Open Journal of Urology, 2011, 1, 76-80
doi:10.4236/oju.2011.14016 Published Online November 2011 (http://www.SciRP.org/journal/oju)
Which Patients Should Be Administrated Prophylactic
Antibacterial Agents ? A Study of Bacte r iuria or Fun g u r i a
by Urine Culture Taken from the Renal Pelvis in Children
with Ureteropelvic Junction Obstruction
Gao-Yan Deng*, Li-Yu Zhang, Zhong-Ming Li, Ying-Quan Wen
Department of Pediatric Surgery, Guang Zhou Women and Children ’s Medical Center, Guang Zhou, China
Received July 21, 2011; revised August 30, 2011; accepted September 9, 2011
Abstract
Objective: To detect bacteriuria or funguria by urine culture taken from the renal pelvis directly before
Anderson-Hynes pyeloplasty. Methods: 290 patients who underwent Anderson–Hynes pyeloplasty for uret-
eropelvic junction obstruction (UPJO) were included in a retrospective analysis. Urine was obtained directly
before the renal pelvis was opened, and was carried to the laboratory for bacterial culture. Clinical features
were analyzed to evaluate risk factors for bacteriuria or funguria by comparing patients whose urine yielded
positive cultures to those whose urine cultures were negative for bacteria or yeast. Results: Eighteen patients
(6.2%) had positive urine cultures, including six cultures positive for Escherichia coli (E. coli), four for
Pseudomonas aeruginosa, three for klebsiella pneumoniae, one for maltophilia monad, one for Enterococcus
faecium, one for Candida albicans, one for Candida parapsilosis, and one for yeast not otherwise specified.
Bacteriuria or funguria was significantly correlated with four clinical features: fever, urinary urgency, and
history of nephrostomy or pyeloplasty. Conclusions: Bacteriuria or funguria was less common in children
with UPJO, and the majority of organisms were identified as Escherichia coli, Pseudomonas aeruginosa, or
Klebsiella pneumoniae. Prophylactic antibacterial agents were probably necessary in those patients who had
signs of urinary tract infection (UTI), or history of nephrostomy or pyeloplasty.
Keywords: Bacteriuria, Ureteropelvic Junction Obstruction, Urine Culture, Pyeloplasty.
1. Introduction
There have been no published reports concerning bacte-
riuria or funguria discovered by urine culture taken di-
rectly from the renal pelvis, despite the extensive li-
terature on urinary tract infection (UTI) in children with
hydronephrosis [1,2] <B5 B6> </C>In our study, the
relationship between clinical features and bacteriuria or
funguria was also analyzed to evaluate risk factors for
these conditions.
2. Methods
2.1. Patients
A total of 300 patients (320 kidneys) underwent Ander-
son-Hynes pyeloplasty between May 1999 and January
2009 in our hospital. Urine was taken directly from the
renal pelvis of 270 patients (290 kidneys) [210 boys and
80 girls, with a median age of 1year (range 1month to 12
years)] for bacterial culture; urine of the remaining 30
cases was not cultured either because the patient had no
signs of UTI, or for other reasons. Prophylactic antibac-
terial agents were administrated to those patients who
had signs and symptoms of UTI (fever and urinary ur-
gency) prior to admission, and the surgical procedure
was postponed until these signs and symptoms disap-
peared. A low dose of antibacterial agents was also pro-
vided to patients who had a history of nephrostomy.
Routinely administer antibiotics was IV just prior to skin
incision.
2.2. Urine Culture Prior to Surgery
Anderson-Hynes pyeloplasty was performed in all cases.
Obtained immediately before the renal pelvis was opened

77 G.-Y. DENG ET AL.
with a needle inserted into the pelvis, urine for culture
was hand-carried in a sterile tube to the laboratory. These
specimens were cultured by inoculation onto cystine lac-
tose electrolyte-deficient agar, and were incubated in an
atmosphere containing 5% carbon dioxide. Cultures were
classified according to the criteria in standard use in the
laboratory, and bacteriuria or funguria was defined as
10/L or more colony-forming units (CFUs). Antibiotic
sensitivity was determined using standard techniques [3].
For our analysis, we classified specimens as resistant or
sensitive, although some organ- isms are not routinely
tested against certain antibiotics.
2.3. Recording of Clinical Features
Clinical features including age, gender, whether ureter-
opelvic junction obstruction (UPJO) was discovered by
prenatal ultrasound, abdominal pain, urgency, previous
nephrostomy and pyeloplasty were recorded in the
medical history. “Fever” was defined as body tempera-
ture higher than 37.5˚, “mass” indicated an abdominal or
lumbar mass on physical examination, and “hematuria”
meant gross hematuria in our study. Ultrasound was per-
formed in all patients, hydronephrotic grade was deter-
mined in accordance with the criteria established by the
Society of Fetal Urology, [4] and “parenchymal thick-
ness” was the averaged measurements of three different
locations(over upper pole, lowest pole and the most di-
lated calyx).
2.4. Statistical Analysis
Sample-size calculations were carried out using Excel
software 2003 (Microsoft corporation, Washington D.C.,
U.S.A.), and statistical analysis was carried out by SPSS
13.0 statistical software (SPSS, Chicago, IL, USA).
Some clinical features (gender, whether UPJO was dis-
covered by prenatal ultrasound, fever, abdominal pain,
hematuria, urgency, mass, previous nephrostomy, and
pyeloplasty) were viewed as univariates, and another
features (age, grade of hydronephrosis and parenchymal
thickness) were numerical variates. Logistic tests were
performed to assess the relationship between the clinical
features and bacteriuria or funguria A P-value of <0.05
was considered to be statistically significant.
3. Results
Eighteen of 300 patients (6.2%) had urine cultures posi-
tive for bacteria or yeast, including six positive for E.
coli, four for pseudomonas aeruginosa, three for kleb-
siella pneumoniae, one for maltophilia monad, one for
Enterococcus faecium, one for Candida albicans, one for
Candida parapsilosis, and one for yeast not otherwise
specified. Most bacteria were sensitive to cefoperazone,
aztreonam, imipenem, vancomycin, and amikacin, whereas
the yeast was sensitive to nysfungin and fluconazole
(Table 1). No patient with a urine culture positive for E.
Table 1. Antibiotic sensitivity of the positive cases.
Sensitivity▲ E.coli. PA KP MM EF CA CP yeast
cases 6 4 3 1 1 1 1 1
CFUs(*107) 7 ± 2 8 ± 3 7 ± 3 6 10 10 5 10
Ampicillin 8/15(53) 4 2 2
Cefazolin 10/15(67) 5 3 2
Cefuroxime 10/15(67) 4 3 1 1 1
Ceftriaxone 12/15(80) 5 4 1 1 1
Ceftazidime 11/15(73) 5 3 1 1 1
Cefepime 10/14(71) 5 3 1 1
cefoperazone 10/10(100) 5 3 1 1
Aztreonam 9/9(100) 4 3 1 1
Imipenem 12/12(100) 6 3 1 1 1
Vancomycin 13/13(100) 5 3 3 1 1
Gentamicin 13/15(87) 6 3 2 1 1
Amikacin 15/15(100) 6 4 3 1 1
Cetotaxime 12/15(80) 5 3 2 1 1
Tobramycin 12/15(80) 5 3 2 1 1
Nitrofurantoin 10/15(67) 4 3 2 1
Trimethoprim 12/15(80) 5 3 2 1 1
Nysfungin 1/3(33) 1
fluconazole 2/3(67) 1 1
▲: Values are expressed as number/total (percentage). PA: Pseudomonas aeruginosa; KP: klebsiella pneumoniae; MM: maltophilia monad; EF: Enterococcus
faecium; CA: Candida albicans; CP: Candida parapsilosis; CFUs: colony-forming unites.
Copyright © 2011 SciRes. *******

G.-Y. DENG ET AL.
78
coli had a history of nephrostomy or pyeloplasty. In con-
trast, some patients with cultures positive for other orga-
nisms cases had a history of nephrostomy or pyeloplasty
(P < 0.05).
The relationship between clinical features and bacteri-
uria or funguria was also analyzed to assess the risk factors
for developing bacteriuria or funguria. Parameter estimation
was obtained by maximum likelihood using the Logistic test.
bacteriuria or funguria had no signify- cant relationship with
some features (age, gender, whether UPJO was discovered
by prenatal ultrasound, pain, hematuria, mass, grade of hy-
dronephrosis, and parenchymal thickness). Another four
clinical features (fever, urgency, previous nephrostomy
history, and history of pyeloplasty) had significant correla-
tion with bacteriuria or funguria (Table 2). These four fac-
tors were identified as risk factors for developing bacteriuria
or funguria by the omnibus tests of model coefficients using
the forward likelihood ratio test (Table 3).
Table 2. Parameter estimation obtained using maximum likelihood test.※
Variables negative group positive group B Exp (B) P value
age(months) 38 ± 5 24 ± 3 –0.0034 0.9966 0.7936
male 230 16 0.4477 1.5647 0.7667
female 42 2
DPU■ 127 7 1.2564 3.5129 0.3770
fever 8 5 3.9694 52.9558 0.0102
pain 48 1 –2.8222 0.0595 0.2026
hematuria 7 0 –15.3862 0.0000 0.9992
urgency 3 1 3.8935 49.0847 0.0109
mass 32 2 2.1741 8.7935 0.2091
hydronephrosis grade –0.7773 0.4596 0.9167
Ⅰ°◆ 0 0
Ⅱ°◆ 0 0
Ⅲ°◆ 125 9 –0.4748 0.6220 0.9366
Ⅳ°◆ 82 5 –0.2796 0.7560 0.9460
Ⅴ°◆ 65 4 –1.5460 0.2131 0.5072
thickness(mm) 5.2 ± 1.1 4.8 ± 1.2 –0.3948 0.6738 0.6731
Nephrostomy history 4 6 6.43555 623.6281 0.0000
pyeloplasty history 2 5 4.9987 148.2242 0.0004
Constant –5.2693 0.0051 0.0217
※: Overall percentage: 96.2%; ■: Discovered by prenatal ultrasound; ◆: hydronephrosis grade according to the Society of Fetal Urology criteria [4]; B: partial
regression coefficient; Exp (B): Odds ratio.
Table 3. Result of likelihood ratio test.
Wald Chi-Square –2 Log Likelihood Nagelkerke R Square P value
Step 1※ Step 24.301 110.620 0.216 0.0000
Model 24.301
Step2△ Step 24.322 86.298 0.415 0.0000
Model 48.624
Step3◇ Step 19.030 67.268 0.559 0.0000
Model 67.654
Step4□ Step 10.095 57.173 0.632 0.0015
Model 77.749
※: urgency, △: previous nephrostomy; ◇: renal surgical history; □: fever.

79 G.-Y. DENG ET AL.
4. Discussion
UPJO is common in children, for which Anderson–Hynes
pyeloplasty is the classic surgical treatment [5,6] < B1
B2 > < /C< B3 B4 > <, and there have been some reports
concerning about UTI in patients with UPJO [1,2].
Unlike previous studies, in the present study, urine taken
directly from the renal pelvis was cultured. The risk fac-
tors for developing bacteriuria or funguria were also
been determined by Logistic test.
As has been described in previous reports, [8] our
study found that E. coli, pseudomonas aeruginosa, and
klebsiella pneumoniae were the most common organisms
cultured from urine. No patient with a urine culture posi-
tive for E. coli had a history of nephrostomy or pye-
loplasty, unlike patients with urine cultures positive for
other organisms. No previous study has reported this. As
has been reported previously, urinary tract abnormalities
increase the incidence of bacteriuria or funguria [7]. Be-
cause E. coli has better adhesive and migratory capabili-
ties, translocation to the renal pelvis is more likely for this
organism, even in patients with no history of nephrostomy
or pyeloplasty. Our study also found that most bacteria
were sensitive for cefoperazone, aztreonam, imipenem,
vancomycin, and amikacin; however, the conclusion was
less persuasive, because some patients had received pro-
phylactic antibacterial agents.
There has been no study of the relationship between
age or gender and bacteriuria or funguria in patients with
UPJO. Our study found that there was no significant re-
lationship between age or gender and bacteriuria or fun-
guria, perhaps because boys were more likely develop
UPJO, [5,6] whereas girls were much more likely to de-
velop a UTI after 1 year [9]. Increasingly, UPJO is dis-
covered by prenatal ultrasound [6], but being diagnosed
by prenatal ultrasound was not a risk factor to develop
bacteriuria or funguria. In addition, hydronephrosis [5],
abdominal pain, hematuria, and mass were not risk fac-
tors.
As has been discussed in some studies [1], the degree
of hydronephrosis was not a risk factor for developing
bacteriuria or funguria in our study. No previous study
has reported a relationship between parenchymal thick-
ness and bacteriuria or funguria. We found that paren-
chymal thickness was not a risk factor, although, deterio-
ration in renal function was associated with exacerbation
of hydronephrosis and decrease in parenchymal thickness
in some patients [5,6]. Prophylactic antibacterial agents
were probably necessary in those patients who had signs
of urinary tract infection (UTI), or history of nephrostomy
or pyeloplasty.
This retrospective study has some limitations. Firstly,
prophylactic antibacterial agents had been administrated
to those patients who had a nephrostomy catheter or had
signs/symptoms of UTI (fever and urgency) prior to the
operation. Some patients with negative urine cultures had
had signs/symptoms of UTI (Table 2). Thus, some im-
portant information might have been lost [2]. In addition,
one of the main important factors responsible for resis-
tance or sensitivity of uropathogens is antibiotic pre-
treatment and the use of prophylactic antibacterial agents
[10]. In this study, five patients (27%) in the bacteri-
uria/funguria group and eight patients (2.9%) in the cul-
ture-negative group had received antibiotics before the
urine was cultured. The antibiotic sensitivity results were
less reliable when urine culture was not performed prior
to administration of prophylactic antibacterial agents,
This issue is a topic for future.
5. Conclusions
In the present study, we found that four factors: fever,
urgency, and history of nephrostomy or pyeloplasty,
were significantly correlated with bacteriuria or funguria.
Fever and urgency were signs of UTI, and were thus in-
dications for urine culture. History of nephrostomy or
renal surgical history increased the risk of bacteriuria or
funguria contamination. These four factors had high
positive predictive value for positive urine cultures
(overall percentage: 96.2%, Table 2). In conclusion,
Prophylactic antibacterial agents were probably neces-
sary in those patients who had signs of urinary tract in-
fection (UTI), or history of nephrostomy or pyeloplasty.
6. References
[1] C. C. Roth, J. M. Hubanks, B. C. Bright, et al., “Occur-
rence of Urinary Tract Infection in Children with Sig-
nificant Upper Urinary Tract Obstruction,” Urology, Vol.
73, No. 1, 2009, pp. 74-78.
doi:10.1016/j.urology.2008.05.021
[2] S. H. Song, S. B. Lee, Y. S. Park and K. S. Kim, “Is Anti-
biotic Prophylaxis Necessary in Infants with Obstructive
Hydronephrosis?” Journal of Urology, Vol. 177, 2007, pp.
1098-1101.
[3] A. W. Bauer, W. M. Kirby, J. C. Sherris and M. Turck.
“Antibiotic Susceptibility Testing by a Standardized Sin-
gle Disk Method,” American Journal of Clinical Pathol-
ogy, Vol. 45, 1996, pp. 493-496.
[4] C. Delaned, “Antenatal Hydronephrosis: Trends and Ma-
nagement,” Urological Nursing, Vol. 25, No. 3, 2005, pp.
179-183.
[5] C. Mami, A. Paolata, A. Palmara, et al., “Outcome and
Management of Isolated Moderate Renal Pelvis Dilata-
tion Detected at Postnatal Screening,” Pediatric Nephrol-
ogy, Vol. 24, No. 10, 2009, pp. 2005-2008.
[6] B. Chertin, A. Pollack, D. Koulikov, et al., “Conservative
Copyright © 2011 SciRes. *******

G.-Y. DENG ET AL.
80
Treatment of Ureteropelvic Junction Obstruction in Chil-
dren with Antenatal Diagnosis of Hydronephrosis: Less-
ens Learned after 16 Years of Follow-up,” European Uro-
logy, Vol. 49, 2006, pp. 734-739.
[7] J. J. Zorc, D. A. Kiddoo and N. Kathy, “Shaw: Diagnosis
and Management of Pediatric Urinary Tract Infections,”
Clinical Microbiology Reviewa, Vol. 18, No. 2, 2005, pp.
417-422. doi:10.1128/CMR.18.2.417-422.2005
[8] J. H. Lee, H. S. Choi, J. K. Kim, et al., “Nonrefluxing
Neonatal Hydronephrosis and the Risk of Urinary Tract
Infection,” Journal of Urology, Vol. 179, No. 4, 2008, pp.
1524-1528.
[9] A. Hellstrom, E. Hanson, S. Hansson, K. Hjalmas, et al.,
“Associatin between Urinary Symptoms at 7 Years Old
and Previous Urinary Tract Infection,” Archives of Dis-
ease in Childhood, Vol. 66, No. 2, 1991, pp. 323-334.
[10] S. A. Lutter, M. L. Currie, L. B. Mitz, et al., “Antibiotic
Resistance Patterns in Children Hospitalized for Urinary
Tract Infection,” Archives of Pediatrics & Adolescent
Medicine, Vol. 159, 2005, pp. 924-927.