Pediatric Urology

Risk Factors for Breakthrough Infection in Children With Primary

Vesicoureteral Reflux

Koji Shiraishi, Kaoru Yoshino,* Masato Watanabe, Hideyasu Matsuyama

and Saburo Tanikaze
From the Department of Urology, Aichi Childrens Health and Medical Center, Obu (KS, KY, MW, ST) and Department of Urology, Graduate
School of Medicine, Yamaguchi University, Yamaguchi (KS, HM), Japan

Purpose: Despite the widespread application of endoscopic therapy and the Abbreviations
debate surrounding the use of prophylactic antibiotics to treat children with and Acronyms
vesicoureteral reflux, many pediatric urologists still favor medical management. DMSA dimercapto-succinic
Breakthrough infection is one of the absolute indications for surgery. Data to acid
predict breakthrough infection are warranted to manage cases of primary reflux.
UTI urinary tract infection
Materials and Methods: We reviewed medical records of 72 girls and 138 boys
VCUG voiding cystourethrogram
(mean SD age at diagnosis 2.66 3.23 years) with primary vesicoureteral
reflux who were followed with antibiotic prophylaxis at Aichi Childrens Health VUR vesicoureteral reflux
and Medical Center. We examined multiple factors by univariate/multivariate
analysis to elucidate risk factors for breakthrough infection. Submitted for publication July 14, 2009.
Results: Breakthrough infection developed in 59 children (28%). On univariate * Correspondence: Department of Urology, Ai-
chi Childrens Health and Medical Center, 1-2
analysis higher reflux grade (p 0.05) and abnormal renal scan determined by Osakada, Morioka-cho, Obu 474-8710, Japan
99m
technetium dimercapto-succinic acid (p 0.0001) were significantly associated (telephone: 0562-43-0500; FAX: 0562-43-0513).
with breakthrough infection. On multivariate analysis abnormal renal scan was
an independent risk factor for breakthrough infection (OR 11.08, 95% CI
0.76 1.72, p 0.0001).
Conclusions: Abnormal renal scan is an independent risk factor for break-
through infection. Parents and physicians should remain aware that these pa-
tients are at high risk for breakthrough infection, which potentially could lead to
renal damage.

Key Words: antibiotic prophylaxis, urinary tract infections,

vesico-ureteral reflux

VESICOURETERAL reflux has a major antibiotic resistant bacteria. Despite

role in the pathogenesis of urinary
tract infection in children. The main
therapeutic options considered for
primary reflux are administration of
prophylactic antibiotics, endoscopic in-
jection and antireflux surgery.1 The
purpose of prophylactic administration
of low dose antibiotics is to minimize bac-
terial growth to prevent pyelonephritis
and renal scarring. Breakthrough in-
fection represents significant morbid-
ity, and treatment is difficult because of
the side effects of long-term antibiotic
prophylaxis, certain children may ben-
efit from this therapy.2 4 It is impor-
tant to provide an estimate of the like-
lihood of breakthrough infection to
allow effective treatment of children
with reflux.
Information regarding risk factors
for breakthrough infection is limited.
Mingin et al reported that children
with an abnormal renal scan are at
increased risk for breakthrough infec-

0022-5347/10/1834-1527/0 Vol. 183, 1527-1532, April 2010

THE JOURNAL OF UROLOGY Printed in U.S.A.
www.jurology.com 1527
2010 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC. DOI:10.1016/j.juro.2009.12.039
1528 RISK FACTORS FOR BREAKTHROUGH INFECTION IN VESICOURETERAL REFLUX

tion.5 The American Urological Association panel on was used to compare a numerical data set. A p value 0.05
pediatric VUR guidelines revealed that the presence was considered significant. Fishers exact test and unpaired
of renal scarring has a major role in determining t test were done using InStat. Multivariate analysis was
treatment options in children with VUR.1 If a renal performed using JMP, version 4.
scar is present, aggressive management is recom-
mended to prevent further loss or renal parenchy-
mal damage, which can cause hypertension and end- RESULTS
stage renal disease in some patients. Table 1 lists patient demographics for the study.
Breakthrough infection is caused by a number of None of the 25 patients with grade I reflux under-
clinical factors. Using univariate and multivariate went DMSA scintigraphy or received prophylactic
analyses, we tested whether abnormal renal scan antibiotics. Breakthrough infection developed in 59
and other factors are predictive of breakthrough in- children (28%). Among the patients 40 of 138 boys
fection. (29%) and 19 of 72 girls (26%) experienced break-
through infection. There was not a significant differ-
ence between the genders. A total of 118 children
MATERIALS AND METHODS (56%) had abnormal renal scans and 47 (40%)
We retrospectively reviewed the medical records of 343 showed breakthrough infection. Of these 118 pa-
children with primary vesicoureteral reflux presenting at
tients 36 (31%) were categorized as having a small
Aichi Childrens Health and Medical Center between May
kidney. There were no significant differences in age,
2003 and December 2007. Exclusion criteria were inability
to undergo 99mtechnetium DMSA scan at more than 3 presentation, laterality, reflux grade or history of
months after febrile urinary tract infection (66 patients) UTI between the groups with and without a small
and lack of prophylactic antibiotic use (42). Patients with kidney.
apparent voiding dysfunction and neurogenic bladder de- Children with an abnormal renal scan were cat-
termined by urodynamic study were also excluded. A total egorized as having renal scarring only (82 patients),
of 210 children were enrolled in the study. VCUG was a small kidney only (19) or renal scarring with a
used to diagnose reflux. Degree was graded according to
the International Reflux Study classification,1 and the
highest grade during followup was used. DMSA scans Table 1. Patient characteristics
were performed at least 3 months after resolution of a
febrile UTI. Characteristics No. Pts (%)
In this study abnormal renal scan is defined as positive Initial presentation:
by presence of 1 or several parenchymal lesions and/or a Febrile UTI 170 (81)
small kidney on the refluxing side (split uptake less than Screening 30 (14)
40%). Clinical parameters included form of presentation, Prenatal ultrasound 5 (2)
gender, age, VUR grade and laterality, presence of abnor- Enuresis 4 (2)
mal renal scan by DMSA, presence of intrarenal reflux by Hematuria 1 (0.5)
VCUG, type of first line prophylactic antibiotics and his- Gender:
M 138 (66)
tory of breakthrough infection. If febrile UTI developed
F 72 (34)
while the patient was receiving prophylactic antibiotics, it Age:*
was considered a breakthrough infection. Diagnostic cri- Younger than 1 yr 105 (50)
teria of breakthrough infection included acute onset of 1 Yr or Older 105 (50)
high grade fever (38C or greater), pyuria (white blood cell Reflux laterality:
count greater than 100 per high power field in urine sam- Rt 48 (23)
ple sediment), positive urine bacterial culture, blood C- Lt 51 (24)
reactive protein 4 mg/ml or greater and white blood cell Bilat 111 (53)
count 10,000/ml or greater. All boys in the study remained Reflux grade:
uncircumcised. II 23 (11)
III 51 (24)
Urine was bag collected if voiding was not independent.
IV 74 (35)
Several patients presented with a urine white blood cell V 62 (30)
count of less than 100 per high power field but the other Abnormal renal scan 118 (56)
focuses of inflammation were excluded. In patients receiv- Intrarenal reflux 21 (10)
ing prophylactic antibiotics repeat VCUG was performed First line prophylactic antibiotics:
annually until resolution. The primary outcome was oc- First generation cephalosporins 84 (40)
currence of breakthrough infection. The majority of chil- Third generation cephalosporins 61 (29)
dren who experienced breakthrough infection underwent Trimethoprim-sulfamethoxazole 48 (23)
antireflux surgery. Penicillin 11 (5)
Prognostic factors were established by univariate and Fosfomycin 4 (2)
Norfloxacin 2 (1)
multivariate analyses. Logistic regression multivariate anal-
Breakthrough infection 59 (28)
ysis was performed only for the factors with a p value 0.1
on univariate analysis (Fishers exact test). Unpaired t test * Mean SD age 2.66 3.23 years (range 1 month to 14 years) at diagnosis.
 RISK FACTORS FOR BREAKTHROUGH INFECTION IN VESICOURETERAL REFLUX 1529

small kidney (17). Corresponding incidence of break- Table 3. Univariate and multivariate analyses of factors
through infection in these groups was 33 (28%), 7 predicting breakthrough infection by gender
(37%) and 7 (41%), which was not significantly differ- Univariate Multivariate
ent. Overall mean SD interval for breakthrough Analysis Analysis
infection was 4.0 4.1 months after starting pro- p Value p Value
phylactic antibiotics. In many cases first line pro- Males:
phylactic antibiotics were prescribed by the primary Febrile vs afebrile presentation 0.239
physician, and compliance was confirmed every 1 or Age younger than 1 vs 1 yr or older 0.8513
2 months at the office visit. The interval for break- Unilat vs bilat reflux 0.3458
Reflux grade II/III vs IV/V 0.0001* 0.05* (OR 0.203,
through infection for girls (mean SD 6.6 5.6 95% CI 1.670.11)
months) was significantly longer than for boys (3.2 Abnormal vs normal renal scan 0.0001* 0.001* (OR 8.31,
3.2 months, p 0.05). 95% CI 0.521.71)
On univariate analysis patients with breakthrough Present vs absent intrarenal reflux 0.3690
infection were at risk for higher reflux grade (p 0.05) Cefem vs noncefem prophylactic 0.1619
antibiotics
and abnormal renal scan (p 0.0001, table 2). On Females:
multivariate analysis abnormal renal scan was an in- Febrile vs afebrile presentation 0.4913
dependent risk factor for breakthrough infection (OR Age younger than 1 vs 1 yr or older 0.2881
11.08, 95% CI 0.76-1.92, p 0.0001). Abnormal renal Unilat vs bilat reflux 0.5936
scan was more prevalent at higher reflux grades Reflux grade II/III vs IV/V 1.000
Abnormal vs normal renal scan 0.0001* 0.01* (OR 21.46,
(p 0.001). Regarding initial symptoms breakthrough 95% CI 0.653.02)
infection was observed in 56 of 170 patients (33%) Present vs absent intrarenal reflux 0.1836
presenting with febrile UTI and in 7 (18%) with other Cefem vs noncefem prophylactic 1.000
presentations, demonstrating that febrile UTI is a antibiotics
risk factor for breakthrough infection. Multivariate * Difference was statistically significant.
analysis did not reveal any statistical difference.
Breakthrough infection was not associated with age,
gender, reflux laterality, presence of intrarenal re- was significantly less than without breakthrough
flux or choice of prophylactic antibiotics (table 3). infection (3.83 3.44, p 0.01).
Among boys higher reflux grade and presence of
abnormal renal scan were independent risk factors
DISCUSSION
for breakthrough infection. In girls the presence of
abnormal renal scan was the only independent risk Our results support the findings of Mingin et al, who
factor. reported that abnormal renal scan detected by
In patients with abnormal renal scans break- DMSA is a prominent risk factor for breakthrough
through infection was not associated with presenta-
tion, gender, laterality, grade or prophylactic anti- Table 4. Characteristics of patients with abnormal renal scan
biotics (table 4). Although there was no significant
With Without
difference between patients younger than 1 year and Breakthrough Breakthrough
those 1 year old or older, mean SD age at diagno- Characteristics UTI UTI p Value
sis of VUR with breakthrough infection (1.92 3.06)
No. initial presentation: 0.7295
Febrile 48 61
Afebrile 5 4
Table 2. Univariate and multivariate analyses of factors No. age at diagnosis: 0.2670
predicting breakthrough infection Younger than 1 yr 21 33
Univariate Analysis Multivariate Analysis 1 Yr or older 32 32
p Value p Value Mean SD yrs age at diagnosis 1.92 3.06 3.83 3.44 0.0021*
No. gender: 0.8472
Febrile vs afebrile presentation 0.1188 M 35 41
Age younger than 1 vs 1 yr or 0.2192 F 18 24
older No. laterality: 1.0000
Male vs female gender 0.7480 Unilat 22 28
Unilat vs bilat reflux 0.1668 Bilat 31 37
Reflux grade II/III vs IV/V 0.05* 0.2639 No. grade: 0.8528
Abnormal vs normal renal scan 0.0001* 0.0001* (OR 11.08, II/III 11 15
95% CI 0.761.72) IV/V 42 50
Present vs absent intrarenal reflux 0.1278 No. prophylactic antibiotics: 0.3159
Cefem vs noncefem prophylactic 0.2531 Cephalosporin 34 48
antibiotics Other 19 17

* Difference was statistically significant. * Difference was statistically significant.

1530 RISK FACTORS FOR BREAKTHROUGH INFECTION IN VESICOURETERAL REFLUX

infection in patients with VUR.5 The close associa-
tion between abnormal renal scan and high reflux
grade (p 0.001) is consistent with previous reports
that high reflux grade is frequently associated with
renal damage.6,7
The fact that abnormal renal scan rather than
reflux grade is a prominent risk factor for break-
through infection indicates that renal damage itself,
more than reflux grade, is associated with the onset
of breakthrough infection. Mean age of patients with
an abnormal renal scan and breakthrough infection
was significantly less compared to those without
infection (table 4), indicating that some as yet un-
identified host factors may also be involved in sus-
ceptibility to and extent of pyelonephritis. Previ-
ously Kanematsu et al showed that secretor status,
determined using the hemagglutination inhibition
assay, is associated with abnormal renal scan, sug-
gesting an unrecognized host disposition that affects
the clinical course of primary VUR.8 Another mea-
surable host factor predisposing children with VUR
to abnormal renal scan is angiotensin-converting
enzyme polymorphism.9 Our results have demon-
strated that abnormal renal scan is an important
marker for predicting breakthrough infection, and
further investigation into the mechanisms of recur-
rent UTIs and abnormal renal scan is warranted.
Higher grades of reflux are also an independent
risk factor for breakthrough infection in boys but not
in girls (table 3). Potential differences due to gender
are apparent, and include phimosis in boys and dys-
functional elimination in girls.10,11 The presence of
nonretractile prepuces is another independent risk
factor for UTI.12 Herndon et al reported a 53% inci-
dence of breakthrough infection in male infants with
prenatally diagnosed reflux and intact preputial
skin despite use of prophylactic antibiotics, which
was significantly higher than in circumcised in-
fants.10 The fact that all of the boys in our series
were uncircumcised may have influenced reflux
grade and breakthrough infection.
In our study the incidence of intrarenal reflux on
VCUG between boys and girls did not differ signifi-
cantly but was higher in boys (11.6% vs 6.9%). In
contrast, reflux grade was not a risk factor for break-
through infection in girls. Girls with reflux appear to
present with concomitant dysfunctional elimination.13
Although a precise causal relationship between consti-
pation and UTI is lacking, careful evaluation of bowel
and voiding habitus may also be beneficial in prevent-
ing breakthrough infection and renal damage.
The limitations of this study are the bias of back-
ground patients, timing of DMSA scintigraphy, eval-
uations of voiding dysfunction and exact number of
previous UTIs. Because the cases referred to our
center from primary physicians were relatively high
risk (eg breakthrough infection, renal insufficiency),
those with small kidneys and abnormal renal scan
obtained within 6 months after pyelonephritis were
included in the study. As a result, the frequency of
abnormal renal scan (56%) was high. Agras et al
proposed that DMSA scintigraphy be performed
more than 6 months after pyelonephritis is diag-
nosed.14 Permanent cortical defect and transient de-
fect caused by acute pyelonephritis would coexist
with variable extent, and longer followup can de-
crease the incidence of resulting cortical defect. Re-
gardless of the causes of cortical defect, abnormal
renal scan more than 3 months after pyelonephritis
diagnosis is considered a risk factor for break-
through infection.
Voiding dysfunction is an apparent risk factor for
breakthrough infection, which depends on manage-
ment of the lower urinary tract. In this study we
sought to evaluate only patients with primary VUR
rather than all patients with VUR. Evaluation for a
history of UTI is important to interpret the results of
DMSA scintigraphy. However, several referred cases
remained indeterminate with regard to whether fever
was caused by UTI.
Several recent studies have shown no significant
difference in rates of recurrent UTI or abnormal
renal scan between patients receiving and those not
receiving prophylactic antibiotics.2 4 However, suf-
ficient numbers of patients are lacking if patients
are subdivided by reflux grade. Up to now, the con-
tinued use of antibiotics has been deemed prudent.
In this study patients with abnormal renal scan
were at high risk for breakthrough infection, which
is a reasonable option for selecting candidates for
antibiotic prophylaxis or surgery.
In conclusion, although the options to treat and
follow children with abnormal renal scans vary by
individual, parents and physicians should remain
aware that these patients are at high risk for break-
through urinary tract infection, potentially leading
to renal damage.

REFERENCES
1. Elder JS, Peters CA, Arant BS Jr et al: Pediatric
Vesicoureteral Reflux Guidelines Panel summary
report on the management of primary vesi-
coureteral reflux in children. J Urol 1997; 157:
1846.
2. Garin EH, Olavarria F, Nieto VG et al: Clinical
significance of primary vesicoureteral reflux and
urinary antibiotic prophylaxis after acute pyelo-
nephritis: a multicenter, randomized, controlled
study. Pediatrics 2006; 117: 626.
3. Pennesi M, Travan L, Peratoner L et al: Is anti-
biotic prophylaxis in children with vesicoureteral
reflux effective in preventing pyelonephritis and
renal scar? A randomized, controlled trial. Pedi-
atrics 2008; 121: 1489.
 RISK FACTORS FOR BREAKTHROUGH INFECTION IN VESICOURETERAL REFLUX 1531

4. Roussey-Kesler G, Gadjos V, Idres N et al: Anti-
biotic prophylaxis for the prevention of recurrent
urinary tract infection in children with low grade
vesicoureteral reflux: results from a prospective
randomized study. J Urol 2008; 179: 674.
5. Mingin GC, Nguyen HT, Baskin LS et al: Abnor-
mal dimercapto-succinic acid scans predict an
increased risk of breakthrough infection in chil-
dren with vesicoureteral reflux. J Urol 2004; 172:
1075.
6. Caione P, Villa M, Capozza N et al: Predictive risk
factors for chronic renal failure in primary high
grade vesico-ureteric reflux. BJU Int 2004; 93:
1309.
7. Soylu A, Demir BK, Turkmen M et al: Predictors
of renal scar in children with urinary infection
and vesicoureteral reflux. Pediatr Nephrol 2008;
23: 2227.
8. Kanematsu A, Yamamoto S, Yoshino K et al:
Renal scarring is associated with nonsecretion of
blood type antigen in children with primary vesi-
coureteral reflux. J Urol 2005; 174: 1594.
9. Cho SJ and Lee SJ: ACE gene polymorphism and
renal scar in children with acute pyelonephritis.
Pediatr Nephrol 2002; 17: 491.
10. Herndon CD, McKenna PH, Kolon TF et al: A
multicenter outcomes analysis of patients with
neonatal reflux presenting with prenatal hydrone-
phrosis. J Urol 1999; 162: 1203.
11. Koff SA, Wagner TT and Jayanthi VR: The rela-
tionship among dysfunctional elimination syn-
dromes, primary vesicoureteral reflux and urinary
tract infections in children. J Urol 1998; 160:
1019.
12. Shin YH, Lee JW and Lee SJ: The risk factors of
recurrent urinary tract infection in infants with
normal urinary systems. Pediatr Nephrol 2009;
24: 309.
13. Snodgrass W: The impact of treated dysfunc-
tional voiding on the nonsurgical management of
vesicoureteral reflux. J Urol 1998; 160: 1823.
14. Agras K, Ortapamuk H, Naldoken S et al: Reso-
lution of cortical lesions on serial renal scans in
children with acute pyelonephritis. Pediatr Radiol
2007; 37: 153.

EDITORIAL COMMENTS
The usefulness of renal scans in evaluating children
with febrile UTI and vesicoureteral reflux is becom-
ing increasingly apparent. The authors conclude
that an abnormal renal scan in a child with reflux
more than 3 months after a febrile UTI constitutes
an independent risk factor for breakthrough infec-
tion. As they indicate, the limitations of this study
include uncontrolled potential selection biases of
other known risk factors for UTI, such as a history of
UTI or voiding dysfunction. Despite these limita-
tions, their conclusions are reasonable and consis-
tent with previous studies demonstrating a higher
chance of breakthrough UTIs in children with renal
scars (reference 5 in article). Abnormal renal scans
are also a negative predictor of reflux resolution
independent of reflux grade.1,2 Additionally renal
scan data improve the assessment of risk, since
scars are associated with a risk of further scarring,
hypertension and renal insufficiency.
Christopher S. Cooper
Department of Pediatric Urology
University of Iowa
Iowa City, Iowa

REFERENCES
1. Nepple KG, Knudson MJ, Austin JC et al: Abnormal renal scans and decreased early resolution of low grade vesicoureteral reflux. J Urol, part 2, 2008; 180: 1643.

2. Nepple KG, Knudson MJ, Austin JC et al: Adding renal scan data improves the accuracy of a computational model to predict vesicoureteral reflux resolution. J Urol, part
2, 2008; 180: 1648.

Shiraishi et al provide us with more data suggesting
that the presence of renal scars in children with
VUR predicts breakthrough infection. Intuitively we
can all surmise that a child with renal scars already
must be at increased risk for scarring and the infec-
tions that lead to this condition. However, well de-
rived data supporting this notion are lacking, and
like most previous studies examining renal scarring
and VUR, the devil here is in the details. First, the
timing of the initial DMSA scan is not well-defined.
The authors report that DMSA scans were obtained
at least 3 months after resolution of febrile UTI.
Given the recent data showing resolution of a signifi-
cant number of cortical defects 6 months and even 1
year after initial DMSA imaging (reference 14 in arti-
cle), we need precise and standardized reporting. Ab-
normal scans at 3 months might not be significant.
The reported 56% rate of renal scarring is ex-
ceptionally high and is inconsistent with the rates
of 15% to 25% found in most other series. Consid-
ering that this incidence followed the first febrile
UTI, it is even more inconsistent with the litera-
ture. Possible explanations include selection bias,
inaccurate DMSA timing as discussed and/or in-
accurate DMSA interpretation. This final possibil-
ity may be an important one, given the definition
here of an abnormal DMSA scan, which included a
small kidney. I wonder if these kidneys are simply
hypoplastic, and not scarred. The presence of se-
verely scarred kidneys is difficult to reconcile with
the fact that these DMSA scans were obtained
following the first febrile UTI.
This study is predicated on the assumption that
UTIs were breakthrough infections. However, the
1532 RISK FACTORS FOR BREAKTHROUGH INFECTION IN VESICOURETERAL REFLUX
authors did not attempt to examine compliance
with antibiotic prophylaxis. I recognize that com-
pliance is almost impossible to study accurately.
However, the alarmingly poor compliance data re-
ported by Hensle et al indicate that some break-
through infections may be misclassified.1 Overall,
the authors provide an interesting article that
REFERENCE
1. Hensle TW, Hyun G, Grogg AL et al: Part 2: examining pediatric vesicoureteral reflux: a real-world evaluation of treatment patterns and outcomes. Curr Med Res Opin,
suppl, 2007; 23: S7.
In this retrospective study the authors confirm a
previously reported finding, that children with an
abnormal renal scan are at risk for breakthrough
infection (reference 5 in article). Although not novel,
this work is important in narrowing the research
focus, hopefully leading to the design of future pro-
spective studies.
Selection bias probably explains several of the
inconsistencies in this study. For proper interpreta-
tion of renal scarring there must be a well-defined
nomenclature. In this study there is no defined grad-
ing system and thus, for example, no definition of
what constitutes a small kidney. The authors do not
clearly describe how cases were selected, except to
say that the severest cases were referred to them.
The timing of the scans is unclear, and there is ever
increasing literature to support resolution of a renal
defect out to 6 months (reference 14 in article). This
fact may also explain why these patients had more
REPLY BY AUTHORS
There are limitations to our retrospective chart
review and, as with most retrospective reviews, a
prospective standardized cohort study would help
confirm our methods and results. Despite these lim-
itations, our data are consistent with previous re-
ports and, given their strong statistical power, dem-
onstrate that renal scan abnormalities in children
with VUR after a UTI strongly predict a subsequent
REFERENCES
1. Knudson MJ, Austin JC, Wald M et al: Computational models for predicting the chance of resolution in children with vesicoureteral reflux. J Urol 2007; 178: 1824.
2. Shiraishi K, Matsuyama H, Nepple KG et al: Validation of a prognostic calculator for prediction of early vesicoureteral reflux resolution in children. J Urol 2009; 182: 687.
supports our intuition but we still need prospec-
tive data collected in a standardized fashion to be
able to believe ourselves completely.
Carlos R. Estrada
Department of Urology
Harvard University
Boston, Massachusetts
than double the rates of renal scarring seen in the
literature (56% vs 20% to 25%).
Nonetheless, their conclusions are consistent
with the literature. The results differ in that higher
grades of reflux are an independent risk factor for
breakthrough infection on univariate analysis. This
finding may be attributed to a disproportionate
number of males, who present with renal dysplasia
and concomitant higher grades of reflux. Future
studies may confirm DMSA scanning as a reliable
tool for predicting breakthrough infection. However,
the manner in which clinicians use this information
to treat their patients with VUR will continue to
engender controversy.
Gerald Mingin
Section of Pediatric Urology
University of Vermont
Burlington, Vermont
breakthrough UTI. This predictive factor combined
with other characteristics may help determine risk
and impact treatment decisions. By adding this risk
factor to other characteristics previously used in a
validated computational model predicting spontane-
ous reflux resolution,1,2 we may be able to better
predict the risk of a breakthrough UTI on a more
individualized basis.