Pediatrics International (2004) 46, 2125

Original Article

Neonatal urinary tract infections: Analysis of the patients and

recurrences
NESE KARAASLAN BIYIKLI,1 HARIKA ALPAY,1 EREN OZEK,2 IPEK AKMAN2 AND
HULYA BILGEN2
Departments of 1Pediatric Nephrology and 2Neonatology, Marmara University School of Medicine,
Istanbul, Turkey
Abstract

Background: Early diagnosis and proper treatment, including long-term follow up, are very important for
neonatal urinary tract infections (UTI).
Methods: The present study reports the analysis and long-term follow-up results of 71 newborns treated for UTI.
Results: Forty-one per cent of patients were preterm babies. Suspected sepsis and hyperbilirubinemia were the
main presenting features. Community-acquired and nasocomial UTI accounted for 63% and 37% of cases,
respectively. The leading causative agents were Escherichia coli for community-acquired UTI and Klebsiella
pneumoniae for nasocomial UTI. The urosepsis rate was 5%. Abnormal ultrasonography findings were present
in 23% and vesicoureteral reflux was present in 15% of babies. A total of 23% of patients showed renal
photopenic areas on dimercaptosuccinic acid scan. The recurrence rate was 28% occurring between 1.5 and
12 months, in particular in the first 6 months. Most of the recurrences developed in patients with no
predisposing abnormalities.
Conclusion: Pediatric nephrologic follow-up of babies experiencing UTI in the neonatal period is very
important to identify the predisposing congenital abnormalities and scarred kidneys, to diagnose and to treat
the recurrences earlier.

Key words

causative agents, newborn, recurrences, urinary tract infections.

The diagnosis of urinary tract infection (UTI) is difficult in

the newborn period because signs and symptoms of UTI are
non-specific in this age group. UTI in infancy can lead to
renal scars and if it is underdiagnosed it can cause permanent
renal damage causing hypertension or end-stage renal disease.
The incidence of UTI in newborn babies is 0.11% and it can
be as high as 10% in low-birthweight and preterm babies.1
In the present study the clinical presentation, causative
agents, imaging findings and recurrence rates of UTI in
newborns were analyzed.

Methods
A retrospective analysis was undertaken of 71 patients
treated for neonatal UTI between 1999 and 2002 at Marmara
University Hospital, Istanbul, Turkey. All patients were
Correspondence: Nese Karaaslan Biyikli MD, Kozyatagi, Sinanercan
cd, Oztor sitesi, C Blok, 38/38, 34736 Istanbul, Turkey.
Email: nesebiyikli@superonline.com, nesekaraaslan@yahoo.com
Received 1 April 2003; revised 1 July 2003; accepted 7 July 2003.

followed by the Pediatric Nephrology Unit for at least

6 months. Clinical findings, causative bacteria, predisposing
urinary tract abnormalities, renal scan data and recurrence
rates were analyzed. Patients with spina bifida were not
included in the study. The neonatal period is defined as the
first 4 weeks of life after birth. The diagnostic criteria for
UTI was the growth of the microorganism over 10 000 c.f.u./mL
in the catheterized urine specimen. Nasocomial UTI was
defined as a positive urine culture detected 48 h after
admission.2 UTI in babies without a history of hospitalization
were accepted as community-acquired UTI. All babies
underwent a full sepsis work up including blood culture.
Patients were initially treated with broad spectrum antibiotics
(ampicilline and aminoglycosides). After exclusion of urosepsis,
the patients were managed with appropriate antibiotics
according to the urine culture results. Following UTI treatment,
all patients were put on antibiotic prophylaxis with 10 mg/kg
per day amoxicilline during imaging studies, which were
planned for the detection of predisposing urinary tract abnormalities. Prophylaxis was continued if reflux or any other
predisposing factor was present and discontinued in patients
without predisposing abnormality or after at least 36 months

22

N Karaaslan Bykl

Table 1

Characteristics of the patients

Sex distribution (M/F)

Gestational age (term/preterm)
SGA babies (term/preterm)
Mean birth weight (g)

Findings

54/17
42/29
1/3
2824 868
(range 7444400)

76/24
59/41

33
29
9
45

47
40
13
63

26

37

Clinical signs
Suspected sepsis
Jaundice
Asymptomatic
Community-acquired urinary
tract infection
Nasocomial urinary tract infection

Table 3

Table 2 The clinical presentation, occurrence of pyuria and the

frequency of Escherichia coli and Klebsiella pneumonia for
preterm and term babies and their statistical analysis results

No. patients
Hyperbilirubinemia
Suspected sepsis
Pyuria
Escherichia coli
Klebsiella pneumonia

Preterm

Term

29
8 (26%)
15 (53%)
12 (41%)
14 (48%)
12 (41%)

42
24 (57%)
15 (36%)
29 (69%)
16 (38%)
15 (36%)

0.014
0.17
0.02
0.88
0.62

The distribution of causative agents in community acquired and nasocomial UTI for preterm and term babies

Nasocomial urinary tract infection

Escherichia coli
Klebsiella pneumonia
Pseudomonas aeruginosa
Proteus vulgaris
Klebsiella oxytoca
Community-acquired urinary tract infection
Escherichia coli
Klebsiella pneumonia
Enterobacter cloacae
Candida albicans
Enterococcus fecalis
Staphylococcus aureus
Group B Streptococcus
Serratia marcessens
Proteus vulgaris

No. preterm babies

No. term babies

Total

20
9 (45%)
9 (45%)
1 (5%)
1 (5%)
0
9
5 (56%)
3 (33%)
1 (11%)
0
0
0
0
0
0

6
1 (17%)
4 (66%)
0
0
1 (17%)
36
15 (42%)
11 (30%)
0
1 (3%)
4 (11%)
2 (5%)
1 (3%)
1 (3%)
1 (3%)

26
10 (38%)
13 (50%)
1 (4%)
1 (4%)
1 (4%)
45
20 (44%)
14 (31%)
1 (3%)
1 (3%)
4 (9%)
2 (4%)
1 (3%)
1 (3%)
1 (3%)

of infection-free period. Renal ultrasonography (USG) and

voiding cystouretrography (VCUG) were planned for all
babies and dimercaptosuccinic acid (DMSA) scan was
performed to most of the babies in the study group.
According to our follow-up protocol monthly urine cultures
were obtained for 3 months and then once every 3 months for
the next 2 years. Additional urine cultures were taken for
UTI symptomatology such as fever, feeding problems or
vomiting, abdominal distention, malodorous urine, abnormal
crying. Parents were informed about the symptomatology of
UTI, the purpose of routine urine cultures and the importance
of follow up.
The data were compared by 2 test.

Results
The mean age of the babies at the time of diagnosis was
18.8 11.2 days (range 230 days), and the duration of
nephrologic follow-up was 12 6 months varying between

6 months and 36 months. There were 54 boys (76%) and 17

girls (24%) in the study group. None of the boys had been
circumcised. Twenty-nine patients (41%) in the study group
were preterm babies with 2737 weeks gestational age
(34 3.4 weeks). One term and three preterm babies were
small for gestation age. Mean birthweight of the study group
was 2824 868 g (range 7444400 g). The characteristics of
the study group are shown in Table 1. Clinical signs at
presentation for preterm babies were signs of suspected
sepsis in 15 infants (53%), and hyperbilirubinemia in eight
infants (26%) and clinical signs at presentation for term
babies were hyperbilirubinemia in 24 infants (57%) and signs
of suspected sepsis 15 infants (36%) (Table 2). The signs of
suspected sepsis were irritability in 11 infants (15%), fever or
hypothermia in six (8%), respiratory distress in six (8%),
feeding problems in four (5%), vomiting in three (4%),
abnormal crying in three (4%), poor weight gain in two (2%),
and rash in one infant (1%). Hyperbilirubinemia as a presenting
feature was significantly higher in term babies. Nine infants
were asymptomatic (13%). As regards asymptomatic cases,

Neonatal UTI
five were preterm hospitalized babies in whom routine urine
culture was a screening test for nasocomial infections and
four of them were term babies with a history of prenatal
dilatation of the renal pelvis. Community-acquired and
nasocomial UTI accounted for 45 (63%) and 26 (37%) cases,
respectively. The most frequent causative agents differed as
Escherichia coli (44%) for community-acquired and Klebsiella
pneumonia (50%) for nasocomial UTI. The distribution of
causative agents in community-acquired and nasocomial
UTI for preterm and term babies are shown in Table 3. The
statistical analyses were not significant in respect to causative
agents for preterm and term babies. The statistical difference
was significant between term and preterm babies for
nasocomial and community-acquired infections with a predilection of nasocomial UTI for preterm babies and a predilection of community-acquired UTI for term babies. Four
of our patients, of which three were preterm, were diagnosed
clinically as urosepsis with the support of sepsis work up.
These babies were hypotonic with depressed newborn
reflexes and hypotensive. Full blood counts showed deep
neutropenia in favor of Gram negative septisemia but the
blood cultures of these patients were negative for the agents
causing UTI. We treated these patients with broad spectrum
antibiotics (ampicilline and aminoglycosides) for 21 days.
Pyuria was detected in 12 preterm (41%) and 29 term
(69%) patients. The statistical analyses were significant for
pyuria in favor of term infants. Ultrasound and VCUG was
planned routinely for all newborn babies with UTI, but five
patients for USG and 25 patients for VCUG were lost to
follow up. Of the 66 patients who completed their USG scan
15 (23%) had an abnormality such as renal pelvis dilatation,
hydronephrosis, or hyperecogenic kidney. VCUG was
performed in 46 babies and DMSA scan in 36 babies. Seven
patients had vesicoureteral reflux (VUR) with an incidence of
15%. Five patients with VUR were term and two of them
were preterm. Five term and three preterm patients (23%)
showed renal photopenic areas on DMSA. The presence of
VUR and the occurrence of renal scars were not statistically
significant between the preterm and term babies.
99mTc mercaptoacetylglycine renography or diethylenetriaminepentaacetic acid renography were performed if required.
Two patients had ureteropelvic junction (UPJ) stenosis and
two patients had posterior uretral valve (PUV). History of
prenatal dilatation of the renal pelvis was present in four
patients. Two patients were diagnosed with PUV and one
patient was diagnosed with VUR. The last patients prenatal
dilatation history was not confirmed postnatally.
During follow-up, 32 UTI episodes were recorded in 20
patients (28%); six were preterm babies. The causative
agents for preterm babies were Klebsiella pneumonia in four
cases, Proteus in one case and Klebsiella oxytoca in one case.
The causative agents for term babies were E. coli in 13
episodes, Klebsiella pneumonia in seven episodes, Candida

23

albicans in three episodes, Enterococcus fecalis in two

episodes, and Proteus in one episode. In the patient group
with recurrences, three patients had VUR and two patients
had UPJ stenosis. The remaining 15 did not have any predisposing factor. The recurrences occurred between 1.5 and
12 months. Of the 32 episodes, 20 (65%) were in 15 infants
and occurred during the first 6 months. Only five cases
experienced the first UTI recurrence during the second
6 months. Eight patients had two or more recurrences. Blood
urea nitrogen and creatinine concentrations of all our patients
were normal.

Discussion
The morbidity of UTI in infancy is very high, resulting in
permanent renal damage causing hypertension or end-stage
renal disease. Routine follow-up with repeated urine cultures
and detailed urinary tract investigations after the first UTI in
neonates are recommended in order to diagnose predisposing
factors and for early identification of recurrent UTI. It is
known that UTI is more frequent in boys in the first 3 months
of life.1 The sex distribution is reported to be 5 : 1 with male
predominance.3 Our results showed male dominancy of 4 : 1.
The diagnosis of UTI in the neonate is confirmed only by
the examination and culture of a properly obtained urine
specimen. False positive results with urine bags and cleancatch specimens are frequent, therefore it is recommended to
obtain urine specimens for culture by bladder catheterization
or suprapubic aspiration. All our urine samples were taken by
bladder catheterization. Absence of pyuria does not rule out
UTI. We detected pyuria in 41% of preterm and 69% of term
infants with a significant difference between gestational ages.
We recommend that especially in term infants the presence
of pyuria can direct the doctor to a diagnosis of UTI, but the
long list of differential diagnoses of pyuria should be kept in
mind. A high index of suspicion is required for the diagnosis
of neonatal UTI as clinical presentation during infancy is
usually non-specific. Poor weight gain, feeding problems,
fever or hypothermia, color changes of the skin, abnormal
crying, irritability, abdominal distention, malodorous urine,
vomiting, diarrhea, rash, jaundice and hepatosplenomegaly
can be presenting features of UTI.1
Falcao et al. reported that the most relevant clinical
symptoms were fever and weight loss in 61 full-term newborn
infants.4 Although jaundice is reported to be a seldom finding,
prolonged jaundice is one of the most frequent presenting
symptoms in our term patient group with the maximum total
bilirubin concentration of 16 mg/dL. Garcia and Nager reported
the incidence of UTI in asymptomatic, afebrile, jaundiced
infants as 7.5%.5 Surprisingly, none of the suspected sepsis
cases in our study group presented with jaundice. Prenatal
dilatation history was present in four patients. Two patients

24

N Karaaslan Bykl

had PUV and one had VUR. The fourth patients prenatal
dilatation was not confirmed postnatally. E. coli and
Klebsiella are the most commonly involved bacteria in
neonatal UTI with the incidence of 75.3%, 13.4%, respectively.1,6 In our series E. coli and Klebsiella pneumonia were
the predominate causative bacteria of community-acquired
and nasocomial UTI, respectively. The overall incidence in
our study group was 43% for E. coli and 38% for Klebsiella
pneumonia. Low-birthweight and prematurity are patient
groups that are prone to all infections including UTI.
Twenty-nine babies (41%) in our study group were preterm
and three preterm babies and one term baby were small for
gestational ages.
Although we were not able to identify the same microorganism in urine culture and blood culture, four of our
patients (5%) were diagnosed clinically as urosepsis with the
support of sepsis work up. The reported incidence for
urosepsis in newborn UTI is 30%.7
It has been stated that up to 30% of infants and children
with a UTI have a urinary tract abnormality.8 USG is the first
investigation method that should be performed in patient
groups experiencing UTI to evaluate the kidneys. It provides
information about renal structural anomalies and hydronephrosis, renal parenchymal abnormalities, perinephric
collections, ureteral dilatation or calculi.9
Ultrasonography was performed in 65 patients in our
group and abnormalities were observed in only 15 patients
(23%). Because reflux is found in as many as 50% of
children with UTI, the diagnosis and initial evaluation of the
child with UTI is often crucial in determining long-term
management and in preventing ultimate renal damage.10,11
Vesicoureteral reflux incidence was 15% in our series
which seems to be less than the reported series. This is
possibly due to the limited number of cases. Previous studies
reporting DMSA abnormalities associated with acute UTI
showed an incidence of 2065%.8,1214 Experimental studies
show that DMSA changes occurring during an acute UTI
correlate well with histologically proven pyelonephritis in
site and severity.15 DMSA scintigraphy is a sensitive
technique for the detection of upper tract involvement with
acute UTI. Recent studies using DMSA to determine the
presence of UTI have revealed that more than 75% of
children under 5 years of age with febrile UTI have pyelonephritis.16,17 Pyelonephritis leads to renal scarring in 27% to
64% of children in this age group.18 A study from Sweden
showed that focal renal scarring caused by pyelonephritis in
children carried a 23% risk of hypertension, a 10% risk of
end-stage renal disease, and 13% risk of toxemia during
pregnancy as an adult.19 DMSA scan is suggested to be
performed in the third month of the UTI for detection of
photopenic areas. We performed DMSA in the third month of
UTI to show renal scars. It was performed in 36 babies, in
which renal photopenic areas were detected in eight (23%).

Recurrence rates was reported to be 25% in neonatal UTI,

which clearly emphasizes the importance of close follow-up
in children with their first UTI.20
Two series reported the UTI recurrence as 26% and 19%,
respectively; the second episode usually occurred during the
first few months after initial infection.21,22 During follow-up
we observed 32 recurrences in 20 subjects. Our recurrence
rate was 28%. Six of them were preterm babies, two being
borderline preterms of 37 weeks gestational age, and 14
were full-term babies.
Most of the recurrences (20 episodes in 15 cases) occurred
in the first 6 months (62%), and these data support our
follow-up programme with the rationality to put newborn
UTI cases on prophylaxis. Only five cases showed recurrences
after the first 6 months.
This is the first article in the English literature that
discusses follow-up results of newborn UTI. We conclude
that UTI are not rare in the neonatal period. Investigation of
the urinary tract is essential for the early detection of predisposing factors. Renal tissue is prone to scarring in infancy
therefore early diagnosis and treatment is critical and can
prevent permanent renal damage of the growing kidney.
Susceptibility of the individual appears to play a major role
in the etiology of neonatal UTI, especially in the absence of
predisposing congenital anomalies.
Hyperbilirubinemia and signs of suspected sepsis were the
main presenting symptoms. Routine screening for nasocomial
UTI in preterm babies should occur in neonatal nurseries.
Pediatric nephrologic follow-up of babies experiencing UTI
in the neonatal period is very important because of the risk of
long-term sequelae.

References
1 Long SS, Klein JO. Bacterial infections of the urinary tract. In:
Remington JS, Klein JO (eds). Infectious Diseases of the Fetus
and Newborn Infant, 5th edn. W.B. Saunders, Philadelphia, 2001.
2 Langley JM, Hanakowski M, Leblanc JC. Unique epidemiology of nasocomial urinary tract infection in children. Am. J.
Infect. Control 2001; 29: 948.
3 Maherzi M, Guignard JP, Torrado A. Urinary tract infection in
high-risk newborn infants. Pediatrics 1978; 62: 5213.
4 Falcao MC, Leone CR, DAndrea RA, Berardi R, Ono NA,
Vaz FA. Urinary tract infections in full-term newborn infants:
value of urine culture by bag specimen collection. Rev. Hosp.
Clin. Fac. Med. Sao Paulo 1999; 54: 916.
5 Garcia FJ, Nager AL. Jaundice as an early sign of urinary tract
infection in infancy. Pediatrics 2002; 109: 84651.
6 Cataldi L, Cianfoni A. Infections of the urinary tract in newborn
and infants. Pediatria Med. E Chirurgica 1997; 19: 319.
7 Ginsburg CM, McCracker GH. Urinary tract infections in
young infants. Pediatrics 1982; 69: 409.
8 Zderic SA. Infection of the urinary tract and vesicoureteral
reflux. In: Taeusch HW, Ballard RA (eds). Averys Diseases of
the Newborn, 7th edn. W.B. Saunders, Philadelphia, 1998.

Neonatal UTI
9 Johnson CE, DeBaz BP, Shurin PA. Renal ultrasound
evaluation of urinary tract infections in children. Pediatrics
1986; 78: 871.
10 Rolleston GL, Shanon FT, Utley WLF. Relationship of
infantile vesicoureteric reflux to renal damage. BMJ 1970; 1:
460.
11 Nuutinen M, Uhari M. Recurrence and follow-up after urinary
tract infection under the age of 1 year. Pediatr. Nephrol. 2001;
16: 6972.
12 Verber IG, Strudley MR, Meller ST. 99m Tc dimercaptosuccinic acid (DMSA) scan as first investigation of urinary
tract infection. Arch. Dis. Child. 1988; 63: 1320.
13 Sty JR, Wells RG, Starshak RJ, Schoeder BA. Imaging in
acute renal infection in children. AJR 1987; 148: 471.
14 Rosenberg AR, Rossleigh MA, Brydon MP, Bass SJ,
Leighton DM, Farnworth RH. Evaluation of acute urinary tract
infection in children by dimercaptosuccinic acid scintigraphy:
a prospective study. J Urol. 1992: 148: 17469.
15 Rushton HG, Majd M, Chandra R, Yim D. Evaluation of 99m
technetium dimercaptosuccinic acid renal scan in experimental
acute pyelonephritis in piglets. J Urol. 1988; 140 (Pt 2): 1169.

25

16 Benador D, Benador N, Slosman DO et al. Cortical scintigraphy in the evaluation of renal parenchymal changes in
children with pyelonephritis. J. Pediatr. 1994; 124: 1720.
17 Jakopson B, Soderlund H, Berg U. Diagnostic significance of
99mTc-dimercaptosuccinic acid (DMSA) scintigraphy in urinary
tract infection. Arch. Dis. Child 1992; 67: 133842.
18 Berg UB. Long-term follow-up of renal morphology and
function in children with recurrent pyelonephritis. J. Urol.
1992; 148: 171520.
19 Jakobson SH, Eklof O, Eriksson CG et al. Development of
hypertension and uraemia after pyelonephritis in childhood:
27-year follow-up. BMJ 1989; 299: 7036.
20 Hamburger EK. Urinary tract infections in infants and children.
Postgraduate Med. 1986; 80: 23541.
21 Bergstrm T, Larson H, Lincoln K. Neonatal urinary tract
infections. J. Pediatr. 1972; 80: 859.
22 Littlewood JM, Kite P, Kite BA. Incidence of neonatal urinary
tract infection. Arch. Dis. Child 1969; 44: 617.

Copyright of Pediatrics International is the property of Wiley-Blackwell 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.