You are on page 1of 5

+ MODEL

Pediatrics and Neonatology (2017) xx, 1e5

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: http://www.pediatr-neonatol.com

Original Article

Urinary tract infections in neonates with


unexplained pathological indirect
hyperbilirubinemia: Prevalence and
significance
Elif Bahat Ozdogan a,*, Mehmet Mutlu b,
Secil Arslansoyu Camlar a, Gülcin Bayramoglu c,
Sebnem Kader b, Yakup Aslan b

a
Department of Pediatrics, Division of Pediatric Nephrology, Karadeniz Technical University, Faculty
of Medicine, Trabzon, Turkey
b
Department of Pediatrics, Division of Neonatology, Karadeniz Technical University, Faculty of
Medicine, Trabzon, Turkey
c
Department of Microbiology, Karadeniz Technical University, Faculty of Medicine, Trabzon, Turkey

Received Jan 30, 2017; received in revised form Aug 14, 2017; accepted Oct 25, 2017
Available online - - -

Key Words Abstract Background: It is controversial to test for urinary tract infection (UTI) in patients
follow-up; with unexplained indirect hyperbilirubinemia in the first 2 weeks of life. We aimed to study
hyperbilirubinemia; the prevalence and significance of UTIs in such neonates who were requiring phototherapy.
neonate; Methods: Subjects were 2- to 14-day-old neonates with indirect bilirubin levels above photo-
urinary tract therapy limit with no other abnormality in their bilirubinaemia-related etiologic workup. UTI
infection; was diagnosed by 2 consecutive positive cultures obtained by catheterisation, documenting
prevalence growth of >10,000 colonies of the same microorganism with consistent antibiograms. The
UTI (þ) patients were evaluated by renal ultrasonography (US), and some were followed up
for possible recurrent UTI.
Results: 262 neonates were included in the study. UTI prevalence was 12.2%, and bacteraemia
was 6.2% among UTI (þ) patients. The two most common pathogens (81.2%) were Escherichia
coli and Klebsiella. pneumonia. All UTI (þ) patients had undergone US, revealing 12.5% pelvi-
caliectasis, other 12.5% increased renal parenchymal echogenicity, 3.1% concurrent pelvica-
liectasis and increased renal parenchymal echogenicity. 53.1% of UTI (þ) patients had
undergone follow-up, after which 23.5% recurrent UTI were found at the end of a mean of
52 months.

* Corresponding author. Department of Pediatrics, Division of Pediatric Nephrology, Karadeniz Technical University, Faculity of Medicine,
61080, Trabzon, Turkey. Fax: þ90 462 2235199.
E-mail address: elifbahat@yahoo.com (E. Bahat Ozdogan).

https://doi.org/10.1016/j.pedneo.2017.10.010
1875-9572/Copyright ª 2017, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-
NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: Bahat Ozdogan E, et al., Urinary tract infections in neonates with unexplained pathological indirect
hyperbilirubinemia: Prevalence and significance, Pediatrics and Neonatology (2017), https://doi.org/10.1016/j.pedneo.2017.10.010
+ MODEL
2 E. Bahat Ozdogan et al

Conclusion: We suggest that the neonates with unexplained pathological jaundice should be
tested for possible UTI. Consequently, all newborns with UTI shall be evaluated by the urinary
US and followed up for recurrent UTI.
Copyright ª 2017, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/4.0/).

1. Introduction test, complete blood count, peripheral blood smear,


reticulocyte count, serum total/direct bilirubin, glucose,
Neonatal hyperbilirubinemia is seen in the first week of life blood urea nitrogen, creatinine, sodium and urinalysis were
in 60% of the term and 80% of the preterm newborns. In performed. If the aetiology of jaundice could not be
these cases, bilirubin levels may remain in the physiologic explained by blood group incompatibility, glucose-6-
range or may exceed the level requiring phototherapy or phosphate dehydrogenase and pyruvate kinase enzyme
blood exchange. The aetiology of pathologic hyper- levels, urinary reducing substances, thyroid hormones and
bilirubinemia could not always be determined. urine cultures were analysed. Polycythaemia was defined as
Urinary tract infection (UTI) presents with variable the haemoglobin concentration greater than two standard
symptoms in the neonatal period. Nonspecific complaints deviations above the normal value for gestational and
(i.e., poor weight gain, vomiting, diarrhoea, fever, irrita- postnatal age. Urine samples for urinalysis and urine cul-
bility, lethargy, jaundice) may be the only symptoms,1 ture were obtained using catheterisation. The samples
among which jaundice is documented as one of the most were analysed with Clinitek 100 (Bayer) or IQ 200 (IRIS)
common.2,3 UTIs are well-known to cause prolonged jaun- automatic analysers. Pyuria was defined as the presence of
dice and investigated as a standard of care. Along with >5 leukocytes in one high-power field in the sediment of
prolonged jaundice, UTIs may underlie the unexplained centrifuged urine.5 Presence of 1 microorganism in one
pathologic jaundice presenting in the first 2 weeks of life. It immersion field by Gram stain was accepted as bacteriuria.
is still controversial to test for UTI in patients with unex- In the urinary culture, 10,000 colonies of a single path-
plained jaundice presenting in the first 2 weeks of life.3e9 ogen were regarded as significant.6 In patients with a pos-
Hence, some authors suggest investigating these neonates itive urine culture, before initiation of related-
requiring treatment for pathologic indirect hyper- antibiotherapy, blood cultures and confirmatory catheter
bilirubinemia for UTIs,3,4,6,7,9 while some do not.5,8 urine culture were obtained. Immature/total neutrophil
In this report, we aimed to study the prevalence and ratio was determined using a peripheral blood smear. C-
significance of UTIs in neonates presenting with indirect reactive protein >5 mg/L and procalcitonin >0.5 mg/L were
hyperbilirubinemia in the first 2 weeks of life requiring regarded as significant. In patients who had elevated acute-
phototherapy who were found to have no other abnormality phase reactant levels or positive blood cultures, lumbar
in their aetiologic workup. puncture was also performed.
UTI was diagnosed in patients with confirmatory urine
culture 10,000 colonies of the same microorganism with
2. Material and methods the first culture and related antibiogram. Patients with
bilirubin levels above the phototherapy cutoff level and
This study was performed between November 2004 and negative cultures were accepted as the UTI () group.
November 2007 in the Karadeniz Technical University Farabi Patients whose second urine culture was negative, positive
Hospital neonatal intensive care unit, Turkey. Some pa- for a different organism, or yielded multiple microorgan-
tients were followed up until December 2013 by the Pae- isms or had no parent-given informed were excluded from
diatric Nephrology Department. Subjects were 2- to 14-day- the study. UTI (þ) and () patients were compared for the
old neonates with indirect bilirubin levels above the pho- above parameters (i.e., birth weight, gestational age, sex,
totherapy limit with no other abnormality in their delivery method, weight at presentation, weight loss, onset
bilirubinaemia-related aetiologic workup for systemic of jaundice, time of presentation, total/direct bilirubin at
infection, isoimmunisation, erythrocyte enzyme defect, presentation, blood exchange, duration of phototherapy,
erythrocyte structural defect, hypothyroidism, seques- urine density, presence of hypoglycaemia and
trated blood, polycythaemia or metabolic disease. Photo- hypernatraemia).
therapy limits of bilirubin levels were determined As secondary analyses, all UTI (þ) patients were evalu-
according to the guidelines of American Pediatrics ated by renal ultrasonography (US). All were offered to
Academy.10 have voiding cystourethrogram (VCUG) at least 1 month
Method of delivery, sex, birth weight, weight at pre- after the sterilisation of urine cultures and dimercapto-
sentation, gestational age, postnatal age at presentation, succinic acid scintigraphy (DMSA). Their follow-up data
weight loss, breast feeding data, onset of jaundice by his- were evaluated.
tory, bilirubin level at presentation, duration of photo- Informed consent was obtained from the parents of all
therapy and requirement of blood exchange were recorded. patients, and the study was approved by the local institu-
Tests for blood group, mother’s blood group, direct Coombs tional ethical board (approval number: 200691).

Please cite this article in press as: Bahat Ozdogan E, et al., Urinary tract infections in neonates with unexplained pathological indirect
hyperbilirubinemia: Prevalence and significance, Pediatrics and Neonatology (2017), https://doi.org/10.1016/j.pedneo.2017.10.010
+ MODEL
Urinary tract infections in neonatal hyperbilirubinemia 3

2.1. Statistical analysis comparative demographic data and the jaundice-related


data of the UTI (þ) vs. UTI () patients are outlined in
The variables were investigated to determine whether they Table 2 and Table 3, respectively.
were normally distributed. Numerical variables were given Weight loss was evaluated as percentage of weight loss
as mean  standard deviation for normally distributed after birth (birth weighteweight at presentation [*100/
variables and as median for skew-distributed continuous birth weight]). The weight loss percentage and rebound
variables. Categorical variables were shown as frequencies. bilirubin levels were significantly different between the UTI
Two groups were compared with independent sample t-test (þ) and () patients (p Z 0.02 and p Z 0.01, respectively).
or Mann Whitney U test when necessary. Chi-square test UTI (þ) patients had significantly less weight loss but higher
with Yates correction and Fisher’s exact test were used for rebound bilirubin levels than UTI () patients. Other study
2  2 contingency tables when appropriate for non- parameters were not different between groups.
numerical data. SPSS (statistical package for social sci- All UTI (þ) patients were evaluated by renal ultraso-
ences) for Windows 15.0 was used for statistical analysis. nography (USG) during hospitalisation which revealed
p < 0.05 was accepted as statistically significant. abnormal finding in 9 patients (28.1%) as follows: 4 patients
(12.5%) had only pelvicaliectasis, 4 (12.5%) patients had
3. Results only increased renal parenchymal echogenicity and 1 pa-
tient (3.1%) had both pelvicaliectasis and increased renal
A total of 482 neonates presented to the neonatal intensive parenchymal echogenicity.
care unit with indirect hyperbilirubinemia above the pho- Among the total 32 UTI (þ) patients, 20 (62.5%) con-
totherapy limit. Among them, 262 (54%) fulfilled our in- sented to VCUG (the remaining 37.5% of the UTI (þ) pa-
clusion criteria. Of the 262, 47% (123) were female, and 53% tients either did not consent to VCUG regarding the invasive
(139) were male. All males were uncircumcised. Mean nature of the procedure and radiation exposure or they
gestational age was 37.6  2.1 (31e41) weeks. Mean age at were lost to follow-up). Two VCUG-evaluated patients (10%)
presentation was 5.3  2.2 (2e14) days. Sixty percent (158) had more than grade 1 vesicoureteral reflux (VUR) (one
were born via vaginal delivery, while 40% (104) were born grade 2 and one grade 3 VUR). Both VUR (þ) patients had
via caesarean section. Mean birth weight was 3021  676 abnormal renal US findings; one had pelvicaliectasis, and
(1500e5500) grams, and mean weight at presentation was the other had increased renal echogenicity.
2881  646 (1320e5570) grams. 90.8% (238) were fed only Among the 32 UTI (þ) patients, 6 (18.7%) consented to
with breast milk, 3.8% (10) were fed formula, and 5.3% (14) DMSA. The remaining 81.3% of the UTI (þ) patients either
were fed both formula and breast milk. Mean elapsed time did not consent to DMSA regarding the invasive nature of
to recognition of jaundice was 3.3  1.0 (2e8) days. Mean the procedure and radiation exposure or they were lost
bilirubin level of the patients at presentation was follow-up. One (16.6%) of them was found to have renal
20.9  6.1 (7.6e45.1) mg/dL. 13.4% (35) were treated with scarring. This patient also had grade 2 VUR and increased
blood exchange and phototherapy; the rest were treated renal parenchymal echogenicity.
with only phototherapy. Mean duration of phototherapy Among the 32 UTI (þ) patients, 17 (53.1%) could have
was 64.2  21 (24e120) hours. None of the subjects had been followed up more than 12 months since the others
fever. UTI was diagnosed in 12.2% (32), but not in 87.8%
(230). In UTI (þ) patients, 37.5% (12) had pyuria and
bacteriuria by Gram stain, 9.4% (3) had only pyuria, 18.7% Table 2 Comparative demographic data of UTI (þ) vs UTI
(6) had only bacteriuria and 34.4% (11) had neither pyuria () patients.
nor bacteriuria. The bacteriuria by Gram stain were Gram Parameter UTI (þ) UTI () p
() bacilli in 88.9% (16) and Gram (þ) cocci in 11.1% (2). patients patients
The two most common microorganisms in urine culture (n Z 32) (n Z 230)
were Escherichia coli (50%) and Klebsiella pneumoniae Birth weight (gram) 3099  534 3010  694 0.5
(31.2%) (Table 1). Of UTI (þ) patients, 6.2% (2) had positive Gestational age (week) 37.9  1.6 37.5  2.2 0.5
blood culture for the microorganism positive in the urine Sex 0.6
cultures (i.e., K. pneumoniae and Staphylococcus aureus). Male (n) (%) 19 (59%) 120 (52%)
These cases were confirmed to not have meningitis. C- Female (n) (%) 13 (41%) 110 (48%)
reactive protein, procalcitonin levels and renal function Delivery 0.9
tests were within normal levels in all patients. The Vaginal (n) (%) 20 (63%) 138 (60%)
c/s (n) (%) 12 (37%) 92 (40%)
Table 1 Microorganisms identified in urine cultures Weight (presentation) 3014  510 2863  661 0.2
(n Z 32). (gram)
Microorganism n (%)
a
Weight loss % 6  2.9 8.4  5.5 0.02
(190  89) (290  367)
Escherichia coli 16 (50%)
Klebsiella pneumoniae 10 (31.2%) Data are given as mean  standard deviation.
Enterobacter aerogenes 2 (6.2%) n: number.
Staphylococcus aureus 2 (6.2%) c/s: caesarean section.
a
Weight loss % was calculated as percentage of weight loss
Kluyvera ascorbata 1 (3.1%)
after birth (birth weighteweight at presentation [*100/birth
Enterococcus faecalis 1 (3.1%)
weight]).

Please cite this article in press as: Bahat Ozdogan E, et al., Urinary tract infections in neonates with unexplained pathological indirect
hyperbilirubinemia: Prevalence and significance, Pediatrics and Neonatology (2017), https://doi.org/10.1016/j.pedneo.2017.10.010
+ MODEL
4 E. Bahat Ozdogan et al

Table 3 Comparative jaundice-related data of UTI (þ) vs UTI () patients.


Parameter UTI (þ) patients (n Z 32) UTI () patients (n Z 230) p
Onset of jaundice (day) 3.4  1.0 3.3  1.0 0.4
Time of presentation (day) 5.3  3.8 5.2  2.3 0.4
Total bilirubin (presentation) (mg/dL) 21.2  5.0 20.8  6.3 0.4
Direct bilirubin (presentation) (mg/dL) 0.6  0.3 0.6  0.3 0.7
Blood exchange (n) (%) 4 (13%) 31 (13%) 0.9
Duration of phototherapy (hour) 70.9  19.8 63.3  21.0 0.06
Rebound bilirubin (mg/dL) 11.1  1.3 10  1.7 0.01
Urine density 1014  7 1014  7 1
Hypoglycaemia (n) (%) 3 (9%) 40 (18%) 0.4
Hypernatraemia (n) (%) 2 (6%) 27 (12%) 0.5
Data are given as mean  standard deviation.
n: number.
Hypoglycaemia: serum glucose level< 50 mg/dL.
Hypernatraemia: serum Na level >145 mg/dL.

were lost to follow-up. Mean follow-up duration was look for UTI in all pathologic jaundice cases, including cases
52  21 months; median was 53 months. In this follow-up, 4 with predefined aetiology, UTI % would be lower and more
(23.5%) had recurrent UTI, of whom one had grade 2 VUR similar to the aforementioned reports in the literature.4,11
and one had concurrent pelvicaliectasis and increased renal Two other studies from Turkey examined UTI prevalence in
parenchymal echogenicity. <2-week-old newborns with asymptomatic, unexplained
indirect hyperbilirubinemia.6,9 Bilgen et al.6 examined 102
patients and reported UTI as 8%. Most recently, Mutlu et al.9
4. Discussion
studied 104 patients and reported UTI as 18%. The first
study was reported from a more developed city (Istanbul)6
Jaundice is one of the most common symptoms of UTIs in than the later one (Erzurum).9 The comparable but
neonates2,3 and may be its sole sign. UTIs are well-known to different UTI rates from the same country may be due to
cause prolonged jaundice and are investigated routinely. the socioeconomic status of the study region affecting
On the other hand, some authors suggest investigating UTI proper baby care. All these studies pointed out the signif-
also in the neonates requiring treatment for pathologic icance of UTIs in the neonatal infants with unexplained
indirect hyperbilirubinaemia,3,4,6,7,9 while some do not.5,8 hyperbilirubinemia, and they suggested urine culture
Therefore, it is controversial to look for UTI in these should be considered in the bilirubin workup of such pa-
patients.3e9 tients.4,6,7,9,11 The counter argument is that most of the
In a similar study, Garcia et al.4 included 160 neonates neonates become jaundiced in the first 2 weeks of life, and
<8 weeks old with unexplained jaundice and reported UTI UTI is not significantly prevalent in these patients and its
in 7.5%. Chen at al.11 studied 217 neonatal infants <8 work-up is too invasive, expensive and with high false-
weeks old with the initial presentation of hyper- positive rates. Therefore, they suggest it is unnecessary
bilirubinemia and reported UTI in 5.5%. Shahian et al.7 to look for UTI in all jaundiced neonates.2,5 However, we
studied 120 neonates <4 weeks old with asymptomatic limited UTI examination to the neonates with unexplained
jaundice and reported UTI in 12.5%. Our UTI prevalence pathologic indirect hyperbilirubinemia instead of all jaun-
(12.2%) was higher than that of Garcia et al.’s (7.5%) and diced neonates, and we had no false-positive UTI, possibly
Chen et al.’s 5.5% studies and comparable to that of Sha- due to our study design. Therefore, our UTI prevalence of
hian et al.’s (12.5%) study. Garcia et al. noted that most 12.2% suggests it is important to investigate UTI in neonates
(75%) of their UTI (þ) patients were younger than 2 weeks with unexplained pathologic indirect hyperbilirubinemia.
old, suggesting a higher UTI prevalence in the <2-week-old The two most common microorganisms in urine culture
neonates with unexplained jaundice. This is similar to the were E. coli (50%) and K. pneumoniae (31.2%) in our study;
result suggested by our study. On the other hand, Chen together, they were the responsible pathogens in 81.2% of
et al. studied patients with the initial presentation of the UTIs, which is a well-expected finding in the neonates.
hyperbilirubinemia. Those patients did not have unex- Of UTI (þ) patients, 6.2% had documented bacteraemia
plained, isolated jaundice. Therefore, it is not surprising in this study. This finding is also in accordance with the
that they had higher prevalence of other jaundice-related results of other studies in the literature, as the prevalence
pathologies and lower percentage of UTI. Additionally, of bacteraemia and sepsis in children with UTI has been
Chen et al. did not report the prevalence in a <2-week-old reported between 3.2% and 31% in various studies.12e15 We
patient subpopulation. Garcia et al. might have had higher suggest that the relatively earlier detection of UTI in these
UTI prevalence in the 2-week-old patients than in our patients than in the standard practice might have resulted
study. Also, a possibility exists that co-existing UTI was in earlier and more timely treatment of the UTI in our
present in our excluded cases, which had predefined study. This hypothesis is supported by lack of inflammatory
pathologic jaundice aetiology. When there is a predefined response in the cases with documented bacteraemia.
cause of pathologic jaundice, the probability of co-existing Hence, our patients with UTI had relatively lower
UTI would be expected to be low. Accordingly, if we would

Please cite this article in press as: Bahat Ozdogan E, et al., Urinary tract infections in neonates with unexplained pathological indirect
hyperbilirubinemia: Prevalence and significance, Pediatrics and Neonatology (2017), https://doi.org/10.1016/j.pedneo.2017.10.010
+ MODEL
Urinary tract infections in neonatal hyperbilirubinemia 5

prevalence of bacteraemia/sepsis than reported in the References


literature. Nevertheless, UTIs may cause urosepsis, renal
scarring, hypertension and chronic renal failure if treated 1. Littlewood JM. Sixty-six infants with urinary tract infection in
late or untreated. Hence, our findings show the benefits of first month of life. Arch Dis Child 1972;47:218e26.
obtaining urine cultures for the diagnosis of possible UTI in 2. Olusanya O, Owa JA, Olusanya OI. The prevalence of bacteri-
neonates with unexplained pathologic hyperbilirubinemia uria among high risk neonates in Nigeria. Acta Paediatr Scand
requiring phototherapy. 1989;78:94e9.
All UTI (þ) patients were further evaluated by renal 3. Biyikli NK, Alpay H, Ozek E, Akman I, Bilgen H. Neonatal urinary
USG, documenting abnormal finding in 28.1%. The American tract infections: analysis of the patients and recurrences.
Pediatr Int 2004;46:21e5.
Academy of Pediatrics (AAP) UTI guideline recommends
4. Garcia FJ, Nager AL. Jaundice as an early diagnostic sign of
performing US in all >2-month-old infants with UTI urinary tract infection in infancy. Pediatrics 2002;109:846e51.
accompanied by fever but does not make any recommen- 5. Maisels MJ, Newman TB. Neonatal jaundice and urinary tract
dation for the <2-month-old infants.16 As our study docu- infections. Pediatrics 2003;112:1213e4.
mented a significant portion of USG abnormality in the non- 6. Bilgen H, Ozek E, Unver T, Biyikli N, Alpay H, Cebeci D. Urinary
pyretic UTI (þ) < 2-week-old neonates, we suggest that the tract infection and hyperbilirubinemia. Turk J Pediatr 2006;48:
USG evaluation shall also be performed in the non-pyretic 51e5.
UTIs of <2-month-old infants. 7. Shahian M, Rashtian P, Kalani M. Unexplained neonatal jaun-
Almost 2/3 of the patients were evaluated by VCUG, dice as an early diagnostic sign of urinary tract infection. Int J
documenting 10% (2 patients) having VUR more than grade 1 Infect Dis 2012;16:e487e490.
8. Abourazzak S, Bouharrou A, Hida M. Jaundice and urinary tract
severity. Both of these patients had abnormal renal US
infection in neonates: simple coincidence or real consequence?
finding. This finding suggests performing VCUG if the pa- Arch Pediatr 2013;20:974e8.
tients have an abnormal finding in USG, which is in line with 9. Mutlu M, Cayır Y, Aslan Y. Urinary tract infections in neonates
AAP recommendations.16 with jaundice in their first two weeks of life. World J Pediatr
Our study has several strengths and some limitations. 2014;10:164e7.
First, the diagnosis of UTI in our study was evidence-based. 10. Stoll JB, Kliegman RM. Jaundice and hyperbilirubinemia in the
Pyuria is not a specific diagnostic criterion for UTI in neo- newborn. In: Berhman RE, Kliegman RM, Jenson HB, editors.
nates. Urine dipstick analysis in infants has low sensitivity, Nelson textbook of Pediatrics. 19th ed. Philadelphia: WB
so urine cultures should be used for diagnosis in suspicious Saunders; 2000. p. 513e9.
cases.17,18 Bacteria seen in Gram-stained urine sediment is 11. Chen HT, Jeng MJ, Soong WJ, Yang CF, Tsao PC, Lee YS, et al.
Hyperbilirubinemia with urinary tract infection in infants
a finding that supports the diagnosis. However, this is not
younger than eight weeks old. J Chin Med Assoc 2011;74:
sufficient for the diagnosis of UTI, and urine culture is 159e63.
indicated. In this study, UTI was diagnosed by 2 consecutive 12. Linder N, Yatsiv I, Tsur M, Matoth I, Mandelberg A, Hoffman B,
positive cultures obtained by the catheterisation doc- et al. Unexplained neonatal jaundice as an early diagnostic
umenting the growth of >10,000 colonies of the same sign of septicemia in the newborn. J Perinatol 1988;8:325e7.
microorganism with consistent antibiograms. Obtaining 13. Marom R, Sakran W, Antonelli J, Horovitz Y, Zarfin Y, Koren A,
urine samples from a urine bag is not suitable in neonates. et al. Quick identification of febrile neonates with low risk for
Rather, urine samples are obtained by USG-guided supra- serious bacterial infection: an observational study. Arch Dis
pubic aspiration or bladder catheterisation,19,20 and the Child Fetal Neonatal Ed 2007;92:F15e8.
latter procedure was applied in this study. Furthermore, we 14. Craig JC, Hodson EM. Treatment of acute pyelonephritis in
children. BMJ 2004;328:179e80.
obtained two consecutive urine cultures for each patient,
15. Ginsburg CM, McCracken Jr GH. Urinary tract infections in
representing one of two studies in this aspect.9 This young infants. Pediatrics 1982;69:409e12.
approach supports the validity of UTI diagnosis, making the 16. Subcommittee on Urinary Tract Infection, Steering Committee
reliability of our study high. Secondly, our study is also the on Quality Improvement and Management, Roberts KB. Urinary
first with follow-up. Lastly, to our knowledge, this is the tract infection: clinical practice guideline for the diagnosis and
largest case series on this subject. On the other hand, one management of the initial UTI in febrile infants and children 2
should consider that all the cases could not have undergone to 24 months. Pediatrics 2011;128:595e610.
VCUG, DMSA analysis and follow-up. So, this study provides 17. Hoberman A, Chao HP, Keller DM, Hickey R, Davis HW, Ellis D.
important data but should be examined in larger case Prevalence of urinary tract infection in febrile infants. J
series. Pediatr 1993;123:17e23.
18. Crain EF, Gershel JC. Urinary tract infections in febrile infants
In conclusion, in view of our findings, we suggest that
younger than 8 weeks of age. Pediatrics 1990;86:363e7.
neonates with unexplained pathologic jaundice should be 19. Kapur R, Yoder MC, Polin RA. Urinary tract infection. In:
examined for possible UTI as a standard of care. Conse- Martin RJ, Fanoroff AA, Walsh MC, editors. Neonatal-perinatal
quently, urinary USdan inexpensive and noninvasive im- medicine diseases of the fetus and infant volume 2. 8th ed.
aging techniquedshould be performed in all newborns with Philadelphia: Mosby Elsevier; 2006. p.815e816.
UTI, and these patients should undergo follow-up. 20. Kozer E, Rosenbloom E, Goldman D, Lavy G, Rosenfeld N,
Goldman M. Pain in infants who are younger than 2 Months
during suprapubic aspiration and transurethral bladder cathe-
Conflict of interest terization: a randomized, controlled study. Pediatrics 2006;
118:e51e6.
No conflict of interest.

Please cite this article in press as: Bahat Ozdogan E, et al., Urinary tract infections in neonates with unexplained pathological indirect
hyperbilirubinemia: Prevalence and significance, Pediatrics and Neonatology (2017), https://doi.org/10.1016/j.pedneo.2017.10.010