Clinical Research

Nutrition in Clinical Practice

Volume 00 Number 0
Improving Nutrition Outcomes for Infants < 1500 Grams xxx 2018 1–9

C 2018 American Society for

With a Progressive, Evidenced-Based Enteral Feeding Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10081
Protocol wileyonlinelibrary.com

Melissa K. Thoene, MS, RD1 ; Elizabeth Lyden, MS2 ;

and Ann Anderson-Berry, MD, PhD3

Abstract
Background: Growth is essential for very low birth weight infants. The purpose of this retrospective chart review was to evaluate the
impact of a new standardized, evidenced-based feeding protocol for infants born < 1500 g in correlation with growth and clinical
outcomes. Methods: Growth and nutrition data was reviewed from 2 groups of infants born < 1500 g within a level III newborn
intensive care unit (NICU). Epoch 1 infants (N = 32) received care following initial implementation of a standardized enteral
feeding protocol. Epoch 2 infants (N = 32) received care following aggressive modification of this initial protocol based on newly
available literature that promotes earlier initiation and advancement of enteral feedings. Results: Epoch 2 infants weighed more at
36 weeks (2562 vs 2304 g) with higher discharge weight percentiles (32nd vs 15th percentile). Epoch 2 infants started and achieved
full enteral feedings earlier (day of life 1 vs 4; 7 vs 22, P < 0.0001) and required less days of parenteral nutrition (5.5 vs 17.5 days,
P < 0.0001), with indwelling central line for parenteral access (6 vs 17.5). There were no differences in retinopathy of prematurity
(17% control vs 19% study), oxygen requirement at 36 weeks (22% epoch 1 vs 43%), necrotizing enterocolitis (3% epoch 1 vs 0%),
intraventricular hemorrhage grade 3–4, periventricular leukomalacia, or death. Conclusion: In this sample of very low birth weight
infants, a progressive standardized, evidence-based feeding protocol was associated with improved growth without increased risk
for necrotizing enterocolitis. (Nutr Clin Pract. 2018;00:1–9)

Keywords
clinical protocols; enteral nutrition; growth; infants, very low birth weight infants

Introduction ratory support remain an immediate forefront priority, it is

It remains well recognized that adequate growth is critical
for every preterm infant, but most specifically in very
low birth weight infants born < 1500 g. These infants
have exceedingly high nutrient needs; therefore, providing
adequate nutrition is necessary to meet full potential for
growth and neurological outcomes. Despite this knowledge,
recent data reports that 50% of North American infants
born < 1500 g are still experiencing postnatal growth failure,
with discharge weights plotting below the 10th percentile on
their respective growth charts.1
Unaggressive, dated nutrition practices contribute to
energy and protein deficits, all resulting in subsequent
poor growth. These include but are not limited to a low
optimization of parenteral nutrition (PN), delay in enteral
feeding advancement, and delay in human milk fortifica-
tion. Poor nutrition management can lead to osteopenia,
cholestasis, chronic lung disease, and a multitude of other
issues that further deplete nutrient stores and affect growth.
Nutrition practices vary significantly among neonatal units,
but achieving adequate growth must remain a common and
consistent priority. Nutrition management must be regarded
as a true medical therapy. Whereas therapies such as respi-
the persistent adequate nutrition support that will influence
these long-term adjunctive therapy requirements.
From the 1 Department of Pharmacy and Nutrition, Nebraska
Medicine, Omaha, Nebraska, USA; the 2 College of Public Health,
University of Nebraska Medical Center, Omaha, Nebraska, USA;
and the 3 Department of Pediatrics, University of Nebraska Medical
Center, Omaha, Nebraska, USA.
Financial disclosure: None declared.
Conflicts of interest: Anderson-Berry has received financial
compensation as a speaker for Mead Johnson and Abbott Nutrition,
as well as a monetary grant from the Gerber Foundation. Mead
Johnson, Abbott Nutrition, and the Gerber Foundation had no role in
the funding or design of the study; in the collection, analyses, or
interpretation of the data; in the writing of the manuscript; and in the
decision to publish the results. The remaining authors declare no
conflicts of interest.
This article originally appeared online on February 28, 2018.
Corresponding Author:
Melissa K. Thoene, MS, RD, Department of Pharmacy and
Nutrition, Nebraska Medicine, 981200 Nebraska Medical Center,
Omaha, NE 68198, USA.
Email: mthoene@nebraskamed.com
2 Nutrition in Clinical Practice 00(0)
In 2007, a standardized feeding protocol was imple-
mented in our level III newborn intensive care unit (NICU),
focusing on a standardized earlier start and quicker ad-
vancement to full enteral feedings.2 This feeding protocol
was applied only to the NICU, not hospital-wide. This
change resulted in a significant reduction in day of life
(DOL) to achieve full enteral feedings (22 vs 30 days),
less days receiving PN (18 vs 27 days), a lower incidence
of chronic lung disease at 36 weeks gestational age (GA)
(22% vs 51%), and fewer infants discharged with weights
plotting < 10th percentile on their respective growth curve
(28% vs 57%), all P < 0.05. Because our feeding protocol
has subsequently been significantly updated based on newly
available published literature that promotes even faster
enteral feeding initiation and advancement, we continue to
experience further improved outcomes. The purpose of this
study is to assess the outcomes from our current feeding
practices (epoch 2) to those in the postimplementation
group following changes in 2007 (epoch 1).
Patients and Methods
Participants and Data Collection
The institutional review board at the affiliated hospital’s
university medical center approved this study with a waiver
of consent, and all study procedures followed were in accor-
dance with their ethical standards. Data was retrospectively
collected from inpatient electronic medical records of study
infants admitted to the NICU between June 2015 and May
2016 who were born < 1500 g. This time period was selected
because this is when the modified feeding protocol was im-
plemented. There was no cutoff for minimum birth weight.
Thirty-two infants were included in the control group
(epoch 1). These infants met birth weight criteria and were
admitted to the same level III hospital NICU during Octo-
ber 2007–July 2008. Data had been previously collected on
these infants by 4 investigators providing direct medical care
to these eligible infants. Thirty-two study infants from an
eligible pool (born between June 2015–May 2016; N = 44)
were selected through randomization and matched 1:1 with
control infants based on GA and weight plotting < 10th per-
centile on the Fenton growth curve at birth. Data for epoch
2 was collected by 1 investigator who had provided direct
medical care to these eligible infants. All infants in epoch 1
and epoch 2 were cared for in the same level III NICU.
Demographics, Growth, and Clinical Outcomes
Demographic information was collected for all infants,
including gender, GA at birth, and DOL at discharge.
Additional outcomes collected included the incidence of
bronchopulmonary dysplasia (BPD), defined as oxygen use
at 36 weeks GA; retinopathy of prematurity (ROP) stage
2 or greater; intraventricular hemorrhage (IVH) grades 3–
4; periventricular leukomalacia (PVL); necrotizing entero-
colitis (NEC) based on Bell’s staging criteria; number of
packed red blood cell transfusions; and use of steroids
(dexamethasone).
Weight was measured daily on a gram scale, and head
circumference was measured weekly using a measuring tape.
Weight with associated growth percentiles on the Fenton
growth curve were assessed at birth, 36 weeks GA, and
discharge. We also evaluated how many infants had weights
plotting < 10th percentile on the Fenton growth curve at
each of these time points. Head circumference size with
associated growth percentiles was assessed at 36 weeks GA
and at discharge.
Enteral feeding progression was evaluated, including
DOL feedings started, DOL full enteral feedings were
achieved, time interval between start of feedings to full
achievement, number of days receiving PN, number of days
with indwelling central line for PN access, and number of
abdominal films taken due to feeding tolerance concerns.
Both highest blood urea nitrogen (BUN) and highest base
deficit were assessed during the first week of life while
receiving PN. Highest BUN was also collected while on
full enteral feedings. Full enteral feedings were defined as
an infant receiving a 150 mL/kg/d of either preterm infant
formula or maternal breast milk fully fortified to 24 cal/oz
using human milk fortifier and a protein modular.
Feeding Regimen
Similar strategies between groups consisted of initiating
PN immediately following birth at 80 mL/kg/d. Total fluids
were typically advanced toward 150 mL/kg/d over 4 days.
Following these initial goals, both PN and enteral nutrition
(EN) management differed between groups.
For epoch 1, parenteral protein goals were 2.5 g/kg/d
with starter PN, followed by advancement to 4.0 g/kg/d
over 3 days. Intralipids were started at 2 g/kg/d on DOL
2 and advanced over 3 days to 3.5 g/kg/d. Enteral feedings
were initiated by DOL 3 at 20 ml/kg/d. This small feeding
volume was maintained for 5–7 days and then advanced by
20 ml/kg/d toward goal 150 mL/kg/d after this initial trophic
period. Fortification of milk with a powdered human milk
fortifier was started when enteral volume reached 100–120
ml/kg/d. Fortification was started at 22 cal/oz for 24 hours,
increased to 24 cal/oz for 24 hours, and then a powdered
protein modular was added to achieve goal of 4.0 g/kg/d per
120 cal/kg/d. Because donor milk was not available during
this time, formula was used as a supplement to mother’s milk
supply. If an infant was formula fed, feedings were initiated
at 20 cal/oz and increased to 24 cal/oz when enteral feeding
volume reached 100–120 ml/kg/d. The caloric density of
feedings was increased to 27 or 30 cal/oz at any point on
full enteral feedings that growth was deemed inadequate, at
the discretion of the medical team.
Thoene et al
Epoch 2 received a more aggressive nutrition regimen
for both EN and PN management. Parenteral protein goals
were minimum of 3.0 g/kg/d with starter PN, followed by
increase to minimum of 4.0 g/kg/d with first bag of custom
PN. Intralipids were started at 2 g/kg/d with the first bag of
custom PN, then increased to goal minimum of 3.0 g/kg/d
with the second bag of custom PN. Enteral feedings were
initiated within the first 24 hours of life (ideally within the
first 4 hours of life) at 30–35 mL/kg/d. Trophic feedings were
continued for 48 hours for infants born < 28 weeks GA.
Infants born > 28 weeks did not receive trophic feedings.
Feedings were then advanced by 30–35 mL/kg daily toward
goal of 150 ml/kg/d. Fortification of milk to 24 cal/oz with a
nonacidified liquid human milk fortifier was initiated when
enteral feedings reached 50–60 mL/kg/d. After 24 hours, a
liquid protein modular was added to achieve goal protein
of 4.2 g/kg/d per 120 cal/kg/d in infants > 1250 g and
goal protein of 4.4 g/kg/d per 120 cal/kg/d in infants <
1250 g. Enteral volume advancement and milk fortification
were completed simultaneously. Donor milk was used as a
bridge to mother’s own milk supply if parents consented.
Donor milk was advanced and fortified similarly to mother’s
own milk to meet nutrition goals. Donor milk was not
used longer than 14 DOL, and some infants transitioned
off donor milk before 14 days of life if early growth
was inadequate to maintain growth percentiles after initial
diuresis. If an infant was formula fed, feedings were initiated
at 24 cal/oz. The caloric density of feedings were increased
to 27 or 30 cal/oz at any point on full enteral feedings
when growth was deemed inadequate to maintain growth
percentiles, at the discretion of the medical team.
Data Analysis
Comparison of continuous variables between groups was
completed using the Wilcoxon rank sum test, which com-
pares the medians of 2 groups. The Fisher’s exact test
was used to compare categorical data. A P value < 0.05
was considered statistically significant. Data was analyzed
between epoch 1 vs epochs 2 as well as between birth weight
categories (< 1000 g and 1000–1500 g). Additional analyses
were completed, as reported in the Discussion section, such
as when comparing head circumference growth for infants
without high-grade IVH. Growth in g/kg/d (from DOL
birth weight regained until first weight ࣙ 2 kg) and g/d (from
first weight ࣙ 2 kg until discharge) was analyzed for epoch
2 infants for further analysis of growth velocity (GV). The
calculation for GV in g/kg/d was calculated as follows:3
GV = [1000 × ln (Wn/W1)]/ (Dn − D1)
Results
There were 32 infants in each group. Baseline characteristics
and incidence of disease are displayed in Table 1. Growth
3
outcomes are displayed in Table 2. Feeding progression and
treatment outcomes are displayed in Table 3.
There were no differences in the number of infants
plotting < 10th percentile for age at birth or GA between
groups because this was matching criteria for study and
control infants. There were no differences in gender among
groups. There were no statistical differences in the incidence
of BPD, ROP, IVH grade 3–4, PVL, NEC, or death between
combined groups. More infants in epoch 1 received steroids
compared with epoch 2 (81% vs 22%, P < 0.0001). Epoch
1 infants were discharged by younger days of life (DOL 64
vs 84), although P = 0.08.
Whereas birth weight and percentile were similar be-
tween combined groups, weight and percentile at 36 weeks
GA were higher in the epoch 2 infants (2563 vs 2304 g; 32nd
vs 15th percentile). This was also similar at discharge, with
epoch 2 infants having higher weight percentiles compared
with the epoch 1 group (32nd vs 15th percentile). There were
no statistical differences between groups in the number of
infants, with head circumferences plotting < 10th percentile
at 36 weeks GA and at discharge.
Epoch 2 infants started enteral feedings earlier than
epoch 1 infants for all weight categories (DOL 1 vs 4, P <
0.0001) and achieved full feedings earlier (DOL 7 vs 22, P <
0.0001). Epoch 2 infants also required less days PN (5.5 vs
17.5 days, P < 0.0001) and required less days with indwelling
central line for PN access (6 vs 17.5 days). Epoch 2 infants
also required fewer abdominal films to assess for feeding
tolerance (median 0.5 vs 19 films) and received fewer blood
transfusions (median 0 vs 3). Highest base deficit for all
infants in the first week of life was similar between groups.
Epoch 1 infants had higher BUN levels while on full enteral
feedings.
Discussion
Evaluation of Growth
Growth was better maintained for epoch 2 infants, given a
higher trend in percentiles compared with the control group.
This trend was evident in most all comparisons between
groups for every weight category at both 36 weeks GA and
at discharge. The only similar comparison was for infants
< 1000 g, who had nonsignificant differences between
weight or percentiles at 36 weeks GA. Epoch 2 infants were
discharged at older days of life, which approached statistical
significance (median DOL 64 vs 84, P = 0.08). It is of more
clinical significance to assess trends in weight percentiles
from birth until discharge for combined weight categories
between groups. Control infants were born with median
weights plotting at the 45th percentile on the Fenton growth
chart, followed by the 15th percentile at both 36 weeks
GA and discharge. Study infants were born with median
weights plotting at the 48th percentile, followed by the
4 Nutrition in Clinical Practice 00(0)

Table 1. Baseline Characteristics and Outcome Data.

Reported Median
(reported percentage or Epoch 2
range) Category Epoch 1 (n = 32) IQR (n = 32) IQR P Value

Weight category < 1000 g 18/32 (56%) 12/32 (38%)

1000–1500 g 14/32 (44%) 20/32 (62%) 0.21
GA at birth All infants 28ˆ0 (26,29) 28ˆ1 (26.7,29.5) 0.42
< 1000 g 26ˆ1 (23ˆ5–28ˆ2) (24.5,28) 26ˆ1 (23ˆ2–30ˆ3) (82,115) 0.67
1000–1500 g 29ˆ2 (27ˆ0–32ˆ2) (28.5,30.2) 29ˆ0 (26ˆ4–30ˆ6) (27.8,29.6) 0.29
DOL at discharge All infants 64 days (49, 90) 84 days (68,98.5) 0.08
< 1000 g 88.5 (29–140) (64,113) 96.5 (70–134) (82, 115) 0.34
1000–1500 g 51.5 (43–80) (46,59) 70 (36–112) (59.5,85.5) 0.0051
Gender All infants 34% female, 34% female, 1.00
66% male 66% male
BPD All infants 6/27 (22%) 13/30 (43%) 0.16
< 1000 g 6/16 (38%) 11/12 (92%) 0.006
1000–1500 g 0/11 (0%) 2/18 (11%) 0.51
ROP All infants 5/30 (17%) 6/32 (19%) 1.00
< 1000 g 4/16 (25%) 5/12 (42%) 0.43
1000–1500 g 1/14 (7%) 1/20 (5%) 1.00
IVH grade 3–4 All infants 0% 3/32 (9%) 0.24
< 1000 g 3/12 (25%) 0.06
1000–1500 g 0%
PVL All infants 2/32 (6%) 1/32 (3%) 1.00
< 1000 g 2/18 (11%) 1/12 (8%) 1.00
1000–1500 g 0% 0%
NEC All infants 1/32 (3%) 0% 1.00
< 1000 g 1/18 (6%)
1000–1500 g 0%
Use of steroids All infants 26/32 (81%) 7/32 (22%) <0.0001
< 1000 g 17/18 (94%) 6/12 (50%) 0.0086
1000–1500 g 9/14 (64%) 1/20 (5%) 0.0003
Death All infants 2/32 (6%) 0% 0.49
< 1000 g 2/18 (11%) 0.50
1000–1500 g 0%

BPD, bronchopulmonary dysplasia; DOL, day of life; GA, gestational age; IQR, interquartile range; IVH, intraventricular hemorrhage; NEC,
necrotizing enterocolitis; PVL, periventricular leukomalacia; ROP, retinopathy of prematurity.
ˆ indicates GA (ie, 28ˆ0 = 28 weeks, 0 days).

32nd percentile at 36 weeks GA and the 32nd percentile at
discharge. These trends demonstrate better growth main-
tenance in the study group to mimic that of intrauterine
growth. This becomes most significant in the high-risk
infants born < 1000 g (ELBW). For this group of epoch
1 infants, median weight percentiles plotted at the 45th
percentile at birth, followed by the 12th percentile at both
36 weeks GA and discharge. For the ELBW epoch 2 infants,
median weight percentiles plotted at the 23rd percentile
at birth, followed by the 16th percentile at 36 weeks and
the 26nd percentile at discharge. One may argue that study
infants had an additional week for catch-up growth if
needed, but it remains more likely that growth was better
maintained for study infants throughout hospitalization
considering a lower drop in percentiles from birth to 36
weeks GA. Epoch 1 infants did receive more dexamethasone
than study infants, which may partially account for altered
growth4,5 ; however, the drop from birth to 36 weeks GA
appears more significant than clinically expected.
Nineteen percent of epoch 1 infants and 9% of
epoch 2 infants were discharged with weights plotting
< 10th percentile. Baseline rates of infants plotting
< 10th percentile on the Fenton growth curve for weight
at birth were 6%, which means there were 4 infants in
the control group and 1 infant in the study group who
were not born < 10th percentile for weight but became
extrauterine growth-restricted (13% and 3% per group).
These rates remain below the nationwide Vermont Oxford
Network average in 2013, reporting 50.3% of infants born
501–1500 g discharging with weights < 10th percentile.1
The Vermont Oxford Network also reported severe growth
restriction with weight plotting < 3rd percentile at discharge
in 27.5% of infants.1 None of the 32 study infants in epoch
2 discharged with weights < 3rd percentile. In further
Thoene et al 5

Table 2. Growth Outcomes.

Epoch 1 (n = 32) Epoch 2 (n =

Reported Median Category Median IQR 32) Median IQR P Value

Birth weight, grams All infants 968 (762,1323) 1125 (868,1360) 0.30
< 1000 g 804 (685,915) 820 (680,905) 0.90
1000–1500 g 1354 (1276,1428) 1310 (1150,1420) 0.44
Birth weight % All infants 45th percentile (20,50) 48th percentile (28,63) 0.34
< 1000 g 45th percentile (20,50) 23rd percentile (13,55) 0.75
1000–1500 g 50th percentile (45,50) 51st percentile (45,66) 0.42
36-week weight, grams All infants 2304 (N = 29) (2196, 2439) 2563 (N = 30) (2344,2751) 0.023
< 1000 g 2247 (1866,2398) 2341 (2104,2577) 0.32
All infants 2389 (2292,2734) 2684 (2542,2920) 0.065
36-week weight % All infants 15th percentile (10,25) 32nd percentile (16,44) 0.027
< 1000 g 12th percentile (3,20) 16th percentile (7,31) 0.49
1000–1500 g 23rd percentile (12,50) 41st percentile (29,58) 0.044
Discharge weight, grams All infants 2638 (2356.5,3049) 3321 (2902,3601) < 0.001
< 1000 g 2588 (2269,3003) 3197 (2951,3601) 0.0046
1000–1500 g 2662 (2377,3094) 3356 (2839,3662) 0.0028
Discharge weight % All infants 15th percentile (10,15) 32nd percentile (23,55) 0.0004
< 1000 g 12th percentile (9,15) 26th percentile (13,36) 0.042
1000–1500 g 23rd percentile (15,50) 43rd percentile (27,60) 0.019
36-week HC, cm All infants 32 (N = 26) (31,33) 32.4 (N = 30) (31,33) 0.88
< 1000 g 32 (30.5, 33) 31.1 (30.5,31.9) 0.32
1000–1500 g 32 (31,33) 32.8 (32,34) 0.45
36-week HC % All infants 33rd percentile (12,50) 37th percentile (11,63) 0.48
< 1000 g 15th percentile (10,50) 16th percentile (8,34) 0.49
1000–1500 g 40th percentile (20,50) 49th percentile (27,77) 0.27
Discharge HC, cm All infants 32.5 (31.9,34.8) 34.8 (33.4, 35.9) 0.002
< 1000 g 32.5 (32,33.5) 34.3 (33.3,35.9) 0.019
1000–1500 g 32.8 (31.8,35.5) 34.8 (33.4,35.9) 0.071
Discharge HC % All infants 25th percentile (14,50) 39th percentile (24,71) 0.031
< 1000 g 15th percentile (10,50) 32nd percentile (18,42) 0.38
1000–1500 g 33rd percentile (25,50) 55th percentile (32,78) 0.10
Weight < 10th percentile All infants 2/32 (6%) 2/32 (6%) 1.00
at birth < 1000 g 2/18 (11%) 2/12 (17%) 1.00
1000–1500 g 0/14 (0%) 0/20 (0%)
Weight < 10th percentile All infants 6/29 (21%) 5/30 (17%) 0.75
at 36 weeks GA < 1000 g 5/15 (33%) 5/12 (42%) 0.71
1000–1500 g 1/14 (7%) 0/18 (0%) 0.44
Weight < 10th percentile All infants 6/32 (19%) 3/32 (9%) 0.47
at discharge < 1000 g 5/18 (28%) 3/12 (25%) 1.00
1000–1500 g 1/14 (7%) 0/20 (0%) 0.41
HC < 10th percentile at All infants 1/26 (4%) 5/30 (17%) 0.20
36 weeks < 1000 g 1/15 (7%) 4/12 (33%) 0.14
1000–1500 g 0% 1/20 (6%) 1.00
HC < 10th percentile at All infants 4/32 (13%) 2/32 (6%) 0.67
discharge < 1000 g 4/18 (22%) 1/12 (8%) 0.62
1000–1500 g 0% 1/20 (5%) 1.00

GA, gestational age; HC, head circumference; IQR, interquartile range.

analyzing the study infant who became growth-restricted,
birth weight plotted at the 12th percentile and subsequently
discharged at the 6th percentile, therefore demonstrating a
less significant growth failure pattern than expected.
The Vermont Oxford Network also reported GV for
> 360,000 infants, as reported in g/kg/d from birth until
discharge.1 In 2013, the interquartile range for GV was 12.3–
13.4 g/kg/d, which they estimate would be an additional 3
g/kg/d higher if initial days of diuresis were not included
in the calculation.1 For comparison, we calculated the GV
in our groups. The average for epoch 1 was 12.8 g/kg/d,
with epoch 2 being 13.3 g/kg/d. Epoch 2 (median DOL
at discharge = 84 days) demonstrated a higher average
GV compared with the 2013 estimate of infants discharged
6 Nutrition in Clinical Practice 00(0)

Table 3. Feeding Progression and Treatment Outcomes.

Epoch 1 (n = 32) Epoch 2 (n = 32)

Reported Median Category Median Range IQR Median Range IQR P Value

DOL birth weight regained All infants 8 (5,9) 7 (6.10) 0.90

< 1000 g 8 2–17 (5,12) 7 3–21 (4.5,9) 0.64
1000–1500 g 8 2–11 (5,9) 7 5–12 (6,10.5) 0.50
DOL feedings started All infants 4 (3,7.5) 1 (1,1) < 0.0001
< 1000 g 5 3–22 (3,10) 1 1–4 (1,1) < 0.0001
1000–1500 g 3 2–11 (2.4) 1 1–2 (1,1) < 0.0001
DOL full enteral feedings All infants 22 (16,25) 7 (6,8) < 0.0001
achieved < 1000 g 22 13–50 (16,25) 8.5 6–16 (7,11.5) < 0.0001
1000–1500 g 21 9–39 (13,25) 6 5–10 (5,7) < 0.0001
Time interval to full feedings All infants 15 (11.5,19) 6 (5,7) < 0.0001
< 1000 g 14.5 9–36 (12,18) 7 5–15 (6,9.5) 0.0001
1000–1500 g 15.5 7–37 (11,23) 5 4–9 (4,6) < 0.0001
Days receiving parenteral All infants 17.5 (12,22.5) 5.5 (5,7.5) < 0.0001
nutrition < 1000 g 19 10–49 (13,23) 7.5 5–16 (6.5,9.5) < 0.0001
1000–1500 g 16.5 7–35 (9,22) 5 3–10 (4,6) < 0.0001
Days of indwelling central line All infants 17.5 (10.5,25) 6 (4,8) < 0.0001
< 1000 g 19 0–56 (13,36) 7.5 2–16 (6,9.5) 0.0009
1000–1500 g 16.5 0–38 (0,24) 5 0–10 (2,7) 0.066
Highest base deficit week 1 All infants 7.2 (5,8.3) 7.2 (4.1,8.6) 0.90
< 1000 g 7.8 0–14.2 (7,8.4) 8.4 4.7–22.5 (7.4,10.2) 0.26
1000–1500 g 5 2.9–11.3 (4,7.7) 6 0.9–13.3 (2.4,7.8) 0.62
Highest BUN week 1 All infants 34 (27,43.5) 39 (28,49.5) 0.23
< 1000 g 38 16–55 (28,48) 38 22–95 (28.5,54) 0.75
1000–1500 g 29 12–48 (19,41) 40 16–59 (28,46) 0.089
Highest BUN on full feedings All infants 35 (27,43.5) 20 (16,25.5) < 0.0001
< 1000 g 38 17–66 (31,50) 22 17–60 (18.5,30.5) 0.0092
1000–1500 g 29 12–48 (19,41) 20 7–28 (13,22.5) 0.010
Number of abdominal films All infants 19 (11,33) 0.5 (0,2) < 0.0001
< 1000 g 29 4–59 (17,38) 2 0–5 (0.5,3.5) < 0.0001
1000–1500 g 14 1–26 (4,19) 0 0–5 (0,1.5) < 0.0001
Number of packed red blood All infants 2.5 (0.5,6.5) 0 (0,0.5) < 0.0001
cell transfusions < 1000 g 6 1–12 (2,11) 1 0–5 (0,4) 0.0015
1000–1500 g 0 0–3 (0,2) 0 0–1 (0,0) 0.011

BUN, blood urea nitrogen; DOL, day of life; IQR, interquartile range.

between 78–174 DOL at 12.9 g/kg/d. These small differences
in GV are likely quite clinically significant when evaluated
at a more detailed level by birth weight groupings. When
comparing weight categories, Horbar et al reported mean
GVs from birth to discharge for infants born 501–1000 g to
be between 13.4–13.6 g/kg/d.1 The median velocity for our
epoch 2 group of infants < 1000 g was 14.5 g/kg/d (and 15.3
g/kg/d for infants < 750 g). Horbar et al also reported the
mean GV for infants 1000–1500 g to range between 12.1–
13.1 g/kg/d.1 Our epoch 2 group infants had a median GV
of 13.1 g/kg/d.
We decided to analyze growth in more detail for epoch 2,
given the different growth goals dependent on kilogram size.
Between first weight exceeding birth weight to first weight
exceeding 2 kg, growth averaged 18.3 g/kg/d (median). From
the first weight of exceeding 2 kg until discharge, growth
averaged 31.7 g/d (median). Our goal is to provide average
GV around minimum of 18 g/kg/d for infants <2 kg because
this has been correlated with improved neurodevelopmental
outcomes.6 Intrauterine growth has been estimated to range
between mean of 12.2–18.6 g/kg/d from 24–33 weeks GA
(and weight < 2 kg).7 Our estimations of growth in the study
group achieved the upper range of these goals. Growth
from 2 kg to discharge remained appropriate, with standard
growth goals for infants < 3 months being 25–35 g/d.7
Analyzing trends in median head circumference
percentiles at 36 weeks GA and at discharge remained
statistically similar, except for the discharge percentile
between combined weight groups (25th vs 39th percentile,
P = 0.0004). However, it remains notable that the epoch
1 infants either maintained or decreased in percentiles,
whereas epoch 2 infants maintained or increased across
all groups. The incidence of high-grade IVH approached
significance in the epoch 2 group of infants < 1000 g
Thoene et al
(P = 0.06) but not in any other weight category group.
After re-analyzing head circumference size and percentiles
for this weight category when excluding infants with
high-grade IVH, results remained similar to initial results
at both 36 weeks and at discharge. The median growth
percentiles for head circumference were the 15th percentile
(control) vs the 15th percentile (study) at 36 weeks GA and
the 15th percentile vs 30th percentile at discharge. There
were no differences in the number of infants per group who
discharged with head circumferences plotting < the 10th
percentile for age (13% control vs 6% study). In review of
epoch 2 infants, this baseline at birth was 13% of the group.
We do not have baseline statistics on head size for epoch
2 group due to no data being previously collected on these
infants for this data point. Compared with percentages
at 36 weeks GA, this discharge trend demonstrates a
decreased rate of small head circumference in epoch 2
and an increased rate in epoch 1. These comparisons are
most prominent in the category of smallest infants born <
1000 g. Discharge rates of small head circumference was
22% in epoch 1 vs 0% in epoch 2. This is an important
consideration because sustaining or catching up in head
growth has been linked to improved neurodevelopment
outcomes in infants born < 750 g,8 born < 1500 g,9 and
born < 1500 g and weight-plotting < 10 percentile at birth.10
Rationale for Feeding Protocol Modifications
and Assessing Clinical Outcomes
Enteral feedings were initiated earlier in epoch 2 infants
(DOL 1 vs 4) and were achieved sooner (DOL 7 vs 22)
than epoch 1 infants. Available literature supports early
feeding introduction because low enteral volumes are well
tolerated, and late introduction of feedings does not reduce
NEC risk.11,12 Length of trophic enteral feedings before
advancement is not well defined, but evidence demonstrates
that delayed feeding only contribute to a prolonged time to
achieve full enteral feedings while requiring longer PN use.13
Our outcomes support this: the quicker advancement in the
epoch 2 group eliminated the prolonged need for PN and
an indwelling central line for access (5.5 vs 17.5 days). Also
unique is the significant decrease in abdominal X-rays to
specifically assess for perceived feeding intolerance (average
of 19 vs 0.5 throughout hospitalization). Not only does
faster enteral advancement optimize nutrition, but it further
reduces the risk of PN-associated complications. There were
no differences in NEC between groups, and there were
no cases of NEC in epoch 2 infants despite faster enteral
feeding advancement. This concurs with a recent Cochrane
review, which demonstrates the ability of very low birth
weight infants to tolerate more rapid enteral advancement
of 30–40 ml/kg/d, further leading to a reduced time on
PN without an increased risk for NEC.13 Additionally, an
advancement of 30 ml/kg can still be utilized in infants
7
weighing as little as 750 g,14 but limited data is available
for infants < 750 g. Our epoch 2 group had 4 infants born
weighing < 750 g (range 618–730 g). All infants began
feedings on the first DOL. The range to achieve full enteral
feeding was 7–16 days of life, with median of 11 days. Birth
weight was regained by median DOL 5. None of these
infants developed NEC. This contrasts data by Viswanathan
et al, who reported that a slow standardized enteral feeding
protocol reduced the incidence of NEC in infants < 750 g.15
Infants began enteral feedings at roughly 14 days of life and
were then slowly advanced to full enteral feedings by 44–
52 days of life.15 Based on our data, study infants < 750 g
achieved enteral feedings earlier than the collective group of
epoch 1 infants (median of 22 days).
Attainment of full enteral feedings (defined as the infant
receiving a minimum of 150 mL/kg/d of preterm infant for-
mula, or breast milk fully fortified to 24 cal/oz using human
milk fortifier and a protein modular) were also achieved
faster by initiating milk fortification earlier. Rationale for
this is exemplified by Miller et al, who demonstrated that
the transition from full PN to full EN was a determinant
of later growth failure.16 A partial reason for this was
the lower protein provision with decreasing supplemental
PN use.16 Therefore, it is important to recognize that PN
will be more limited with faster enteral advancement. This
supports early milk fortification because more nutrition
is required from enteral feedings alongside supplemental
PN to meet overall nutrition goals for growth. Likewise,
past studies have demonstrated tolerance with early milk
fortification at enteral volumes of 20 mL/kg/d or at first
feeding initiation.17,18 Furthermore, enteral protein goals
for epoch 2 were slightly higher; the American Academy of
Pediatrics recommends a range of 3.8–4.4 g/kg/d for infants
< 1000 g, followed by a recommended dose of 3.4–4.2 g/kg/d
for infants 1000–1500 g.19,20 Safe upper end protein doses
are reported as 4.9 g/kg/d.21
Incidence rates of BPD at 36 weeks GA remained sta-
tistically similar in our analysis between combined groups.
However, infants < 1000 g had a higher rate in epoch 2
compared with control (92% vs 38%, P = 0.006). There
are multiple factors to consider when assessing these differ-
ences. Firstly, a higher portion of epoch 1 infants receiving
dexamethasone (81% vs 22%, P < 0.0001), which has been
linked to a lower risk of having chronic lung disease at
36 weeks GA.22 Secondly, epoch 2 infants experienced
improved growth compared with epoch 1, which we would
expect would reduce incidence of BPD.
There were no statistical differences in rates of ROP be-
tween groups (17% control vs 19% study). Dexamethasone
has been linked to higher rates of severe ROP, but this is not
significant in survivors.22 One contributing factor among
multiple variables to consider in ROP development is an
early adequate delivery of essential nutrients followed by
sufficient early growth.23 Equations and models to identify
8 Nutrition in Clinical Practice 00(0)
infants at high risk for developing ROP include growth
analysis24 ; therefore, growth must be closely monitored
from a clinical standpoint to promote best outcomes.
Strength and Limitations
The primary strength of this study is that it analyzes
growth and feedings outcomes in very low birth weight
infants cared for in a high level NICU. Our high inclusion
encompasses various acuity levels and is more indicative
of a true population of NICU infants. Although we have
a standard protocol for feeding advancement, it may vary
based on an infant’s unique clinical status. Again, this is
more reflective of real life scenarios, further allowing our
feeding methods to be highly applicable to other neonatal
units.
Limitations of this study include that it is retrospective;
thus, there were limitations on data to be analyzed for epoch
1 insofar that it was previously collected. Evaluation of head
circumference may vary among nursing staff due to differ-
ences in measuring tape placement. Growth measurements
were unavailable for infants discharged prior to 36 weeks
GA. Although many epoch 2 infants demonstrated catch-
up growth for head size, it remains unknown if this led to
improved neurodevelopmental outcomes.
Conclusions
Data from our current feeding practices with a progressive,
standardized, and evidenced-based enteral feeding protocol
demonstrated that early introduction and rapid advance-
ment of enteral feedings are feasible in very low birth weight
infants without an increased risk for NEC. This aggressive
feeding practice is associated with shorter duration of
central line access, less PN use, and improved weight gain
and head growth.
Statement of Authorship
M. Thoene and A. Anderson-Berry equally contributed to the
conception and design of the research. M. Thoene contributed
to the acquisition of the data. E. Lyden contributed to the
analysis of the data. M. Thoene, E. Lyden, and A. Anderson-
Berry contributed to the interpretation of the data. M. Thoene
and A. Anderson-Berry drafted the manuscript. All authors
critically revised the manuscript, agree to be fully accountable
for the integrity and accuracy of the work, and approved the
final manuscript.
References
1. Horbar JD, Ehrenkranz RA, Badger GJ, et al. Weight growth velocity
and postnatal growth failure in infants 501 to 1500 grams: 2000–2013.
Pediatrics 2015;136:e84–e92.
2. Hanson C, Sundermeier J, Dugick L, Lyden E, Anderson-Berry AL.
Implementation, process, and outcomes of nutrition best practices for
infants <1500 g. Nutr Clin Pract 2011;26:614–624.
3. Patel AL, Engstrom JL, Meier PP, Kimura RE. Accuracy of methods
for calculating postnatal growth velocity for extremely low birth weight
infants. Pediatrics 2005;116:1466–1473.
4. Bartholomew J, Martin CR, Allred E, et al. Risk factors and correlates
of neonatal growth velocity in extremely low gestational age newborns:
the ELGAN Study. Neonatology 2013;104:298–304.
5. Stark AR, Carlo WA, Tyson JE; National Institute of Child Health
and Human Development Neonatal Research Network, et al. Adverse
effects of early dexamethasone in extremely-low-birth-weight infants.
National Institute of Child Health and Human Development Neonatal
Research Network. N Engl J Med 2001;344:95–101.
6. Ehrenkranz R, Dusick A, Vohr B, Wright L, Wrage, L, Poole W.
Growth in the neonatal intensive care unit influences neurodevelop-
mental and growth outcomes of extremely low birth weight infants.
Pediatrics 2006;117:1253–1261.
7. Corkins MR, Balint J, Bobo E, Plogsted S, Yaworski JA, eds.
The A.S.P.E.N. Pediatric Nutrition Suppport Core Curriculum. Silver
Spring, MD: American Society for Parenteral and Enteral Nutrition;
2010.
8. Claas M, de Vries L, Koopman C, et al. Postnatal growth of preterm
born children ࣘ 750g at birth. Early Hum Dev, 2011;87(7):495–
507.
9. Sammallahti S, Pyhälä R, Lahti M, et al. Infant growth after preterm
birth and neurocognitive abilities in young adulthood. Pediatrics
2014;165:1109–1115.
10. Leppänen M, Lapinleimu H, Lind A; PIPARI Study Group, et al.
Antenatal and postnatal growth and 5-year cognitive outcome in very
preterm infants. Pediatrics 2014;133:63–70.
11. Morgan J, Bombell S, McGuire W. Early trophic feeding versus enteral
fasting for very preterm or very low birth weight infants. Cochrane
Database Syst Rev 2013;3:CD000504.
12. Morgan J, Young L, McGuire W. Delayed introduction of progressive
enteral feeds to prevent necrotising enterocolitis in very low birth weight
infants. Cochrane Database Syst Rev 2014;12:CD001970.
13. Morgan J, Young L, McGuire W. Slow advancement of enteral feed
volumes to prevent necrotising enterocolitis in very low birth weight
infants. Cochrane Database Syst Rev 2015;10:CD001241.
14. Karagol BS, Zenciroglu A, Okumus N, Polin RA. Randomized con-
trolled trial of slow vs rapid enteral feeding advancements on the
clinical outcomes of preterm infants with birth weight 750–1250 g.
JPEN J Parenter Enteral Nutr 2013;37:223–228.
15. Viswanathan S, McNelis K, Super D, Einstadter D, Groh-
Wargo S, Collin, M. Standardized slow enteral feeding protocol
and the incidence of necrotizing enterocolitis in extremely low
birth weight infants. JPEN J Parenter Enteral Nutr 2015;39:644–
654.
16. Miller M, Vaidya R, Rastogi D, Bhutada A, Rastogi S. From parenteral
to enteral nutrition: a nutrition-based approach for evaluating postna-
tal growth failure in preterm infants. JPEN J Parenter Enteral Nutr
2014;38:489–497.
17. Shah SD, Dereddy N, Jones TL, Dhanireddy R, Talati AJ. Early
versus delayed human milk fortification in very low birth weight
infants: a randomized controlled trial. J Pediatr 2016;174:126–
131.e1.
18. Tillman S, Brandon DH, Silva SG. Evaluation of human milk fortifi-
cation from the time of the first feeding: effects on infants of less than
31 weeks gestational age. J Perinatol 2012;32:525–531.
19. NCM. Pediatric Nutrition Care Manual. Academy of Nutrition and
Dietetics. Available at: https://www.nutritioncaremanual.org/topic.
cfm?ncm_category_id=32&lv1=273269&lv2=145079&lv3=269737&
ncm_toc_id=269737&ncm_heading=Preterm%20Infants
20. Kleinman RE, Greer FR, AAP Committee on Nutrition. Pediatric
Nutrition Handbook, 7th ed. Elk Grove Village, IL: American Academy
of Pediatrics; 2014.
Thoene et al
21. Wagner J, Hanson C, Anderson-Berry A. Considerations in meeting
protein needs of the human milk-fed preterm infant. Adv Neonatal Care
2014;14:281–289.
22. Halliday HL, Ehrenkranz RA, Doyle LW. Late (>7 days) postnatal
corticosteroids for chronic lung disease in preterm infants. Cochrane
Database Syst Rev 2009;1:CD001145.
9
23. VanderVeen DK, Martin CR, Mehendale R, Allred EN, Dammann O,
Leviton A; ELGAN Study Investigators. Early nutrition and weight
gain in preterm newborns and the risk of retinopathy of prematurity.
PLoS One 2013;8:e64325.
24. Binenbaum, G. Algorithms for the prediction of retinopathy of prema-
turity based on postnatal weight gain. Clin Perinatol 2013;40;261–270.