ORIGINAL RESEARCH

Protein Feeding in Pediatric Acute Kidney

Injury Is Not Associated With a Delay
in Renal Recovery
Ursula G. Kyle, MS, RD/LD,* Ayse Akcan-Arikan, MD,*, Jaime C. Silva, MD,*
Michelle Goldsworthy, MSN,* Lara S. Shekerdemian, MD,* and Jorge A. Coss-Bu, MD*

Objective: Critically ill children with acute kidney injury (AKI) are at high risk of underfeeding. Newer guidelines for nutrition support
recommend higher protein intake. Therefore, the study evaluated the effects of protein feeding on the resolution of AKI and compared
energy and protein intake in patients with and without AKI after implementation of Nutrition Support guidelines.
Design: Retrospective study.
Subjects: Five hundred twenty critically ill children from October 2012 to June 2013 and October to December 2013.
Main Outcome Measure: Energy and protein intake in patients with no AKI, resolved, or persistent AKI. Energy and protein intake was
documented for days 1-8 of Pediatric Intensive Care Unit stay and in the postimplementation versus preimplementation period of nutri-
tion support guidelines. AKI was defined by modified pRIFLE. Persistent AKI was defined as patients who did not resolve their AKI during
the study period.
Results: A higher percentage of patients with resolved and persistent AKI met $ 80% of protein needs versus no AKI. After adjustment
for Pediatric Risk of Mortality Score, the odds ratio for protein intake of $ 80% compared to ,80% of estimated protein needs was not
significant, which suggests that higher protein intake was not associated with nonresolution of AKI. There were significant improvements
in the cumulative protein gap in patients with no AKI in the postimplementation (21.0 [21.7 to 20.6] g/kg/day) compared to preimple-
mentation period (21.3 [21.7 to 20.9] g/kg/day, P 5 .001) and persistent AKI in the postimplementation (20.8 [21.4 to 20.1] g/kg/day)
compared to preimplementation (21.3 [21.7 to 20.9] g/kg/day, P 5 .03).
Conclusions: Higher protein intake was not associated with a delay in renal recovery in patients with AKI after adjustment for severity
of illness. Protein intake was improved in critically ill children with no AKI, resolved, and persistent AKI after implementation of Nutrition
Support Guidelines, but underfeeding persisted in these patients.
2016 by the National Kidney Foundation, Inc. All rights reserved.

Introduction their Pediatric Intensive Care Unit (PICU) stay,6 and this

H OSPITAL UNDERNUTRITION IS a known risk

factor for morbidity and mortality in children, ado-
lescents, and adults,1-3 and malnutrition has been shown to
adversely affect patient outcome. In addition, suboptimal
nutrient provision contributes to the deterioration of
nutritional status and has been shown to increase the risk
of multiorgan failure, length of stay (LOS), and
mortality.4,5 Critically ill children are underfed early in
*
Section of Critical Care Medicine, Department of Pediatrics, Baylor College
of Medicine/Texas Childrens Hospital, Houston, Texas.
Section of Nephrology, Department of Pediatrics, Baylor College of Medicine/
Texas Childrens Hospital, Houston, Texas.
Financial Disclosure: All authors of this research have no conflict of interest to
declare related with employment, consultancies, stock ownership, honoraria, and
grants or funding.
Support: See Acknowledgments on page XXX.
Address correspondence to Jorge A. Coss-Bu, MD, Section of Critical Care
Medicine, Department of Pediatrics, Baylor College of Medicine/Texas Chil-
drens Hospital, 6621 Fannin WT6-006, Houston, TX 77030. E-mail:
jorgec@bcm.edu, jacossbu@texaschildrens.org
2016 by the National Kidney Foundation, Inc. All rights reserved.
1051-2276/$36.00
http://dx.doi.org/10.1053/j.jrn.2016.09.009
may contribute to worse outcome.
Acute kidney injury (AKI) occurs in 10% of all PICU ad-
missions and in up to 3/4 of patients with cardiorespiratory
failure.7 The risk of acute and chronic malnutrition is high
in patients with AKI, and the presence of malnutrition
in the context of AKI has been associated with more
severe clinical deterioration and organ dysfunction.8-11
Derangements in substrate metabolism and body
composition12 as a result of physiological changes due to
AKI can lead to hypercatabolism,13 hypoalbuminemia,
loss of lean body mass, and depletion of adipose tissue,
even with adequate ingestion of nutrients.14 Maintenance
of protein balance in such conditions requires that at least
adequate energy and protein intake be provided during
acute illness.
Nutrition support is often deferred in critically ill chil-
dren, until they are medically stabilized, which delays
adequate nutrition support for several days due to fluid re-
striction, digestive intolerance, and interruption of feeding
for diagnostic and therapeutic procedures.15 Pediatric pa-
tients are highly dependent on nutritional support due to
their intrinsic high anabolic drive and lower nutrient re-
serves, when compared to adults. AKI patients are at higher

Journal of Renal Nutrition, Vol -, No - (-), 2016: pp 1-8 1

2 KYLE ET AL
risk due to fluid restriction and concerns about protein
loading leading to worsening uremia. In fact, due to the
catabolic nature of the illness associated with AKI, it is
now suggested that standard recommendations for energy
and protein are appropriate for intensive care patients
with AKI.16-18 Nutrition Support Guidelines were
developed and implemented for Texas Childrens PICU
because of previous reports6,19-21 of underprescription
and underdelivery of nutrition support among critically ill
children, including patients with AKI.22
There is evidence that increased protein provision might
protect the kidney from ischemic injury and may preserve
glomerular filtration rate in critical illness23 and may lead
to faster recovery from severe AKI.24-26 Studies linking
achievement of protein requirements to outcome are
lacking in children with AKI. The purpose of this study
was to evaluate the effect of protein feeding on the
resolution of AKI and to compare energy and protein
intake in AKI compared to no AKI patients after
implementation of Nutrition Support Guidelines.
Subjects and Methods
Five hundred ninety patients aged 0-18 years admitted to
the Texas Childrens Hospital PICU for .48 hours be-
tween October 2012-June 2013 (preimplementation) and
October-December 2013 (postimplementation) were
eligible. Patients were excluded if they were admitted to
the PICU more than 72 hours after hospital admission
(n 5 51), had more than one PICU admission (n 5 18),
or had a diagnosis of end-stage renal disease, requiring renal
replacement therapy or had received a kidney transplant
(n 5 10).
Data collected included age, sex, height, admission
weight, admission diagnosis, duration of mechanical venti-
lation, medications, intravenous (IV) fluids, duration of
time in PICU and hospital LOS, severity of illness, and sur-
vival. Recumbent length of children younger than
24 months was measured using a length board. For patients
who were unable to stand due to severity of illness, knee/
heel measurements were obtained by using knee/heel cal-
ipers and converted to height using the previously validated
formulas.27 Weight was obtained with digital scales (hoist if
nonambulatory; Scaletronix Inc, Wheaton, IL) and digital
infant scales.
Patients were classified as AKI, and further stratified as
risk (R), injury (I), and failure (F) by pRIFLE creatinine
criteria.7 Urine output data were not available. Resolution
of AKI was defined as any stage of AKI present during the
period of observation and subsequently transitioning to no
AKI. Persistent AKI was defined as patients who did not
resolve their AKI during the first 8 days of PICU stay.
Weight wasting (moderate/severe) was defined as
weight-for-height $ 22 z-scores and height stunting
(moderate/severe) as height-for-age $ 22 z-scores, using
World Health Organization28 and 2000 Center for Disease
Control growth charts29 for children ,2 and $2 years,
respectively. Severity of illness was assessed on the day of
admission with the Pediatric Risk of Mortality Score
(PRISM III)30,31 and organ dysfunction with the
Pediatric Logistic Organ Dysfunction score (PELOD).32,33
Estimated energy and protein needs were determined ac-
cording to the Schofield prediction equation (basal meta-
bolic rate)34 (without correction factors) and American
Society for Parenteral and Enteral Nutrition guidelines
for critically ill children,35 respectively (protein recommen-
dations 0-2 years: 2-3 g/kg; 2-13 years: 1.5-2 g/kg;
.13 years: 1.5 g/kg).
Enteral, parenteral nutrition, and IV-glucose containing
solutions were collected. For enteral nutrition, enteral for-
mula provided and the volume given for each 24 hours was
collected for each day. For IV and parenteral nutrition, the
volume and concentration of all IV fluids were collected.
Continuous renal replacement therapy (CRRT) fluids
were not included in the calorie and protein calculations.
The actual energy and protein intakes were calculated in a
spreadsheet for each 24 hours for the first 8 days or until
oral intake was initiated. The cumulative gap or deficit
was defined as the differences between actual energy
(kcal/kg/day) or protein (g/kg/day) intake minus recom-
mended intake based on Schofield equation3 or American
Society for Parenteral and Enteral Nutrition guidelines,4
respectively, for the time the patients remained in the
PICU.
In July and August 2013, the Texas Childrens Hospital
Nutrition Support Team implemented Nutrition Support
Guidelines to improve feeding practices in our PICU.
The implementation of Nutrition Support Guidelines con-
sisted of a 45-minute training session for all PICU clinical
practitioners as well as written guidelines, made available
on the hospital intranet. The changes in energy and protein
intakes were also evaluated in the postimplementation
compared to preimplementation of Nutrition Support
Guidelines. In order to assess the effects of the energy and
protein intake on the resolution of AKI, a cutoff of $
80% was used.
The study was approved by the Baylor College of Med-
icine Institutional Review Board, with a waiver of consent.
Statistics
Continuous variables were classified as normally or non-
normally distributed based on an analysis of frequency dis-
tribution graph. Normally distributed continuous
variables, reported as mean (standard deviation), were
compared using paired and unpaired t test. Nonnormally
distributed variables were expressed as median with inter-
quartile ranges. For dependent samples, Wilcoxon Signed
Rank tests and for independent samples Mann-Whitney
U nonparametric tests were performed. Pearsons chi-
square or Fishers exact test was used for independent obser-
vations to determine differences between groups of
 PROTEIN FEEDING IN CHILDREN WITH AKI 3

categorical variables. Odds ratios with 95% confidence in-

terval were determined by simple and multiple logistic
regression models.
Statistical significance was established a priori P , .05.
Statistical analysis was performed with Statview version
5.0.1 (SAS Institute Inc, Inc, Cary, NC).

Results
One hundred fifty-six of 511 patients (30%) had AKI
(Table 1). Of the AKI patients, 72% of patients had R,
19% had I, and 9% had F. AKI burden was highest on day
2 with 15% (79) of patients classified as AKI, which
decreased to 6% (30) of patients by day 8 (Fig. 1). Nine pa-
tients received CRRTat sometime during the first 8 days of
PICU stay.
Patients with AKI were significantly shorter and weighed
less than no AKI patients. Weight-for-age and weight-for-
height z-scores were not significantly different. However,
height-for-age z-scores were significantly lower in AKI
compared to non-AKI patients (Table 1), which suggests
that they had more height stunting or chronic malnutrition.
There were no significant differences in diagnostic cate-
gories between no AKI and AKI patients, with 46% and
42% having a respiratory diagnosis, 20% and 25% sepsis/
septic shock/infection, 6% and 10% cancer, 6% and 5%
seizure disorders, respectively. AKI patients had higher
PELOD and PRISM scores and longer PICU and hospital
LOS than no AKI (Table 1).
AKI resolved after 1 (1-2.5) days (median [interquartile
range]) in 72% (n 5 112) of patients. Patients with persis-
tent AKI (n 5 44/156) had a median duration of AKI of
4.5 (2.5-6) days of the first 8 days of PICU stay. Patients
with chronic disease (AKI 55.8%, no AKI 44.2%,
P 5 .41) versus acute disease (AKI 44.2%, no AKI 48.1%,
Figure 1. Incidence of risk (R), injury (I), or failure (F) in critically
ill children by admission day to the Pediatric Critical Care Unit,
as determined by modified pRIFLE (Akcan-Arikan, 2007).
P 5 .41) did not have a higher incidence of AKI during
the PICU stay. In patients with persistent AKI during the
observation period, there was a significant association
with moderate/severe height stunting, PRISM score $
11, PELOD score $ 20, PICU LOS $ 10 days, hospital
LOS $ 20 days, and decreased survival compared to pa-
tients with no AKI (Table 2).
Median energy and protein intake was less than recom-
mended in the first 8 days of PICU stay in all patient
groups (Table 3). Patients with no AKI, resolved AKI,
and persistent AKI received 73%, 80%, and 80% of rec-
ommended energy needs, but only 39%, 42%, and 51%
of protein needs, respectively. There was no significant
difference in energy and protein intake between patients
with resolved and persistent AKI. The energy gap be-
tween the patients with resolved and persistent was signif-
icantly smaller AKI compared to no AKI. In addition, the
protein gap between patients with persistent versus no
AKI was also significantly smaller. However, the protein

Table 1. Characteristics of Patients Without and With Acute Kidney Injury (AKI)
Basic Characteristics No AKI AKI P

Gender (male/female) % (n) 54.6/45.4 (194/161) 57.1/42.9 (89/67) .615*

AKI (R/I/F) % (n) 72.4/18.6/9.0 (113/29/14)
Age (y) 1.5 (0.4-6.2) 1.2 (0.2-5.8) .100
Height (cm) 81.0 (64.1-116.7) 76.3 (58.0-109.5) .023
Height-for-age z-score 20.34 (2.0) 20.86 (2.3) .009
Weight (kg) 11.0 (6.6-21.3) 9.3 (4.8-20.9) .049
Weight-for-age z-score 20.47 (1.75) 20.72 (1.99) .157
Weight-for-height z-score 20.22 (1.7) 20.0 (2.00) .253
PELOD score 11 (1-11) 11 (2-12) .010
PRISM score 4 (1-8) 7 (3-12.5) ,.001
PICU length of stay (d) 5 (4-9) 7 (3.5-12) ,.002
Hospital length of stay (d) 11 (8-19) 15 (10-24) ,.003
Survival (survived/died) % (n) 95.8/4.2 (340/15) 91.7/8.3 (143/13) .060*
F, failure; I, injury; R, risk; PELOD, Pediatric Logic Organ Dysfunction Score, score 0-71; PICU, Pediatric Critical Care Unit; PRISM, Pediatric
Risk of Mortality score, score 1-76 points; SD, standard deviation.
Median (25-75th percentile) or mean 6 SD; AKI versus non-AKI.
*Chi-square.
Median (25-75th percentile) MannWhitney rank test.
Mean 6 SD, unpaired t test.
4 KYLE ET AL

Table 2. Odds Ratio (OR) in Patients With Resolved and Persistent Acute Kidney Injury (AKI) Compared To No AKI
Basic Characteristics % (n) % (n) OR (CI) P

Weight wasting No/mild Moderate/severe

No AKI 87.0 (309) 13.0 (46) 1
AKI resolved 92.0 (103) 9.0 (9) 0.6 (0.3-1.2) .428
AKI persistent 77.3 (34) 22.7 (10) 1.9 (0.9-4.3) .083
Height stunting No/mild Moderate/severe
No AKI 82.0 (291) 18.0 (64) 1
AKI resolved 78.6 (88) 21.4 (24) 1.2 (0.7-2.1) .422
AKI persistent 65.9 (29) 34.1 (15) 2.4 (1.2-4.6) .014
PRISM score ,10 $11
No AKI 84.0 (274) 16.0 (52) 1
AKI resolved 77.8 (74) 28.2 (29) 2.1 (1.2-3.5) .050
AKI persistent 55.5 (24) 41.5 (17) 3.7 (1.9-7.4) ,.001
PELOD score ,20 $21
No AKI 86.7 (281) 13.3 (43) 1
AKI resolved 83.0 (83) 17.0 (17) 1.1 (0.5-2.2) .351
AKI persistent 65.9 (27) 34.1 (14) 3.3 (1.1-9.4) ,.001
PICU LOS (d) ,10 $10
No AKI 76.9 (273) 23.1 (55) 1
AKI resolved 69.6 (78) 30.4 (34) 1.5 (0.9-2.3) .651
AKI persistent 52.3 (3) 47.7 (21) 3.0 (1.6-5.8) ,.001
Hospital LOS (d) ,15 $15
No AKI 64.5 (229) 35.5 (126) 1
AKI resolved 53.6 (60) 46.4 (52) 1.6 (1.0-2.4) .039
AKI persistent 38.6 (17) 61.4 (27) 2.9 (1.5-5.5) .001
Survival Alive Died
No AKI 95.8 (340) 4.2 (15) 1
AKI resolved 96.8 (105) 6.2 (7) 1.5 (0.6-3.8) .381
AKI persistent 86.4 (38) 13.6 (6) 3.6 (1.3-9.8) .013
CI, confidence interval; LOS, length of hospital stay; PELOD, pediatric logistic organ dysfunction; PICU, Pediatric Critical Care Unit; PRISM,
Pediatric Risk of Mortality Score.
Moderate/severe stunting: height-for-age .2 z-scores below normal, 2000 CDC growth charts; moderate/severe wasting: weight of height .2
z-scores below normal.

and energy gaps were not significant between patients
with resolved and persistent AKI.
Patients (n 5 9) who required CRRT during the PICU
stay received 64.1% (16.6-80.5) of recommended energy
needs and 49.8% (10.1-86.3) of recommended protein
needs compared to patients (n 5 147) with AKI without
CRRT: 80.8% (56.6-112.0, P 5 .06) and 43.3% (23.3-
67.9, P 5 .91), respectively. Patients with AKI requiring
CRRT were sicker than those without AKI as shown by
higher PRISM and PELOD scores and longer PICU and
hospital LOS. Energy and protein provided by CRRT
was not included in the calculations.
A higher percentage of patients with resolved and persis-
tent AKI met $ 80% of protein needs than patients with no
AKI (Table 4). However, after adjustment for PRISM
score, the odds ratio for protein intake of $ 80 compared
to ,80% of estimated protein needs was not significant,
which suggests that higher protein intake was not associated
with nonresolution of AKI.
There were significant differences in the actual energy
and protein intakes in the postimplementation compared
to preimplementation period (Table 5) in all groups, except
in patients with persistent AKI, which suggests that patients
with persistent AKI were better fed in the postimplementa-
tion period. For the period of observation, there were sig-
nificant improvements in the cumulative protein gap in
patients with no AKI and persistent AKI, but not in the cu-
mulative energy gap in the postimplementation compared
to preimplementation period for any of the patient groups
(Table 5). Furthermore, significantly more patients
received $ 80% compared to ,80% of protein in the post-
implementation compared to preimplementation period:
no AKI patients: 16% versus 9%, P 5 .045; persistent
AKI: 38% versus 11%, P 5 .045; and resolved AKI: 24%
versus 14%, P 5 .232. Thus, in general, the actual energy
and protein intakes were improved in all patient groups after
implementation of Nutrition Support Guidelines. For all
patients, there were significant improvements in the actual
protein, percentage of protein intake of recommended
needs and protein gap in the postimplementation compared
to preimplementation period.
Discussion
The cumulative energy and protein gap in patients with
persistent AKI and the energy gap in patients with resolved
AKI were significantly lower than in patients with no AKI;
and the actual protein intake, although not significant, was
higher in resolved and persistent AKI than no AKI. A
 PROTEIN FEEDING IN CHILDREN WITH AKI 5

Table 3. Recommended and Actual Energy and Protein Intake and Gap in Patients With No Acute Kidney Injury (AKI) and With
Resolved and Persistent AKI in the First 8 Days of PICU Stay
AKI Persistent
Basic Characteristics No AKI (n 5 355) AKI Resolved (n 5 112) P* (n 5 44) P P

BMR kcal/kg/d 51.2 (43.0-57.7) 48.9 (41.8-54.1) 49.2 (37.6-56.1)

Actual energy intake kcal/kg/d 35.7 (21.6-50.5) 41.3 (24.2-53.4) .18 38.9 (24.0-54.1) .46 .98
% BMR 73.1 (50.2-100.3) 80.0 (52.0-110.6) .05 79.8 (57.2-119.3) .11 .82
Cumulative energy gap kcal/kg/d 213.0 (224.2 to 0.1) 29.5 (220.0-4.8) .02 26.9 (219.1 to 8.9) .05 .66
Recommended protein intake g/kg/d 2.5 (1.8-2.5) 2.5 (1.8-2.5) 2.5 (1.8-2.5)
Actual protein intake g/kg/d 0.81 (0.4-1.3) 0.96 (0.4-1.4) .29 1.02 (0.5-1.9) .04 .20
% protein needs 39.0 (19.0-62.6) 42.4 (21.2-64.8) .54 51.3 (25.6-82.8) .07 .17
Cumulative protein gap g/kg/d 21.3 (21.8 to 20.8) 21.3 (21.6 to 0.8) .47 21.1 (21.6 to 20.4) .04 .15
BMR, basal metabolic rate, estimated by Schofield; IQR, interquartile range; PICU, Pediatric Critical Care Unit.
Energy or protein gap 5 actual minus recommended intake; MannWhitney test.
Values are median (IQR).
*AKI resolved versus no AKI.
AKI persistent versus no AKI.
AKI persistent versus resolved.
Wilcoxon Signed Rank test between actual and recommended energy and protein intake, P , .05.

higher percentage of patients with AKI met $ 80% of pro-
tein needs. However, protein intake $ 80% of estimated
needs, after adjustment for severity of illness, was not signif-
icantly associated with persistent AKI. This suggests that
higher protein intake was not associated with a delay in
renal recovery in patients with AKI. Overall, there was a
significant improvement in the protein intake in the post-
implementation period.
It is well known that children with chronic disease are
more likely to have chronic malnutrition/stunting
(height-for-age z-score). Although patients with chronic
disease did not have a higher incidence of AKI during
PICU stay, patients with AKI had lower height-for-age z-
score (chronic malnutrition/stunting), and there was a sig-
nificant association in patients with persistent AKI with
moderate/severe height stunting. It is likely that height
stunting and AKI are related, but no causality can be in-
ferred in this retrospective study.
Patients with persistent AKI were sicker as assessed by
PRISM III and PELOD, which might have lead to delayed
enteral feeding but does not fully explain the underfeeding
and underprescription of energy and protein. They also had
longer PICU and hospital LOS and lower survival.
Overall, energy and protein intake did not reach the feeding
goals. Mehta et al.36 reported an association between higher
protein intake (.60% of prescribed) and lower mortality.
The current study suggests that higher protein intake was
not associated with a delay in renal recovery in patients with
AKI. These two studies suggest that protein intake of at least
half of recommended intake may have had beneficial effects
without associated negative outcome. Prospective random-
ized studies are necessary to determine the benefits of higher
protein intake in critically ill children with and without AKI.
Because recommendations for nutrition support were
recently updated, the Nutrition Support Team at Texas
Childrens Hospital PICU decided to implement the new
recommendations. The implementation of Nutrition Sup-
port Guidelines improved feeding practices in the PICU,
but underfeeding persisted in critically ill children. The
percentage of protein intake increased significantly in no
AKI, but not in patients with AKI, but the intake remained
insufficient after the implementation of the Nutrition

Table 4. Odds Ratio (OR) in Patients With Resolved and Persistent Acute Kidney Injury (AKI) Compared To No AKI
Basic Characteristics % (n) % (n) P* OR (CI) P

Calorie intake ,80% $80%

No AKI 57.7 (205) 42.3 (150) 1
AKI resolved 50.0 (56) 50.0 (56) 1.3 (0.8-2.1) .238
AKI persistent 54.5 (24) 45.5 (20) .350 1.2 (0.6-2.3) .633
Protein intake ,80% $80%
No AKI 88.2 (313) 11.8 (42) 1
AKI resolved 83.0 (93) 17.0 (19) 1.1 (0.6-2.1) .820
AKI persistent 72.7 (32) 27.3 (12) .014 1.9 (0.8-4.2) .135
CI, confidence interval; PRISM, Pediatric Risk of Mortality Score.
*Unadjusted chi-square.
Adjusted for PRISM score.
6 KYLE ET AL

Table 5. Recommended and Actual Energy and Protein Intake and Gap in Patients With No AKI and With Resolved or Persistent
AKI in the First 8 Days of PICU Stay
All Patients No AKI
Implementation Pre (n 5 328) Post (n 5 183) Pre (n 5 227) Post (n 5 128)

BMR kcal/kg/d 50.1 (42.6-56.1) 50.7 (42.1-57.6) 51.3 (43.0-57.6) 51.1 (43.1-57.9)
Actual energy intake kcal/kg/d 36.6 (22.9-50.3) 40.3 (21.9-54.9) 35.3 (21.8-48.6)* 38.1 (21.5-53.0)*
% BMR 73.9 (49.9-102.1) 79.2 (52.4-106.8) 69.3 (48.8-97.4) 77 (51.6-105.4)
Energy gap kcal/kg/d 212.2 (222.9 to 1.1) 210.1 (222.7 to 3.4) 214.0 (224.2 to 21.4) 211.0 (224.7 to 1.9)
Recommended protein 2.5 (1.8-2.5) 2.5 (1.8-2.5) 2.5 (1.8-2.5) 2.5 (1.8-2.5)
intake g/kg/d
Actual protein intake g/kg/d 0.8 (0.4-1.2) 1.1 (0.5-1.6) 0.74 (0.4-1.2)* 0.96 (0.5-1.6)*
% protein needs 35.6 (18.9-54.4) 52.3 (23.4-75.4) 34.5 (18.7-53.2) 48.6 (20.7-71.1)
Protein gap g/kg/d 21.3 (21.8 to 20.9) 21.0 (21.6 to 20.5) 21.3 (21.7 to 20.9) 21.0 (21.7 to 20.6)
Resolved AKI Persistent AKI
Pre (n 5 83) Post (n 5 29) Pre (n 5 18) Post (n 5 26)
BMR 48.9 (41.9-53.6) 49 (38.2-58.1) 50.3 (41.4-55.4) 46.9 (34.3-57.4)
Actual energy intake 40.8 (24.8-53.4)* 41.8 (19.5-52.1)* 32.9 (24.5-49.2)* 45.8 (23.7-66.2)
% BMR 80.9 (53.8-111.3) 77.2 (37.9-103.4) 72.8 (49.8-90.3) 90.4 (74.5-120.0)
Energy gap 29.1 (218.7 to 25.0) 212.4 (220.9 to 1.7) 213.1 (220.7 to 25.3) 22.8 (211.9 to 12.1)
Recommended protein intake 2.5 (1.8-2.5) 2.5 (1.8-2.5) 2.5 (1.8-2.5) 2.5 (1.8-2.5)
Actual protein intake 0.89 (0.4-1.3)* 1.25 (0.4-1.5)* 0.8 (0.4-1.4)* 1.39 (0.6-2.1)*
% protein needs 40.6 (21.5-59.1) 50 (18.7-79.2) 33.8 (18.6-54.4) 61.5 (32.3-96.1)
Protein gap 21.3 (21.7 to 20.9) 21.2 (21.5 to 20.4) 21.3 (21.7 to 20.9) 20.8 (21.4 to 20.1)
AKI, acute kidney injury; BMR, basal metabolic rate, estimated by Schofield; IQR, interquartile range.
Values are median (IQR).
Energy or protein gap 5 actual minus recommended intake.
*Wilcoxon Signed Rank test between actual and recommended energy and protein intake; MannWhitney test.
Postimplementation versus preimplementation, P , .05.
AKI resolved versus no AKI.

Support Guidelines. In addition, the cumulative energy and
protein gap for the 8-day period improved in the
postimplementation compared to preimplementation,
with protein and energy provision remaining below
recommended intakes. The implementation of Nutrition
Support Guidelines did decrease the gap between energy
and protein delivery and needs.
Previous studies have shown that critically ill children
who received better nutritional support had significant
improvement in physiological stability and outcome.37
Furthermore, recent studies have shown that higher protein
intake might protect the kidney from injury and may pre-
serve glomerular filtration rate in critically ill patients23
and could decrease the time to recovery from AKI.24-26
Studies linking achievement of protein requirements to
outcome are lacking in children with AKI.
The goals of nutrition support in AKI are identical to
those suggested for other critically ill patients in the
PICU17,38 and include prevention of protein energy
wasting, preservation of lean body mass and nutritional
status, avoidance of metabolic derangements and
complications, improvement of wound healing, and
support of immune function.11,38 Protein restriction with
the goal of delaying the need for renal replacement
therapy should be avoided.39 Since protein energy malnu-
trition is associated with increased mortality and morbidity
in children37 and adults,40 with a higher risk for infection,
poor wound healing, longer dependency on mechanical
ventilation, and increased length of hospital stay in
adults,41,42 adequate nutrition support, especially protein
provision, is essential for improving patient outcomes.
Furthermore, adequate protein feeding seems to be
beneficial in AKI as systemic administration of protein in
the form of amino acids early in the course of
experimental AKI in small animal models of nephrotoxic
and septic AKI had anti-inflammatory and antiapoptotic ef-
fects in the renal tubular epithelium.43,44 Renal recovery
was faster and survival higher in adults with oliguric AKI
who were given IV amino acid infusions than in patients
who were given isocaloric carbohydrates alone.24
Several studies have shown beneficial effects of higher
protein provision.23-26 The current study was not able to
demonstrate a beneficial effect of protein feeding on renal
function. Although the amount of protein provided was
suboptimal and less than recommended in critically ill
children, the higher protein intake was not associated
with a delay in renal recovery in patients with AKI.
Limitations of this study were several; data were collected
retrospectively from patient charts and were limited single
center and, therefore, may not be generalized to other
PICUs. Fluid overload and oliguria are both well-
recognized risk factors in the PICU.45-47 There was no
 PROTEIN FEEDING IN CHILDREN WITH AKI
data available on the fluid balance and urine volume of
patients. PRISM and PELOD scores were obtained on
PICU admission day only, which might not reflect the
progression of organ dysfunction during the entire study
period. Another limitation of this study was that energy
expenditure was calculated from prediction equations
instead of using indirect calorimetry.
In conclusion, higher protein intake ($80% of estimated
needs) was not associated with a delay in renal recovery in
patients with AKI after adjustment for severity of illness.
Protein intake was improved in critically ill children with
no AKI, resolved, and persistent AKI after implementation
of Nutrition Support Guidelines, but underfeeding per-
sisted in these patients.
Practical Application
Adequate nutrition support has been shown to improve
outcome in critically ill children. Higher protein intake was
not associated with a delay in renal recovery in patients with
AKI after adjustment for severity of illness. The implemen-
tation of Nutrition Support Guidelines improved feeding
practices in the PICU, but underfeeding persisted in criti-
cally ill children with AKI.
Acknowledgments
Support: The study was carried out at Texas Childrens Hospital, with
internal funding by Baylor College of Medicine.
U.G.K. designed and carried out the study, carried out the data ana-
lyses, and drafted the manuscript. A.A.-A. conceived and J.C.S., M.G.,
and L.S.S. assisted with the data analysis and the draft of the manuscript.
J.A.C.-B. participated in the design of the study, contributed to the data
analysis, and helped to draft the manuscript. All authors read and
approved the final manuscript.
References
1. Coss-Bu JA, Klish WJ, Walding D, et al. Energy metabolism, nitrogen
balance, and substrate utilization in critically ill children. Am J Clin Nutr.
2001;74:664-669.
2. Delgado AF, Okay TS, Leone C, et al. Hospital malnutrition and inflam-
matory response in critically ill children and adolescents admitted to a tertiary
intensive care unit. Clinics (Sao Paulo). 2008;63:357-362.
3. Kyle UG, Genton L, Pichard C. Low phase angle determined by
bioelectrical impedance analysis is associated with malnutrition and nutri-
tional risk at hospital admission. Clin Nutr. 2013;32:294-299.
4. Cahill NE, Dhaliwal R, Day AG, Jiang X, Heyland DK. Nutrition ther-
apy in the critical care setting: what is best achievable practice? An interna-
tional multicenter observational study. Crit Care Med. 2010;38:395-401.
5. Bartlett RH, Dechert RE, Mault JR, et al. Measurement of metabolism
in multiple organ failure. Surgery. 1982;92:771-779.
6. Kyle UG, Jaimon N, Coss-Bu JA. Nutrition support in critically ill chil-
dren: underdelivery of energy and protein compared with current recommen-
dations. J Acad Nutr Diet. 2012;112:1987-1992.
7. Akcan-Arikan A, Zappitelli M, Loftis LL, et al. Modified RIFLE criteria
in critically ill children with acute kidney injury. Kidney Int. 2007;71:1028-
1035.
8. Alkandari O, Eddington KA, Hyder A, et al. Acute kidney injury is
an independent risk factor for pediatric intensive care unit mortality,
longer length of stay and prolonged mechanical ventilation in critically
7
ill children: a two-center retrospective cohort study. Crit Care.
2011;15:R146.
9. Wooley JA, Btaiche IF, Good KL. Metabolic and nutritional aspects of
acute renal failure in critically ill patients requiring continuous renal replace-
ment therapy. Nutr Clin Pract. 2005;20:176-191.
10. Fiaccadori E, Lombardi M, Leonardi S, et al. Prevalence and clinical
outcome associated with preexisting malnutrition in acute renal failure: a pro-
spective cohort study. J Am Soc Nephrol. 1999;10:581-593.
11. Fiaccadori E, Cremaschi E, Regolisti G. Nutritional assessment
and delivery in renal replacement therapy patients. Semin Dial.
2011;24:169-175.
12. Maursetter L, Kight CE, Mennig J, Hofmann RM. Review of the
mechanism and nutrition recommendations for patients undergoing contin-
uous renal replacement therapy. Nutr Clin Pract. 2011;26:382-390.
13. National Kidney Foundation. KDOQI Clinical Practice Guidelines
and Clinical Practice Recommendations for Anemia in Chronic Kidney Dis-
ease. Am J Kidney Dis. 2006;47(Suppl 3):S11-S146.
14. Berbel MN, Pinto MP, Ponce D, Balbi AL. Nutritional aspects in acute
kidney injury. Rev Assoc Med Bras. 2011;57:600-606.
15. Lambe C, Hubert P, Jouvet P, Cosnes J, Colomb V. A nutritional sup-
port team in the pediatric intensive care unit: changes and factors impeding
appropriate nutrition. Clin Nutr. 2007;26:355-363.
16. Kalista-Richards M. The kidney: medical nutrition therapyyesterday
and today. Nutr Clin Pract. 2011;26:143-150.
17. McClave SA, Martindale RG, Vanek VW, et al. Guidelines for the Pro-
vision and Assessment of Nutrition Support Therapy in the Adult Critically Ill
Patient: Society of Critical Care Medicine (SCCM) and American Society for
Parenteral and Enteral Nutrition (A.S.P.E.N.). JPEN J Parenter Enteral Nutr.
2009;33:277-316.
18. Kyle UG, Akcan-Arikan A, Orellana RA, Coss-Bu JA. The adequacy
of nutritional support in critically ill children with acute kidney injury. In:
Rajendram R, Preedy VR, Patel VB, eds. Diet and nutrition in critical care. Lon-
don, UK: Springer; 2015:885-896.
19. Briassoulis G, Zavras N, Hatzis T. Malnutrition, nutritional indices,
and early enteral feeding in critically ill children. Nutrition. 2001;17:
548-557.
20. Mehta NM, McAleer D, Hamilton S, et al. Challenges to optimal
enteral nutrition in a multidisciplinary pediatric intensive care unit. JPEN J
Parenter Enteral Nutr. 2010;34:38-45.
21. Taylor RM, Preedy VR, Baker AJ, Grimble G. Nutritional support in
critically ill children. Clin Nutr. 2003;22:365-369.
22. Kyle UG, Akcan-Arikan A, Orellana RA, Coss-Bu JA. Nutrition sup-
port among critically ill children with AKI. Clin J Am Soc Nephrol.
2013;8:568-574.
23. Doig GS, Simpson F, Bellomo R, et al. Intravenous amino acid therapy
for kidney function in critically ill patients: a randomized controlled trial.
Intensive Care Med. 2015;41:1197-1208.
24. Abel RM, Beck CH Jr, Abbott WM, et al. Improved survival from
acute renal failure after treatment with intravenous essential L-amino acids
and glucose. Results of a prospective, double-blind study. N Engl J Med.
1973;288:695-699.
25. Singer P. High-dose amino acid infusion preserves diuresis and im-
proves nitrogen balance in non-oliguric acute renal failure. Wien Klin Wo-
chenschr. 2007;119:218-222.
26. Doig GS, Simpson F, Finfer S, et al. Effect of evidence-based feeding
guidelines on mortality of critically ill adults: a cluster randomized controlled
trial. JAMA. 2008;300:2731-2741.
27. Chumlea WC, Guo SS, Steinbaugh ML. Prediction of stature from
knee height for black and white adults and children with application to
mobility-impaired or handicapped persons. J Am Diet Assoc. 1994;94:1385-
1388. 1391; quiz 1389-1390.
28. World Health Organization Multicentre Growth Reference Study
Group. WHO Child Growth Standards: length/height-for age, weight-for-age,
weight-for-length, weight-for-height and body mass index-for-age: methods and devel-
opment. Geneva, Switzerland: World Health Organization; 2006.
8 KYLE ET AL
29. Kuczmarski RJ, Ogden CL, Guo SS, et al. 2000 CDC Growth Charts
for the United States: Methods and Development. Vital Health Stat 11.
2002;246:1-190.
30. De Le on AL, Romero-Gutierrez G, Valenzuela CA, Gonzalez-
Bravo FE. Simplified PRISM III score and outcome in the pediatric intensive
care unit. Pediatr Int. 2005;47:80-83.
31. Pollack MM, Patel KM, Ruttimann UE. PRISM: an updated pediatric
risk of mortality score. Crit Care Med. 1996;24:743-752.
32. Leteurtre S, Duhamel A, Grandbastien B, Lacroix J, L F. Paediatric logistic
organ dysfunction (PELOD) score. Lancet. 2006;367:897. author reply 900-902.
33. Leteurtre S, Martinot A, Duhamel A, et al. Validation of the paediatric
logistic organ dysfunction (PELOD) score: prospective, observational, multi-
centre study. Lancet. 2006;362:192-197.
34. Schofield WN. Predicting basal metabolic rate, new standards and re-
view of previous work. Hum Nutr Clin Nutr. 1985;39 Suppl 1:5-41.
35. Mehta NM, Compher C. A.S.P.E.N. Clinical Guidelines: nutrition
support of the critically ill child. JPEN J Parenter Enteral Nutr.
2009;33:260-276.
36. Mehta NM, Bechard LJ, Zurakowski D, Duggan CP, Heyland DK.
Adequate enteral protein intake is inversely associated with 60-d mortality
in critically ill children: a multicenter, prospective, cohort study. Am J Clin
Nutr. 2015;102:199-206.
37. Pollack MM, Ruttimann UE, Wiley JS. Nutritional depletions in crit-
ically ill children: associations with physiologic instability and increased quan-
tity of care. JPEN J Parenter Enteral Nutr. 1985;9:309-313.
38. Cano NJ, Aparicio M, Brunori G, et al. ESPEN Guidelines on Paren-
teral Nutrition: adult renal failure. Clin Nutr. 2009;28:401-414.
39. Sabatino A, Regolisti G, Maggiore U, Fiaccadori E. Protein/energy
debt in critically ill children in the pediatric intensive care unit: acute kidney
injury as a major risk factor. J Ren Nutr. 2014;24:209-218.
40. Biolo G. Can we increase protein synthesis by anabolic factors? Am J
Kidney Dis. 2001;37(1 Suppl 2):S115-S118.
41. Pichard C, Kyle UG, Morabia A, et al. Nutritional assessment: lean
body mass depletion at hospital admission is associated with increased length
of stay. Am J Clin Nutr. 2004;79:613-618.
42. Heyland DK. Making inferences from scientific research. Nutr Clin
Pract. 1998;13:S1-S7.
43. Hu YM, Pai MH, Yeh CL, Hou YC, Yeh SL. Glutamine administra-
tion ameliorates sepsis-induced kidney injury by downregulating the high-
mobility group box protein-1-mediated pathway in mice. Am J Physiol Ren
Physiol. 2011;302:F150-F158.
44. Kim YS, Jung MH, Choi MY, et al. Glutamine attenuates tubular cell
apoptosis in acute kidney injury via inhibition of the c-Jun N-terminal kinase
phosphorylation of 14-3-3. Crit Care Med. 2009;37:2033-2044.
45. Macedo E, Malhotra R, Bouchard J, Wynn SK, Mehta RL. Oliguria is
an early predictor of higher mortality in critically ill patients. Kidney Int.
2011;80:760-767.
46. Arikan AA, Zappitelli M, Goldstein SL, et al. Fluid overload is associ-
ated with impaired oxygenation and morbidity in critically ill children. Pediatr
Crit Care Med. 2011;13:253-258.
47. Sutherland SM, Zappitelli M, Alexander SR, et al. Fluid overload and
mortality in children receiving continuous renal replacement therapy: the
prospective pediatric continuous renal replacement therapy registry. Am J Kid-
ney Dis. 2010;55:316-325.