487926

research-article2013
PENXXX10.1177/0148607113487926Journal of Parenteral and Enteral Nutrition / Vol. XX, No. X, Month XXXXMiller et al

Original Communication
Journal of Parenteral and Enteral
Nutrition
From Parenteral to Enteral Nutrition: A Nutrition-Based Volume 38 Number 4
May 2014 489­–497
Approach for Evaluating Postnatal Growth Failure in © 2013 American Society

Preterm Infants for Parenteral and Enteral Nutrition

DOI: 10.1177/0148607113487926
jpen.sagepub.com
hosted at
online.sagepub.com

Malki Miller, MS, RD, CNSC1,2; Ruben Vaidya, MBBS2; Deepa Rastogi, MBBS,
MS3; Alok Bhutada, MD2; and Shantanu Rastogi, MD, MMM2

Abstract
Background: Nutrition practices for preterm infants include phases of parenteral nutrition (PN), full enteral nutrition (EN), and the
transitional phase in between. Our aim was to identify the nutrition phases during which infants are most likely to exhibit poor growth
that would affect risk for growth failure (GF) at discharge and to examine factors associated with GF. Methods: A retrospective chart
review was conducted on infants born <32 weeks’ gestation. The neonatal intensive care unit stay was divided into 3 nutrition phases:
(1) full PN, (2) transitional PN + EN, and (3) full EN. Weekly growth rates were calculated, and for each growth velocity <10 g/kg/d,
the coinciding phase was recorded. GF was defined as a discharge weight below the 10th percentile. The nutrition phases during which
growth inadequacy predicted GF at discharge were determined, correcting for other clinical factors associated with GF. Results: In total,
156 eligible infants were identified. Seventy-six infants (49%) were discharged with weights <10%. Incidence of poor growth was highest
during the transitional phase (46%) and was predictive of GF when adjusted for gestational age, birth weight, and severity of illness.
Although energy intakes during the transitional phase were comparable to baseline parenteral provision, protein intakes progressively
decreased (P < .0001), consistently providing <3 g/kg/d as PN was weaned. Serum urea nitrogen also declined and was correlated with
protein intake (r = −0.32, P < .001). Conclusion: Growth was compromised during the transitional phase, likely related to decreased
protein intake. Optimizing protein provision while PN is weaned is an important strategy to prevent postnatal growth failure. (JPEN J
Parenter Enteral Nutr. 2014;38:489-497)

Keywords
preterm infant; postnatal growth failure; parenteral nutrition; protein; blood urea nitrogen

Clinical Relevancy Statement in cognitive development and neurological outcomes,5-10 and

Since poor growth in the neonatal period negatively affects
cognitive development while rapid “catch-up” growth may
have deleterious effects on later metabolic outcomes, consis-
tently maintaining targeted growth velocities throughout the
nutrition therapy spectrum of preterm infants is essential. Our
study identifies the nutrition inadequacy of the transitional
period during which parenteral nutrition is being weaned with
advancing enteral feeds, leading to poor growth during this
phase, and highlights the need for more aggressive nutrition
strategies to better simulate intrauterine growth patterns.
Introduction
The goal of nutrition management of very low birth weight
(VLBW) infants is to achieve postnatal growth velocity that
mimics intrauterine growth rates of the developing fetus.1
However, postnatal growth failure (GF), as defined by dis-
charge weights below the 10th percentile for corrected gesta-
tional age, has been reported at rates as high as 97% for VLBW
infants.2-4 GF is associated with long-term impairments
promoting accelerated weight gain in preterm infants after a
period of slow growth to “catch up” on standardized growth
curves is associated with later risk for visceral adiposity, insu-
lin resistance, and hypertension.11-13 Thus, optimizing growth
velocity to maintain intrauterine growth patterns and to avoid
the need for catch-up growth is an integral component of the
nutrition management of preterm infants.
From the 1Department of Nutrition and 2Division of Neonatology,
Maimonides Infants and Children’s Hospital, Brooklyn, New York, and
3
Children’s Hospital at Montefiore, Albert Einstein College of Medicine,
Bronx, New York.
Financial disclosure: None declared.
Received for publication September 11, 2012; accepted for publication
March 29, 2013.
This article originally appeared online on May 14, 2013.
Corresponding Author:
Shantanu Rastogi, MD, MMM, Maimonides Infants and Children’s
Hospital, 4802 Tenth Ave, G-103, Brooklyn, NY 11219, USA.
Email: srastogi@maimonidesmed.org
490 Journal of Parenteral and Enteral Nutrition 38(4)
Multiple factors contribute to postnatal GF. Nutrition dur-
ing the first week of life is often compromised; despite evi-
dence that early aggressive nutrition support of VLBW infants
optimizes growth,14,15 metabolic immaturity in glucose and
lipid tolerance often makes this a difficult goal. Nutrition fac-
tors such as energy and protein deficit,16,17 prolonged time on
parenteral nutrition (PN),3,18 and delayed enteral feeds3 are
independent predictors of GF, as are medical conditions such
as low birth weight, chronic lung disease, corticosteroid expo-
sure, and sepsis.3,4,16,18
Although nutrition practices vary across neonatal intensive
care units (NICUs),19 the nutrition course of all preterm infants
involves phases of PN, enteral feeds, and the transitional period
in between. PN is administered early to provide the initial
nutrient requirement for growth. Subsequent initiation and
advancement of enteral feeds are frequently interrupted by
abdominal distension, gastric residuals, blood transfusions,
and medical conditions such as sepsis, necrotizing enterocolitis
(NEC), and indomethacin treatment.20 Thus, the amount of
time spent on parenteral, enteral, and transitional phases varies
significantly from infant to infant. Since barriers to optimal
nutrient provision differ throughout these different phases of
the nutrition, with PN complicated by metabolic intolerance
and enteral nutrition (EN) phases interrupted by bouts of “feed-
ing intolerance,” different interventions would be required in
these nutrition phases to optimize nutrition intake.
Previous studies examining growth outcomes of VLBW
infants were chronologically designed, reporting growth veloc-
ities and nutrient intakes by week without distinguishing
between these different phases of nutrition. Categorizing poor
growth velocities by nutrition phases, rather than chronologi-
cally, offers additional insight to the clinician to allow for a
more targeted approach in optimizing nutrition status during
vulnerable phases. Thus, the aim of our study was to identify
the phases within the nutrition therapy timeline that were most
associated with poor growth and predictive of postnatal GF, as
well as to identify contributory nutrition factors.
Methods
Study Population
A retrospective chart review was conducted on infants born
<32 weeks gestation, admitted to the NICU at Maimonides
Infants and Children’s Hospital from July 2008 through June
2010. Gestational age was estimated by obstetrical records
obtained by early pregnancy ultrasound. Infants who were
small for gestational age (birth weight below the 10th percen-
tile when plotted on Fenton growth curves21) or diagnosed with
short bowel syndrome were excluded from the study. Since
growth trends were monitored only until infants achieved a
weight of 2 kg or were discharged home, infants with birth
weights >2 kg and those who transferred from the NICU before
reaching that point were also excluded. Infants were classified
with GF if discharge weights were below the 10th percentile
for corrected gestational age. The study was approved by the
institutional review board at Maimonides Medical Center.
Nutrition Management
PN was initiated within 24 hours of birth, containing dextrose
at a glucose infusion rate (GIR) of 4–6 mg/kg/min, amino acids
(Trophamine; McGaw, Irvine, CA) at 2–3 g/kg/d, and lipids
(Intralipid; Clintec, Deerfield, IL) at 0.5–1 g/kg/d. Calories
were advanced daily based on metabolic tolerance, with a goal
GIR of 11–14 mg/kg/min, amino acids of 3–3.5 g/kg/d, and
lipids of 3 g/kg/d. Trophic feeds of expressed breast milk
(EBM) were initiated 24–72 hours after birth if the infant was
clinically stable and maintained at volumes <20 mL/kg/d for
3–10 days, based on gestational age, and thereafter advanced
by 10–20 mL/kg/d. When enteral feeds exceeded 20 mL/kg/d,
the rate of PN infusion was decreased to maintain total fluids at
140 mL/kg/d, with the concentration of the nutrients in the PN
fluids maintained as before. At EN volumes of 100 mL/kg/d,
PN was discontinued and powder human milk fortifier (Similac
HMF; Abbott Nutrition, Abbott Park, IL) was added to breast
milk at standard fortification levels. EN volumes were
advanced by 10–20 mL/kg/d to goal feeds of 160 mL/kg/d. If
no EBM was available, preterm infant formula was used, initi-
ated at 20 kcal/oz and increased to 24 kcal/oz at an EN volume
of 100 mL/kg/d.
Classification of Poor Growth by Nutrition
Phase
Body weights were measured 3 times weekly, on Mondays,
Wednesdays, and Fridays, using an electronic digital scale.
Weekly growth rates were calculated automatically by the
Neodata computer program (Isoprime, Lisle, IL) and reported
on a gram per kilogram basis. Each incidence of growth veloc-
ity <10 g/kg/d for a week was considered poor growth. This
value was chosen since accepted standards of growth for pre-
term infants are 15–20 g/kg/d,22 and rates <10 g/kg/d would
unequivocally indicate inadequate growth. We then examined
the nutrition intakes that coincided with the timing of each
observed incidence of poor growth and classified them as
nutrition phases of (a) PN only, with or without trophic feeds;
(b) transitional phase of weaning of PN with concomitant
advancement of EN; or (c) goal EN of 160 mL/kg/d. If 2 dif-
ferent nutrition phases overlapped a phase of poor growth,
each phase was counted unless its duration was <2 days.
Furthermore, when weekly growth rates averaged <10 g/kg/d
but were not consistent for the entire week (ie, the infant
exhibited poor growth for the first 5 days but not the remain-
ing 2 days of the 7-day period), only the nutrition phase asso-
ciated with the days of poor growth was counted. This method
was used to quantify GF rather than comparing growth veloci-
ties between each nutrition phase since many infants
Miller et al 491
experienced more than one of each nutrition phase with the
discontinuation or reinitiation of PN due to clinical conditions
such as sepsis and NEC.
Growth velocity was calculated from 1 week postnatally (to
account for postbirth diuresis) until the infant was discharged
or achieved a weight of 2 kg. The end point of 2 kg was chosen
to eliminate potential growth inadequacies associated with
more chronic medical problems in older infants necessitating a
more prolonged NICU stay and also since standards of growth
velocity once this weight is achieved are different compared
with early postnatal life.22
Poor growth that coincided with the use of corticosteroids,
sepsis (determined by a positive blood or urine culture), and
NEC (diagnosed radiologically by Pneumatosis intestinalis)
was excluded, as our aim was to identify nutrition factors asso-
ciated with poor growth. We also excluded phases with concur-
rent renal failure, resolving edema, and the use of furosemide
since weight changes likely reflected acute fluid shifts. These
cases were not included in the denominator when calculating
the incidence of poor growth during the nutrition phase
involved; thus, the incidence of poor growth reflects only the
infants for whom an accurate evaluation of nutrition-related
growth during each nutrition phase was possible, without con-
founding medical problems. Similarly, infants of an older ges-
tational age had no PN phase during which to evaluate growth
since EN was initiated quickly and advanced into the transi-
tional phase soon after the first week of life; for these infants,
the PN phase was not included in growth evaluation.
Calculation of Nutrition Intake During the
Transitional Phase
For infants exhibiting poor growth during the transitional
phase, energy and protein intakes and protein/energy ratios
were calculated for all nutrition phases. For the PN phase,
intakes were reported as an average for the 3 days immediately
prior to the weaning phase, as a baseline comparison for intakes
during the transitional phase. PN nutrient intake during the
transitional phase was calculated at the volume-adjusted rate
used to supplement enteral feeds to maintain total fluids at 140
mL/kg/d, from enteral intakes of 30 mL/kg (when PN was
likely to already be volume adjusted) until the infant reached
160 mL/kg. For enteral feeds, EBM nutrient provision was
estimated at 20 kcal/oz and 1.4 g protein/dL, accepted stan-
dards for preterm breast milk nutrient composition,23 and for-
mula intakes were calculated based on kcal and protein
composition of formula provided by manufacturers. To allow
for a more accurate comparison of energy intakes between PN
and EN, enteral caloric intakes were adjusted for predicted
energy losses due to incomplete absorption and specific
dynamic action, calculated at 85% of their original caloric
value. Protein/energy ratios were calculated based on grams of
protein provided per 100 kcal.
Blood urea nitrogen (BUN) levels were also recorded. For
the PN phase, since BUN values obtained in early life most
likely reflect catabolism and fluid shifts expected of preterm
infants and are less correlated with nutrition intake, only the
last available BUN level prior to PN weaning was documented,
to most closely reflect baseline protein intake prior to the tran-
sitional phase. All available BUN laboratory levels were docu-
mented during the PN portion of the transitional phase; since
blood draws were infrequently done after PN was discontin-
ued, limited BUN data were available for the late transitional
and full EN phases and were therefore not analyzed for these
time points. Since changes in BUN may be related to nutrient
intake, we calculated total energy and protein intakes for 1 day
prior to each documented BUN value, as BUN reflects 24-hour
nutrient intake.24 ΔBUN was calculated as the difference
between BUN during the transitional phase and baseline BUN
during the preceding PN phase. Since multiple values were
obtained for BUN during the transitional phase, multiple
ΔBUN values are reported.
Clinical Factors
Clinical factors associated with GF, such as bronchopulmonary
dysplasia (need for supplemental oxygen at 36 weeks’ cor-
rected gestational age), sepsis (confirmed by a positive blood
or urine culture), NEC (presence of Pneumatosis intestinalis),
and intraventricular hemorrhage (diagnosed by ultrasound and
if ≥ grade 3), were also studied.
Statistical Methods
Univariate analysis was conducted using the t test or analysis
of variance for continuous variables and χ2 test for categorical
variables to study their association with GF and when compar-
ing the energy, protein, and protein-energy ratios with increas-
ing enteral feeds during the transitional phase. Spearman
correlation was used to study the associations between ΔBUN,
protein intake, energy intake, and enteral feeds volume in the
transitional nutrition phase.
Linear regression analysis was done to adjust for predictor
variables. Variables that reached statistical significance (P <
.05) on univariate analysis with GF were included in the model.
Regression diagnostics were performed to ensure that assump-
tions of linear regression analysis were not violated. Statistical
analysis was done using STATA version 10 (StataCorp LP,
College Station, TX).
Results
A total of 196 eligible infants were admitted to the NICU; 4
infants had short bowel syndrome resulting from NEC, 4
infants were born >2 kg, and 8 were transferred out of the unit.
Twenty-four died during the study period (Figure 1). Growth
492 Journal of Parenteral and Enteral Nutrition 38(4)
Figure 1.  Breakdown of all infants <32 weeks’ gestation. BW, birth weight; GF, growth failure; SBS, short bowel syndrome; SGA,
small for gestational age.
velocity data of the remaining 156 infants were examined for
this study. Demographics and clinical factors of the study
group are shown in Table 1.
The incidence of poor growth was highest during the transi-
tional phase, with growth rates <10 g/kg/d in 46% of infants
(Table 2). Compared with infants without GF at discharge,
infants with GF were more likely to exhibit poor growth during
the PN and transitional, but not EN, phases. They were also of
lower gestational age and birth weight percentiles, had higher
incidences of requiring mechanical ventilation on the first day
of life and NEC, and were less likely to have regained birth
weight by 2 weeks of life (Table 1).
Multivariate regression analysis of factors affecting GF at
discharge is shown in Table 3. When poor growth during nutri-
tion phases was corrected for clinical factors, the transitional
phase remained a significant predictor of GF. This analysis
excluded poor growth data during the transitional phase for 9
infants who were discharged with weights below the 10th per-
centile since they had concurrent medical issues that would
have confounded growth evaluation as it related to nutrition
intake. From the clinical factors evaluated, birth weight per-
centile, mechanical ventilation, NEC, and delayed return to
birth weight were independent predictors of GF at discharge.
Since the transitional phase significantly affected growth
outcomes, energy, protein, and protein/energy ratios during
this nutrition phase were calculated and are shown in Figure 2.
The mean ± SD duration of the transitional phase was 10.5 ±
4.9 days. From the parenteral to transitional phase, there were
progressive decreases in both protein and protein/energy ratios
as enteral feeds progressed by 10 mL/kg/d, reaching a nadir of
2.1 ± 0.5 g/kg and 2.3 ± 0.2 g/100 kcal, respectively (P < .05).
Adjusted energy intakes remained comparable until PN was
discontinued. With the discontinuation of PN, protein and
energy intakes progressively increased as enteral feeds were
advanced to goal (P < .05). Enteral feeds volume during the PN
portion of the transitional phase negatively correlated with
total protein intake (r = −0.69; P < .0001).
BUN also declined significantly from the parenteral to tran-
sitional phase (15.8 ± 5.9 vs 8.7 ± 4.6; P < .0001), and ΔBUN
from the PN to transitional phase negatively correlated with
protein intake (r = −0.32, P < .001) and positively correlated
with enteral feeds volume (r = 0.27, P < .005; Figure 3). Since
protein intake was related individually to both enteral volume
and ΔBUN, multivariate regression analysis showed that the
association between protein intake and ΔBUN remained sig-
nificant (P = .04) when corrected for volume intake.
Miller et al 493

Table 1.  Demographic and Clinical Factors in Growth Failure (GF) and Non-GF Groups.

Discharge Weight <10% Discharge Weight ≥10%

Demographic / Clinical Factors Total (n = 156) (n = 76) (n = 80) P Value
Male sex 81 (52.0) 37 (49.0) 44 (55.0) .43
Gestational age, wk, mean ± SD 28.8 ± 2.1 27.9 ± 2.4 29.6 ± 1.6 <.0001
BW, percentile, mean ± SD 40 ± 17.8 30.5 ± 15.9 49.1 ± 14.6 <.0001
Ethnicity .1
 Asian 33 (21.1) 17 (22.4) 16 (20.0)  
  African American 18 (11.5) 10 (13.1) 8 (10.0)  
 White 39 (25.0) 19 (25.0) 20 (25.0)  
 Hispanic 43 (27.5) 24 (31.5) 19 (23.7)  
 Other 23 (14.7) 6 (3.8) 17 (10.8)  
Postnatal corticosteroids 4 (2.6) 3 (3.9) 1 (1.2) .287
IVH ≥ stage 3 9 (5.8) 4 (5.2) 5 (6.2) .792
BPD 12 (7.7) 8 (10.5) 4 (5.0) .187
Sepsis 30 (19.2) 18 (23.6) 12 (15.0) .169
NEC ≥ stage 2 10 (6.4) 8 (10.5) 2 (2.5) .041
Respiratory support on DOL 1
  Mechanical ventilation 44 (28.2) 30 (39.4) 14 (17.5) .002
 CPAP 100 (64.1) 44 (57.8) 56 (70.0) .115
  Room air 12 (7.7) 2 (2.6) 10 (12.5) .021
Delayed return to BW >14 d 40 (25.6) 26 (34.2) 14 (17.5) .014

Values are presented as number (percentage) except where indicated otherwise. BPD, bronchopulmonary dysplasia; BW, birth weight; CPAP, continuous
positive airway pressure; DOL, day of life; IVH, intraventricular hemorrhage; NEC, necrotizing enterocolitis.

Table 2.  Incidence of Poor Growth by Nutrition Phase in Infants With and Without Growth Failure.

Nutrition Phase Total, No. (%) Discharge Weight <10%, No. (%) Discharge Weight ≥10%, No. (%) P Value
Full PN 29/129 (22.5) 18/58 (31.0) 11/71 (15.5) .035
Transitional phase 65/141 (46.1) 41/66 (62.1) 24/75 (32.0) <.001
Full EN 25/146 (17.1) 16/70 (22.8) 9/76 (11.8) .078

EN, enteral nutrition; PN, parenteral nutrition.

Table 3.  Multivariate Analysis of Nutrition and Clinical adequate on full PN and full enteral feeds, weight gain is most
Predictors of Postnatal Growth Failure. compromised in the transitional phase and is likely related to a
Predictors of Postnatal Growth protein rather than a caloric deficit. Furthermore, growth inad-
Failure Odds Ratio 95% CI equacy during the transitional phase makes infants 5 times
more likely to be discharged with weights below the 10th per-
Poor growth: PN phase 0.84 0.23-3.04 centile for corrected gestational age.
Poor growth: transitional phase 5.4 1.66-17.52 The design of this study in examining GF was different from
Delayed BW regain >2 wk 4.61 1.29-16.53 other published studies3,14,16,18,25-29 as we did not follow neonatal
Birth weight percentile 0.88 0.84-0.92
growth chronologically but rather by nutrition phases. This
NEC ≥ stage 2 40.95 3.55-472.92
method better identifies the phases across the nutrition therapies
Mechanical ventilation DOL 1 5.36 1.4-20.56
spectrum during which preterm infants are most vulnerable to GF
Room air DOL 1 0.59 0.05-7.01
and allows for a more targeted approach in optimizing nutrition
CI, confidence interval; BW, birth weight; DOL, day of life; NEC, necro- management of this population. Previous studies examined nutri-
tizing enterocolitis; PN, parenteral nutrition. tion and growth outcomes of preterm infants and reported longer
duration of PN and days to reach full enteral feeds as predictors of
Discussion growth failure,3,18,25,26 with the worst growth velocities reported
when nutrient provision was predominantly provided via the par-
We describe poor growth patterns during the nutrition phases enteral route.3,27 In our study, 22% of infants did not grow ade-
of preterm infants and show that although overall growth is quately while on full PN, but we did not find this phase to be a
494 Journal of Parenteral and Enteral Nutrition 38(4)

Figure 2.  Combined parenteral and enteral (a) protein, (b) energy, and (c) protein/energy ratios during the transition from parenteral
to enteral feeds. Enteral feeds volume of “0” represents mean 3-day parenteral intakes prior to feed advancement; enteral volumes ≥30
mL/kg represent transitional phase. *Significantly lower vs baseline parenteral intakes (P < .05). #Significantly higher vs baseline
parenteral intakes (P < .05). **Significantly higher vs enteral feed volumes (P < .05).

predictor of GF. Since the link between PN duration and poor
growth was found in studies designed to examine nutrient intake
chronologically by week, rather than nutrition phase, the duration
of PN described in these studies, when translated into the catego-
ries used in our study, actually encompasses both the PN-only and
the early portion of the transitional phase when PN is weaned.
Thus, the association reported previously between duration of PN
and GF may not reflect poor growth during PN alone but also
throughout a prolonged transitional phase, similar to our findings.
Also, the lack of association between PN and GF in our study may
simply reflect the evolution of neonatal PN management over the
years, providing for more aggressive protein and energy provi-
sion.19 It is also possible that although we excluded periods of
poor weight gain that coincided with medical conditions that neg-
atively affect growth to more accurately relate growth trends to
nutrition phase, we may have inadvertently selected out the PN
period, the nutrition phase that most likely coincided with active
sepsis or NEC.
To better determine the causes of the poor growth observed,
we examined nutrition intakes during the transitional phase
and found protein intake significantly lower as compared with
the PN-only phase, providing <3 g/kg/d throughout the entire
PN weaning process. Energy intakes during the early transi-
tional phase were within guidelines until PN was discontinued,
at which point they fell below targeted goals until full enteral
feeds were established. Taken together, our results suggest that
although both protein and energy deficits were responsible for
poor growth observed once the infants were transitioned off
PN until reaching full enteral feeds, protein intake alone was a
limiting factor for growth during the early transitional phase
while the infant was still receiving PN and continued to be
throughout the transitional phase.
This finding is consistent with previous observations, as
protein intake has been implicated repeatedly as the limiting
factor for growth. Protein is positively correlated with weight
and head circumference gain,27,30 and retrospective evaluation
of factors affecting growth velocity reported protein intake as
an independent predictor of weight gain.16 Similar to our find-
ings, the strength of association of protein intake and growth is
stronger as compared with total energy intake,16,27,30,31
Miller et al 495
Figure 3.  Relationships between nutrition parameters during
parenteral nutrition portion of the transitional phase. (A) Total
protein intake vs ΔBUN. (B) Enteral feeds volume vs ΔBUN.
(ΔBUN = difference in serum urea nitrogen from the parenteral
to transitional phase.)
supported by multiple randomized controlled trials in which
varying energy intakes resulted in similar weight gain when
protein intakes were fixed.32-36
The change in BUN that we observed across the nutrition
phases may be a biochemical reflection of protein insuffi-
ciency, with the severity of protein deficit related to the magni-
tude of the decline of BUN during the transitional phase and
progressively worsening with continued advancement of
enteral feeds. BUN levels are affected by renal function, nitro-
gen intake, nitrogen quality, and hydration status. Although it
can be argued that the higher quality of enteral protein pro-
motes better nitrogen retention, which may lower circulating
urea levels, we demonstrated that the decrease in BUN is
related to decreased protein intake and not to increased enteral
volume intake with resultant improvement in protein quality.
In addition, total fluid allowance for the infant during the early
transitional phase remained constant at 140 mL/kg/d, and time
periods of renal dysfunction were excluded from this analysis.
Collectively, these observations suggest that the decline in
BUN observed during the transitional phase is most likely a
function of reduced protein intake. This association has been
described in previous studies, although not with consistent
results.35-38 The correlation between BUN and nitrogen intake
appears stronger in enterally fed infants, with variable associa-
tion found during the PN phase. We did not find any correla-
tion between BUN and protein intake during either of the PN
nutrition phases (data not shown), suggesting that BUN trends,
rather than the isolated BUN levels, may be more sensitive
indicators of protein status. Although the interpretation of
BUN as a nutrition marker carries limitations since levels may
be affected by non-nutrition factors, levels are obtained as part
of routine biochemical monitoring on neonates during the par-
enteral and transitional phases, and our results contribute to the
body of evidence that supports serial monitoring of BUN lev-
els as part of the nutrition evaluation. Indeed, it has been dem-
onstrated that adjusting protein intake in human milk–fed
infants based on BUN improved weight and head circumfer-
ence growth.30
Although we excluded the first week of life from our analy-
sis due to expected diuresis-related weight loss, optimal nutri-
tion support during this critical window of early life is
important to stem the early catabolic phase and rapidly transi-
tion the infant toward positive nitrogen balance and continued
growth. We looked at failure to regain birth weight by 2 weeks
of life as an indirect measure of poor nutrition status during
this phase and found this parameter to be an independent pre-
dictor of GF, similar to previous reports.25,26 Martin et al,28
when prospectively evaluating nutrient intakes by week in a
multicenter study, found suboptimal nutrient intakes only dur-
ing the first week to be associated with lower growth velocity,
and infants requiring the most days to regain birth weight had
higher decreases in z scores from birth until discharge.26
Paradoxically, nutrition intakes are perhaps the most limited
during this phase due to early metabolic intolerance. More
recently, studies have demonstrated the safety and efficacy of
early aggressive nutrient provision, with early amino acid
administration soon after birth associated with better glucose
and lipid tolerance, positive nitrogen balance, and better
growth during this critical phase, and more clinicians are
adapting these practices.39-41
Clinical factors and comorbidities are shown to be associated
with postnatal GF.3,4,16,18 Although infants with NEC and those
requiring mechanical ventilation on the first day of life had a
higher risk of GF in our study, we did not find GF to be associ-
ated with corticosteroid use, bronchopulmonary dysplasia
(BPD), or sepsis. This is likely due to the small number of infants
with corticosteroid use and BPD, as our unit seldom uses corti-
costeroids for prolonged oxygen requirement due to deleterious
neurocognitive effects and our lower rates of BPD.42
There are several limitations to our study. The retrospective
design in itself carries limitations. Furthermore, our definition
of poor growth captured only severely growth-restricted
infants; growth velocities of 10–14 g/kg/d are generally inad-
equate for extremely low birth weight infants, and time frames
496 Journal of Parenteral and Enteral Nutrition 38(4)
with these growth rates were not included in our study. We
used this restrictive definition to target those infants who had
poor growth unequivocally, although perhaps a more gradu-
ated approach (ie, defining different categories of growth
restrictions based on weight and/or gestational age) would be
more appropriate. In addition, although absolute criteria were
established by which we excluded periods of poor weight gain
that coincided with medical conditions, all nutrition phases
may not have been excluded uniformly, creating a potential
selection bias. Another limitation of our study involves our
definition of GF. We defined GF as a discharge weight below
the 10th percentile, the common clinical definition used by the
National Institute of Child Health and Human Development
Neonatal Research Network, the New York State Neonatal
Quality Collaborative, and various large multicenter clinical
studies.3,4,18,26 Not surprisingly, we found that infants born at a
higher birth weight percentile were less likely to be discharged
less than the 10th percentile for corrected gestational age.
Larger babies are obviously less likely to reach the cutoff per-
centile considered as “growth failure,” even with significant
growth inadequacy and poor nutrient provision that may affect
their clinical course as well as later neurodevelopment.
Although other definitions of GF have been used (ie, z score
deviation from birth25), we included birth weight percentile in
the logistic regression model to compensate for the effect of
variation in birth weight.
Our results support the use of PN weaning strategies that do
not compromise total protein intake as an important interven-
tion to optimize growth patterns in the neonatal population.
Although some units customize PN orders during the transi-
tional phase to complement enteral feeds calorie and protein
provision, most others practice a volume-based weaning proto-
col similar to ours that does not adequately maintain targeted
protein intakes, since unfortified breast milk is significantly
lower in protein (per milliliter) as compared with typical PN
admixtures. Minimizing the time the infant experiences in this
phase by encouraging more rapid feeds advancement or earlier
human milk fortification may negatively affect feeds tolerance;
thus, optimizing the enteral portion of the transitional phase is
impractical. More aggressive PN management during the early
transitional phase, however, would more safely minimize
nutrient deficit.
In conclusion, this study brings to attention the nutrition
inadequacy of the time frame during which infants are transi-
tioned from full PN to full enteral feeds, when poor growth is
a predictor of postnatal growth failure even when corrected for
severity of illness. The nutrition factors most likely responsible
for limiting growth are consistent with decreased protein
intake. The increased likelihood to be discharged with weights
below the 10th percentile for corrected gestational age in
infants experiencing growth delay during this time frame
should prompt the clinician involved in neonatal nutrition
management to more closely examine nutrition practices dur-
ing the transitional phase to prevent postnatal GF with its asso-
ciated long-term outcomes.
References
1. Nutritional needs of preterm infants. In: Kleinman R, ed. Pediatric
Nutrition Handbook. 6th ed. Elk Grove Village, IL: American Academy
of Pediatrics; 2009:79-112.
2. Dusick AM, Poindexter BB, Ehrenkranz RA, Lemons JA. Growth fail-
ure in the preterm infant: can we catch up? Sem Perinatol. 2003;24(4):
302-310.
3. Ehrenkranz RA, Younes N, Lemons JA, et al. Longitudinal growth of hos-
pitalized very low birth weight infants. Pediatrics. 1999;104(2):280-289.
4. Clark RH, Thomas P, Peabody J. Extrauterine growth restriction
remains a serious problem in prematurely born neonates. Pediatrics.
2003;111(5):986-990.
5. Ehrenkranz RA, Dusick AM, Vohr BR, Wright LL, Wrage LA, Poole K.
Growth in the neonatal intensive care unit influences neurodevelopmental
and growth outcomes of extremely low birth weight infants. Pediatrics.
2006;117(4):1253-1261.
6. Georgieff MK, Hoffman JS, Pereira GR, Bernbaum J, Hoffman-Williamson
M. Effect of neonatal caloric deprivation on head growth and 1-year
developmental status in preterm infants. J Pediatr. 1985;107(4):581-587.
7. Brandt I, Sticker EJ, Lentze MJ. Catch-up growth of head circumference
of very low birth weight, small for gestational age preterm infants and
mental development to adulthood. J Pediatr. 2003;142:463-468.
8. Poindexter BB, Langer JC, Dusick AM, Ehrenkranz RA. Early provision
of parenteral amino acids in extremely low birth weight infants: relation to
growth and neurodevelopmental outcome. J Pediatr. 2006;148:300-305.
9. Hayakawa M, Okumura A, Hayakawa F, et al. Nutritional state and
growth and functional maturation of the brain in extremely low birth
weight infants. Pediatrics. 2003;111(5):991-995.
10. Tan MJ, Cooke RW. Improving head growth in very preterm infants:
a randomized control trial I: neonatal outcomes. Arch Dis Child Fetal
Neonatal Ed. 2008;93:F337-F341.
11. Uthaya S, Thomas EL, Hamilton G, Dore CJ, Bell J, Modi N. Altered
adiposity after extremely preterm birth. Pediatr Res. 2005;57(2):211-215.
12. Hofman PL, Regan F, Cutfield WS. Prematurity—another example of
perinatal metabolic programming? Hormone Res. 2006;66:33-39.
13. Thureen PJ. The neonatologist’s dilemma: catch-up growth or beneficial
undernutrition in very low birth weight infants—what are optimal growth
rates? J Pediatr Gastroenterol Nutr. 2007;S152-S154.
14. Ziegler EE, Thureen PJ, Carlson SJ. Aggressive nutrition of the very low
birthweight infant. Clin Perinatol. 2002;29:225-244.
15. Thureen PJ, Hay WW. Early aggressive nutrition in preterm infants. Semin
Neonatol. 2001;6:403-415.
16. Olsen IE, Richardson DK, Schmid CH, Ausman LM, Dwyer JT. Intersite
differences in weight growth velocity of extremely premature infants.
Pediatrics. 2002;110(6):1125-1132.
17. Embleton NE, Pang N, Cooke RJ. Postnatal malnutrition and growth retar-
dation: an inevitable consequence of current recommendations in preterm
infants? Pediatrics. 2001;107:270-273.
18. Radmacher PG, Looney SW, Rafail ST, Adamkin DH. Prediction of
extrauterine growth retardation (EUGR) in VVLBW infants. J Perinatol.
2003;23:392-395.
19. Hans DM, Pylipow M, Long JD, Thureen PJ, Georgieff MK. Nutritional
practices in the neonatal intensive care unit: analysis of a 2006 neonatal
nutrition survey. Pediatrics. 2009;123(1):51-57.
20. Gregory KE, Connolly TC. Enteral feeding practices in the NICU: results
from a 2009 Neonatal Enteral Feeding Survey. Adv Neonatal Care.
2012;12(1):46-55.
21. Fenton TR. A new growth chart for preterm babies: Babson and Benda’s
chart updated with recent data and a new format. BMC Pediatr. 2003;3:13.
22. Anderson DM, Eichenwald EC, Chan SW, et al, eds. Guidelines for Acute
Care of the Neonate. 18th ed. Houston, TX: Section of Neonatology,
Department of Pediatrics, Baylor College of Medicine; 2010.
23. Sapsford AL. Composition of human milk and selected enteral products.
In: Groh-Wargo S, Thompson M, and Cox JH, eds. Nutritional Care for
High-Risk Newborns. 3rd ed. Chicago, IL: Precept Press; 2000.
Miller et al 497
24. Nutrition support for premature infants. In: Bunting KD, Mills J, Phillips
S, Ramsey E, Rich S, and Trout S, eds. Pediatric Nutrition Reference
Guide. 9th ed. Houston: Texas Children’s Hospital; 2010:84.
25. Shah PS, Wong KY, Merko S, Bishara R, Dunn M, Asztalos E, Darling
PB. Postnatal growth failure in preterm infants: ascertainment and relation
to long-term outcome. J Perinat Med. 2006;34(6):484-489.
26. Steward DK, Pridham KF. Growth patterns of extremely low-birth-
weight hospitalized preterm infants. J Obstet Gynecol Neonatal Nurs.
2002;31(1):57-65.
27. Carlson SJ, Ziegler EE. Nutrient intakes and growth of very low birth
weight infants. J Perinatol. 1998;18(4):252-258.
28. Martin CR, Brown YF, Ehrenkranz RA, et al. Nutritional practices and
growth velocity in the first month of life in extremely premature infants.
Pediatrics. 2009;124(2):649-657.
29. Berry MA, Abrahamowicz M, Usher RH. Factors associated with growth
of extremely premature infants during initial hospitalization. Pediatrics.
1997;100(4):640-646.
30. Arsanoglu S, Moro GE, Ziegler EE. Adjustable fortification of human
milk fed to preterm infants: does it make a difference? J Perinatol.
2006;26:614-621.
31. Brumberg HL, Kowalski L, Troxell-Dorgan A, et al. Randomized trial of
enteral protein and energy supplementation in infants less than or equal to
1250 g at birth. J Perinatol. 2010;30(8):517-521.
32. Kashyap S, Forsyth M, Zucker C, Ramakrishnan R, Dell RB, Heird WC.
Effects of varying protein and energy intakes on growth and metabolic
response in low birth weight infants. J Pediatr. 1986;108:955-963.
33. Kashyap S, Schulze K, Forsyth M, et al. Growth, nutrient retention, and
metabolic response in low birth weight infants fed varying intakes of pro-
tein and energy. J Pediatr. 1988;113:713-721.
34. van Goudoever JB, Sulkers EJ, Lafeber HN, Sauer PJ. Short-term
growth and substrate use in very-low-birth-weight infants fed for-
mulas with different energy contents. Am J Clin Nutr. 2000;71(3):
816-821.
35. Cooke R, Embleton N, Rigo J, Carrie A, Haschke F, Ziegler E. High pro-
tein preterm infant formula: effect on nutrient status, metabolic status and
growth. Ped Res. 2006;59(2):265-270.
36. Embleton ND, Cooke RJ. Protein requirements in preterm infants: effect
of different levels of protein intake on growth and body composition. Ped
Res. 2005;58(5):855-860.
37. Fairey AK, Butte NF, Mehta N, Thotathuchery M, Schanler RJ, Heird
WC. Nutrient accretion in preterm infants fed formula with differ-
ent protein:energy ratios. J Pediatr Gastroenterol Nutr. 1997;25(1):
37-45.
38. Roggero P, Gianni ML, Morlacchi L, et al. Blood urea nitrogen concen-
trations in low-birth-weight preterm infants during parenteral and enteral
nutrition. J Pediatr Gastroenterol Nutr. 2010;51(2):213-215.
39. te Braake FW, van den Akker CH, Wattimena DJ, Huijmans JG, van
Goudoever JB. Amino acid administration to premature infants directly
after birth. J Pediatr. 2005;147(4):457-461.
40. Mahaveer A, Grime C, Morgan C. Increasing early protein intake is asso-
ciated with a reduction in insulin-treated hyperglycemia in very preterm
infants. Nutr Clin Pract. 2012;27(3):399-405.
41. Valentine CJ, Fernandez S, Rogers LK, et al. Early amino-acid admin-
istration improves preterm infant weight. J Perinatol. 2009;29(6):
428-432.
42. Rastogi S, Rajasekhar H, Gupta A, Bhutada A, Rastogi D, Wung JT.
Factors affecting the weaning from nasal CPAP in preterm neonates. Int J
Pediatr. 2012;2012:416073.