CLINICAL

STUDY

, Turkey
R
E
ABSTRACT
N
A
L Acute renal failure (ARF) is a common problem in the neonatal intensive care unit
F (NICU). In most cases, ARF is associated with a primary condition such as sepsis,
metabolic diseases, perinatal asphyxia and/or prematurity. This retrospective study
A
investigated the course of illness, therapeutic interventions, early prognosis and risk
I factors associated with development of ARF in the neonatal period. A total of 1311
L neonates were treated in our NICU during the 42-month study period, and 45 of these
U babies had ARF. This condition was defined as serum creatinine level above 1.5 mg/dL
R despite normal maternal renal function. The data collected for each ARF case were
E contributing condition, cause and clinical course of ARF, gestational age and birth
Vol. 26, weight, age at the time of diagnosis, treatment, presence of perinatal risk factors and
need for mechanical ventilation. The frequency of ARF in the NICU during the study
No. 3, pp.
period was 3.4%. Premature newborns constituted 31.1% of the cases. The mean birth
305309,
weight in the group was 2863 1082 g, and the mean age at diagnosis was
2004
6.2 7.4 days. The causes of ARF were categorized as prerenal in 29 patients (64.4%),
renal in 14 patients (31.1%) and postrenal in 2 patients (4.4%). Forty-seven percent of
the cases were nonoliguric ARF. Asphyxia was the most common condition that
contributed to ARF (40.0%), followed by sepsis/metabolic disease (22.2%) and
feeding problems (17.8%). Therapeutic interventions were supportive in 77.8% of the
cases, and dialysis was required in the other 22.2%. The mortality rate in the 45 ARF
cases was 24.4%. Acute renal failure of renal origin, need for dialysis, and need for
mechanical ventilation were associated with significantly increased mortality
(p < 0.05). There were no statistical correlations between mortality rate and perinatal
Renal
risk factors, oliguria, prematurity or blood urea nitrogen and creatinine levels. The

Acute
Failure in the
Neonatal Period

*Correspondence: Pinar Isik Agras, M.D., 6. Cadde No: 72/3, Bahcelievler, Ankara, 06490 Turkey; Fax: 090-

Pinar Isik Agras, M.D., *

Aylin Tarcan, M.D., Esra
Baskin, M.D.,
Nurcan Cengiz,
M.D., Berkan
Gurakan, M.D., and
U
mi
t
Sa
atc
i,
M.
D.
1,

312215-75-97;
E-mail: pinaris2001@yahoo.com.

305

1Department

of Pediatric
Nephrology and Department
of Neonatalogy,
Baskent
University
, Ankara,
Turkey
3Department of
Pediatric Nephrology,
Baskent University,
A
d
a
n
a
2

0886-022X (Print); 1525-6049 (Online)

DOI: 10.1081/JDI-200026749
Copyright D 2004 by Marcel Dekker, Inc.

www.dekker.com

Agras et al.
306
study showed that, at our institution, ARF in the neonatal period is frequently
associated with preventable conditions, specifically asphyxia, sepsis and feeding
problems. Supportive therapy is effective in most cases of neonatal ARF. Acute renal
failure of renal origin, need for dialysis, and need for mechanical ventilation were
identified as indicators of poor prognosis in these infants. Early recognition of risk
factors and rapid effective treatment of contributing conditions will reduce mortality
in neonatal ARF.
Key Words: Acute renal failure (ARF); Neonatal intensive care; Sepsis; Perinatal
asphyxia; Neonate; Breast-feeding; Neonatal ARF.

to
primary
renal
INTROD
disease,
UCTION
most of
these
De
cases are
spi
related to
te
underlyin
sig
g
condition
nifi
s such as
can
sepsis or
t
metabolic
ad
diseases,
va
which
nce
cause
s
hypovole
in
mia,
our
hypoxemi
a and
un
hypotensi
der
on.[1 3]
sta
A
ndi
newborn
ng
is
of the
diagnosed
pathoph
with ARF
ysiology
when
of acute
urinary
renal
output
failure
is less
(ARF)
than 0.5
and its
1
treatmen
mL/kg/hr
t, this
for a 24conditio
hr period,
n is still
and/or
common
when the
in
infants
newborn
serum
s.
creatinine
Publishe
is above
d data
1. 5
indicate
mg/dL
that 8
despite
24% of
normal
neonates
kidney
admitted
function
to the
in the
neonatal
mother.
intensiv
[2,3] Still,
e care
it can be
unit
difficult
(NICU)
to
develop
diagnose
ARF.
this
As
condition
opposed
in
[1]

group results were compared

neonate
using the chi-square test. Correlations
s
Table 1.
Results for the 45 newborns with
presence/abse were assessed
for
acute renal
using Spearmans correlation coefficient. failure (ARF ).
several nce of
reasons perinatal risk P
Number
values
: 1) the factors
matern (preeclampsi < 0.05 were taken to indicate statistical
a,
significance.
al
(%)
serum oligohydram
Female/Male
23/22
31
the course of illness, treatments
creatini nios,
Prematurity
14
maternal
used, short-term
ne
40.0
prognosis and risk factors related to the Contributing conditions
level infection,
Asphyxia
18
22.2
development
affects gestational
Metabolic
10
17.8
diabeof ARF in neonates at our
the
disease/Sepsis
8
11.1
institution.
babys tes), blood
Feeding problems
5
8.9
creatini urea nitrogen
Congenital heart
4
(BUN) and
ne
disease
64.4
MATERIALS AND
29
Congenital renal
31.1
level; serum
disease
14
4.5
2) the creatinine
METHODS
Cause of ARF*
2
time levels at the
Prerenal
57.8
time of
that
A total of 1311 infants were
Renal
26
42.2
urine diagnosis,
treated in our hospitals
Postrenal
19
output treatment
NICU in the 42-month period from
Clinical course
77.8
starts administered,
January 1999
35
Oliguric
22.2
after early
through July 2002. Forty-five of these
Nonoliguric
10
24.4
birth is prognosis
newborns were
Treatment
11
variabl and early
diagnosed with ARF, and we
Supportive
Dialysis
e; and outcome.
retrospectively investigatWe
3)
ed these cases. Acute renal failure was Mortality
grouped
nonoli
defined as serum
*ARF: acute renal
the
guric
creatinine level above 1.5 mg/dL
failure.
patients
ARF
despite normal
according
is
maternal renal function. Babies who
to various
comm
had developed

on in parameters
this which are as
age follows:
group. contributing
Thcone ditions,
ai clinical
m course
of (oliguric vs.
thi nonoliguric)
s of
ret ARF, cause
ro of ARF
sp (prerenal/ren
ec al/postrenal),
tiv presence
e of perinatal
st risk factors
ud and need for
y mechanical
w ventilation,
as treatment
to modality and
re early
ve outcome.
al

For
statistical
analysis,

ARF after cardiac surgery were

excluded from the
study. The data collected for each case
were sex,
gestation period, birth weight, postnatal
age at diagnosis
of ARF, contributing diseases, cause of
ARF (prerenal,
renal or postrenal), oliguric/nonoliguric
course of ARF,

307
Acute Renal Failure in the Neonatal Period

RESULTS
The frequency of ARF in the NICU during the study
period was 3.4%. The data collected from these cases are

Table 3.

Results for the 10 patients who required dialysis.

Number

Contributing conditions

summarized in Table 1.
The patients were 23
(51.1%)

Asphyxia
Metabolic
disease/Sepsis
Feeding
female and 22 (48.9%)
problems
male newborns.
Congenital heart
Gestational age
disease
was less than 37 weeks in
Congenital renal
31.1% of the cases. The
disease
mean
Cause of ARF*
birth weight was 2863
Prerenal
1082 g, and the mean
Renal
postnatal
Postrenal
age at diagnosis was 6.2 Clinical course
Oliguric
7.4 days. The mean BUN
Nonoliguric
and
serum creatinine levels at Perinatal risk
factors
diagnosis were 60 50
Mechanical
mg/dL
ventilation
and 2.6 2.3 mg/dL,
Premature
respectively. Asphyxia was
Mortality
the most
*ARF: acute renal
frequent contributing
condition (n = 18, 40.0%) failure.
followed
by sepsis/metabolic disease
(n = 10, 22.2%) and
feeding
problems (n = 8, 17.8%).
The cause of ARF was
prerenal
in 29 (64.4%) cases, renal
in 14 (31.1%) cases, and
postrenal in 2 (4.5%) cases.
The course of the illness
was
nonoliguric in 26 (57.8%)
patients and oliguric in 19
(42.2%) patients. Perinatal
risk factors were present in
10
newborns and they were as
follows: preeclampsia in 2,
oligohydramnios in 4,
urinary tract infection in
one,

Table 2.

3
5
1
1

30
50
10
10

1
9

10
90

10

1
3
1
5

100

10
30
10
50

de
tot

Results for links

between parameters

Contributing conditions

investigated

and mortality.
No.

intrauterine growth retardation in

supportive treatment
alone, and 10 (22.2%) required
peritoneal dialysis.

one, uterine anomaly in

Eleven patients died, thus overall
one and gestational diabetes in one of
mortality was
them. Thirty-five
24.4%. The mortality rate in the
(77.8%) of the infants responded to nonoliguric cases
Asphyxia 4/18
22.2 (n = 2, 10.5%) was significantly
Metabolic 1/10
10.0 lower than that in the
disease/Seps 1/8
12.5 oliguric cases (n = 9, 34.6%; p <
is
3/5
60.0 0.05). Mortality rate
Feeding
2/4
50.0
was higher in cases that needed
problems
mechanical ventilation
Congenital 3/29
10.3
heart disease 8/14
57.1 and dialysis ( p < 0.05). Renal
cause of ARF was
Congenital 0/2
associated with significant increase
renal disease
9/26
34.6 in mortality rate
Cause of
ARF*
2/19
10.5 than prerenal and postrenal causes
Prerenal
( p < 0.05). Mortality
Renal**
6/35
17.1 rate was not increased in presence
Postrenal 5/10
50.0 of perinatal risk
Clinical course
factors, oliguric course of disease,
Oliguric** 8/21
38.0
prematurity and it
Nonoliguric 2/24
8.3
was not correlated with BUN and
Treatment
Supportive 2/10
20.0 serum creatinine
Dialysis** 9/35
25.7 levels at the time of diagnosis
Mechanical 5/14
35.7 ( p>0.05; Table 2).
ventilation
6/31
Present**
Absent
Perinatal risk
factors
Present
Absent
Premature
Term
*ARF: acute
renal failure.
**p < 0.05.

19.3

The results for the 10 patients who

required

peritoneal dialysis are summarized

in Table 3. The
most frequent contributing
conditions in this group
were sepsis/metabolic disease, and
the course was
oliguric in all cases. The cause of
ARF was prerenal in
1 case and renal in the other 9
newborns. The mortality
rate in the dialysis group was 50%.

DISCUSSION
Acute renal failure in childhood
usually develops

due to causes in the renal

parenchyma, including acute

Agras et al.
308

tubular necrosis, hemolytic uremic syndrome, glomerulonephritis and urinary tract obstruction.[4,5]In
contrast, most cases of ARF in the NICU tend to be
associated with conditions other than primary renal
disease.[1,2,6]
Pulmonary conditions like acute respiratory distress syndrome (RDS) and use of positivepressure ventilation are common in NICU cases. Both
these factors are associated with reduced cardiac
output, decreased renal blood flow, and decreased
glomerular filtration rate. In neonates who have RDS
and ARF, the severity of renal dysfunction is usually
correlated with the severity of RDS. Perinatal hypoxia,
sepsis, metabolic illness and congenital (renal/cardiac)
disease are other conditions in NICU patients that
frequently contribute to ARF.[1,2,7]
In addition to these
problems, invasive therapeutic procedures, such as
umbilical vein/artery catheterization, often play a role
in development of ARF because associated microemboli and renal vein thrombosis can cause renal
ischemia. Drug toxicity, which occurs relatively
frequent in the neonatal period, can lead to both
[2]

prerenal ARF and direct renal damage.

[7]

In our patients with neonatal ARF, the most

frequent contributing condition was asphyxia, followed
by sepsis/metabolic disease. Although feeding problems
are not frequently reported in relation to ARF in
previous reports, we identified this as the third most
common contributing condition in our NICU ARF cases.
Although breast-feeding has many well-known benefits
for mother and infant, breast-feeding malnutrition is a
potential problem in the first weeks. This problem can
lead to severe complications, including dehydration and
ARF. Factors associated with feeding problems
include shorter postpartum hospital stay, inconsistent
follow-up breast-feeding support, lower socioeconomic
status, and general lack of health care support. It is
therefore not surprising that feeding problems and
breast-feeding malnutrition commonly play roles in
ARF in developing countries. In contrast to the above
conditions, we observed that congenital renal or cardiac
problems contributed to ARF in only 20% of our cases.
The cause of ARF was prerenal in origin in 64.4% of the
neonates we studied, and the condition was corrected by
supportive treatment in 77.8% of cases. All these
findings indicate that most cases of neonatal ARF at
our institution are linked to preventable conditions.
Improvement of perinatal conditions, better infection
control, and enhanced maternal education and support in
relation to breast-feeding will be key to preventing
[8]

[9]

this illness.
As noted above, the reported rates of ARF in the
NICU range from 8 24%.[1 3]
According to our data,
approximately 3.4% of the newborns in our unit are

diagnosed with ARF. The variation in the reported

frequencies may be attributed to discrepancies in
defining and recognizing the clinical findings of ARF
in newborns, some of which are as follows: a significant
proportion of neonatal ARF patients do not become
oliguric. In our NICU, 57.8% of the ARF cases were in
this category. Three percent of premature newborns
may show a delay in the time of first urine output, with
no urinary output in the first 24 hours. Also, the
maternal serum creatinine level influences that of the
newborn during the first 24 hours. In premature newborns, the physiological decline in serum creatinine
occurs gradually over the first 2 3 weeks after birth.
[3]

[1]

Interestingly, although approximately 31% of the

babies we studied were premature, RDS was not a

common contributor to ARF in these cases. This may
also be explained by discrepancies in diagnosing
neonatal ARF, as mentioned above, which may be
causing us to under diagnose some of the neonatal ARF
cases. Determination of newborns in risk of developing ARF may provide close follow-up urine output and
serum creatinine levels of these cases. This would enable
physicians to differentiate the pathological findings
from the physiologic features of first days of life and
to avoid under diagnosing ARF in the neonatal period.
Ten of our 45 infants with ARF required dialysis

treatment in the NICU. Gong et al. investigated 66

patients who developed ARF during childhood and
needed dialysis.[10] The diagnoses in the majority of
these cases were acute tubular necrosis, hepatorenal
syndrome and acute glomerulonephritis. In contrast, in
our neonatal ARF cases, the most common contributing
conditions in the patients who required dialysis were
sepsis/metabolic disease. Nine of the 10 babies that
needed dialysis had ARF of renal origin, and all 10
cases followed an oliguric course.
The overall mortality rate in our 45 NICU ARF

cases was 24.4%. The rate in the nonoliguric cases was

significantly lower than that in the oliguric group. This
finding is in keeping with previous reports of excellent
survival rates in nonoliguric patients versus mortality
rates of 25% 78% in oliguric ones.[1,11] The mortality
rate in our neonates with ARF who required dialysis
(50%) was significantly higher than the overall mortality rate. However, this 50% rate was still lower than
the 68.8% mortality rate reported in the abovementioned study of ARF patients who underwent
dialysis during childhood.[10]
We also found that the need for mechanical

ventilation was associated with increased mortality in

our neonates with ARF. Close monitoring of renal
function and urinary output in newborns on ventilator
support would help achieve early diagnosis.

Copyright of Renal Failure is the property of Taylor & Francis Ltd and its content may not be copied or emailed
to multiple sites or posted to a listserv without the copyright holder's express written permission. However,
users may print, download, or email articles for individual use.