Sunteți pe pagina 1din 18

Urinary Tract Infection: Clinical Practice Guideline for the Diagnosis and

Management of the Initial UTI in Febrile Infants and Children 2 to 24 Months


SUBCOMMITTEE ON URINARY TRACT INFECTION and STEERING
COMMITTEE ON QUALITY IMPROVEMENT AND MANAGEMENT
Pediatrics; originally published online August 28, 2011;
DOI: 10.1542/peds.2011-1330

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://pediatrics.aappublications.org/content/early/2011/08/24/peds.2011-1330

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly


publication, it has been published continuously since 1948. PEDIATRICS is owned,
published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point
Boulevard, Elk Grove Village, Illinois, 60007. Copyright 2011 by the American Academy
of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from pediatrics.aappublications.org by guest on September 2, 2011


FROM THE AMERICAN ACADEMY OF PEDIATRICS

CLINICAL PRACTICE GUIDELINE

Urinary Tract Infection: Clinical Practice Guideline for


the Diagnosis and Management of the Initial UTI in
Febrile Infants and Children 2 to 24 Months
SUBCOMMITTEE ON URINARY TRACT INFECTION, STEERING
COMMITTEE ON QUALITY IMPROVEMENT AND MANAGEMENT abstract
KEY WORDS
urinary tract infection, infants, children, vesicoureteral reux,
OBJECTIVE: To revise the American Academy of Pediatrics practice
voiding cystourethrography parameter regarding the diagnosis and management of initial urinary
ABBREVIATIONS tract infections (UTIs) in febrile infants and young children.
SPAsuprapubic aspiration METHODS: Analysis of the medical literature published since the last
AAPAmerican Academy of Pediatrics
UTIurinary tract infection
version of the guideline was supplemented by analysis of data provided
RCTrandomized controlled trial by authors of recent publications. The strength of evidence supporting
CFUcolony-forming unit each recommendation and the strength of the recommendation were
VURvesicoureteral reux
WBCwhite blood cell
assessed and graded.
RBUSrenal and bladder ultrasonography RESULTS: Diagnosis is made on the basis of the presence of both
VCUGvoiding cystourethrography
pyuria and at least 50 000 colonies per mL of a single uropathogenic
This document is copyrighted and is property of the American organism in an appropriately collected specimen of urine. After 7 to 14
Academy of Pediatrics and its Board of Directors. All authors
have led conict of interest statements with the American days of antimicrobial treatment, close clinical follow-up monitoring
Academy of Pediatrics. Any conicts have been resolved through should be maintained to permit prompt diagnosis and treatment of
a process approved by the Board of Directors. The American recurrent infections. Ultrasonography of the kidneys and bladder
Academy of Pediatrics has neither solicited nor accepted any
commercial involvement in the development of the content of
should be performed to detect anatomic abnormalities. Data from the
this publication. most recent 6 studies do not support the use of antimicrobial prophy-
The recommendations in this report do not indicate an exclusive laxis to prevent febrile recurrent UTI in infants without vesicoureteral
course of treatment or serve as a standard of medical care. reux (VUR) or with grade I to IV VUR. Therefore, a voiding cystoure-
Variations, taking into account individual circumstances, may be thrography (VCUG) is not recommended routinely after the rst UTI;
appropriate.
VCUG is indicated if renal and bladder ultrasonography reveals hydro-
All clinical practice guidelines from the American Academy of
Pediatrics automatically expire 5 years after publication unless
nephrosis, scarring, or other ndings that would suggest either high-
reafrmed, revised, or retired at or before that time. grade VUR or obstructive uropathy and in other atypical or complex
clinical circumstances. VCUG should also be performed if there is a
recurrence of a febrile UTI. The recommendations in this guideline do
not indicate an exclusive course of treatment or serve as a standard of
care; variations may be appropriate. Recommendations about antimi-
crobial prophylaxis and implications for performance of VCUG are
based on currently available evidence. As with all American Academy of
Pediatrics clinical guidelines, the recommendations will be reviewed
www.pediatrics.org/cgi/doi/10.1542/peds.2011-1330 routinely and incorporate new evidence, such as data from the Ran-
doi:10.1542/peds.2011-1330
domized Intervention for Children With Vesicoureteral Reux (RIVUR)
study.
PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).
Copyright 2011 by the American Academy of Pediatrics CONCLUSIONS: Changes in this revision include criteria for the diag-
COMPANION PAPERS: Companions to this article can be found
nosis of UTI and recommendations for imaging. Pediatrics 2011;128:
on pages 572 and e749, and online at www.pediatrics.org/cgi/ 595610
doi/10.1542/peds.2011-1818 and www.pediatrics.org/cgi/doi/10.
1542/peds.2011-1332.

PEDIATRICS Volume 128, Number 3, September 2011 595


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
INTRODUCTION for use in a variety of clinical settings ness of prophylactic antimicrobial
Since the early 1970s, occult bactere- (eg, ofce, emergency department, or therapy to prevent recurrence of fe-
mia has been the major focus of con- hospital) by clinicians who treat in- brile UTI/pyelonephritis in children
cern for clinicians evaluating febrile fants and young children. This text is a with vesicoureteral reux (VUR). The
infants who have no recognizable summary of the analysis. The data on latter was based on the new and grow-
source of infection. With the introduc- which the recommendations are ing body of evidence questioning the
tion of effective conjugate vaccines based are included in a companion effectiveness of antimicrobial prophy-
against Haemophilus inuenzae type technical report.4 laxis to prevent recurrent febrile UTI in
b and Streptococcus pneumoniae Like the 1999 practice parameter, this children with VUR. To explore this par-
(which have resulted in dramatic de- revision focuses on the diagnosis and ticular issue, the literature search was
creases in bacteremia and meningi- management of initial urinary tract in- expanded to include trials published
tis), there has been increasing appre- fections (UTIs) in febrile infants and since 1993 in which antimicrobial pro-
ciation of the urinary tract as the most young children (224 months of age) phylaxis was compared with no treat-
frequent site of occult and serious bac- who have no obvious neurologic or an- ment or placebo treatment for chil-
terial infections. Because the clinical atomic abnormalities known to be as- dren with VUR. Because all except 1 of
presentation tends to be nonspecic in sociated with recurrent UTI or renal the recent randomized controlled tri-
infants and reliable urine specimens damage. (For simplicity, in the remain- als (RCTs) of the effectiveness of pro-
for culture cannot be obtained without der of this guideline the phrase fe- phylaxis included children more than
invasive methods (urethral cathe- brile infants is used to indicate febrile 24 months of age and some did not
terization or suprapubic aspiration infants and young children 224 provide specic data according to
[SPA]), diagnosis and treatment may months of age.) The lower and upper grade of VUR, the authors of the 6 RCTs
be delayed. Most experimental and age limits were selected because stud- were contacted; all provided raw data
clinical data support the concept that ies on infants with unexplained fever from their studies specically ad-
delays in the institution of appropriate generally have used these age limits dressing infants 2 to 24 months of age,
treatment of pyelonephritis increase and have documented that the preva- according to grade of VUR. Meta-
the risk of renal damage.1,2 lence of UTI is high (5%) in this age analysis of these data was performed.
This clinical practice guideline is a re- group. In those studies, fever was de- Results from the literature searches
vision of the practice parameter pub- ned as temperature of at least 38.0C and meta-analyses were provided to
lished by the American Academy of (100.4F); accordingly, this denition committee members. Issues were
Pediatrics (AAP) in 1999.3 It was devel- of fever is used in this guideline. Ne- raised and discussed until consensus
oped by a subcommittee of the Steer- onates and infants less than 2 was reached regarding recommenda-
ing Committee on Quality Improvement months of age are excluded, because tions. The quality of evidence support-
and Management that included physi- there are special considerations in ing each recommendation and the
cians with expertise in the elds of ac- this age group that may limit the ap- strength of the recommendation were
ademic general pediatrics, epidemiol- plication of evidence derived from assessed by the committee member
ogy and informatics, pediatric the studies of 2- to 24-month-old chil- most experienced in informatics and
infectious diseases, pediatric nephrol- dren. Data are insufcient to deter- epidemiology and were graded ac-
ogy, pediatric practice, pediatric radi- mine whether the evidence gener- cording to AAP policy5 (Fig 1).
ology, and pediatric urology. The AAP ated from studies of infants 2 to 24 The subcommittee formulated 7 rec-
funded the development of this guide- months of age applies to children ommendations, which are presented
line; none of the participants had any more than 24 months of age. in the text in the order in which a clini-
nancial conicts of interest. The cian would use them when evaluating
guideline was reviewed by multiple METHODS and treating a febrile infant, as well as
groups within the AAP (7 committees, 1 To provide evidence for the guideline, 2 in algorithm form in the Appendix. This
council, and 9 sections) and 5 external literature searches were conducted, clinical practice guideline is not in-
organizations in the United States and that is, a surveillance of Medline-listed tended to be a sole source of guidance
Canada. The guideline will be reviewed literature over the past 10 years for for the treatment of febrile infants with
and/or revised in 5 years, unless new signicant changes since the guideline UTIs. Rather, it is intended to assist clini-
evidence emerges that warrants revi- was published and a systematic re- cians in decision-making. It is not in-
sion sooner. The guideline is intended view of the literature on the effective- tended to replace clinical judgment or to

596 FROM THE AMERICAN ACADEMY OF PEDIATRICS


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
FROM THE AMERICAN ACADEMY OF PEDIATRICS

tainer, because they may be contami-


nated by bacteria in the distal urethra.
Cultures of urine specimens collected
in a bag applied to the perineum have
an unacceptably high false-positive
rate and are valid only when they yield
negative results.6,1416 With a preva-
lence of UTI of 5% and a high rate of
false-positive results (specicity:
63%), a positive culture result for
FIGURE 1 urine collected in a bag would be a
AAP evidence strengths. false-positive result 88% of the time.
For febrile boys, with a prevalence of
UTI of 2%, the rate of false-positive re-
establish an exclusive protocol for the administered, because the antimicro- sults is 95%; for circumcised boys,
care of all children with this condition. bial agents commonly prescribed in with a prevalence of UTI of 0.2%, the
such situations would almost certainly rate of false-positive results is 99%.
DIAGNOSIS obscure the diagnosis of UTI.
Therefore, in cases in which antimicro-
Action Statement 1 SPA has been considered the standard bial therapy will be initiated, catheter-
If a clinician decides that a febrile method for obtaining urine that is un- ization or SPA is required to establish
infant with no apparent source for contaminated by perineal ora. Vari- the diagnosis of UTI.
the fever requires antimicrobial able success rates for obtaining urine
Aggregate quality of evidence: A (diag-
therapy to be administered be- have been reported (23%90%).68
nostic studies on relevant populations).
cause of ill appearance or another When ultrasonographic guidance is
pressing reason, the clinician used, success rates improve.9,10 The Benets: A missed diagnosis of UTI
should ensure that a urine speci- technique has limited risks, but tech- can lead to renal scarring if left un-
men is obtained for both culture nical expertise and experience are treated; overdiagnosis of UTI can
and urinalysis before an antimicro- required, and many parents and phy- lead to overtreatment and unneces-
bial agent is administered; the sicians perceive the procedure as sary and expensive imaging. Once an-
specimen needs to be obtained unacceptably invasive, compared timicrobial therapy is initiated, the op-
through catheterization or SPA, be- with catheterization. However, there portunity to make a denitive
cause the diagnosis of UTI cannot may be no acceptable alternative to diagnosis is lost; multiple studies of
be established reliably through cul- SPA for boys with moderate or se- antimicrobial therapy have shown
ture of urine collected in a bag vere phimosis or girls with tight la- that the urine may be rapidly
(evidence quality: A; strong bial adhesions. sterilized.
recommendation). Urine obtained through catheteriza- Harms/risks/costs: Catheterization
When evaluating febrile infants, clini- tion for culture has a sensitivity of 95% is invasive.
cians make a subjective assessment of and a specicity of 99%, compared Benet-harms assessment: Prepon-
the degree of illness or toxicity, in ad- with that obtained through SPA.7,11,12 derance of benet over harm.
dition to seeking an explanation for the The techniques required for catheter-
Value judgments: Once antimicro-
fever. This clinical assessment deter- ization and SPA are well described.13
bial therapy has begun, the opportu-
mines whether antimicrobial therapy When catheterization or SPA is being
attempted, the clinician should have a nity to make a denitive diagnosis is
should be initiated promptly and af-
sterile container ready to collect a lost. Therefore, it is important to
fects the diagnostic process regarding
UTI. If the clinician determines that the urine specimen, because the prepara- have the most-accurate test for UTI
degree of illness warrants immediate tion for the procedure may stimulate performed initially.
antimicrobial therapy, then a urine the child to void. Whether the urine is Role of patient preferences: There is
specimen suitable for culture should obtained through catheterization or is no evidence regarding patient pref-
be obtained through catheterization or voided, the rst few drops should be erences for bag versus catheterized
SPA before antimicrobial agents are allowed to fall outside the sterile con- urine. However, bladder tap has

PEDIATRICS Volume 128, Number 3, September 2011 597


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
been shown to be more painful than
urethral catheterization.
Exclusions: None.
Intentional vagueness: The basis of
the determination that antimicro-
bial therapy is needed urgently is
not specied, because variability in
clinical judgment is expected; con-
siderations for individual patients,
such as availability of follow-up
care, may enter into the decision, FIGURE 2
and the literature provides only gen- Probability of UTI Among Febrile Infant Girls28 and Infant Boys30 According to Number of Findings
Present. aProbability of UTI exceeds 1% even with no risk factors other than being uncircumcised.
eral guidance.
Policy level: Strong recommendation.
be obtained through catheteriza- than 1% for 10.4% of academicians and
Action Statement 2
tion or SPA and cultured; if urinaly- 11.7% for practitioners26; when the
If a clinician assesses a febrile infant sis of fresh (<1 hour since void) threshold was increased to 1% to 3%,
with no apparent source for the fever urine yields negative leukocyte es- 67.5% of academicians and 45.7% of
as not being so ill as to require imme- terase and nitrite test results, then practitioners considered the yield suf-
diate antimicrobial therapy, then the it is reasonable to monitor the clin- ciently high to warrant urine culture.
clinician should assess the likelihood of ical course without initiating anti- Therefore, attempting to operational-
UTI (see below for how to assess microbial therapy, recognizing that ize low likelihood (ie, below a thresh-
likelihood). negative urinalysis results do not old that warrants a urine culture) does
rule out a UTI with certainty. not produce an absolute percentage;
Action Statement 2a If the clinician determines that the de- clinicians will choose a threshold de-
If the clinician determines the febrile gree of illness does not require imme- pending on factors such as their con-
infant to have a low likelihood of UTI diate antimicrobial therapy, then the dence that contact will be maintained
(see text), then clinical follow-up likelihood of UTI should be assessed. through the illness (so that a specimen
monitoring without testing is suf- As noted previously, the overall preva- can be obtained at a later time) and com-
cient (evidence quality: A; strong lence of UTI in febrile infants who have fort with diagnostic uncertainty. Fig 2 in-
recommendation). no source for their fever evident on the dicates the number of risk factors as-
basis of history or physical examina- sociated with threshold probabilities
Action Statement 2b tion results is approximately 5%,17,18 of UTI of at least 1% and at least 2%.
If the clinician determines that the but it is possible to identify groups In a series of studies, Gorelick, Shaw,
febrile infant is not in a low-risk with higher-than-average likelihood and colleagues2729 derived and vali-
group (see below), then there are 2 and some with lower-than-average dated a prediction rule for febrile in-
choices (evidence quality: A; strong likelihood. The prevalence of UTI fant girls on the basis of 5 risk factors,
recommendation). Option 1 is to ob- among febrile infant girls is more than namely, white race, age less than 12
tain a urine specimen through cath- twice that among febrile infant boys months, temperature of at least 39C,
eterization or SPA for culture and (relative risk: 2.27). The rate for uncir- fever for at least 2 days, and absence
urinalysis. Option 2 is to obtain a cumcised boys is 4 to 20 times higher of another source of infection. This
urine specimen through the most than that for circumcised boys, whose prediction rule, with sensitivity of
convenient means and to perform a rate of UTI is only 0.2% to 0.4%.1924 The 88% and specicity of 30%, permits
urinalysis. If the urinalysis results presence of another, clinically obvious some infant girls to be considered in
suggest a UTI (positive leukocyte source of infection reduces the likeli- a low-likelihood group (Fig 2). For ex-
esterase test results or nitrite test hood of UTI by one-half.25 ample, of girls with no identiable
or microscopic analysis results In a survey asking, What yield is re- source of infection, those who are non-
positive for leukocytes or bacte- quired to warrant urine culture in fe- white and more than 12 months of age
ria), then a urine specimen should brile infants?, the threshold was less with a recent onset (2 days) of low-

598 FROM THE AMERICAN ACADEMY OF PEDIATRICS


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
FROM THE AMERICAN ACADEMY OF PEDIATRICS

grade fever (39C) have less than a TABLE 1 Sensitivity and Specicity of Components of Urinalysis, Alone and in Combination
1% probability of UTI; each additional Test Sensitivity (Range), % Specicity (Range), %
risk factor increases the probability. It Leukocyte esterase test 83 (6794) 78 (6492)
Nitrite test 53 (1582) 98 (90100)
should be noted, however, that some Leukocyte esterase or 93 (90100) 72 (5891)
of the factors (eg, duration of fever) nitrite test positive
may change during the course of the Microscopy, WBCs 73 (32100) 81 (4598)
Microscopy, bacteria 81 (1699) 83 (11100)
illness, excluding the infant from a Leukocyte esterase test, 99.8 (99100) 70 (6092)
low-likelihood designation and nitrite test, or
prompting testing as described in microscopy positive
action statement 2a.
As demonstrated in Fig 2, the major denitive urine specimen through Action Statement 3
risk factor for febrile infant boys is catheterization initially.
whether they are circumcised. The prob- To establish the diagnosis of UTI,
Aggregate quality of evidence: A (diag- clinicians should require both uri-
ability of UTI can be estimated on the ba-
nostic studies on relevant populations). nalysis results that suggest infec-
sis of 4 risk factors, namely, nonblack
race, temperature of at least 39C, fever Benets: Accurate diagnosis of UTI tion (pyuria and/or bacteriuria)
can prevent the spread of infection and the presence of at least 50 000
for more than 24 hours, and absence of
and renal scarring; avoiding overdi- colony-forming units (CFUs) per mL
another source of infection.4,30
agnosis of UTI can prevent over- of a uropathogen cultured from a
If the clinician determines that the in- treatment and unnecessary and ex- urine specimen obtained through
fant does not require immediate anti- pensive imaging. catheterization or SPA (evidence
microbial therapy and a urine speci- quality: C; recommendation).
Harms/risks/costs: A small propor-
men is desired, then often a urine
tion of febrile infants, considered at
collection bag afxed to the perineum Urinalysis
low likelihood of UTI, will not receive
is used. Many clinicians think that this
timely identication and treatment General Considerations
collection technique has a low contam- of their UTIs.
ination rate under the following cir- Urinalysis cannot substitute for urine
Benet-harms assessment: Prepon- culture to document the presence of
cumstances: the patients perineum is
derance of benet over harm. UTI but needs to be used in conjunction
properly cleansed and rinsed before
application of the collection bag, the Value judgments: There is a risk of with culture. Because urine culture re-
UTI sufciently low to forestall fur- sults are not available for at least 24
urine bag is removed promptly after
ther evaluation. hours, there is considerable interest
urine is voided into the bag, and the
Role of patient preferences: The
in tests that may predict the results of
specimen is refrigerated or processed
choice of option 1 or option 2 and the urine culture and enable presump-
immediately. Even if contamination
the threshold risk of UTI warranting tive therapy to be initiated at the rst
from the perineal skin is minimized,
obtaining a urine specimen may be encounter. Urinalysis can be per-
however, there may be signicant con- formed on any specimen, including
tamination from the vagina in girls or inuenced by parents preference
to avoid urethral catheterization (if one collected from a bag applied to the
the prepuce in uncircumcised boys, perineum. However, the specimen
the 2 groups at highest risk of UTI. A a bag urine sample yields negative
urinalysis results) versus timely must be fresh (1 hour after voiding
positive culture result from a speci- with maintenance at room tempera-
evaluation (obtaining a denitive
men collected in a bag cannot be used ture or 4 hours after voiding with re-
specimen through catheterization).
to document a UTI; conrmation re- frigeration), to ensure sensitivity and
quires culture of a specimen collected Exclusions: Because it depends on a
specicity of the urinalysis. The tests
through catheterization or SPA. Be- range of patient- and physician-
that have received the most atten-
cause there may be substantial delay specic considerations, the precise
tion are biochemical analyses of leu-
waiting for the infant to void and a sec- threshold risk of UTI warranting ob-
kocyte esterase and nitrite through a
ond specimen, obtained through cath- taining a urine specimen is left to
rapid dipstick method and urine
eterization, may be necessary if the the clinician but is below 3%.
microscopic examination for white
urinalysis suggests the possibility of Intentional vagueness: None. blood cells (WBCs) and bacteria
UTI, many clinicians prefer to obtain a Policy level: Strong recommendation. (Table 1).

PEDIATRICS Volume 128, Number 3, September 2011 599


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
Urine dipsticks are appealing, because 64%92%]) generally is not as good as during infancy. In a study of infants 2 to
they provide rapid results, do not re- the sensitivity, which reects the non- 24 months of age, 0.7% of afebrile girls
quire microscopy, and are eligible for specicity of pyuria in general. Accord- had 3 successive urine cultures with
a waiver under the Clinical Laboratory ingly, positive leukocyte esterase test 105 CFUs per mL of a single uropatho-
Improvement Amendments. They indi- results should be interpreted with cau- gen.26 Asymptomatic bacteriuria can
cate the presence of leukocyte es- tion, because false-positive results are be easily confused with true UTI in a
terase (as a surrogate marker for common. With numerous conditions febrile infant but needs to be distin-
pyuria) and urinary nitrite (which is other than UTI, including fever result- guished, because studies suggest that
converted from dietary nitrates in the ing from other conditions (eg, strepto- antimicrobial treatment may do more
presence of most Gram-negative enteric coccal infections or Kawasaki dis- harm than good.36 The key to distin-
bacteria in the urine). The conversion of ease), and after vigorous exercise, guishing true UTI from asymptomatic
dietary nitrates to nitrites by bacteria re- WBCs may be found in the urine. There- bacteriuria is the presence of pyuria.
quires approximately 4 hours in the fore, a nding of pyuria by no means
bladder.31 The performance characteris- Microscopic Analysis for Bacteriuria
conrms that an infection of the uri-
tics of both leukocyte esterase and ni- nary tract is present. The presence of bacteria in a fresh,
trite tests vary according to the deni- Gram-stained specimen of uncentri-
The absence of pyuria in children with
tion used for positive urine culture fuged urine correlates with 105 CFUs
true UTIs is rare, however. It is theoret-
results, the age and symptoms of the per mL in culture.37 An enhanced uri-
ically possible if a febrile child is as-
population being studied, and the nalysis, combining the counting
sessed before the inammatory re-
method of urine collection. chamber assessment of pyuria noted
sponse has developed, but the
previously with Gram staining of drops
Nitrite Test inammatory response to a UTI pro-
of uncentrifuged urine, with a thresh-
duces both fever and pyuria; therefore,
A nitrite test is not a sensitive marker old of at least 1 Gram-negative rod in
for children, particularly infants, who children who are being evaluated be-
10 oil immersion elds, has greater sen-
empty their bladders frequently. cause of fever should already have
sitivity, specicity, and positive predic-
Therefore, negative nitrite test results WBCs in their urine. More likely expla-
tive value than does the standard urinal-
have little value in ruling out UTI. More- nations for signicant bacteriuria in
ysis33 and is the preferred method of
over, not all urinary pathogens reduce culture in the absence of pyuria in-
urinalysis when appropriate equipment
nitrate to nitrite. The test is helpful clude contaminated specimens, insen-
and personnel are available.
when the result is positive, however, sitive criteria for pyuria, and asymp-
because it is highly specic (ie, there tomatic bacteriuria. In most cases, Automated Urinalysis
are few false-positive results).32 when true UTI has been reported to oc- Automated methods to perform uri-
cur in the absence of pyuria, the de- nalysis are now being used in many
Leukocyte Esterase Test nition of pyuria has been at fault. The hospitals and laboratories. Image-
The sensitivity of the leukocyte es- standard method of assessing pyuria based systems use ow imaging
terase test is 94% when it used in the has been centrifugation of the urine analysis technology and software to
context of clinically suspected UTI. and microscopic analysis, with a classify particles in uncentrifuged
Overall, the reported sensitivity in var- threshold of 5 WBCs per high-power urine specimens rapidly.38 Results
ious studies is lower (83%), because eld (25 WBCs per L). If a counting correlate well with manual methods,
the results of leukocyte esterase tests chamber is used, however, the nding especially for red blood cells, WBCs,
were related to culture results without of at least 10 WBCs per L in uncentri- and squamous epithelial cells. In the
exclusion of individuals with asymp- fuged urine has been demonstrated to future, this may be the most common
tomatic bacteriuria. The absence of be more sensitive33 and performs well method by which urinalysis is per-
leukocyte esterase in the urine of indi- in clinical situations in which the stan- formed in laboratories.
viduals with asymptomatic bacteriuria dard method does not, such as with
is an advantage of the test, rather than very young infants.34 Culture
a limitation, because it distinguishes An important cause of bacteriuria in The diagnosis of UTI is made on the ba-
individuals with asymptomatic bacte- the absence of pyuria is asymptomatic sis of quantitative urine culture re-
riuria from those with true UTI. bacteriuria. Asymptomatic bacteriuria sults in addition to evidence of pyuria
The specicity of the leukocyte es- often is associated with school-aged and/or bacteriuria. Urine specimens
terase test (average: 72% [range: and older girls,35 but it can be present should be processed as expediently as

600 FROM THE AMERICAN ACADEMY OF PEDIATRICS


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
FROM THE AMERICAN ACADEMY OF PEDIATRICS

possible. If the specimen is not pro- spp are not considered clinically rele- Harms/risks/costs: Stringent diag-
cessed promptly, then it should be re- vant urine isolates for otherwise nostic criteria may miss a small
frigerated to prevent the growth of or- healthy, 2- to 24-month-old children.) number of UTIs.
ganisms that can occur in urine at Reducing the threshold from 100 000 Benet-harms assessment: Prepon-
room temperature; for the same rea- CFUs per mL to 50 000 CFUs per mL derance of benet over harm.
son, specimens that require transpor- would seem to increase the sensitivity
Value judgments: Treatment of
tation to another site for processing of culture at the expense of decreased
asymptomatic bacteriuria may be
should be transported on ice. A prop- specicity; however, because the pro-
harmful.
erly collected urine specimen should posed criteria for UTI now include evi-
dence of pyuria in addition to positive Role of patient preferences: We as-
be inoculated on culture medium that
culture results, infants with positive sume that parents prefer no action
will allow identication of urinary tract
culture results alone will be recog- in the absence of a UTI (avoiding
pathogens.
nized as having asymptomatic bacteri- false-positive results) over a very
Urine culture results are considered small chance of missing a UTI.
uria rather than a true UTI. Some labo-
positive or negative on the basis of the
ratories report growth only in the Exclusions: None.
number of CFUs that grow on the cul-
ture medium.36 Denition of signicant
following categories: 0 to 1000, 1000 to Intentional vagueness: None.
10 000, 10 000 to 100 000, and more Policy level: Recommendation.
colony counts with regard to the
than 100 000 CFUs per mL. In such
method of collection considers that
cases, results in the 10 000 to 100 000 MANAGEMENT
the distal urethra and periurethral
CFUs per mL range need to be evalu-
area are commonly colonized by the Action Statement 4
ated in context, such as whether the
same bacteria that may cause UTI;
urinalysis ndings support the diagno- Action Statement 4a
therefore, a low colony count may be
sis of UTI and whether the organism is
present in a specimen obtained When initiating treatment, the clini-
a recognized uropathogen.
through voiding or catheterization cian should base the choice of
when bacteria are not present in blad- Alternative culture methods, such as route of administration on practi-
der urine. Denitions of positive and dipslides, may have a place in the of- cal considerations. Initiating treat-
negative culture results are opera- ce setting; sensitivity is reported to ment orally or parenterally is
tional and not absolute. The time the be in the range of 87% to 100%, and equally efcacious. The clinician
urine resides in the bladder (bladder specicity is reported to be 92% to should base the choice of agent on
incubation time) is an important deter- 98%, but dipslides cannot specify the local antimicrobial sensitivity pat-
minant of the magnitude of the colony organism or antimicrobial sensitivi- terns (if available) and should ad-
ties.41 Practices that use dipslides just the choice according to sensi-
count. The concept that more than
should do so in collaboration with a tivity testing of the isolated
100 000 CFUs per mL indicates a UTI
certied laboratory for identication uropathogen (evidence quality: A;
was based on morning collections of
and sensitivity testing or, in the ab- strong recommendation).
urine from adult women, with compar-
sence of such results, may need to per-
ison of specimens from women with-
form test of cure cultures after 24 Action Statement 4b
out symptoms and women considered
hours of treatment. The clinician should choose 7 to 14
clinically to have pyelonephritis; the
transition range, in which the propor- Aggregate quality of evidence: C (ob- days as the duration of antimicrobial
tion of women with pyelonephritis ex- servational studies). therapy (evidence quality: B;
ceeded the proportion of women with- Benets: Accurate diagnosis of UTI recommendation).
out symptoms, was 10 000 to 100 000 can prevent the spread of infection The goals of treatment of acute UTI are
CFUs per mL.39 In most instances, an and renal scarring; avoiding overdi- to eliminate the acute infection, to pre-
appropriate threshold to consider agnosis of UTI can prevent over- vent complications, and to reduce the
bacteriuria signicant in infants and treatment and unnecessary and ex- likelihood of renal damage. Most chil-
children is the presence of at least pensive imaging. These criteria dren can be treated orally.4244 Patients
50 000 CFUs per mL of a single urinary reduce the likelihood of overdiagno- whom clinicians judge to be toxic or
pathogen.40 (Organisms such as sis of UTI in infants with asymptom- who are unable to retain oral intake
Lactobacillus spp, coagulase-negative atic bacteriuria or contaminated (including medications) should re-
staphylococci, and Corynebacterium specimens. ceive an antimicrobial agent parenter-

PEDIATRICS Volume 128, Number 3, September 2011 601


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
TABLE 2 Some Empiric Antimicrobial Agents TABLE 3 Some Empiric Antimicrobial Agents for Oral Treatment of UTI
for Parenteral Treatment of UTI Antimicrobial Agent Dosage
Antimicrobial Dosage Amoxicillin-clavulanate 2040 mg/kg per d in 3 doses
Agent Sulfonamide
Ceftriaxone 75 mg/kg, every 24 h Trimethoprim-sulfamethoxazole 612 mg/kg trimethoprim and 30-60 mg/kg sulfamethoxazole
Cefotaxime 150 mg/kg per d, per d in 2 doses
divided every 68 h Sulsoxazole 120150 mg/kg per d in 4 doses
Ceftazidime 100150 mg/kg per d, Cephalosporin
divided every 8 h Cexime 8 mg/kg per d in 1 dose
Gentamicin 7.5 mg/kg per d, Cefpodoxime 10 mg/kg per d in 2 doses
divided every 8 h Cefprozil 30 mg/kg per d in 2 doses
Tobramycin 5 mg/kg per d, Cefuroxime axetil 2030 mg/kg per d in 2 doses
divided every 8 h Cephalexin 50100 mg/kg per d in 4 doses
Piperacillin 300 mg/kg per d,
divided every 68 h

the total course of therapy should be 7 distinguishes the benet of treating


ally (Table 2) until they exhibit clinical to 14 days. The committee attempted to 7 vs 10 vs 14 days, and the range is
improvement, generally within 24 to 48 identify a single, preferred, evidence- allowable.
hours, and are able to retain orally ad- based duration, rather than a range, but Policy level: Strong recommendation/
ministered uids and medications. In a data comparing 7, 10, and 14 days di-
study of 309 febrile infants with UTIs, recommendation.
rectly were not found. There is evidence
only 3 (1%) were deemed too ill to be that 1- to 3-day courses for febrile UTIs Action Statement 5
assigned randomly to either paren- are inferior to courses in the recom-
teral or oral treatment.42 Parenteral Febrile infants with UTIs should
mended range; therefore, the minimal
administration of an antimicrobial undergo renal and bladder ultra-
duration selected should be 7 days.
agent also should be considered when sonography (RBUS) (evidence
Aggregate quality of evidence: A/B quality: C; recommendation).
compliance with obtaining an antimi-
crobial agent and/or administering it (RCTs).
The purpose of RBUS is to detect ana-
orally is uncertain. The usual choices Benets: Adequate treatment of UTI tomic abnormalities that require fur-
for oral treatment of UTIs include can prevent the spread of infection ther evaluation, such as additional im-
a cephalosporin, amoxicillin plus and renal scarring. Outcomes of aging or urologic consultation. RBUS
clavulanic acid, or trimethoprim- short courses (13 d) are inferior to also provides an evaluation of the re-
sulfamethoxazole (Table 3). It is essen- those of 7- to 14-d courses. nal parenchyma and an assessment of
tial to know local patterns of suscepti- Harms/risks/costs: There are mini- renal size that can be used to monitor
bility of coliforms to antimicrobial mal harm and minor cost effects of renal growth. The yield of actionable
agents, particularly trimethoprim- antimicrobial choice and duration ndings is relatively low.45,46 Wide-
sulfamethoxazole and cephalexin, be- of therapy. spread application of prenatal ultra-
cause there is substantial geographic Benet-harms assessment: Prepon- sonography clearly has reduced the
variability that needs to be taken into derance of benet over harm. prevalence of previously unsuspected
account during selection of an antimi- obstructive uropathy in infants, but the
Value judgments: Adjusting antimi-
crobial agent before sensitivity results consequences of prenatal screening
crobial choice on the basis of avail-
are available. Agents that are excreted with respect to the risk of renal abnor-
able data and treating according to
in the urine but do not achieve thera- malities in infants with UTIs have not
best evidence will minimize cost and
peutic concentrations in the blood- yet been well dened. There is consid-
consequences of failed or unneces-
stream, such as nitrofurantoin, should erable variability in the timing and
sary treatment.
not be used to treat febrile infants with quality of prenatal ultrasonograms,
UTIs, because parenchymal and serum Role of patient preferences: It is as-
and the report of normal ultrasono-
antimicrobial concentrations may be sumed that parents prefer the
graphic results cannot necessarily be
insufcient to treat pyelonephritis or most-effective treatment and the
relied on to dismiss completely the
urosepsis. least amount of medication that en-
possibility of a structural abnormality
sures effective treatment.
Whether the initial route of administra- unless the study was a detailed ana-
tion of the antimicrobial agent is oral Exclusions: None. tomic survey (with measurements),
or parenteral (then changed to oral), Intentional vagueness: No evidence was performed during the third tri-

602 FROM THE AMERICAN ACADEMY OF PEDIATRICS


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
FROM THE AMERICAN ACADEMY OF PEDIATRICS

mester, and was performed and inter- children with reduced renal function. Action Statement 6
preted by qualied individuals.47 The radiation dose from dimercapto-
Action Statement 6a
The timing of RBUS depends on the succinic acid is additive with that of
VCUG should not be performed
clinical situation. RBUS is recom- VCUG when both studies are per-
routinely after the rst febrile
mended during the rst 2 days of treat- formed.50 The radiation dose from
UTI; VCUG is indicated if RBUS re-
ment to identify serious complications, VCUG depends on the equipment that
veals hydronephrosis, scarring,
such as renal or perirenal abscesses is used (conventional versus pulsed
or other ndings that would sug-
or pyonephrosis associated with ob- digital uoroscopy) and is related di-
gest either high-grade VUR or ob-
structive uropathy when the clinical ill- rectly to the total uoroscopy time. structive uropathy, as well as in
ness is unusually severe or substantial Moreover, the total exposure for the other atypical or complex clinical
clinical improvement is not occurring. child will be increased when both circumstances (evidence quality
For febrile infants with UTIs who dem- acute and follow-up studies are ob- B; recommendation).
onstrate substantial clinical improve- tained. The lack of exposure to radi-
ment, however, imaging does not need ation is a major advantage of RBUS, Action Statement 6b
to occur early during the acute infec- even with recognition of the limita- Further evaluation should be con-
tion and can even be misleading; ani- tions of this modality that were de- ducted if there is a recurrence of fe-
mal studies demonstrate that Esche- scribed previously. brile UTI (evidence quality: X;
richia coli endotoxin can produce recommendation).
Aggregate quality of evidence: C (ob-
dilation during acute infection, which
servational studies). For the past 4 decades, the strategy to
could be confused with hydronephro-
Benets: RBUS in this population
protect the kidneys from further dam-
sis, pyonephrosis, or obstruction.48
age after an initial UTI has been to de-
Changes in the size and shape of the will yield abnormal results in 15%
tect childhood genitourinary abnor-
kidneys and the echogenicity of renal of cases, and 1% to 2% will have ab-
malities in which recurrent UTI could
parenchyma attributable to edema normalities that would lead to ac-
increase renal damage. The most com-
also are common during acute infec- tion (eg, additional evaluation, re-
mon of these is VUR, and VCUG is used
tion. The presence of these abnormal- ferral, or surgery).
to detect this. Management included
ities makes it inappropriate to con- Harms/risks/costs: Between 2% continuous antimicrobial administra-
sider RBUS performed early during and 3% will be false-positive results, tion as prophylaxis and surgical inter-
acute infection to be a true baseline leading to unnecessary and invasive vention if VUR was persistent or recur-
study for later comparisons in the as- evaluations. rences of infection were not prevented
sessment of renal growth. with an antimicrobial prophylaxis reg-
Benet-harms assessment: Prepon-
Nuclear scanning with technetium- derance of benet over harm. imen; some have advocated surgical
labeled dimercaptosuccinic acid has intervention to correct high-grade re-
Value judgments: The seriousness
greater sensitivity for detection of ux even when infection has not re-
acute pyelonephritis and later scar- of the potentially correctable abnor-
curred. However, it is clear that there
ring than does either RBUS or voiding malities in 1% to 2%, coupled with are a signicant number of infants
cystourethrography (VCUG). The scan- the absence of physical harm, was who develop pyelonephritis in whom
ning is useful in research, because it judged sufciently important to tip VUR cannot be demonstrated, and the
ensures that all subjects in a study the scales in favor of testing. effectiveness of antimicrobial prophy-
have pyelonephritis to start with and it Role of patient preferences: Be- laxis for patients who have VUR has
permits assessment of later renal cause ultrasonography is noninva- been challenged in the past decade.
scarring as an outcome measure. The sive and poses minimal risk, we as- Several studies have suggested that
ndings on nuclear scans rarely affect sume that parents will prefer RBUS prophylaxis does not confer the de-
acute clinical management, however, over taking even a small risk of sired benet of preventing recurrent
and are not recommended as part of missing a serious and correctable febrile UTI.5155 If prophylaxis is, in fact,
routine evaluation of infants with their condition. not benecial and VUR is not required
rst febrile UTI. The radiation dose to for development of pyelonephritis,
Exclusions: None.
the patient during dimercaptosuccinic then the rationale for performing
acid scanning is generally low (1 Intentional vagueness: None. VCUG routinely after an initial febrile
mSv),49 although it may be increased in Policy level: Recommendation. UTI must be questioned.

PEDIATRICS Volume 128, Number 3, September 2011 603


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
RCTs of the effectiveness of prophy- TABLE 4 Recurrences of Febrile UTI/Pyelonephritis in Infants 2 to 24 Months of Age With and
Without Antimicrobial Prophylaxis, According to Grade of VUR
laxis performed to date generally in-
Reux Prophylaxis No Prophylaxis P
cluded children more than 24 months
Grade
of age, and some did not provide com- No. of Total N No. of Total N
Recurrences Recurrences
plete data according to grade of VUR.
None 7 210 11 163 .15
These 2 factors have compromised I 2 37 2 35 1.00
meta-analyses. To ensure direct com- II 11 133 10 124 .95
parisons, the committee contacted the III 31 140 40 145 .29
IV 16 55 21 49 .14
6 researchers who had conducted the
most recent RCTs and requested raw
data from their studies.5156 All com- the initial UTI, those who have the Benets: This avoids, for the vast
plied, which permitted the creation of greatest likelihood of having high- majority of febrile infants with UTIs,
a data set with data for 1091 infants 2 grade VUR. Unfortunately, there are no radiation exposure (of particular
to 24 months of age according to grade clinical or laboratory indicators that concern near the ovaries in girls),
of VUR. A 2 analysis (2-tailed) and a have been demonstrated to identify in- expense, and discomfort.
formal meta-analysis did not detect a fants with high-grade VUR. Indications Harms/risks/costs: Detection of a
statistically signicant benet of pro- for VCUG have been proposed on the small number of cases of high-
phylaxis in preventing recurrence of basis of consensus in the absence of grade reux and correctable abnor-
febrile UTI/pyelonephritis in infants data57; the predictive value of any of the
malities is delayed.
without reux or those with grades I, II, indications for VCUG proposed in this
manner is not known. Benet-harms assessment: Prepon-
III, or IV VUR (Table 4 and Fig 3). Only 5 derance of benet over harm.
infants with grade V VUR were in- The level of evidence supporting rou-
tine imaging with VCUG was deemed Value judgments: The risks associ-
cluded in the RCTs; therefore, data for
insufcient at the time of the 1999 ated with radiation (plus the ex-
those infants are not included in Table
practice parameter to receive a rec- pense and discomfort of the proce-
4 or Fig 3.
ommendation, but the consensus of dure) for the vast majority of infants
The proportion of infants with high- outweigh the risk of delaying the de-
the subcommittee was to strongly en-
grade VUR among all infants with fe- tection of the few with correctable
courage imaging studies. The position
brile UTIs is small. Data adapted from abnormalities until their second UTI.
of the current subcommittee reects
current studies (Table 5) indicate that,
the new evidence demonstrating anti- Role of patient preferences: The
of a hypothetical cohort of 100 infants
microbial prophylaxis not to be effec- judgment of parents may come into
with febrile UTIs, only 1 has grade V
tive as presumed previously. More- play, because VCUG is an uncomfort-
VUR; 99 do not. With a practice of wait-
over, prompt diagnosis and effective able procedure involving radiation
ing for a second UTI to perform VCUG,
treatment of a febrile UTI recurrence exposure. In some cases, parents
only 10 of the 100 would need to un- may be of greater importance regard-
dergo the procedure and the 1 with may prefer to subject their children
less of whether VUR is present or the to the procedure even when the
grade V VUR would be identied. (It child is receiving antimicrobial pro-
also is possible that the 1 infant with chance of benet is both small and
phylaxis. A national study (the Ran- uncertain. Antimicrobial prophy-
grade V VUR might have been identied domized Intervention for Children With
after the rst UTI on the basis of abnor- laxis seems to be ineffective in pre-
Vesicoureteral Reux study) is cur- venting recurrence of febrile UTI/py-
mal RBUS results that prompted VCUG rently in progress to identify the ef-
to be performed.) Data to quantify ad- elonephritis for the vast majority of
fects of a prophylactic antimicrobial
ditional potential harm to an infant infants. Some parents may want to
regimen for children 2 months to 6
who is not revealed to have high-grade avoid VCUG even after the second
years of age who have experienced a
VUR until a second UTI are not precise UTI. Because the benet of identify-
UTI, and it is anticipated to provide ad-
but suggest that the increment is in- ing high-grade reux is still in some
ditional important data58 (see Areas
sufcient to justify routinely subject- doubt, these preferences should be
for Research).
ing all infants with an initial febrile UTI considered. It is the judgment of the
to VCUG (Fig 4). To minimize any harm Action Statement 6a committee that VCUG is indicated af-
incurred by that infant, attempts have Aggregate quality of evidence: B ter the second UTI.
been made to identify, at the time of (RCTs). Exclusions: None.

604 FROM THE AMERICAN ACADEMY OF PEDIATRICS


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
FROM THE AMERICAN ACADEMY OF PEDIATRICS

TABLE 5 Rates of VUR According to Grade in


Hypothetical Cohort of Infants After
First UTI and After Recurrence
Rate, %
After First After
UTI Recurrence
(N 100) (N 10)
No VUR 65 26
Grades IIII VUR 29 56
Grade IV VUR 5 12
Grade V VUR 1 6

FIGURE 4
Relationship between renal scarring and num-
ber of bouts of pyelonephritis. Adapted from
Jodal.59

Intentional vagueness: None.


Policy level: Recommendation.

Action Statement 6b
Aggregate quality of evidence: X (ex-
ceptional situation).
Benets: VCUG after a second UTI
should identify infants with very
high-grade reux.
Harms/risks/costs: VCUG is an un-
comfortable, costly procedure that
involves radiation, including to the
ovaries of girls.
Benet-harms assessment: Prepon-
FIGURE 3 derance of benet over harm.
A, Recurrences of febrile UTI/pyelonephritis in 373 infants 2 to 24 months of age without VUR, with and
without antimicrobial prophylaxis (based on 3 studies; data provided by Drs Craig, Garin, and Mon- Value judgments: The committee
tini). B, Recurrences of febrile UTI/pyelonephritis in 72 infants 2 to 24 months of age with grade I VUR, judged that patients with high-
with and without antimicrobial prophylaxis (based on 4 studies; data provided by Drs Craig, Garin,
Montini, and Roussey-Kesler). C, Recurrences of febrile UTI/pyelonephritis in 257 infants 2 to 24
grade reux and other abnormali-
months of age with grade II VUR, with and without antimicrobial prophylaxis (based on 5 studies; data ties may benet from interventions
provided by Drs Craig, Garin, Montini, Pennesi, and Roussey-Kesler). D, Recurrences of febrile UTI/ to prevent further scarring. Further
pyelonephritis in 285 infants 2 to 24 months of age with grade III VUR, with and without antimicrobial
prophylaxis (based on 6 studies; data provided by Drs Brandstrm, Craig, Garin, Montini, Pennesi, and studies of treatment for grade V
Roussey-Kesler). E, Recurrences of febrile UTI/pyelonephritis in 104 infants 2 to 24 months of age with VUR are not underway and are un-
grade IV VUR, with and without antimicrobial prophylaxis (based on 3 studies; data provided by Drs likely in the near future, because the
Brandstrm, Craig, and Pennesi). M-H indicates Mantel-Haenszel; CI, condence interval.
condition is uncommon and ran-
domization of treatment in this
group generally has been consid-
ered unethical.

PEDIATRICS Volume 128, Number 3, September 2011 605


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
Role of patient preferences: As men- Aggregate quality of evidence: C (ob- rst UTI is not recommended; VCUG is
tioned previously, the judgment of servational studies). indicated if RBUS reveals hydrone-
parents may come into play, be- Benets: Studies suggest that early phrosis, scarring, or other ndings
cause VCUG is an uncomfortable treatment of UTI reduces the risk of that would suggest either high-grade
procedure involving radiation expo- renal scarring. VUR or obstructive uropathy, as well as
sure. In some cases, parents may in other atypical or complex clinical
Harms/risks/costs: There may be
prefer to subject their children to circumstances. VCUG also should be
additional costs and inconvenience
the procedure even when the performed if there is a recurrence of
to parents with more-frequent visits
chance of benet is both small and febrile UTI.
to the clinician for evaluation of
uncertain. The benets of treatment
fever.
of VUR remain unproven, but the AREAS FOR RESEARCH
point estimates suggest a small po- Benet-harms assessment: Prepon-
derance of benet over harm. One of the major values of a compre-
tential benet. Similarly, parents hensive literature review is the identi-
may want to avoid VCUG even after Value judgments: None.
cation of areas in which evidence is
the second UTI. Because the benet Role of patient preferences: Parents lacking. The following 8 areas are pre-
of identifying high-grade reux is will ultimately make the judgment to sented in an order that parallels the
still in some doubt, these prefer- seek medical care. previous discussion.
ences should be considered. It is the Exclusions: None. 1. The relationship between UTIs in in-
judgment of the committee that
Intentional vagueness: None. fants and young children and re-
VCUG is indicated after the second
Policy level: Recommendation. duced renal function in adults has
UTI.
been established but is not
Exclusions: None.
CONCLUSIONS well characterized in quantitative
Intentional vagueness: Further eval- terms. The ideal prospective cohort
The committee formulated 7 key action
uation will likely start with VCUG but study from birth to 40 to 50 years of
statements for the diagnosis and
may entail additional studies de- age has not been conducted and is
treatment of infants and young chil-
pending on the ndings. The details unlikely to be conducted. There-
dren 2 to 24 months of age with UTI and
of further evaluation are beyond the fore, estimates of undesirable
unexplained fever. Strategies for diag-
scope of this guideline. outcomes in adulthood, such as
nosis and treatment depend on
Policy level: Recommendation. whether the clinician determines that hypertension and end-stage renal
antimicrobial therapy is warranted im- disease, are based on the mathe-
Action Statement 7 matical product of probabilities
mediately or can be delayed safely un-
After conrmation of UTI, the cli- til urine culture and urinalysis results at several steps, each of which is
nician should instruct parents or are available. Diagnosis is based on subject to bias and error. Other
guardians to seek prompt medical the presence of pyuria and at least attempts at decision analysis and
evaluation (ideally within 48 50 000 CFUs per mL of a single uro- thoughtful literature review have
hours) for future febrile ill- pathogen in an appropriately collected recognized the same limitations.
nesses, to ensure that recurrent specimen of urine; urinalysis alone Until recently, imaging tools avail-
infections can be detected and does not provide a denitive diagnosis. able for assessment of the effects
treated promptly (evidence qual- After 7 to 14 days of antimicrobial of UTIs have been insensitive. With
ity: C; recommendation). treatment, close clinical follow-up the imaging techniques now avail-
Early treatment limits renal damage monitoring should be maintained, with able, it may be possible to identify
better than late treatment,1,2 and the evaluation of the urine during subse- the relationship of scarring to re-
risk of renal scarring increases as the quent febrile episodes to permit nal impairment and hypertension.
number of recurrences increase (Fig prompt diagnosis and treatment of re- 2. The development of techniques that
4).59 For these reasons, all infants who current infections. Ultrasonography of would permit an alternative to inva-
have sustained a febrile UTI should the kidneys and bladder should be per- sive sampling and culture would be
have a urine specimen obtained at the formed to detect anatomic abnormali- valuable for general use. Special at-
onset of subsequent febrile illnesses, ties that require further evaluation tention should be given to infant
so that a UTI can be diagnosed and (eg, additional imaging or urologic girls and uncircumcised boys, be-
treated promptly. consultation). Routine VCUG after the cause urethral catheterization may

606 FROM THE AMERICAN ACADEMY OF PEDIATRICS


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
FROM THE AMERICAN ACADEMY OF PEDIATRICS

be difcult and can produce con- microbial resistance. To overcome the infants will have recurrent UTIs;
taminated specimens and SPA now these issues, evidence of effective- some will be identied as having VUR
is not commonly performed. Incu- ness with a well-tolerated, safe or other abnormalities. Further re-
bation time, which is inherent in the product would be required, and search addressing the optimal
culture process, results in delayed parents would need sufcient edu- course of management in specic sit-
treatment or presumptive treat- cation to understand the value and uations would be valuable.
ment on the basis of tests that lack importance of adherence. A urinary 8. The optimal duration of antimicro-
the desired sensitivity and specic- antiseptic, rather than an antimi- bial treatment has not been deter-
ity to replace culture. crobial agent, would be particularly mined. RCTs of head-to-head com-
3. The role of VUR (and therefore of desirable, because it could be taken parisons of various duration would
VCUG) is incompletely understood. indenitely without concern that be valuable, enabling clinicians to
It is recognized that pyelonephritis bacteria would develop resistance. limit antimicrobial exposure to
(dened through cortical scintigra- Another possible strategy might be what is needed to eradicate the of-
phy) can occur in the absence of the use of probiotics. fending uropathogen.
VUR (dened through VCUG) and 5. Better understanding of the ge-
that progressive renal scarring nome (human and bacterial) may LEAD AUTHOR
provide insight into risk factors Kenneth B. Roberts, MD
(dened through cortical scintigra-
phy) can occur in the absence of (VUR and others) that lead to in- SUBCOMMITTEE ON URINARY TRACT
demonstrated VUR.52,53 The pre- creased scarring. Blood specimens INFECTION, 2009 2011
sumption that antimicrobial pro- will be retained from children en- Kenneth B. Roberts, MD, Chair
Stephen M. Downs, MD, MS
phylaxis is of benet for individuals rolled in the Randomized Interven-
S. Maria E. Finnell, MD, MS
with VUR to prevent recurrences of tion for Children With Vesi- Stanley Hellerstein, MD
UTI or the development of renal scars coureteral Reux study, for future Linda D. Shortliffe, MD
examination of genetic determinants Ellen R. Wald, MD
is not supported by the aggregate of J. Michael Zerin, MD
data from recent studies and cur- of VUR, recurrent UTI, and renal scar-
rently is the subject of the Random- ring.58 VUR is recognized to run in OVERSIGHT BY THE STEERING
ized Intervention for Children With families,60,61 and multiple investiga- COMMITTEE ON QUALITY
tors are currently engaged in re- IMPROVEMENT AND MANAGEMENT,
Vesicoureteral Reux study.58
search to identify a genetic basis for 2009 2011
4. Although the effectiveness of anti-
microbial prophylaxis for the pre- VUR. Studies may also be able to dis- STAFF
tinguish the contribution of congeni- Caryn Davidson, MA
vention of UTI has not been demon-
tal dysplasia from acquired scarring
strated, the concept has biological ACKNOWLEDGMENTS
attributable to UTI.
plausibility. Virtually all antimicro- The committee gratefully acknowl-
bial agents used to treat or to pre- 6. One of the factors used to assess edges the generosity of the research-
vent infections of the urinary tract the likelihood of UTI in febrile in- ers who graciously shared their data
are excreted in the urine in high fants is race. Data regarding rates to permit the data set with data for
concentrations. Barriers to the ef- among Hispanic individuals are lim- 1091 infants aged 2 to 24 months ac-
fectiveness of antimicrobial pro- ited and would be useful for predic- cording to grade of VUR to be com-
phylaxis are adherence to a daily tion rules. piled, that is, Drs Per Brandstrm,
regimen, adverse effects associ- 7. This guideline is limited to the initial Jonathan Craig, Eduardo Garin, Gio-
ated with the various agents, and management of the rst UTI in febrile vanni Montini, Marco Pennesi, and
the potential for emergence of anti- infants 2 to 24 months of age. Some of Gwenaelle Roussey-Kesler.
REFERENCES
1. Winter AL, Hardy BE, Alton DJ, Arbus GS, 3. American Academy of Pediatrics, Commit- 4. Finnell SM, Carroll AE, Downs SM, et al. Technical
Churchill BM. Acquired renal scars in chil- tee on Quality Improvement, Subcommittee report: diagnosis and management of an initial
dren. J Urol. 1983;129(6):1190 1194 on Urinary Tract Infection. Practice urinary tract infection in febrile infants and
2. Smellie JM, Poulton A, Prescod NP. Retro- parameter: the diagnosis, treatment, and young children. Pediatrics. 2011;128(3):e749
spective study of children with renal scar- evaluation of the initial urinary tract infec- 5. American Academy of Pediatrics, Steering
ring associated with reux and urinary in- tion in febrile infants and young children. Committee on Quality Improvement and
fection. BMJ. 1994;308(6938):11931196 Pediatrics. 1999;103(4):843 852 Management. Classifying recommenda-

PEDIATRICS Volume 128, Number 3, September 2011 607


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
tions for clinical practice guidelines. Pedi- 21. Wiswell TE, Hachey WE. Urinary tract infec- 35. Kunin C. A ten-year study of bacteriuria in
atrics. 2004;114(3):874 877 tions and the uncircumcised state: an up- schoolgirls: nal report of bacteriologic,
6. Leong YY, Tan KW. Bladder aspiration for date. Clin Pediatr (Phila). 1993;32(3): urologic, and epidemiologic ndings. J In-
diagnosis of urinary tract infection in in- 130 134 fect Dis. 1970;122(5):382393
fants and young children. J Singapore Pae- 22. Wiswell TE, Smith FR, Bass JW. Decreased 36. Kemper K, Avner E. The case against screen-
diatr Soc. 1976;18(1):43 47 incidence of urinary tract infections in cir- ing urinalyses for asymptomatic bacteri-
7. Pryles CV, Atkin MD, Morse TS, Welch KJ. cumcised male infants. Pediatrics. 1985; uria in children. Am J Dis Child. 1992;146(3):
Comparative bacteriologic study of urine 75(5):901903 343346
obtained from children by percutaneous 23. Ginsburg CM, McCracken GH Jr. Urinary 37. Wald E. Genitourinary tract infections: cysti-
suprapubic aspiration of the bladder and tract infections in young infants. Pediatrics. tis and pyelonephritis. In: Feigin R, Cherry
by catheter. Pediatrics. 1959;24(6):983991 1982;69(4):409 412 JD, Demmler GJ, Kaplan SL, eds. Textbook of
8. Djojohadipringgo S, Abdul Hamid RH, Thahir 24. Craig JC, Knight JF, Sureshkumar P, Mantz Pediatric Infectious Diseases. 5th ed. Phila-
S, Karim A, Darsono I. Bladder puncture in E, Roy LP. Effect of circumcision on inci- delphia, PA: Saunders; 2004:541555
newborns: a bacteriological study. Paediatr dence of urinary tract infection in pre- 38. Mayo S, Acevedo D, Quiones-Torrelo C,
Indones. 1976;16(1112):527534 school boys. J Pediatr. 1996;128(1):2327 Cans I, Sancho M. Clinical laboratory auto-
9. Gochman RF, Karasic RB, Heller MB. Use of 25. Levine DA, Platt SL, Dayan PS, et al. Risk of mated urinalysis: comparison among auto-
portable ultrasound to assist urine collec- serious bacterial infection in young fe- mated microscopy, ow cytometry, two test
tion by suprapubic aspiration. Ann Emerg brile infants with respiratory syncytial vi- strips analyzers, and manual microscopic
Med. 1991;20(6):631 635 rus infections. Pediatrics. 2004;113(6): examination of the urine sediments. J Clin
10. Buys H, Pead L, Hallett R, Maskell R. Supra- 1728 1734 Lab Anal. 2008;22(4):262270
pubic aspiration under ultrasound guid- 26. Roberts KB, Charney E, Sweren RJ, et al. Uri- 39. Kass E. Asymptomatic infections of the uri-
ance in children with fever of undiagnosed nary tract infection in infants with unex- nary tract. Trans Assoc Am Phys. 1956;69:
cause. BMJ. 1994;308(6930):690 692 plained fever: a collaborative study. J Pedi- 56 64
atr. 1983;103(6):864 867 40. Hoberman A, Wald ER, Reynolds EA, Pen-
11. Kramer MS, Tange SM, Drummond KN, Mills
EL. Urine testing in young febrile children: a 27. Gorelick MH, Hoberman A, Kearney D, Wald chansky L, Charron M. Pyuria and bacteri-
risk-benet analysis. J Pediatr. 1994;125(1): E, Shaw KN. Validation of a decision rule uria in urine specimens obtained by cathe-
6 13 identifying febrile young girls at high risk ter from young children with fever. J
for urinary tract infection. Pediatr Emerg Pediatr. 1994;124(4):513519
12. Bonadio WA. Urine culturing technique in fe-
Care. 2003;19(3):162164
brile infants. Pediatr Emerg Care. 1987;3(2): 41. Downs SM. Technical report: urinary tract infec-
7578 28. Gorelick MH, Shaw KN. Clinical decision rule tions in febrile infants and young children. Pedi-
to identify febrile young girls at risk for uri- atrics. 1999;103(4). Available at: www.pediatrics.
13. Lohr J. Pediatric Outpatient Procedures.
nary tract infection. Arch Pediatr Adolesc org/cgi/content/full/103/4/e54
Philadelphia, PA: Lippincott; 1991
Med. 2000;154(4):386 390
14. Taylor CM, White RH. The feasibility of 42. Hoberman A, Wald ER, Hickey RW, et al. Oral
29. Shaw KN, Gorelick M, McGowan KL, Yakscoe versus initial intravenous therapy for uri-
screening preschool children for urinary
NM, Schwartz JS. Prevalence of urinary nary tract infections in young febrile chil-
tract infection using dipslides. Int J Pediatr
tract infection in febrile young children in
Nephrol. 1983;4(2):113114 dren. Pediatrics. 1999;104(1):79 86
the emergency department. Pediatrics.
15. Srensen K, Lose G, Nathan E. Urinary tract 1998;102(2). Available at: www.pediatrics. 43. Hodson EM, Willis NS, Craig JC. Antibiotics
infections and diurnal incontinence in girls. org/cgi/content/full/102/2/e16 for acute pyelonephritis in children. Co-
Eur J Pediatr. 1988;148(2):146 147 chrane Database Syst Rev. 2007;(4):
30. Shaikh N, Morone NE, Lopez J, et al. Does this
16. Shannon F, Sepp E, Rose G. The diagnosis of CD003772
child have a urinary tract infection? JAMA.
bacteriuria by bladder puncture in infancy 2007;298(24):28952904 44. Bloomeld P, Hodson EM, Craig JC. Antibiot-
and childhood. Aust Pediatr J. 1969;5(2): ics for acute pyelonephritis in children. Co-
31. Powell HR, McCredie DA, Ritchie MA. Urinary
97100 chrane Database Syst Rev. 2005;(1):
nitrite in symptomatic and asymptomatic
17. Hoberman A, Chao HP, Keller DM, Hickey R, CD003772
urinary infection. Arch Dis Child. 1987;62(2):
Davis HW, Ellis D. Prevalence of urinary tract 138 140 45. Hoberman A, Charron M, Hickey RW, Baskin
infection in febrile infants. J Pediatr. 1993; M, Kearney DH, Wald ER. Imaging studies af-
32. Kunin CM, DeGroot JE. Sensitivity of a nitrite
123(1):1723 indicator strip method in detecting bacteri- ter a rst febrile urinary tract infection in
18. Haddon RA, Barnett PL, Grimwood K, Hogg uria in preschool girls. Pediatrics. 1977; young children. N Engl J Med. 2003;348(3):
GG. Bacteraemia in febrile children present- 60(2):244 245 195202
ing to a paediatric emergency department. 33. Hoberman A, Wald ER, Reynolds EA, Pen- 46. Jahnukainen T, Honkinen O, Ruuskanen O,
Med J Aust. 1999;170(10):475 478 chansky L, Charron M. Is urine culture nec- Mertsola J. Ultrasonography after the rst
19. Wiswell TE, Roscelli JD. Corroborative evi- essary to rule out urinary tract infection in febrile urinary tract infection in children.
dence for the decreased incidence of uri- young febrile children? Pediatr Infect Dis J. Eur J Pediatr. 2006;165(8):556 559
nary tract infections in circumcised male 1996;15(4):304 309 47. Economou G, Egginton J, Brookeld D. The
infants. Pediatrics. 1986;78(1):96 99 34. Herr SM, Wald ER, Pitetti RD, Choi SS. En- importance of late pregnancy scans for re-
20. To T, Agha M, Dick PT, Feldman W. Cohort hanced urinalysis improves identication nal tract abnormalities. Prenat Diagn. 1994;
study on circumcision of newborn boys and of febrile infants ages 60 days and younger 14(3):177180
subsequent risk of urinary-tract infection. at low risk for serious bacterial illness. Pe- 48. Roberts J. Experimental pyelonephritis in
Lancet. 1998;352(9143):18131816 diatrics. 2001;108(4):866 871 the monkey, part III: pathophysiology of ure-

608 FROM THE AMERICAN ACADEMY OF PEDIATRICS


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
FROM THE AMERICAN ACADEMY OF PEDIATRICS

teral malfunction induced by bacteria. In- 53. Montini G, Rigon L, Zucchetta P, et al. Pro- Management: NICE Clinical Guideline 54.
vest Urol. 1975;13(2):117120 phylaxis after rst febrile urinary tract in- London, England: National Institute for
49. Smith T, Evans K, Lythgoe MF, Anderson PJ, fection in children? A multicenter, random- Health and Clinical Excellence; 2007. Avail-
Gordon I. Radiation dosimetry of ized, controlled, noninferiority trial. able at: www.nice.org.uk/nicemedia/live/
technetium-99m-DMSA in children. J Nucl Pediatrics. 2008;122(5):1064 1071 11819/36032/36032.pdf. Accessed March
Med. 1996;37(8):1336 1342 54. Roussey-Kesler G, Gadjos V, Idres N, et al. 14, 2011
50. Ward VL. Patient dose reduction during Antibiotic prophylaxis for the prevention of 58. Keren R, Carpenter MA, Hoberman A, et al.
voiding cystourethrography. Pediatr Radiol. recurrent urinary tract infection in children Rationale and design issues of the Random-
2006;36(suppl 2):168 172 with low grade vesicoureteral reux: re- ized Intervention for Children With Vesi-
51. Pennesi M, Travan L, Peratoner L, et al. Is sults from a prospective randomized study. coureteral Reux (RIVUR) study. Pediatrics.
antibiotic prophylaxis in children with vesi- J Urol. 2008;179(2):674 679
2008;122(suppl 5):S240 S250
coureteral reux effective in preventing py- 55. Craig J, Simpson J, Williams G. Antibiotic
59. Jodal U. The natural history of bacteriuria in
elonephritis and renal scars? A random- prophylaxis and recurrent urinary tract in-
childhood. Infect Dis Clin North Am. 1987;
ized, controlled trial. Pediatrics. 2008; fection in children. N Engl J Med. 2009;
1(4):713729
121(6). Available at: www.pediatrics.org/ 361(18):1748 1759
cgi/content/full/121/6/e1489 60. Eccles MR, Bailey RR, Abbott GD, Sullivan MJ.
56. Brandstrm P, Esbjorner E, Herthelius M,
Swerkersson S, Jodal U, Hansson S. The Unravelling the genetics of vesicoureteric
52. Garin EH, Olavarria F, Garcia Nieto V, Valen-
Swedish Reux Trial in Children, part III: uri- reux: a common familial disorder. Hum
ciano B, Campos A, Young L. Clinical signi-
nary tract infection pattern. J Urol. 2010; Mol Genet. 1996;5(Spec No.):14251429
cance of primary vesicoureteral reux and
urinary antibiotic prophylaxis after acute 184(1):286 291 61. Scott JE, Swallow V, Coulthard MG, Lambert
pyelonephritis: a multicenter, randomized, 57. National Institute for Health and Clinical Ex- HJ, Lee RE. Screening of newborn babies for
controlled study. Pediatrics. 2006;117(3): cellence. Urinary Tract Infection in Children: familial ureteric reux. Lancet. 1997;
626 632 Diagnosis, Treatment, and Long-term 350(9075):396 400

PEDIATRICS Volume 128, Number 3, September 2011 609


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
APPENDIX
Clinical practice guideline algorithm.

610 FROM THE AMERICAN ACADEMY OF PEDIATRICS


Downloaded from pediatrics.aappublications.org by guest on September 2, 2011
Urinary Tract Infection: Clinical Practice Guideline for the Diagnosis and
Management of the Initial UTI in Febrile Infants and Children 2 to 24 Months
SUBCOMMITTEE ON URINARY TRACT INFECTION and STEERING
COMMITTEE ON QUALITY IMPROVEMENT AND MANAGEMENT
Pediatrics; originally published online August 28, 2011;
DOI: 10.1542/peds.2011-1330
Updated Information & including high resolution figures, can be found at:
Services http://pediatrics.aappublications.org/content/early/2011/08/24
/peds.2011-1330
Permissions & Licensing Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
http://pediatrics.aappublications.org/site/misc/Permissions.xht
ml
Reprints Information about ordering reprints can be found online:
http://pediatrics.aappublications.org/site/misc/reprints.xhtml

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly


publication, it has been published continuously since 1948. PEDIATRICS is owned, published,
and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk
Grove Village, Illinois, 60007. Copyright 2011 by the American Academy of Pediatrics. All
rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from pediatrics.aappublications.org by guest on September 2, 2011