteral nutrient supply, combined with failure or with increased cardiovascular risk on a population scale. In contrast,
delay to provide adequate enteral nutrition often result in no significant difference was observed regarding corpulence, blood
extrauterine growth restriction (EUGR). Despite improvement in glucose, or fasting serum insulin [23] nor insulin resistance [24].
nutritional care, EUGR remains common in premature infants. For Whether or not early markers of risk of hypertension and
example, it was observed in 75% of very preterm infants in the early insulin resistance can be found during childhood remains a matter
2000s [6], and still concerns half of the very preterm infants in the of debate. In a British cohort of 209 adults born preterm, the
most recent studies [7,8]. EUGR may be, at least partly, preventable growth rate in the upper quartile during the first 2 weeks of life
in infants devoid of severe morbidities such as severe necrotizing was associated with reduced flow-mediated arterial dilation at
enterocolitis, by optimizing PN from birth onwards [9–13]. More- adulthood [25]. In a French cohort, preterm birth was associated
over, suboptimal nutrition in the first few weeks of life was shown with elevation of systolic blood pressure and arterial stiffness
to be associated with a higher risk of nosocomial infection, (assessed by pulse wave velocity) at adulthood [26], but in another
necrotizing enterocolitis, and hospital admission [14]. study conducted earlier at 6 years of age, no such association was
found [27]. Factors other than nutrition obviously play a role as
1.2. Premature birth and the risk of suboptimal growth and well. Preterm babies undergo significant stress and stress-related
neurodevelopment hormones may also play a major role in these outcomes.

Premature birth is a well-known risk factor for suboptimal 1.4. Risk of disease at adulthood due to rapid catch-up growth
neurodevelopment. At 8 years of age in the EPIPAGE 1 cohort, 77% of
infants born very preterm attended a classroom corresponding to Several studies suggested a deleterious effect of excess growth
their postnatal age, 5% were in special classes, and 18% had repeated in infancy or childhood in preterm infants. In the study by Regan
a class (vs. 95, 1, and 5%, respectively, for children born full term) et al., weight gain between 0 and 4 years was associated with
[15]. In a landmark study, Ehrenkranz et al. showed that increased insulin resistance assessed at 10 years of age [28]. In a
neurodevelopment correlates with the initial growth rate: 600 very more recent study, growth of preterm infants in early infancy did
preterm infants weighing between 500 and 1000 g at birth were not impact metabolic status in adolescence, in contrast to rapid
ranked in four growth rate quartiles during the hospital stay. Infants weight gain in childhood [29]. The most dangerous weight gain
in the 1st quartile gained an average of 12 g/kg/day vs. 21 g/kg/day may in fact be that occurring after 1 year of age. A recent review
for those in the 4th quartile. At 18–22 months of age, the prevalence showed that growth between birth and 12–18 months post-term
of cerebral palsy was 21% in children in the 1st quartile vs. 6% in the has no significant effect on later blood pressure and metabolic
4th quartile [14]. This difference persisted after adjusting for syndrome in adulthood. In contrast, growth during late infancy and
confounding variables affecting neurological outcome. childhood appears to be a major determinant of later metabolic
Many studies confirmed this observation. In the EPIPAGE and cardiovascular well-being [29].
1 study, deficient postnatal growth was found to be associated
with poor neurologic outcome for both AGA and SGA preterm 1.5. Impact of nutrition on body composition
infants. Specifically, risk of cerebral palsy was greater for those
born AGA and who lost more than 1 SD between birth and Fetal weight triples over the 3rd trimester, with a weight
6 months of age compared to those who grew adequately (aOR gain around 15 g/kg/day, consisting of 1.5–2.0 g/kg/day of fat,

Please cite this article in press as: Darmaun D, et al. Parenteral nutrition for preterm infants: Issues and strategy. Archives de Pédiatrie
(2018), https://doi.org/10.1016/j.arcped.2018.02.005
G Model
ARCPED-4621; No. of Pages 9

D. Darmaun et al. / Archives de Pédiatrie xxx (2018) xxx–xxx 3

1.5–2.0 g/kg/day of protein, and 9–12 g/kg/day of water and dingerbroek et al. randomized preterm infants (mean weight,
minerals. Fat mass gain is to be expected over the first 3 months of 862 g; mean gestational age, 27 weeks) into three groups: the
extrauterine life in the preterm infant, but such accretion should be controls received from the 1st day of life 2.4 g of amino acids/kg/
much lower than the 4 g/kg/day observed in breastfed term day, and two other groups received either the same dose of amino
infants: the percent of fat mass in preterm infants at hospital acids combined with 2–3 g lipid/kg/day or 3.6 g amino acids/kg/
discharge is actually higher than in term neonates in several [30– day plus 2–3 g lipid/kg/day. The two intervention groups had a
32] but not all studies [33–35]. Visceral fat mass was found better nitrogen balance, the intake at 3.6 g/kg/day from the outset
increased in one study [36], whereas a more recent study found no did not yield any additional benefit [48]. Morgan et al. compared
evidence for abnormal fat distribution [35]. Of interest is the PN supplying 2.8 g/kg/day amino acids (and 2.8 g/kg/day lipid) vs.
finding that early body composition changes may be associated 3.8 g/kg/day amino acids (and 3.8 g/kg/day lipid) from 6 h of life in
with long-term neurological development: the greater the lean premature infants under 1200 g and 29 weeks of gestational age
body mass accretion, the better the neurodevelopment at 1 year [49]. The growth of head circumference – an index of brain growth
corrected age [37]. Finally, the respective roles of enteral and [50] correlated with neurodevelopment [51] – between day 1 and
parenteral feeding in later body composition are unclear, but most day 28 was higher (with a gain of 0.37 Z-score) with high amino
recent randomized controlled trials suggest that manipulation of acid intake. However, this study does not allow us to delineate the
parenteral nutrition has little effect on lean body mass accretion respective benefits of amino acid and lipid enrichment. Discrepant
[38]. results were obtained in a more recent study: 168 preterm infants
It has long been felt that nutrition in preterm infants may alter under 31 weeks of gestational age received PN supplying either
health outcomes. The ultimate goal is to avoid the occurrence of 3.6 g/kg/day of amino acids from the 1st day (immediate group) or
EUGR, so that no subsequent catch-up growth is needed. The aim of 1.7 g/kg/day on day 1 with a gradual increase up to 2.7 g/kg/day on
the current position paper is to review the evidence supporting this the 3rd day (gradual group), the lipids consisting either of soybean
strategy with regards to the early phase of nutrition, which is oil (Intralipid1) or a mixture of oils (olive, soybean, medium-chain
mainly covered by parenteral nutrition. This narrative review triglyceride, fish oil, SMOF1). No difference was observed on
reflects the opinion of the Committee on Nutrition of the French growth or body composition, as assessed by magnetic resonance
Society of Pediatrics (SFP) and the French Society of Neonatology imaging (MRI), nor on liver fat accumulation, perhaps due to lack of
(SFN), taking into account recommendations from international statistical power. A disturbing finding was the lesser gain in head
expert groups. Although vitamin and micronutrient intake may be circumference in the immediate group [49]. It should be noted that
equally important and are obviously needed in any regimen of in most studies the intervention consists of a modification of amino
parenteral nutrition, it should be remembered that the current acid intakes rather than an overall strategy, and that the effect is
narrative review only addresses issues related to macronutrients, most often only assessed in the short term.
water, electrolytes, and minerals.

2.2. When should amino acid supply be initiated?

2. What strategy should be adopted for parenteral amino acid
supply in preterm infants? In the 1990s, amino acid supply was not started on the 1st day of
life, for fear of hyperammonia. There is in fact no increased risk of
Amino acids, the building blocks of body protein, are the driving acidosis with early amino acid intake [46]. In the retrospective
force for growth. In the absence of nitrogen supply, premature analysis of a cohort of more than 1000 preterm infants, those who
babies lose 1–2% of their protein stores per 24 h [39], and the reached an amino acid intake of 3 g/kg/day prior to day 4 had a
nitrogen balance increases as a function of amino acid intake [40]. twofold lower risk of having a head circumference under the 10th
percentile at discharge; this difference was significant only in boys
2.1. What amount of amino acids is needed? [52]. In a randomized prospective study, Van Goudoever et al.
demonstrated that starting amino acid intake in the early hours of life
In the term fetus, the net transfer of amino acids from mother to does not result in hyperammonemia or acidosis, and improves the net
fetus (measured by infusing labeled leucine during elective amino acid balance as measured by perfusion of 13C-leucine
caesarean sections) ranges between 3.1 [41] and 4.5 g/kg fetal [53]. Compared to PN glucose alone, the initiation of amino acids
weight/day [42]. (2.4 g/kg/day) in the first 2 h of life in 32 premature babies with an
Although carried out in enteral nutrition, the landmark study by average weight of 923 g improved not only protein synthesis and
Lucas et al. deserves to be cited as one of the first randomized nitrogen balance on the 2nd day of life [42], but the synthesis of
artificial nutrition studies in preterm infants: increasing formula glutathione as well [54]. Preserving glutathione is relevant since that
protein content from 1.5 to 2.0 g/dL produced a 12-point benefit on tripeptide is the body’s first antioxidant defense system, and
verbal IQ at 8 years, and 6 points on overall IQ, solely in boys [43]; glutathione deficiency is present at birth in premature infants [55],
this benefit was not confirmed by studies comparing neurodeve- even before they are confronted with high oxygen pressure, artificial
lopment before and after implementation of protocols designed to ventilation, or sepsis, which are all known to exacerbate oxidative
increase protein intake in the 1st month [44,45]. stress. Starting amino acids at 3.6 g/kg/day (rather than 2.4 g/kg/day)
Regarding PN, Thureen et al. randomly assigned 28 preterm did confer additional benefit, provided lipids were started early
infants (with an average weight of 946 g) to PN, providing either [48]. The long-term benefit is less clear: the growth of children
1 or 3 g/kg/day amino acids at 52 h of life. Protein gain was higher observed at 2 years was no different, and boys who had earlier intakes
in the high amino acid group, due to increased rates of protein of amino acids were more likely to achieve optimal neurodevelop-
synthesis, measured using 13C-leucine infusion [46]. In 142 chil- ment (odds ratio, 3.8; 95%CI, 1.3–11.4), but the development index
dren born before 29 weeks of gestational age, in a randomized trial was lower for girls who had survived without disability [56].
Tan et al. compared conventional PN to intensified PN, in an
attempt to approach the recommended intakes as closely as 2.3. What are the risks of an increase in early protein intake?
possible. They observed a correlation between cumulative energy
and protein deficit and head circumference at 36 weeks post- In the short term, increased protein intake exposes the infant to
conceptional age, but no difference between groups [47]. Vlaar- the immediate risk of hypophosphatemia. In 154 preterm infants

Please cite this article in press as: Darmaun D, et al. Parenteral nutrition for preterm infants: Issues and strategy. Archives de Pédiatrie
(2018), https://doi.org/10.1016/j.arcped.2018.02.005
G Model
ARCPED-4621; No. of Pages 9

4 D. Darmaun et al. / Archives de Pédiatrie xxx (2018) xxx–xxx

less than 33 weeks of gestational age, the incidence of hypo- glutamine and proline, arginine depletion was observed in preterm
phosphatemia was 12.5% in those with an amino acid intake in the infants with necrotizing enterocolitis (NEC) [69]. In two studies,
upper tercile vs. 4.6% and 0% in the mid and lower terciles, arginine supplementation reduced the incidence of NEC to five
respectively [57]. Hypophosphoremia is thought to result from the cases out of 76, vs. 21 out of 77 in controls in the first study [70],
rise in phosphorus utilization for glucose conversion to glucose-6- and from 18.6 to 2.5% in the second one [71]. Such evidence is
phosphate and higher ATP utilization for peptide bonds when unconvincing since: (1) arginine depletion could be not the cause,
protein synthesis increases, a mechanism analogous to the but the consequence of NEC since gut ischemia may impair the
refeeding syndrome known to occur upon rapid nutritional intestinal production of citrulline, and, consequently, arginine; (2)
replenishment in severely undernourished patients [58]. Regarding in the trials, the incidence of NEC was unusually high in the control
hypercalcemia, the incidence was 9.8%, 4.8%, and 1.9%, respectively groups.
[57]. In the long run, the potential risks of excessive catch-up Cysteine, a nonessential sulfur-containing amino acid synthe-
growth promoted by an increase in early protein intake are still sized from methionine, serves as a precursor of glutathione (GSH,
being debated. It is not known whether the risk of premature tripeptide glutamyl-cysteinyl-glycine), the most abundant antiox-
infants developing insulin resistance is related to premature birth idant in the body. Whether cysteine is essential for the PN-fed
per se, EUGR during neonatal stay, or subsequent catch-up. preterm infants is unclear: (1) the activity of cystathionase, a key
enzyme for cysteine synthesis, is low in fetal liver; (2) methionine
is less well converted to cysteine when methionine is provided by
2.4. Is there a requirement for specific amino acids? NP than enterally [72]; (3) premature infants experience major
oxidative stress and GSH deficiency; and (4) due to poor stability,
The composition of PN amino acid mixtures designed for little cysteine is included in PN mixtures. However, cysteine
preterm infants differs from that for adults since (1) proteins differ supplementation of PN failed to correct GSH deficiency [73]. Cys-
in amino acid composition between tissues, and the relative teine deficiency may in fact occur prior to birth, since it is observed
proportion of the various organs differs widely between adults and in the blood of mothers delivering preterm infants [55].
infants; (2) due to enzyme immaturity, amino acids deemed Taurine, a sulfo-amino acid not incorporated in proteins, is
nonessential for adults become conditionally essential in preterm essential to conjugate and excrete bile acids. In addition, taurine
infants. The composition of the mixtures was chosen to mimic deficiency alters retinal function, and neurodevelopment at 7 years
either the profile of an ideal protein – human milk protein – in the of age correlates with neonatal taurine supply [74]. Taurine
case of Vaminolact1 (Fresenius Kabi Bad Homburg, Germany) or therefore is essential in preterm PN.
that of umbilical cord blood in term neonates as for Primene1
(Baxter, Guyancourt, France), while taking into account constraints
related to the solubility or stability of amino acids. When the 3. What strategy should be adopted regarding energy supply in
newborn is fed enterally, 50–75% of glutamine and glutamate, 70% preterm PN?
of threonine (synthesis of intestinal mucins) [59–61] are extracted
in the first pass in the splanchnic territory (intestine + liver). As the 3.1. What is the glucose requirement of preterm infants?
latter territory is shunted when the child receives PN, amino acid
requirements may well differ between PN and enteral nutrition. In the neonate, brain accounts for 90% of whole-body glucose
Glutamine, a nonessential amino acid synthesized in muscle, is a utilization [75]. Endogenous (hepatic) glucose production corre-
prominent fuel for rapidly dividing cells such as those of intestinal lates with brain volume across ages, and reaches 6 mg/kg/min
mucosa and the immune system, and becomes conditionally (8.6 g/kg/day) and 3–5 mg/kg/min in the preterm and term infant,
essential under conditions of stress: glutamine stores are depleted, respectively, vs. 2.2 mg/kg/min (3 g/kg/day) in the adult [76]. This
at a time when splanchnic glutamine utilization increases [62] and high brain glucose requirement explains why hypoglycemia is
outstrips glutamine synthesis in skeletal muscle. Because of poor common in preterm infants, and why exposure to even mild or
heat stability, glutamine is not part of any of the PN mixtures, but asymptomatic hypoglycemia (< 45 mg/dL) in early life correlates
can be provided as alanyl-glutamine, a stable dipeptide precursor. with impaired neurodevelopment at 18 months of age [77]. In
Glutamine supplementation is a grade A recommendation for PN in healthy adults who have fasted overnight, two-thirds of hepatic
critically ill adults [63]. Yet a recent study found an increase in glucose production arises from glycogenolysis and one-third from
mortality in a group of septic and critically ill adults receiving gluconeogenesis. Gluconeogenesis does not operate in fetal life.
simultaneous intravenous and enteral glutamine supply [64]. Pre- The physiological drop in blood glucose associated with birth
term infants appear likely candidates for glutamine supplementa- acutely elicits glucagon secretion, and inhibits insulin secretion,
tion since (1) fetal glutamine uptake exceeds that of all other resulting in the induction of PEPCK [78,79], which in turn initiates
amino acids in utero, and the cutoff of umbilical supply results in gluconeogenesis from amino acids released by proteolysis in
abrupt glutamine deprivation at birth; (2) although preterm skeletal muscle. Twenty-four hours after birth, glucose de novo
infants are able to synthesize glutamine [60], low muscle mass synthesis capacity is fully operational in very preterm infants [80].
limits glutamine synthesis capacity, at a time when the developing The lipid supply is crucial for the maintenance of blood glucose
gut avidly takes up glutamine; (3) the preterm infant is exposed to in preterm infants. Though the fat supply does not enhance glucose
major stress; and (4) in addition to its putative role in protein production, the lipid supply decreases glucose utilization [81]. Con-
homeostasis, glutamine plays multiple roles, e.g., in the replenish- versely, over the course of intravenous nutrition delivering amino
ment of substrates in the tricarboxylic acid (Krebs) cycle. acids, glucose, and fat, interrupting lipid supply impairs gluconeo-
Supplementation of PN with glutamine (0.5 g/kg/day) inhibits genesis (mainly from glycerol), whereas interruption of the amino
proteolysis [65,66] but did not reduce the risk of mortality or sepsis acid supply does not [82].
in 1423 preterm infants [67]. In contrast, early life enteral In the adult, intravenous glucose infusion decreases liver
glutamine supplementation was reported to be associated with glucose production in a dose-dependent manner [83]. This brake
improved brain growth [68]. No consensus has been reached to on endogenous glucose production does not operate in preterm
date regarding glutamine supply in preterm infants. infants: residual glucose production arising from gluconeogenesis
Although arginine can be produced in part by the renal persists despite exogenous glucose supply [84–86], even during PN
conversion of citrulline, an amino acid synthesized in the gut from supplying all three macronutrients, including glucose [86]. Even

Please cite this article in press as: Darmaun D, et al. Parenteral nutrition for preterm infants: Issues and strategy. Archives de Pédiatrie
(2018), https://doi.org/10.1016/j.arcped.2018.02.005
G Model
ARCPED-4621; No. of Pages 9

D. Darmaun et al. / Archives de Pédiatrie xxx (2018) xxx–xxx 5

though high rates of glucose delivery are required to meet energy Conversely, although PN-induced liver disease clearly is
needs, such high rates are often poorly tolerated, resulting in multifactorial, excess phospholipid and lipid intake is a contribut-
hyperglycemia. Hyperglycemia is associated with increased ing factor [97]. The mechanisms mediating intravenous lipid liver
mortality [87]. Glucose uptake does not require insulin in brain, toxicity are debated. Plasma phytosterol concentration is elevated
which is by far the main glucose user. Glucose utilization therefore in children receiving long-term PN, and correlates with the severity
is poorly sensitive to insulin in preterm infants. In addition, in of cholestasis. The abundance of unsaturated n-6 fatty acids in soy
preterm infants both insulin sensitivity and insulin secretion are oil could also enhance inflammation given that n-6 PUFA are
impaired, the latter due to insufficient conversion of proinsulin to precursors for the synthesis of pro-inflammatory prostaglandins.
insulin [79]. Accordingly, elevated C-reactive protein, suspicion of sepsis, or
Such relatively poor insulin sensitivity likely explains the lack high bilirubin concentration were long considered contraindica-
of benefit of insulin treatment in situations associated with tions for lipid supply.
hyperglycemia in very preterm infants [88]. It remains a matter of The presumed toxicity of classic lipid admixtures led to the
debate whether (i) initiating insulin treatment, (ii) reducing development of emulsions enriched with medium-chain triglyc-
glucose delivery to match brain glucose requirement (6 mg/kg/ erides, thought to be less toxic for liver. Regarding protein gain,
min), or (iii) supplying medium-chain triglycerides is the best medium-chain triglycerides, however, are less efficient than long-
option to tackle hyperglycemia in PN-fed preterm infants. The chain triglycerides [98].
early supply of amino acids may decrease the incidence of
hyperglycemia in PN-fed preterm infants [89] through as yet 3.5. What specific lipid should be delivered?
unidentified mechanisms (enhanced insulin sensitivity or stimu-
lation of insulin secretion by amino acids). LA (n-6) and ALA (n-3), the two fatty acids essential regardless
of age, probably are not the only essential fatty acids for preterm
3.2. Lipid delivery in PN-fed preterm infants: what is at stake? infants.
Arachidonic (ARA; C20:4n-6) and docosahexaenoic acid (DHA;
In preterm infants, there is a clear rationale for parenteral lipid C22:3n-3) are the most abundant PUFAs in neurons. In utero, the
supply: (1) delivering glucose as a sole source of non-protein fetal LC-PUFA supply is ensured through several mechanisms,
energy, even at a rate of 12 g/kg/day only supplies 48 kcal/kg/day, including preferential placental transfer [90]. Compelling evidence
less than half the overall energy need; adding lipid covers energy suggests long-chain polyunsaturated fatty acids (LC-PUFAs), which
needs, at a lesser cost in terms of osmolarity; (2) excess glucose are synthesized in vivo from LA and ALA, are conditionally essential
delivery exposes the infant to the risk of hyperglycemia (see above) in preterm infants:
or steatosis; (3) a lipid supply is required for the physiological
accretion of body fat that normally occurs during the last trimester  although preterm infants are able to synthesize LC-PUFAs from
of intrauterine life [90]; (4) linoleic acid (LA; C18:2n-6) and a- their 18-carbon precursors (LA and ALA) [99], endogenous DHA
linolenic acid (ALA; C18:3n-3), two 18-carbon fatty acids, are synthesis (13 mg/kg/day in preterm infants born at 32 weeks
indispensable regardless of age, since they are not synthesized in gestational age) is much lower than fetal DHA accretion in the
the human body; (5) the synthesis of longer fatty acids from such 3rd trimester of pregnancy (43 mg/kg/day) [100]; the preterm
precursors is limited; and (6) the long-chain polyunsaturated fatty infant is therefore dependent on an exogenous DHA supply;
acid docosahexaenoic acid (DHA) accounts for 56% of the lipid mass  even though human milk covers the LC-PUFA requirement in
of neuron membranes, which makes it essential for brain and term infants, it likely is insufficient for preterm infants due to
retina development. their fast brain growth;
 even in preterm infants fed human milk or DHA-enriched
3.3. When should parenteral lipid supply be started in preterm preterm formula, intravenous supply accounts for a larger
infants? fraction of overall nutritional support than enteral nutrition over
the first few weeks of life, and most fat emulsions used in PN are
When lipid supply is delayed, the concentration of essential devoid of LC-PUFAs;
fatty acids (LA and ALA) declines by 0.75% per day, and deficiency  in rodents and non-human primates, deficient DHA deposition
appears [91]. Starting lipid delivery before 48 h of life is well in retina and brain impairs vision and neuromotor development
tolerated, but is not associated with any growth benefit, regardless [101–103].
of the emulsion used [92].
Although the PUFA supply appears to be safe, the benefit of LC-
3.4. How much fat should be delivered? PUFA supplementation by intravenous lipids on neurodevelop-
ment has yet to be established [102].
A minimum lipid intake of 0.5–1 g/kg/day is theoretically Finally, supplementation with DHA alone without arachidonic
sufficient to cover the essential fatty acid requirement. A acid (C20:4;n-6) results in impaired growth in terms of weight
retrospective study on 121 preterm infants nevertheless reports gain, length, and head circumference in the 1st year of life [103].
an association between the amount of fat delivered during the 1st
week and growth at 28 days of age, and a negative correlation
between growth in the 1st month and the delay in introducing lipid 3.6. Can n-3 polyunsaturated fatty acids prevent PN-induced
emulsions in preterm PN [93]. In a cohort of 48 very preterm cholestasis in preterm infants?
infants, neurodevelopment assessed by the Brunet-Lézine test at
1 year of age correlated with cumulative fat intake in the first A meta-analysis of three studies including 93 subjects conclud-
14 days, whereas it did not with amino acid supply [94]; of note, ed that PN including fish oil is more effective than PN containing
protein intake (3 g/kg/day) was close to recommended intake in standard lipid emulsion to treat cholestasis when the latter is
the latter study. already present (OR, 6.14; 95% CI, 2.27–16.6; P <0.01) [104]. In a
Regarding protein anabolism, a 50/50 glucose/fat energy ratio meta-analysis of 17 studies conducted in preterm infants or
for non-protein energy supply was more efficient than supplying critically ill neonates with a short duration of intervention, no
100% as glucose [95], and a 66/33 ratio appears optimal [96]. differences in the rate of cholestasis or bilirubin levels were

Please cite this article in press as: Darmaun D, et al. Parenteral nutrition for preterm infants: Issues and strategy. Archives de Pédiatrie
(2018), https://doi.org/10.1016/j.arcped.2018.02.005
G Model
ARCPED-4621; No. of Pages 9

6 D. Darmaun et al. / Archives de Pédiatrie xxx (2018) xxx–xxx

associated with short-term use of different intravenous lipid 4.2. What sodium intake is required?
emulsions [105]. Another meta-analysis concluded that all lipid
emulsions appeared to be safe in preterm infants, but found no The sodium balance is an important issue in the extreme
statistically significant differences in clinically important outco- premature infant. Wide variations in serum sodium are associated
mes such as death, growth, and PN-associated liver disease with with a worse prognosis at 2 years [108]. In the first few days, a
the use of newer alternative emulsions vs. the pure soy oil-based physiological contraction of extracellular volume takes place
emulsion [106]. Pure soybean oil lipid emulsions are considered [109]. Since sodium is the predominant cation in extracellular
less balanced (higher omega-6/omega-3 ratio, higher phytosterol medium, a physiological loss of sodium is therefore present in the
content, lower alpha-tocopherol content) than composite lipid first few days of life. If sodium intake is lower than the excess loss
emulsions; composite lipid emulsions with or without fish oil due to impaired sodium renal reabsorption, there is a risk of
should therefore be the first choice in preterm infants. hyponatremia due to sodium depletion. A randomized trial
Taken together, the literature data suggest that in the showed a benefit of sodium intake from the first few days on
pathophysiology of cholestasis, the omega-6/omega-3 ratio in outcome at 10 years of age [110], even though this strategy is
fatty acid supply could matter less than other factors such as associated with increased initial oxygen requirement [111].
the glucose/fat energy ratio. In a prospective trial on After the first hours to days of life, which are characterized by
31 patients suffering from PN-induced liver disease, a simple initial relative oliguria, a diuretic phase lasting a few days will
reduction of fat supply to 1 g/kg twice a week improved drive the sodium losses. All things considered, a restriction of Na
cholestasis, compared with historical control patients who intake less than 2 mEq/kg/day should be considered from the first
had received the same emulsion at a rate of 3 g/kg/day, albeit day but followed by a rapid increase in sodium intake (4 mEq/Kg/
with a trend toward a rise in the triene/tetraene ratio, an day or above adjusted to sodium losses) when sodium losses
index of subclinical essential fatty acid deficiency [107]. It increase because of increased urine output [110,112,113]. This
should, however, be remembered that lipid reduction may sodium intake must take into account the hidden sodium intake
lead to energy shortage in preterm infants and should not be associated with drugs. In the days following birth, preterm infants
promoted to a large extent. are exposed to two contrasting risks: dilution hyponatremia due to
oligoanuria secondary to drug toxicity (antibiotics, ibuprofen) on
one hand, and hypernatremia through transcutaneous loss of
4. What strategy is best for water and electrolyte supply? extracellular water due to extreme cutaneous immaturity [114].

4.1. What water intake is necessary? 4.3. What chlorine intake is required?

Water and electrolyte balance of the preterm infant is Chlorine intake is mostly not prescribed but passively
influenced by: administered when potassium and phosphorus are prescribed in
the form of chlorides, which often leads to excess intake [115],
 substantial insensible, cutaneous, and respiratory water loss. responsible for hyperchloremic metabolic acidosis [116]. Control-
Cutaneous loss is very high during the first few days of life due to ling this intake using other sodium, potassium, or phosphorus salts
cutaneous immaturity, particularly in case of extreme prema- reduces the risk of metabolic acidosis [117]. As a rule of thumb,
turity, warranting high water intakes. The infant’s environment, total chloride intake should be lower than the addition of sodium
designed to warm the infant, plays an important role: a relative and potassium intakes in order to avoid hyperchloremic metabolic
humidity of 80% or more must be sought, best obtained in an acidosis.
enclosed incubator;
 the contraction of the extracellular compartment associated 4.4. What potassium intake is needed?
with the transition to extrauterine life, implies that a
physiological weight loss should be allowed in the days Potassium is an intracellular cation: 98% of potassium stores are
following birth, but must be strictly controlled; intracellular. The intracellular concentration ranges between
 renal immaturity, responsible for a loss of sodium and water, in 100 and 150 mmol/L between tissues, whereas potassium
fact facilitating the transition to extrauterine life in the first few concentration in the extracellular compartment is 4.5 mmol/L.
days, but requiring close monitoring, followed by high hydro- Tissue catabolism representing 1% of body cell mass transfers
sodium intakes; approximately 2.5 mmol of potassium to the extracellular sector.
 needs related to growth. Thus, kalemia reflects not only (1) potassium intake but also (2)
potassium flux from the intracellular to the extracellular
An excess of water and sodium intake increases the risk of compartment (attrition, catabolism), which increases serum
ductus arteriosus and bronchopulmonary dysplasia. Insufficient potassium, (3) the passage of ions from the extracellular to the
water and sodium intake exposes the child to a risk of dehydration, intracellular compartment (anabolism), and (4) urinary potassium
particularly in the case of carbohydrate intolerance, because loss and potassium secretion into intestinal lumen, which lower
glycosuria increases water loss. serum potassium [118]. PN with higher and earlier amino acid and
Water intake during the first few days of life is determined energy intake will tend to lower serum potassium, avoid
according to an individualized progression, taking into account nonoliguric hyperkalemia, and above all improve the potassium
changes in the infant’s weight, serum sodium, and to a lesser balance [119]. Nevertheless, in a recent study, the increased intake
extent urine output (which is difficult to interpret due to renal of amino acids did not lead to changes in serum electrolyte
immaturity). Most often, initial intakes of 70 mL/kg/24 h (50– concentrations [120]. The recommended intake of potassium is 2–
80 mL/kg/24 h, higher for lower gestational age) are proposed, 4 mEq/kg/day after urine output is well established [112].
followed by a progression adapted depending on the infant’s
characteristics, thermal environment, sodium intakes, type of PN 4.5. What phosphorus and calcium intakes?
solution used, and clinical and biological parameters, up to
approximately 140 mL/kg/day (100–200 mL/kg/24 h, depending The phosphorus/calcium balance requires managing calcium
on the variables above) at the end of the 1st week. and phosphorus intake at the same time. Phosphorus is an

Please cite this article in press as: Darmaun D, et al. Parenteral nutrition for preterm infants: Issues and strategy. Archives de Pédiatrie
(2018), https://doi.org/10.1016/j.arcped.2018.02.005
G Model
ARCPED-4621; No. of Pages 9

D. Darmaun et al. / Archives de Pédiatrie xxx (2018) xxx–xxx 7

intracellular ion. High amino acid and energy intake from the 1st A.L.: occasional conferences: invitation as speaker for Baxter SA,
day can lead to hypophosphatemia, hypercalcemia, and hyper- Fresenius Kabi; occasional consultancies: expert reports for Baxter
calciuria [56]. The calcium balance therefore may become SA, Fresenius Kabi.
negative, whereas most of the calcium accretion observed in the U.S.: occasional conference as invited speaker for Baxter SA.
fetus (30 g over the entire duration of pregnancy) occurs during the A.Bo., A.Br., D.T., E.S., F.F., J.-C.R., J.-P.C., J.-P.G., M.L.F., R.H.
3rd trimester [121]. Intakes of 1–2 mmol/kg/day for phosphorus declare that they have no competing interest.
and 1–2 mmol/kg/day for calcium must be given to accommodate
the enhanced growth rate due to high amino acid intake, as early as
the first few days of life [112,113]. During the first days of life when References
mineral intake is usually low and amino acid intake high, a molar [1] Delnord M, Blondel B, Zeitlin J. What contributes to disparities in the preterm
calcium-to-phosphorus ratio at 1 or below can be used to prevent birth rate in European countries? Curr Opin Obstet Gynecol 2015;27:133–42.
the occurrence of early hypophosphatemia. [2] Blencowe H, Cousens S, Oestergaard MZ, et al. National, regional, and world-
wide estimates of preterm birth rates in the year 2010 with time trends since
1990 for selected countries: a systematic analysis and implications. Lancet
5. Conclusions 2012;379:2162–72.
[3] Vayssière C, Sentilhes L, Ego A, et al. Fetal growth restriction and intra-uterine
For the very preterm infant, optimizing PN is a crucial issue, growth restriction: guidelines for clinical practice from the French College of
Gynaecologists and Obstetricians. Eur J Obstet Gynecol Reprod Biol
since PN is the main source of nutrient intake for the first 2 weeks 2015;193:10–8.
of life. The ultimate goal of nutrition during hospitalization is to [4] Ancel PY, Goffinet F, EPIPAGE-2 Writing Group. et al. Survival and morbidity
ensure a rate of growth close to that of the fetus during the 3rd of preterm children born at 22 through 34 weeks’ gestation in France in 2011:
results of the EPIPAGE-2 Cohort Study. JAMA Pediatr 2015;169:230–8.
trimester of pregnancy in order to ensure optimal brain growth. [5] Karagol BS, Zenciroglu A, Okumus N, et al. Randomized controlled trial of slow
This objective is often difficult to achieve, and a cumulative energy vs rapid enteral feeding advancements on the clinical outcomes of preterm
and protein deficit during the hospital stay can lead to extrauterine infants with birth weight 750-1250 g. JPEN J Parenter Enteral Nutr
2013;37:223–8.
growth restriction (EUGR). Recent studies suggest that more [6] Martin CR, Brown YF, Ehrenkranz RA, et al. Nutritional practices and growth
optimal nutrition through both enteral and intravenous routes velocity in the first month of life in extremely premature infants. Pediatrics
may prevent EUGR. 2009;124:649–57.
[7] Horbar JD, Ehrenkranz RA, Badger GJ, et al. Weight growth velocity and
A striking association between growth rate in the first few postnatal growth failure in infants 501 to 1500 grams: 2000-2013. Pediatrics
weeks and long-term neurodevelopment has been observed in 2015;136:e84–92.
most retrospective observational studies. This correlation is, as [8] Griffin IJ, Tancredi DJ, Bertino E, et al. Postnatal growth failure in very low
birthweight infants born between 2005 and 2012. Arch Dis Child Fetal
such, subjected to multiple sources of bias, and a correlation does
Neonatal Ed 2016;101:F50–5.
not necessarily imply a causal relationship, not to mention the [9] Roggero P, Giannı̀ ML, Orsi A, et al. Implementation of nutritional strategies
paucity of prospective, randomized trials. decreases postnatal growth restriction in preterm infants. PLoS One
Amino acid supply is the driver of growth, and should be started 2012;7:e51166.
[10] Senterre T, Rigo J. L’alimentation parentérale du prématuré : comment la
from the time of birth, if possible at the rates recommended. There prescrire pour optimiser la croissance et le développement. Arch Pediatr
appears to be room for improvement in the composition of 2013;20:986–93.
parenteral amino acid admixtures. [11] Rigo J, Senterre T. Intrauterine-like growth rates can be achieved with
premixed parenteral nutrition solution in preterm infants. J Nutr
Attention should be paid to providing an adequate amount of 2013;143(Suppl.):2066–70.
other nutrients, such as electrolytes and minerals, since inadequa- [12] Loÿs CM, Maucort-Boulch D, Guy B, et al. Extremely low birth weight infants:
cy of these nutrients may alter short- and long-term outcomes. how neonatal intensive care unit teams can reduce postnatal malnutrition
and prevent growth retardation. Acta Paediatr 2013;102:242–8.
The non-protein energy supply should combine glucose and fat [13] Lapointe M, Barrington KJ, Savaria M, et al. Preventing postnatal growth
from the time of birth, and primarily aims at covering brain restriction in infants with birthweight less than 1300 g. Acta Paediatr
requirements for energy (glucose) and building materials (poly- 2016;105:e54–9.
[14] Ehrenkranz RA, Dusick AM, Vohr BR, et al. Growth in the neonatal intensive
unsaturated fatty acids). care unit influences neurodevelopmental and growth outcomes of extremely
In children born preterm, it remains to be elucidated whether low birth weight infants. Pediatrics 2006;117:1253–61.
the risk of developing chronic disease at adulthood is linked to [15] Larroque B, Ancel PY, Marchand-Martin L, et al. Special care and school
difficulties in 8-year-old very preterm children: the Epipage cohort study.
catch-up growth occurring after EUGR or poor quality of growth
PLoS One 2011;6:e21361.
(suboptimal fat/lean body mass accretion). Optimized, personal- [16] Guellec I, Lapillonne A, Marret S, et al. Effect of intra- and extrauterine growth
ized – rather than aggressive – intravenous nutrition may help on long-term neurologic outcomes of very preterm infants. J Pediatr
prevent both EUGR and suboptimal lean/fat mass accretion, and 2016;175:93–9. e1.
[17] Frondas-Chauty A, Simon L, Branger B, et al. Early growth and neurodeve-
may in turn improve long-term health outcomes. lopmental outcome in very preterm infants: impact of gender. Arch Dis Child
Even though uncertainty remains regarding the long-term Fetal Neonatal Ed 2014;99:F366–72.
effect of early nutrition, it appears reasonable to propose more [18] Barker DJ. The developmental origins of adult disease. J Am Coll Nutr
2004;23(Suppl.):588–95.
optimal initial PN and to follow the recommended intakes. More [19] Simeoni U, Bocquet A, Briend A, et al. L’origine précoce des maladies chro-
randomized trials are needed, however, to further support this niques de l’adulte. Arch Pediatr 2016;23:443–6.
approach and to more firmly demonstrate that this type of strategy [20] Lapillonne A, Griffin IJ. Feeding preterm infants today for later metabolic and
cardiovascular outcomes. J Pediatr 2013;162(Suppl.):7–16.
improves short-term and long-term outcomes. [21] Hofman PL, Regan F, Jackson WE, et al. Premature birth and later insulin
resistance. N Engl J Med 2004;351:2179–86.
Disclosure of interest [22] Hovi P, Andersson S, Eriksson JG, et al. Glucose regulation in young adults
with very low birth weight. N Engl J Med 2007;356:2053–63.
[23] Parkinson JR, Hyde MJ, Gale C, et al. Preterm birth and the metabolic
D.D.: occasional conferences: invitation as speaker for Baxter syndrome in adult life: a systematic review and meta-analysis. Pediatrics
SA, B Braun, Bledina, Danone, Fresenius Kabi, Nestlé, Novartis, 2013;131:e1240–63.
[24] Willemsen RH, Leunissen RW, Stijnen T, et al. Prematurity is not associated
Nutricia, Novartis SA; occasional consultancies: expert reports for
with reduced insulin sensitivity in adulthood. J Clin Endocrinol Metab
Baxter SA, B Braun, Bledina, Danone, Fresenius Kabi, Nestlé, 2009;94:1695–700.
Novartis, Nutricia, Novartis SA. [25] Singhal A, Cole TJ, Fewtrell M, et al. Is slower early growth beneficial for long-
D.D.: substantial contribution to research grants from: Baxter term cardiovascular health? Circulation 2004;109:1108–13.
[26] Tauzin L, Rossi P, Grosse C, et al. Increased systemic blood pressure and
SA, B Braun, Bledina, Danone, Fresenius Kabi, Nestlé, Novartis, arterial stiffness in young adults born prematurely. J Dev Orig Health Dis
Nutricia, Novartis SA. 2014;5:448–52.

Please cite this article in press as: Darmaun D, et al. Parenteral nutrition for preterm infants: Issues and strategy. Archives de Pédiatrie
(2018), https://doi.org/10.1016/j.arcped.2018.02.005
G Model
ARCPED-4621; No. of Pages 9

8 D. Darmaun et al. / Archives de Pédiatrie xxx (2018) xxx–xxx

[27] Mohlkert LA, Hallberg J, Broberg O, et al. Preterm arteries in childhood: acid administration in preterm infants. J Pediatr Gastroenterol Nutr
dimensions, intima-media thickness, and elasticity of the aorta, coronaries, 2014;59:714–9.
and carotids in 6-y-old children born extremely preterm. Pediatr Res [57] Bonsante F, Iacobelli S, Latorre G, et al. Initial amino acid intake influences
2017;81:299–306. phosphorus and calcium homeostasis in preterm infants–it is time to
[28] Regan FM, Cutfield WS, Jefferies C, et al. The impact of early nutrition in change the composition of the early parenteral nutrition. PLoS One
premature infants on later childhood insulin sensitivity and growth. Pediat- 2013;8:e72880.
rics 2006;118:1943–9. [58] Stanga Z, Brunner A, Leuenberger M, et al. Nutrition in clinical practice-the
[29] Embleton ND, Korada M, Wood CL, et al. Catch-up growth and metabolic refeeding syndrome: illustrative cases and guidelines for prevention and
outcomes in adolescents born preterm. Arch Dis Child 2016;101:1026–31. treatment. Eur J Clin Nutr 2008;62:687–94.
[30] Roggero P, Giannı̀ ML, Amato O, et al. Is term newborn body composition being [59] Beaufrère B, Fournier V, Salle B, et al. Leucine kinetics in fed low-birth-weight
achieved postnatally in preterm infants? Early Hum Dev 2009;85:349–52. infants: importance of splanchnic tissues. Am J Physiol 1992;263:E214–20.
[31] Ramel SE, Gray HL, Ode KL, et al. Body composition changes in preterm [60] Darmaun D, Roig JC, Auestad N, et al. Glutamine metabolism in very low birth
infants following hospital discharge: comparison with term infants. J Pediatr weight infants. Pediatr Res 1997;41:391–6.
Gastroenterol Nutr 2011;53:333–8. [61] Riedijk MA, van Goudoever JB. Splanchnic metabolism of ingested amino
[32] Simon L, Borrego P, Darmaun D, et al. Effect of sex and gestational age on acids in neonates. Curr Opin Metabol Care 2007;10:58–62.
neonatal body composition. Br J Nutr 2013;109:1105–8. [62] Thibault R, Welch S, Mauras N, et al. Corticosteroids increase glutamine
[33] Roggero P, Giannı̀ ML, Forzenigo L, et al. No relative increase in intra- utilization in human splanchnic bed. Am J Physiol 2008;294:G548–53.
abdominal adipose tissue in healthy unstressed preterm infants at term. [63] Singer P, Berger MM, Van den Berghe G, et al. ESPEN Guidelines on Parenteral
Neonatology 2014;107:14–9. Nutrition: intensive care. Clin Nutr 2009;28:387–400.
[34] Johnson MJ, Wootton SA, Leaf AA, et al. Preterm birth and body composition [64] Heyland D, Muscedere J, Wischmeyer PE, et al. A randomized trial of
at term equivalent age: a systematic review and meta-analysis. Pediatrics glutamine and antioxidants in critically ill patients. N Engl J Med
2012;130:e640–9. 2013;368:1489–97.
[35] Roggero P, Giannı̀ ML, Forzenigo L, et al. No relative increase in intra- [65] Des Robert C, Le Bacquer O, Piloquet H, et al. Acute effects of intravenous
abdominal adipose tissue in healthy unstressed preterm infants at term. glutamine supplementation on protein metabolism in very low birth weight
Neonatology 2015;107:14–9. infants: a stable isotope study. Pediatr Res 2002;51:87–93.
[36] Uthaya S, Thomas EL, Hamilton G, et al. Altered adiposity after extremely [66] Kalhan SC, Parimi PS, Gruca LL, et al. Glutamine supplement with parenteral
preterm birth. Pediatr Res 2005;57:211–5. nutrition decreases whole body proteolysis in low birth weight infants. J
[37] Ramel SE, Gray HL, Christiansen E, et al. Greater early gains in fat-free mass, Pediatr 2005;146:642–7.
but not fat mass, are associated with improved neurodevelopment at 1 year [67] Poindexter BB, Ehrenkranz RA, Stoll BJ, et al. Parenteral glutamine supple-
corrected age for prematurity in very low birth weight preterm infants. J mentation does not reduce the risk of mortality or late-onset sepsis in
Pediatr 2016;173:108–15. extremely low birth weight infants. Pediatrics 2004;113:1209–15.
[38] Uthaya S, Liu X, Babalis D, et al. Nutritional evaluation and optimisation in [68] de Kieviet JF, Vuijk PJ, van den Berg A, et al. Glutamine effects on brain growth
neonates: a randomized, double-blind controlled trial of amino acid regimen in very preterm children in the first year of life. Clin Nutr 2014;33:69–74.
and intravenous lipid composition in preterm parenteral nutrition. Am J Clin [69] Zamora SA, Amin HJ, McMillan DD, et al. Plasma L-arginine concentrations in
Nutr 2016;103:1443–52. premature infants with necrotizing enterocolitis. J Pediatr 1997;131:226–32.
[39] Denne SC, Poindexter BB. Evidence supporting early nutritional support with [70] Amin HJ, Zamora SA, McMillan DD, et al. Arginine supplementation prevents
parenteral amino acid infusion. Semin Perinatol 2007;31:56–60. necrotizing enterocolitis in the premature infant. J Pediatr 2002;140:425–31.
[40] Zello GA, Menendez CE, Rafii M, et al. Minimum protein intake for the [71] Polycarpou E, Zachaki S, Tsolia M, et al. Enteral L-arginine supplementation
preterm neonate determined by protein and amino acid kinetics. Pediatr for prevention of necrotizing enterocolitis in very low birth weight neonates:
Res 2003;53:338–44. a double-blind randomized pilot study of efficacy and safety. JPEN J Parenter
[41] Chien PF, Smith K, Watt PW, et al. Protein turnover in the human fetus studied at Enteral Nutr 2013;37:617–22.
term using stable isotope tracer amino acids. Am J Physiol 1993;265:E31–5. [72] Stegink LD, Den Besten L. Synthesis of cysteine from methionine in normal
[42] Van den Akker CH, Schierbeek H, Minderman G, et al. Amino acid metabolism adult subjects: effect of route of alimentation. Science 1972;178:514–6.
in the human fetus at term: leucine, valine, and methionine kinetics. Pediatr [73] te Braake FW, Schierbeek H, Vermes A, et al. High-dose cysteine administra-
Res 2011;70:566–71. tion does not increase synthesis of the antioxidant glutathione preterm
[43] Lucas A, Morley R, Cole TJ. Randomised trial of early diet in preterm babies infants. Pediatrics 2009;124:e978–84.
and later intelligence quotient. BMJ 1998;317:1481–7. [74] Wharton BA, Morley R, Isaacs EB, et al. Low plasma taurine and later
[44] Cester EA, Bloomfield FH, Taylor J, et al. Do recommended protein intakes neurodevelopment. Arch Dis Child Fetal Neonatal Ed 2004;89:F497–8.
improve neurodevelopment in extremely preterm babies? Arch Dis Child [75] Haymond MW, Sunehag A. Controlling the sugar ball: regulation of glucose
Fetal Neonatal Ed 2015;100:F243–7. homeostasis in children. Endocrinol Metab Clin N Am 1999;28:463–94.
[45] Bellagamba MP, Carmenati E, D’Ascenzo R, et al. One extra gram of protein to [76] Bier DM, Leake RD, Haymond MW, et al. Measurement of ‘‘true’’ glucose
preterm infants from birth to 1800 g: a single blinded randomized clinical production rates in infancy and childhood with 6,6-dideuteroglucose. Dia-
trial. J Pediatr Gastroenterol Nutr 2016;62:879–84. betes 1977;26:1016–23.
[46] Thureen PJ, Melara D, Fennessey PV, et al. Effect of low versus high intrave- [77] Lucas A, Morley R, Cole TJ. Adverse neurodevelopmental outcome of moder-
nous amino acid intake on very low birth weight infants in the early neonatal ate neonatal hypoglycaemia. BMJ 1988;297:1304–8.
period. Pediatr Res 2003;53:24–32. [78] Girard J. Gluconeogenesis in late fetal and early neonatal life. Biol Neonate
[47] Tan MJ, Cooke RW. Improving head growth in very preterm infants–a 1986;50:237–58.
randomised controlled trial I: neonatal outcomes. Arch Dis Child Fetal [79] Mitanchez-Mokhtari D, Lahlou N, Kieffer F, et al. Both relative insulin resis-
Neonatal Ed 2008;93:F337–41. tance and defective islet beta-cell processing of proinsulin are responsible for
[48] Vlaardingerbroek H, Vermeulen MJ, Rook D, et al. Safety and efficacy of early transient hyperglycemia in extremely preterm infants. Pediatrics
parenteral lipid and high-dose amino acid administration to very low birth 2004;113:537–41.
weight infants. J Pediatr 2013;163:638–44. e1-5. [80] Kalhan SC, Parimi P, Van Beek R, et al. Estimation of gluconeogenesis in
[49] Morgan C, McGowan P, Herwitker S, et al. Postnatal head growth in preterm newborn infants. Am J Physiol 2001;281:E991–7.
infants: a randomized controlled parenteral nutrition study. Pediatrics [81] Yunis KA, Oh W, Kalhan S, et al. Glucose kinetics following administration of
2014;133:e120–8. an intravenous fat emulsion to low-birth-weight neonates. Am J Physiol
[50] Lindley AA, Benson JE, Grimes C, et al. The relationship in neonates between 1992;263:E844–9.
clinically measured head circumference and brain volume estimated from [82] Sunehag AL. The role of parenteral lipids in supporting gluconeogenesis in
head CT-scans. Early Hum Dev 1999;56:17–29. very premature infants. Pediatr Res 2003;54:480–6.
[51] Cheong JL, Hunt RW, Anderson PJ, et al. Head growth in preterm infants: [83] Wolfe RR, Allsop JR, Burke JF. Glucose metabolism in man: responses to
correlation with magnetic resonance imaging and neurodevelopmental out- intravenous glucose infusion. Metabolism 1979;28:210–20.
come. Pediatrics 2008;121:e1534–40. [84] Sunehag A, Gustafsson J, Ewald U. Very immature infants ( 30 Wk) respond
[52] Poindexter BB, Langer JC, Dusick AM, et al. Early provision of parenteral to glucose infusion with incomplete suppression of glucose production.
amino acids in extremely low birth weight infants: relation to growth and Pediatr Res 1994;36:550–5.
neurodevelopmental outcome. J Pediatr 2006;148:300–5. [85] Chacko SK, Sunehag AL. Gluconeogenesis continues in premature infants
[53] Van Goudoever JB, Colen T, Wattimena JL, et al. Immediate commencement of receiving total parenteral nutrition. Arch Dis Child Fetal Neonatal Ed
amino acid supplementation in preterm infants: effect on serum amino acid 2010;95:F413–8.
concentrations and protein kinetics on the first day of life. J Pediatr [86] Sunehag AL, Haymond MW, Schanler RJ, et al. Gluconeogenesis in very low
1995;127:458–65. birth weight infants receiving total parenteral nutrition. Diabetes
[54] Te Braake FW, Schierbeek H, de Groof K, et al. Glutathione synthesis rates 1999;48:791–800.
after amino acid administration directly after birth in preterm infants. Am J [87] Hays SP, Smith EO, Sunehag AL. Hyperglycemia is a risk factor for early death
Clin Nutr 2008;88:333–9. and morbidity in extremely low birth-weight infants. Pediatrics
[55] Küster A, Tea I, Ferchaud-Roucher V, et al. Cord blood glutathione depletion in 2006;118:1811–8.
preterm infants: correlation with maternal cysteine depletion. PLoS One [88] Beardsall K, Vanhaesebrouck S, Ogilvy-Stuart AL, et al. Early insulin therapy in
2011;6:e27626. very-low-birth-weight infants. N Engl J Med 2008;359:1873–84.
[56] van den Akker CH, te Braake FW, Weisglas-Kuperus N, et al. Observational [89] Morgan C, Herwitker S, Badhawi I, et al. SCAMP: standardised, concentrated,
outcome results following a randomized controlled trial of early amino additional macronutrients, parenteral nutrition in very preterm infants: a

Please cite this article in press as: Darmaun D, et al. Parenteral nutrition for preterm infants: Issues and strategy. Archives de Pédiatrie
(2018), https://doi.org/10.1016/j.arcped.2018.02.005
G Model
ARCPED-4621; No. of Pages 9

D. Darmaun et al. / Archives de Pédiatrie xxx (2018) xxx–xxx 9

phase IV randomised, controlled exploratory study of macronutrient intake, [106] Kapoor V, Glover R, Malviya MN. Alternative lipid emulsions versus pure soy
growth and other aspects of neonatal care. BMC Pediatr 2011;11:53. oil based lipid emulsions for parenterally fed preterm infants. Cochrane
[90] Haggarty P. Effect of placental function on fatty acid requirements during Database Syst Rev 2015;CD009172.
pregnancy. Eur J Clin Nutr 2004;58:1559–70. [107] Cober MP, Killu G, Brattain A, et al. Intravenous fat emulsions reduction for
[91] Lee EJ, Simmer K, Gibson RA. Essential fatty acid deficiency in parenterally fed patients with parenteral nutrition-associated liver disease. J Pediatr
preterm infants. J Paediatr Child Health 1993;29:51–5. 2012;160:421–7.
[92] Vlaardingerbroek H, Veldhorst MA, Spronk S, et al. Parenteral lipid adminis- [108] Baraton L, Ancel PY, Flamant C, et al. Impact of changes in serum sodium
tration to very-low-birth-weight infants—early introduction of lipids and use levels on 2-year neurologic outcomes for very preterm neonates. Pediatrics
of new lipid emulsions: a systematic review and meta-analysis. Am J Clin 2009;124:e655–61.
Nutr 2012;96:255–68. [109] Modi N, Bétrémieux P, Midgley J, et al. Postnatal weight loss and contraction
[93] Fischer CJ, Maucort-Boulch D, Essomo Megnier-Mbo CM, et al. Early paren- of the extracellular compartment is triggered by atrial natriuretic peptide.
teral lipids and growth velocity in extremely-low-birth-weight infants. Clin Early Hum Dev 2000;59:201–8.
Nutr 2014;33:502–8. [110] Al-Dahhan J, Jannoun L, Haycock GB. Effect of salt supplementation of
[94] Dit Trolli SE, Kermorvant-Duchemin E, Huon C, et al. Early lipid supply and newborn premature infants on neurodevelopmental outcome at 10-13 years
neurological development at one year in very low birth weight (VLBW) of age. Arch Dis Child Fetal Neonatal Ed 2002;86:F120–3.
preterm infants 2012;88(Suppl.):25–9. [111] Hartnoll G, Bétrémieux P, Modi N. Randomised controlled trial of
[95] Bresson JL, Bader B, Rocchiccioli F, et al. Protein-metabolism kinetics and postnatal sodium supplementation on oxygen dependency and body weight
energy-substrate utilization in infants fed parenteral solutions with different in 25–30-week gestational age infants. Arch Dis Child Fetal Neonatal Ed
glucose-fat ratios. Am J Clin Nutr 1991;54:370–6. 2000;82:F19–23.
[96] Kashyap S, Towers HM, Sahni R, et al. Effects of quality of energy on substrate [112] Bolisetty S, Osborn D, Sinn J, et al. Standardised neonatal parenteral nutrition
oxidation in enterally fed, low-birth-weight infants. Am J Clin Nutr formulations - an Australasian group consensus 2012. BMC Pediatr
2001;74:374–80. 2014;14:48.
[97] Colomb V, Jobert-Giraud A, Lacaille F, et al. Role of lipid emulsions in [113] Gargasz A. Neonatal and pediatric parenteral nutrition. AACN Adv Crit Care
cholestasis associated with long-term parenteral nutrition in children. JPEN 2012;23:451–64.
J Parenter Enteral Nutr 2000;24:345–50. [114] Modi N. Management of fluid balance in the very immature neonate. Arch Dis
[98] Liet JM, Piloquet H, Marchini JS, et al. Leucine metabolism in preterm infants Child Fetal Neonatal Ed 2004;89:F108–11.
receiving parenteral nutrition with medium-chain compared with long- [115] Kermorvant-Duchemin E, Iacobelli S, Eleni-Dit-Trolli S, et al. Early chloride
chain triacylglycerol emulsions. Am J Clin Nutr 1999;69:539–43. intake does not parallel that of sodium in extremely-low-birth-weight
[99] Carnielli VP, Simonato M, Verlato G, et al. Synthesis of long-chain polyunsat- infants and may impair neonatal outcomes. J Pediatr Gastroenterol Nutr
urated fatty acids in preterm newborns fed formula with long-chain poly- 2012;54:613–9.
unsaturated fatty acids. Am J Clin Nutr 2007;86:1323–30. [116] Seifter JL. Integration of acid–base and electrolyte disorders. N Engl J Med
[100] Lapillonne A, Groh-Wargo S, Gonzalez CH, et al. Lipid needs of preterm 2014;371:1821–31.
infants: updated recommendations. J Pediatr 2013;162(Suppl.):37–47. [117] Richards CE, Drayton M, Jenkins H, et al. Effect of different chloride infusion
[101] Lauritzen L, Hansen HS, Jørgensen MH, et al. The essentiality of long chain n-3 rates on plasma base excess during neonatal parenteral nutrition. Acta
fatty acids in relation to development and function of the brain and retina. Paediatr 1993;82:678–82.
Prog Lipid Res 2001;40:1–4. [118] Lorenz JM, Kleinman LI, Markarian K. Potassiuium metabolism in
[102] Gould JF, Treyvaud K, Yelland LN, et al. Does n-3 LCPUFA supplementation extremely low birth weight infants in the first week of life. J Pediatr
during pregnancy increase the IQ of children at school age? Follow-up of a 1997;131:81–6.
randomised controlled trial. BMJ Open 2016;6:e011465. [119] Bonsante F, Iacobelli S, Chantegret C, et al. The effect of parenteral nitrogen
[103] Carlson SE. Docosahexaenoic acid and arachidonic acid in infant develop- and energy intake on electrolyte balance in the preterm infant. Eur J Clin Nutr
ment. Semin Neonatol 2001;6:437–49. 2011;65:1088–930.
[104] Park HW, Lee NM, Kim JH, et al. Parenteral fish oil-containing lipid emulsions [120] Elstgeest LE, Martens SE, Lopriore E, et al. Does parenteral nutrition influence
may reverse parenteral nutrition-associated cholestasis in neonates: a sys- electrolyte and fluid balance in preterm infants in the first days after birth?
tematic review and meta-analysis. J Nutr 2015;145:277–83. PLoS One 2010;5:e9033.
[105] Hojsak I, Colomb V, Braegger C, et al. Intravenous lipid emulsions and risk of [121] Abrams SA. In utero physiology: role in nutrient delivery and fetal develop-
hepatotoxicity in infants and children: a systematic review and meta-analy- ment for calcium, phosphorus, and vitamin D. Am J Clin Nutr
sis. J Pediatr Gastroenterol Nutr 2016;62:776–92. 2007;85(Suppl.):604–7.

Please cite this article in press as: Darmaun D, et al. Parenteral nutrition for preterm infants: Issues and strategy. Archives de Pédiatrie
(2018), https://doi.org/10.1016/j.arcped.2018.02.005