In theClinic
In the Clinic
Physician Writer
Kalpana Gupta, MD, MPH
Barbara Trautner, MD, PhD
Affiliations: VA Boston
Healthcare System and
Boston University School of
Medicine, Boston, Massachu-
setts; and Houston VA Health
Services Research Center of
Excellence at the Michael E.
DeBakey VA Medical Center
and Baylor College of Medi-
cine, Houston, Texas.
Section Editors
Deborah Cotton, MD, MPH
Darren Taichman, MD, PhD
Sankey Williams, MD
Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013
Urinary Tract
Infection
Background page ITC3-2
Screening and Prevention page ITC3-3
Diagnosis and Evaluation page ITC3-5
Treatment and Management page ITC3-8
Practice Improvement page ITC3-13
Tool Kit page ITC3-14
Patient Information page ITC3-15
CME Questions page ITC3-16
The content of In the Clinic is drawn from the clinical information and education
resources of the American College of Physicians (ACP), including PIER (Physicians
Information and Education Resource) and MKSAP (Medical Knowledge and Self-
Assessment Program). Annals of Internal Medicine editors develop In the Clinic
from these primary sources in collaboration with the ACPs Medical Education
and Publishing divisions and with the assistance of science writers and physician
writers. Editorial consultants from PIER and MKSAP provide expert review of the
content. Readers who are interested in these primary resources for more detail
can consult http://pier.acponline.org, http://www.acponline.org/products_services/
mksap/15/?pr31, and other resources referenced in each issue of In the Clinic.
CME Objective: To review current evidence for the background, screening and
prevention, diagnosis and evaluation, and treatment and management of urinary
tract infection.
The information contained herein should never be used as a substitute for clinical
judgment.
2012 American College of Physicians
 rinary tract infections (UTIs) are one of the most common infec-

U tions in both outpatient and inpatient settings. The term urinary

tract infection applies to a heterogeneous group of clinical syn-
dromes. Clinical entities encompassed by UTI include asymptomatic
bacteriuria, acute uncomplicated cystitis, recurrent cystitis, complicated
UTI, catheter-associated asymptomatic bacteriuria, catheter-associated
UTI (CAUTI), prostatitis, and pyelonephritis. Appropriate classification
of the UTI syndrome is crucial for optimal diagnosis and management.
The various categories of UTI are distinguished by the presence or absence
of symptoms referable to the urinary tract; the patients sex and comorbid
conditions; and genitourinary history, including the presence of stones or
stents. Because acute cystitis is the most common manifestation of UTI
and is most prevalent in women, most clinical research on UTI has been
done in adult women. Clinicians must carefully consider whether recom-
mendations derived from this evidence base are applicable to their patient
populations.

Background
What patient populations are at In contrast to the predominant
greatest risk for UTI? role of behavioral risk factors in
In the absence of known abnor- premenopausal women, mechani-
1. Hooton TM, Scholes
D, Hughes JP, et al. A
malities of the urinary tract, cal and physiologic factors that
prospective study of women are at higher risk for UTIs affect bladder emptying become
risk factors for symp-
tomatic urinary tract
than are men. Premenopausal important in postmenopausal
infection in young adult women are at especially high women (3). Diabetes may increase
women. N Engl J
Med. 1996;335:468- risk for acute cystitis; incidence is the risk for certain urinary tract
74. [PMID: 8672152]
2. Scholes D, Hooton
0.5 to 0.7 per person-year among disorders, including asymptomatic
TM, Roberts PL, Sta- sexually active women (1). Other bacteriuria, perirenal abscess,
pleton AE, Gupta K,
Stamm WE. Risk fac-
populations at risk for UTI include and emphysematous pyelonephri-
tors for recurrent uri- patients with voiding abnormalities tis (4). In men, risk for UTI is
nary tract infection in
young women. J In- related to diabetes, neurogenic primarily related to the prostatic
fect Dis. 2000;
182:1177-82.
bladder, spinal cord injury, preg- hypertrophy that occurs with ad-
[PMID: 10979915] nancy, prostatic hypertrophy, or vancing age. Temporary instru-
3. Hooton TM. Recurrent
urinary tract infection
urinary tract instrumentation. mentation of the urinary tract is
in women. Int J An- Bacteriuria, with or without ac- the major medical intervention
timicrob Agents.
2001;17:259-68. companying symptoms, is generally that increases the risk for UTI
[PMID: 11295405]
4. Hooton TM. Patho-
considered unavoidable in patients in hospitalized patients. Other
genesis of urinary requiring long-term indwelling comorbid conditions that increase
tract infections: an
update. J Antimicrob catheters. risk in both sexes include stones
Chemother. 2000;46 or foreign bodies, such as ureteral
Suppl 1:1-7. [PMID: What lifestyle factors or comorbid
11051617] stents, in the urinary system, and
5. Scholes D, Hawn TR, conditions are risk factors for diseases associated with a neuro-
Roberts PL, et al.
Family history and UTI? genic bladder.
risk of recurrent cysti- The strongest risk factors for acute
tis and pyelonephritis
in women. J Urol. uncomplicated cystitis in pre- A recent casecontrol study of 1261 fe-
2010;184:564-9. menopausal women include sexual male outpatients 18 to 49 years of age
[PMID: 20639019]
6. Nicolle LE, Bradley S, intercourse, use of spermicides, investigated the role of family history of
Colgan R, Rice JC,
pregnancy, and previous UTI. A UTI as a risk factor for recurrent UTI or
Schaeffer A, Hooton
TM; Infectious Dis- history of maternal UTI and age pyelonephritis. A history of any UTI in the
eases Society of mother conferred an increased risk for
America. Infectious at first UTI are also important risk
both recurrent UTI (odds ratio [OR], 2.5;
Diseases Society of
America guidelines
factors in this group, suggesting a 95% CI, 1.93.4) and pyelonephritis (OR,
for the diagnosis and genetic component to susceptibili- 3.3; CI, 2.44.5). Having a sister or daugh-
treatment of asymp-
tomatic bacteriuria in ty (2). Changes in vaginal micro- ter with UTI also increased the risk for re-
adults. Clin Infect Dis. bial flora in perimenopausal current UTI, with ORs ranging from 2.6 to
2005;40:643-54.
[PMID: 15714408] women may increase risk for UTI. 4.1 (5).

2012 American College of Physicians ITC3-2 In the Clinic Annals of Internal Medicine 6 March 2012

Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013

 Screening and
Is there a role for screening for recommended. Dipstick urinalysis is not Prevention
UTI or asymptomatic bacteriuria? considered sensitive enough to be a
In men and nonpregnant women, stand-alone test in pregnant women.
screening for asymptomatic bacteriuria
is generally not recommended because Screening for and treating asympto-
treatment does not improve clinical matic bacteriuria is also recommended
outcomes (6). Asymptomatic bacteri- in men who are about to have
uria does not lead to hypertension, transurethral resection of the prostate
chronic kidney disease, or decreased (TURP) or other urinary tract instru-
survival (7). Women with asympto- mentation resulting in mucosal bleed-
matic bacteriuria are at increased risk ing. Studies have shown that TURP
for symptomatic UTI, but treatment in bacteriuric men can precipitate
of asymptomatic bacteriuria does not bacteremia, with associated sepsis syn-
decrease the frequency of sympto- drome, and that antimicrobial treat-
matic infection (8). Asymptomatic ment of the bacteriuria can prevent
bacteriuria is a marker for poor overall these complications. The level of risk
health status in diabetic patients, associated with specific invasive uro-
7. Abrutyn E, Mossey J,
noncatheterized women living in logic procedures other than TURP in Berlin JA, et al. Does
patients with preexisting bacteriuria is asymptomatic bac-
retirement homes, and catheterized teriuria predict mor-
inpatients, but asymptomatic bac- not well-defined. However, the Infec- tality and does an-
timicrobial treatment
teriuria itself is not an independent tious Diseases Society of America reduce mortality in
risk factor for mortality. (IDSA) recommends that proce- elderly ambulatory
women? Ann Intern
dures anticipated to cause mucosal Med. 1994;120:827-
Abrutyn and coworkers prospectively fol- 33. [PMID: 7818631]
bleeding warrant preprocedure 8. Hooton TM, Scholes
lowed 1491 female residents of retirement screening by urine culture and D, Stapleton AE, et al.
communities. Clean-catch urine cultures treatment of asymptomatic bacteri-
A prospective study
of asymptomatic
were obtained at enrollment and every 6
uria before the procedure (6). Sim- bacteriuria in sexually
months. Women who had asymptomatic active young women.
bacteriuria were older and sicker than
ple catheter placement does not N Engl J Med. 2000;
343:992-7. [PMID:
those who had never had asymptomatic warrant screening. 11018165]
9. Patterson TF, Andriole
bacteriuria, but no relationship was found VT. Detection, signifi-
with mortality after adjusting for covari- Evidence is inadequate to support cance, and therapy of
ates, such as age at study entry. (7). definitive guidelines for the manage- bacteriuria in preg-
nancy. Update in the
ment of asymptomatic bacteriuria in managed health care
Asymptomatic bacteriuria during renal transplant recipients and neu- era. Infect Dis Clin
North Am. 1997;
pregnancy (4%7% of pregnant tropenic patients. Patients with renal 11:593-608. [PMID:
9378925]
women) is associated with a high transplants who have asymptomatic 10. Smaill F, Vazquez JC.
rate of progression to symptomatic bacteriuria are at higher risk for Antibiotics for
asymptomatic bac-
UTI, including pyelonephritis pyelonephritis, but whether pyelo- teriuria in pregnan-
cy. Cochrane Data-
(20%40% of pregnant women nephritis affects graft function is base Syst Rev.
with untreated asymptomatic bac- controversial (11, 12). The relation- 2007:CD000490.
[PMID: 17443502]
teriuria) (9). Asymptomatic bacteri- ship of asymptomatic bacteriuria to 11. Fiorante S, Fernn-
uria of pregnancy also is associated UTI and sepsis in patients with dez-Ruiz M, Lpez-
Medrano F, et al.
with low birthweight and preterm neutropenia is not well-studied. Bac- Acute graft
pyelonephritis in re-
labor, although a causative relation- teriuria diagnosed in neutropenic pa- nal transplant recipi-
ship has not been established. tients as part of a fever workup by ents: incidence, risk
factors and long-
definition is not asymptomatic. term outcome.
A meta-analysis of 14 studies involving Nephrol Dial Trans-
plant. 2011;26:1065-
2302 pregnant women found that anti- How can UTI be prevented? 73. [PMID: 20805254]
biotic treatment was effective at eradicat- Screening urine cultures should be 12. Fiorante S, Lpez-
Medrano F, Lizasoain
ing asymptomatic bacteriuria and at
obtained only in pregnant women M, et al. Systematic
preventing pyelonephritis (10). The num- screening and treat-
ber needed to treat to prevent 1 episode of and patients about to have an invasive ment of asympto-
matic bacteriuria in
pyelonephritis was 7 (CI, 68). Thus, 1-time urologic procedure. Infection should renal transplant re-
screening for asymptomatic bacteriuria be treated with directed antimicrobial cipients. Kidney Int.
2010;78:774-81.
by urine culture early in pregnancy is therapy based on culture results. [PMID: 20720526]

6 March 2012 Annals of Internal Medicine In the Clinic ITC3-3 2012 American College of Physicians

Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013

 Symptomatic UTI in women with patient-initiated therapy. In this
recurrent episodes can be prevented approach, women are provided with
with antimicrobial prophylaxis. The a course of appropriate antibiotics
criterion for the number of episodes to take at the onset of symptoms.
of UTI per year that justifies preven- The advantages of this approach
tive measures is arbitrary, and this include avoiding long-term use
decision should be individualized. of prophylactic antibiotics. The
13. Stapleton A, Latham Postcoital antibiotic prophylaxis has dosing frequency can be tailored to
RH, Johnson C,
Stamm WE. Post- been shown to be highly effective in an individual patients response,
coital antimicrobial preventing symptomatic recurrences with the goal of achieving UTI
prophylaxis for re-
current urinary tract in women with 3 to 4 episodes of prevention while minimizing an-
infection. A random-
UTI per year, particularly if these are tibiotic exposure. However, rates
ized, double-blind,
placebo-controlled temporally associated with coitus. are decreased only during the active
trial. JAMA. 1990;
264:703-6. [PMID:
prophylaxis period and return to
2197450] A randomized, double-blind, placebo- baseline levels after antimicrobials
14. Albert X, Huertas I,
Pereir II, Sanflix J,
controlled trial found that among women are discontinued.
Gosalbes V, Perrota with at least 2 culture-documented epi-
C. Antibiotics for
preventing recurrent sodes of UTI in the previous year, postcoital In postmenopausal women, daily
urinary tract infec- use of half of a single-strength tablet topical application of intravaginal
tion in non-preg-
nant women. of trimethoprim-sulfamethoxazole (TMP- estriol cream may help reduce the
Cochrane Database SMX) (40 mg TMP plus 200 mg SMX) result- frequency of symptomatic episodes.
Syst Rev. 2004:
CD001209. ed in an infection rate of 0.3 per patient per Estrogen therapy is associated with a
[PMID: 15266443] year, compared with a rate of 3.6 per pa- return of the premenopausal lacto-
15. Perrotta C, Aznar M,
tient per year in the control group (13). Side
Mejia R, Albert X, Ng bacillus-dominated vaginal flora, an
CW. Oestrogens for effects were infrequent and minor. In
preventing recurrent acid vaginal pH, and reduced vaginal
urinary tract infec- women intolerant of or resistant to TMP-
tion in post- SMX, an alternative agent (e.g., nitrofuran-
colonization with Escherichia coli. Two
menopausal
toin macrocrystals or a fluoroquinolone) studies in a 2008 Cochrane system-
women. Cochrane
Database Syst Rev. may be as effective for postcoital prophy- atic review compared vaginal estro-
2008:CD005131.
[PMID: 18425910] laxis as TMP-SMX, although clinical evi- gens with placebo and showed that
16. Bent S, Nallamothu dence specifically related to postcoital use vaginal estrogens reduced the num-
BK, Simel DL, Fihn
SD, Saint S. Does this is not available for these agents (13). ber of UTIs in postmenopausal
woman have an
acute uncomplicat-
women. Response varied according
ed urinary tract in- For women with more frequent re- to the type of estrogen and treatment
fection? JAMA.
2002;287:2701-10.
currences, more frequent coitus, or duration. Oral estrogens did not re-
[PMID: 12020306] recurrences temporally unrelated to duce UTI compared with placebo.
17. Gupta K, Hooton TM,
Roberts PL, Stamm coitus, continuous (daily, 3 times Drug-related events associated with
WE. Patient-initiated
treatment of un-
weekly, or even weekly) prophylaxis vaginal estrogens can include vaginal
complicated recur- may be preferable (14). Another itching, burning, discharge, and
rent urinary tract in-
fections in young approach, which is appropriate for metrorrhagia (15). A discussion of
women. Ann Intern women with recurrent, uncompli- possible cancer risk should be indi-
Med. 2001;135:9-16.
[PMID: 11434727] cated UTI unrelated to coitus, is vidualized for each patient.
18. Hooton TM, Bradley
SF, Cardenas DD, et
al; Infectious Dis-
eases Society of
America. Diagnosis,
prevention, and Screening and Prevention... Inappropriate screening for asymptomatic bacteriuria
treatment of can be detrimental because it can lead to unnecessary antibiotic use. Screening
catheter-associated and treatment of asymptomatic bacteriuria are not recommended in nonpregnant
urinary tract infec-
tion in adults: 2009 women, diabetic women, elderly persons, patients with spinal cord injury, or
International Clinical catheterized patients. Screening and treatment of asymptomatic bacteriuria is
Practice Guidelines
from the Infectious recommended only in pregnant women and patients about to have an invasive
Diseases Society of urologic procedure. Consider postcoital antibiotic prophylaxis for women with 2 or
America. Clin Infect
Dis. 2010;50:625-63.
more episodes of UTI per year, particularly if these are temporally associated with
[PMID: 20175247] coitus. Otherwise, daily or thrice-weekly antibiotic prophylaxis can be used to
19. Huppert JS, Biro F, prevent UTIs. Consider topical intravaginal estrogen therapy to prevent sympto-
Lan D, Mortensen JE,
Reed J, Slap GB. Uri- matic UTIs for postmenopausal women with recurrent UTIs.
nary symptoms in
adolescent females:
STI or UTI? J Adolesc
Health. 2007;40:418- CLINICAL BOTTOM LINE
24. [PMID: 17448399]

2012 American College of Physicians ITC3-4 In the Clinic Annals of Internal Medicine 6 March 2012

Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013

 Diagnosis and
What signs and symptoms should suprapubic pain, or subprapubic Evaluation
raise suspicion of UTI? tenderness are also compatible with
Diagnosis begins with a detailed CAUTI (18). 103 colony-forming
history because in ambulatory adult units/mL of urine or greater is suf-
women, the history provided by the ficient for the diagnosis of CAUTI
patient has high predictive value for with an indwelling or intermittent
the presence or absence of cystitis. catheter. By definition, CAUTI can
The most common symptoms in occur with indwelling urethral (Fo-
noncatheterized individuals include ley) catheters, suprapubic catheters,
dysuria, urinary frequency, and and condom catheters. Since many
urgency. of these defining symptoms are non-
specific, other infections and poten-
A meta-analysis of the accuracy of history
tial causes should be considered
and physical examination for diagnosing
acute, uncomplicated UTI in ambulatory
before attributing the symptoms to
women found that the probability of UTI was catheter-associated bacteriuria. The
approximately 50% in women who present- distinction between CAUTI and
ed with 1 or more symptoms of UTI. Specific catheter-associated asymptomatic 20. Saint S, Scholes D,
Fihn SD, Farrell RG,
combinations of symptoms raised the prob- bacteriuria is challenging but clinically Stamm WE. The ef-
ability to more than 90%. Symptoms that in- relevant, because only the former fectiveness of a clini-
cal practice guide-
creased the probability were dysuria, hema- should be treated with antibiotics. line for the
turia, and costovertebral angle tenderness. A management of pre-
sumed uncomplicat-
history of vaginal discharge or irritation de- What other disorders should be ed urinary tract in-
creased the probability of UTI (16). Longitudi- fection in women.
considered? Am J Med. 1999;
nal studies have shown that ambulatory The main alternative diagnoses to 106:636-41. [PMID:
adult women, particularly those who have 10378621]
consider in young women present- 21. Lachs MS,
had recurrent UTIs, are often able to accu- ing with symptoms of cystitis in- Nachamkin I, Edel-
rately identify symptoms of UTI (17). stein PH, Goldman J,
clude sexually transmitted urethritis Feinstein AR,
Schwartz JS. Spec-
A prospective trial enrolled women who or vaginitis, noninfectious urethritis, trum bias in the
had at least 2 episodes of cystitis in the pri- and early pyelonephritis (Table 1). evaluation of diag-
nostic tests: lessons
or 12 months from a university-based pri- Women who are sexually active are from the rapid dip-
stick test for urinary
mary care clinic. The participants were at risk for both UTIs and sexually tract infection. Ann
given urine collection materials and a transmitted diseases (STDs). Symp- Intern Med.
course of fluoroquinolone antibiotics, both 1992;117:135-40.
toms of STDs may be subtle, and [PMID: 1605428]
to be used if they developed symptoms thus they should always be consid- 22. Geerlings SE. Urinary
of UTI. Overall, 88 of 172 women self- tract infections in
ered in the differential diagnosis for patients with dia-
diagnosed a total of 172 UTIs. Laboratory betes mellitus: epi-
evaluation showed a uropathogen in 144
UTI. A history of vaginal discharge demiology, patho-

cases (84%), sterile pyuria in 19 cases (11%), or irritation decreases the probabili- genesis and
treatment. Int J
and no pyuria or bacteriuria in 9 cases (5%). ty of UTI and warrants a workup Antimicrob Agents.
2008;31 Suppl 1:S54-
Clinical and microbiological cures occurred for STDs and other vaginal condi- 7. [PMID: 18054467]
in 92% and 96%, respectively, of culture- tions, such as candidiasis (16). 23. Meron M, Regua-
Mangia AH, Teixeira
confirmed episodes. In this population, the LM, et al. Urinary
strategy of self-diagnosis and manage- A cross-sectional study screened 296 sexu- tract infections in re-
ally active females aged 14 to 22 years who nal transplant recipi-
ment of acute cystitis was highly effective. ents: virulence traits
were visiting a teen health center for both of uropathogenic Es-
UTI (by urine culture) and STD (by vaginal cherichia coli. Trans-
In catheterized patients, signs and plant Proc.
symptoms compatible with a swab and nucleic-acid amplification test- 2010;42:483-5.
ing). In this population, the prevalences of [PMID: 20304171]
CAUTI include new onset or wors- 24. Nicolle LE. Catheter-
ening fever, rigors, altered mental sta- UTI and STI were 17% and 33% respec- related urinary tract
tively; 4% had both. The presence or ab- infection. Drugs Ag-
tus, malaise or lethargy with no other ing. 2005;22:627-39.
sence of urinary symptoms did not predict [PMID: 16060714]
identified cause, flank pain, costover- STD, indicating that telephone manage- 25. Ulleryd P, Zackrisson
tebral angle tenderness, acute hema- ment may not be appropriate for adoles-
B, Aus G, Bergdahl S,
Hugosson J, Sand-
turia, or pelvic discomfort. In patients cent women with urinary symptoms (19). berg T. Selective uro-
logical evaluation in
whose catheters have been removed men with febrile uri-
within the past 48 hours, dysuria, The treating clinician should also nary tract infection.
BJU Int. 2001;88:15-
urgency, frequent urination, consider whether the patient could 20. [PMID: 11446838]

6 March 2012 Annals of Internal Medicine In the Clinic ITC3-5 2012 American College of Physicians

Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013

 Table 1. Differential Diagnosis of Acute Cystitis in Women
Condition Pathogens History Symptoms
Vaginitis Candida, Trichomonas vaginalis, Possibly new sex partner or unprotected Vaginal discharge, odor, or itching;
Bacteroides species, Gardnerella vaginalis sexual activity; history of vaginitis external dysuria (from urine coming
into contact with inflamed and irritated
vulvar epithelial surfaces)
Urethritis Chlamydia trachomatis, Neisseria New sex partner, unprotected sexual Gradual onset of symptoms (Chlamydia)
gonorrhoeae, or herpes simplex virus activity, history of sexually transmitted vaginal discharge; urinary frequency
diseases or genital herpes simplex or urgency
Irritation None No unusual sexual exposure; possible Vaginal itching or discharge; usually a
chemical or allergen exposures diagnosis of exclusion, unless withdrawal
(e.g., douches, bath products, feminine of a suspected offending substance
hygiene products, spermicides) resolves symptoms
Pyelonephritis Same as acute cystitis Previous UTI (pyelonephritis or cystitis) Constitutional symptoms (fever, malaise,
sweats, headache), gastrointestinal
symptoms (anorexia, nausea, vomiting,
abdominal pain), local renal symptoms
(back, flank or loin pain), voiding
symptoms (as in cystitis)

have pyelonephritis (or in men, organism and its antimicrobial sus-

prostatitis) before initiating therapy ceptibility. It is appropriate to ob-
for acute cystitis. The recom- tain a culture (with susceptibility
mended duration of therapy for testing) of a pretreatment urine
pyelonephritis is longer than that of sample for women with suspected
cystitis; treatment of incipient cystitis if the diagnosis is not clear
pyelonephritis with a short-course from the history and physical
regimen for cystitis could predis- examination, if an unusual or
pose the patient to relapse. antimicrobial-resistant organism is
Pyelonephritis may or may not be suspected, if the episode represents
associated with symptoms of cysti- a suspected relapse or treatment
tis, which in some patients may failure, or if the patients therapeu-
overshadow the renal or systemic tic options are limited by medica-
26. Abarbanel J, Engel- manifestations. In patients present- tion intolerance. Pretreatment urine
stein D, Lask D, Livne ing with symptoms of acute cystitis, culture is also considered standard
PM. Urinary tract in-
fection in men confirm the absence of fever, nau- of care in pregnant women and men.
younger than 45
years of age: is there sea, vomiting, rigors, and flank
a need for urologic pain. In men with urinary symp- Women presenting with classic
investigation? Urolo-
gy. 2003;62:27-9. toms and fever, consider both acute symptoms of cystitis (dysuria, fre-
[PMID: 12837416]
27. Gupta K, Hooton TM,
infectious prostatitis and/or quency) who do not have symptoms
Naber KG, et al; In- pyelonephritis in the differential. of possible alternative diagnoses or
fectious Diseases
Society of America.
International clinical What tests should be done to underlying complicating conditions
practice guidelines
diagnose UTI? may be treated for UTI without
for the treatment of
acute uncomplicat- Urine culture is generally not re- further testing. For such women,
ed cystitis and self-initiated therapy or telephone
pyelonephritis in quired in suspected cases of acute,
women: A 2010
uncomplicated cystitis because the clinic provider-guided presumptive
update by the Infec-
tious Diseases Socie- spectrum of causative organisms is therapy may be appropriate (20). In
ty of America and
the European Socie- predictable and urine culture results a woman presenting with symptoms
ty for Microbiology
are often not available until after of acute cystitis, a positive urine
and Infectious Dis-
eases. Clin Infect Dis. completion of short-course empiri- dipstick result can help to confirm
2011;52:e103-20.
[PMID: 21292654] cal treatments. The role of urine the diagnosis, but a negative dip-
28. Gupta K, Hooton TM,
Roberts PL, Stamm
culture is 2-fold: to provide retro- stick result cannot rule out the di-
WE. Short-course ni- spective confirmation of the pres- agnosis in a woman with a high
trofurantoin for the
treatment of acute ence of bacteriuria, which in the pretest probability of cystitis (16,
uncomplicated cys-
titis in women. Arch
correct clinical setting confirms the 21). Thus, a dipstick test is not
Intern Med. 2007; diagnosis of UTI, and to provide necessary if the history is clearly
167:2207-12. [PMID:
17998493] specific information about the diagnostic of UTI (particularly in

2012 American College of Physicians ITC3-6 In the Clinic Annals of Internal Medicine 6 March 2012

Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013

 women with symptoms akin to pre-
vious episodes).
If the diagnosis is not clear-cut or
if further complicating factors or
alternative diagnoses emerges from
the history, a more thorough and
broad-based diagnostic evaluation
should be conducted.
Blood tests (including cultures) are
not indicated in women with sus-
pected cystitis unless they are needed
to screen for alternative diagnoses
suggested by the history or physical
examination or to assess the status
of a known underlying medical
condition (e.g., diabetes mellitus).
Up to 30% of women with acute
pyelonephritis may have secondary
bacteremia, and identification of the
organism by blood culture can be
helpful in cases where antibiotics are
started in advance of the urine cul-
ture. Diabetic women and renal
transplant recipients have a higher
incidence of secondary bacteremia
with UTI (22, 23).
What organisms are generally
found in UTI?
Escherichia coli is the pathogen iso-
lated in more than 90% of patients
with uncomplicated cystitis and
pyelonephritis. Other coliforms,
including Klebsiella and Proteus, are
less common. Staphylococcus sapro-
phyticus is a coagulase-negative
staphylococcus that causes uncom-
plicated cystitis and pyelonephritis
in a small proportion (5%10%) of
otherwise-healthy women. Es-
cherichia coli is still a predominant
pathogen in complicated UTI, but
other coliforms and enterococci are
also common. CAUTI in patients
with short-term catheters can be
caused by E. coli as well as by a
spectrum of typical hospital-
acquired pathogens, including
Klebsiella, Citrobacter, Enterobacter,
Pseudomonas, coagulase-negative
staphylococci, enterococci, and
Candida. Patients with long-term
catheters typically have polymicro-
bial infections; in addition to the
6 March 2012 Annals of Internal Medicine
Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013
pathogens above, Proteus, Mor-
ganella, and Providencia are com-
mon (24).
Interpretation of urine culture de-
pends on the clinical context and
the urinalysis findings; the thresh-
old of bacteriuria required for the
diagnosis of acute cystitis is not ab-
solute. Even a low concentration
(102 colony-forming units/mL) of
coliform bacteria in a urine sample
from a woman with acute dysuria
and pyuria often represents true
(i.e., from within the urinary tract)
bacteriuria. Such organisms should
not be dismissed as contaminants
in a clinical setting suggestive of
cystitis (e.g., a patient with irrita-
tive voiding symptoms plus pyuria).
On the other hand, even high con-
centrations (>105 colony-forming
units/mL) of nonpathogens may
not reflect true bacteriuria if the
urine specimen was not collected
properly or was allowed to stand at
room temperature before process-
ing. Organisms other than coliform
bacilli, S. saprophyticus, and Entero-
coccus (e.g., lactobacilli, alpha strep-
tococci, and coagulase-negative
staphylococci other than S. sapro-
phyticus) are usually considered con- 29. Falagas ME, Kastoris
taminants in urine cultures from AC, Kapaskelis AM,
women with uncomplicated cysti-
Karageorgopoulos
DE. Fosfomycin for
tis, whereas in complicated UTI al- the treatment of
multidrug-resistant,
most any organism can be causative including extended-
and must be considered seriously if spectrum beta-
lactamase produc-
the patient is symptomatic. ing, Enterobacteri-
aceae infections: a
systematic review.
Is there a role for diagnostic Lancet Infect Dis.
2010;10:43-50.
imaging in diagnosing UTI? [PMID: 20129148]
For uncomplicated bladder infec- 30. Falagas ME, Kotsan-
tis IK, Vouloumanou
tions, expert consensus is that imag- EK, Rafailidis PI. An-
tibiotics versus
ing studies (abdominal radiographs, placebo in the treat-
ultrasonography, computed tomog- ment of women
with uncomplicated
raphy, and excretory urography) add cystitis: a meta-
analysis of random-
little or no benefit but increase cost, ized controlled trials.
cause delays, and carry some poten- J Infect. 2009;58:91-
102. [PMID:
tial risk for the patient. Such studies 19195714]
31. Little P, Moore MV,
should be done only if the pretest Turner S, et al. Effec-
suspicion is high for an alternative tiveness of five dif-
ferent approaches in
diagnosis or an anatomical problem management of uri-
nary tract infection:
(such as bladder obstruction or randomised con-
stone) that would require interven- trolled trial. BMJ.
2010;340:c199.
tion. Although men with acute [PMID: 20139214]
In the Clinic ITC3-7 2012 American College of Physicians
 cystitis should be considered for
further evaluation for urologic ab-
normalities, imaging studies for
acute cystitis in men younger than
45 years or in older men without
any symptoms of voiding difficul-
ties or hematuria may not be fruit-
ful. Unfortunately, the evidence base
for male cystitis is limited (25).
A prospective study of UTI was done
in 29 consecutive men younger than
45 years who were hospitalized for
febrile UTI but were otherwise healthy.
Significant urethral stricture was ex-
cluded in all patients by insertion of a
16-French urethral catheter. Ultra-
sonography and intravenous (IV) urog-
raphy were normal in all patients. Ten
patients with hematuria had cys-
toscopy, and the urinary tract was
healthy in all 10. One patient (3%) had
bladder outflow obstruction detected
by uroflowmetry (26).

Diagnosis and Evaluation... Diagnosis of UTI begins with a detailed history. In am-
bulatory adult women, the history provided by the patient has high predictive value
for the presence or absence of cystitis. Consider the diagnosis of pyelonephritis (or
in men, prostatitis) before starting therapy for acute cystitis. Consider complicating
factors, namely underlying medical or urologic conditions that may predispose to
treatment failure, infection with antibiotic-resistant organisms, or infectious com-
plications that would affect the appropriate diagnostic workup and course of thera-
py. Use urinalysis via dipstick, microscopy, or automated microscopy to confirm the
diagnosis in women with suspected UTI when the history alone is not diagnostic. Al-
ways culture the urine of patients with pyelonephritis, complicated UTI, men, preg-
nant women, or those with a history of failure of initial therapy. Initiate empirical
therapy and make adjustments based on the results of the urine culture, if done.

CLINICAL BOTTOM LINE

Treatment and
Management What are the preferred In addition to the usual concerns
treatments for UTI? for efficacy and safety, the 2010
32. Stapleton AE, Women with cystitis should be IDSA treatment recommendations
Dziura J, Hooton TM,
Cox ME, Yarova- treated with an appropriate antimi- for acute cystitis and pyelonephritis
Yarovaya Y, Chen S,
et al. Recurrent Uri-
crobial agent. Before selecting a were guided by 2 important princi-
nary Tract Infection treatment regimen for women with ples: the increasing prevalence of re-
and Urinary Es-
cherichia coli in presumed cystitis, ask about factors sistant organisms, and the potential
Women Injesting
that may influence the choice of for propagation of resistance (collat-
Cranberry Juice
Daily: A Randomized antimicrobial agent, including eral damage) among normal
Controlled Trial.
Mayo Clin Proc. pregnancy and breast-feeding, other host flora with the use of broad-
2012;87:143-50.
medications being taken, drug spectrum antibiotics. No single
[PMID:22305026]
33. Barbosa-Cesnik C, allergy history, recent antibiotic agent was designated as the pre-
Brown MB, Buxton
M, Zhang L, DeBuss- therapy, other recent infections or ferred regimen. Instead, agents are
cher J, Foxman B.
positive cultures, and recent travel. listed as recommended for first-line
Cranberry juice fails
to prevent recurrent Also, whether there are any compli- therapy (nitrofurantoin, TMP-SMX,
urinary tract infec-
cating factors should be determined pivemecillinam, and fosfomycin
tion: results from a
randomized place- trometamol) and alternative
bo-controlled trial. because management of uncompli-
agents (fluoroquinolones and beta-
Clin Infect Dis. cated cystitis and pyelonephritis
2011;52:23-30. lactams). Treatment regimens for
[PMID: 21148516] differs from that of complicated
34. Beerepoot MA, ter cystitis and pyelonephritis are listed
Riet G, Nys S, et al. UTI. The IDSA has recently pub-
in Tables 2 and 3, respectively.
Cranberries vs an-
tibiotics to prevent
lished updated guidelines for treat-
urinary tract infec- ment of uncomplicated cystitis and Each agent has its own caveats. Nitro-
tions: a randomized
double-blind nonin- pyelonephritis, specifically in pre- furantoin concentrates in the urine
feriority trial in pre- menopausal, nonpregnant women
menopausal
but has little tissue penetration and
women. Arch Intern with no known urologic abnormali- thus should be avoided if there is
Med. 2011;171:1270-
8. [PMID: 21788542 ties or comorbid conditions (27). any possibility of pyelo-nephritis.

2012 American College of Physicians ITC3-8 In the Clinic Annals of Internal Medicine 6 March 2012

Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013

 Table 2. Treatment Regimens for Acute Uncomplicated Cystitis
Drug Dose and Duration
Recommended agents
Nitrofurantoin 100 mg twice
monohydrate/ daily for 5 days
macrocrystals
Trimethoprim- 160/800 mg (one
sulfamethoxazole DS tablet) twice
daily for 3 days
Fosfomycin 3-g single-dose
trometamol sachet
Alternative agents
Fluoroquinolones Dose varies by
agent; 3day
regimen
Beta-lactams Dose varies by
agent; 57 day
regimen
A 5-day course of nitrofurantoin
was as effective as a 3-day course of
TMP-SMX in terms of both clini-
cal and microbiological cure in a
randomized trial (28). The main
limitation to the use of TMP-SMX
is the rising rate of resistance
among uropathogens, especially
outside the United States, and con-
sistent evidence that in vitro resist-
ance correlates with bacterial and
clinical failures in at least 50% of
women. However, the 4 studies re-
viewed by the IDSA guidelines
committee showed that TMP-SMX
remains an appropriate empirical
treatment for acute uncomplicated
cystitis in women when the local
rate of resistance is known or ex-
pected to be < 20%. Fosfomycin is
available in the United States but is
rarely used. The recommended 3-g
single dose may not be as effica-
cious as other recommended agents,
and it should be not be used if
pyelonephritis is suspected. Suscep-
tibility data are not routinely re-
ported for this drug, but surveys
demonstrate that it retains activity
against multidrug resistant uropath-
ogens, such as extended-spectrum
beta-lactamaseproducing gram-
negative organisms (29). Pivmecilli-
nam, which is currently unavailable
in the United States, is an extend-
ed-spectrum penicillin that is active
against gram-negative organisms
6 March 2012 Annals of Internal Medicine
Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013
Common Side Effects Comments
Nausea, headache Resistance rare to date; cost varies; usually well-tolerated;
generally safe in pregnancy
Rash, urticaria, nausea, Excellent efficacy if local resistance less than 20%; resistance
vomiting, hematologic prevalence is increasing; inexpensive; extensive clinical experience;
avoid during pregnancy, particularly first and third trimesters
Diarrhea, nausea, May be useful for multidrug-resistant pathogens; may be less
headache effective than other agents; safe in pregnancy
Nausea, vomiting, diarrhea, Prevalence of resistance increasing; cost varies; excellent
headache, drowsiness, efficacy; high collateral damage; better reserved for more
insomnia serious conditions; avoid in pregnancy; tendon inflammation and
rupture have been reported with use of quinolone antibiotics
Diarrhea, nausea, vomiting, Resistance varies by agent; increased adverse effects compared
rash, urticaria with other options; safe during pregnancy
but is used only for treatment of
UTI. Although efficacy rates are
lower than those of the other rec-
ommended agents, its low resistance
rates have made it a popular first-
line choice in some European coun-
tries. The fluoroquinolones,
ofloxacin, ciprofloxacin, and lev-
ofloxacin, are all highly efficacious
in 3-day regimens but are recom-
mended as alternative agents be-
cause they have a high propensity
for collateral damage of the normal
fecal flora. Increasing rates of fluoro-
quinolone resistance in certain areas
suggest that these agents should be
reserved for conditions other than
acute cystitis. Beta-lactams in gener- 35. Talan DA, Stamm
WE, Hooton TM, et
al have inferior efficacy, high collat- al. Comparison of
ciprofloxacin (7 days)
eral damage, and greater rates of and trimethoprim-
adverse effects than other UTI sulfamethoxazole
(14 days) for acute
antimicrobials and thus are also uncomplicated
pyelonephritis
considered alternative agents. pyelonephritis in
women: a random-
ized trial. JAMA.
Pyelonephritis is a tissue-invasive 2000;283:1583-90.
[PMID: 10735395]
disease, and the initial empirical 36. Saint S, Meddings
regimen should be broad enough to JA, Calfee D, Kowals-
ki CP, Krein SL.
ensure in vitro activity against the Catheter-associated
urinary tract infec-
uropathogen. In all suspected cases, tion and the
a urine culture would ideally be ob- Medicare rule
changes. Ann Intern
tained for susceptibility testing be- Med. 2009;150:877-
84. [PMID:
fore starting therapy so the initial 1952856784]
empirical therapy can be tailored 37. Warye KL, Murphy
DM. Targeting zero
appropriately. The decision points health care-associ-
ated infections [Edi-
in managing acute pyelonephritis torial]. Am J Infect
include ruling out complicated in- Control. 2008;
36:683-4. [PMID:
fection (pregnancy, nephrolithiasis, 19084162]
In the Clinic ITC3-9 2012 American College of Physicians
 Table 3. Oral Treatment Regimens for Acute Uncomplicated Pyelonephritis*
Drug Dose and Duration
Fluoroquinolones
Ciprofloxacin 500 mg twice daily for 7 d
Ciprofloxacin XR 1000 mg once daily for 7 d
Levofloxacin 750 mg once daily for 5 d
Trimethoprim- 160/800 mg (one DS tablet)
sulfamethoxazole twice daily for 14 days
Beta-lactams Dose varies by agent;
10- to 14-d regimen
* For pyelonephritis, urine culture and susceptibility testing should always be done, with empirical treatment modified based on the results.
obstruction) and then determining
whether the patient can take oral an-
tibiotics as an outpatient. If oral thera-
py is feasible, oral ciprofloxacin in a
7-day regimen is the preferred regi-
men if local resistance rates to the
fluoroquinolones do not exceed 10%.
The extended-release formulation of
ciprofloxacin for 7 days or a once-
daily dose of levofloxacin for 5 days
can also be used, albeit the evidence is
not as robust (27). TMP-SMX is also
effective if the pathogen is susceptible,
but in the absence of evidence to sup-
port short-course therapy a 14-day
course is the official recommendation.
If susceptibility of the uro-pathogen is
not known, an initial single IV dose
of ceftriaxone or a long-acting amino-
glycoside is recommended before
outpatient oral therapy. In a study
comparing ciprofloxacin to TMP-
SMX, an initial dose of IV ceftriaxone
resulted in improved outcomes in
women receiving TMP-SMX who
had a resistant uropathogen. Oral
beta-lactam agents are not recom-
mended for treatment of pyelonephri-
tis given inferior efficacy rates.
In women being admitted for IV ther-
apy, a broad-spectrum agent should be
given until the susceptibilities of the
organism are known. This can be
achieved with a carbapenem agent,
an aminoglycoside with or without
ampicillin, or extended spectrum beta-
lactam with or without an aminogly-
coside, or another regimen chosen on
the basis of local resistance patterns.
2012 American College of Physicians
Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013
Comments
If local resistance prevalence is < 10%
If pathogen known to be susceptible;
otherwise give an initial IV agent
Oral beta-lactams are less effective and
should be used cautiously only when
other agents cannot be used
Urinary analgesics for acute cystitis
are appropriate in certain situations
to speed resolution of bladder dis-
comfort. The analgesic phenazopy-
ridine is widely used but may cause
nausea. Combination analgesics
containing urinary antiseptics
(methenamine, methylene blue), a
urine-acidifying agent (sodium phos-
phate), and an antispasmodic agent
(hyoscamine) are also available.
Because these analgesics can mask the
symptoms of antimicrobial failure,
they are best used in patients with a
clear diagnosis of cystitis.
Patients with underlying compli-
cating conditions are more likely
to have a drug-resistant organism,
to exhibit a poor response to anti-
microbial therapy even when the
urine organism is susceptible, and to
develop complications if initial ther-
apy for UTI is suboptimal. Broader-
spectrum empirical therapy with
agents to which resistance is least
common and longer treatment dura-
tions are measures intended to blunt
the negative effects of host compro-
mise on treatment outcomes. In
clinical trials of therapy for compli-
cated UTI, oral fluoroquinolones
were as effective as traditional iv
regimens and as or even more
effective than oral TMP-SMX or
TMP (particularly for organisms
resistant to TMP-SMX). However,
resistance to fluoroquinolones
among uropathogens is increasing
worldwide. If the local prevalence
ITC3-10 In the Clinic
Other
Consider adding an initial 1-time IV dose of a long-
acting parenteral antimicrobial (such as 1 g of
ceftriaxone or a single 24-h dose of an amino-
glycoside) if the patient is borderline for oral
therapy but not meeting admission criteria or if
there may be a delay in starting oral therapy
Same as above
Give an initial IV dose of a long-acting parenteral
antimicrobial when using oral beta-lactams
of fluoroquinolone resistance ex-
ceeds 10%, another broad-spectrum
antimicrobial should be considered,
including an extended-spectrum
cephalosporin with or without an
aminoglycoside or a carbapenem.
Combinations of a beta-lactam and a
beta-lactamase inhibitor (e.g.,
ampicillin-sulbactam, ticarcillin-
clavulanate, and piperacillin-
tazobactam) could also be considered.
The IDSA has issued evidence-based
guidelines on the diagnosis, preven-
tion, and treatment of CAUTI (18).
The goal of limiting exposure to an-
timicrobial therapy and thus limiting
selection pressure for resistant organ-
isms is balanced by the awareness
that microbial eradication requires a
longer duration of therapy in patients
with a urinary catheter. Seven days is
the recommended duration of anti-
microbial treatment for patients
whose symptoms resolve promptly,
and 10 to 14 days is recommended
for patients with a delayed response.
For those with CAUTI who are not
severely ill, a 5-day regimen of lev-
ofloxacin may be considered (18).
In pregnant women with sympto-
matic UTI, a urine culture and
susceptibility testing should be per-
formed. Empirical therapy with an
oral antimicrobial agent that is safe
for use in pregnancy should be
given for 3 to 7 days for cystitis or
7 to 14 days for pyelonephritis. An-
tibiotic therapy should be adjusted
on the basis of culture results.
Annals of Internal Medicine 6 March 2012
 Is there a role for nonpharmcologic
therapies in treating UTI, including
alternative therapies?
There is no defined role for nondrug
therapy in managing acute cystitis. Pa-
tients may have heard of supposed
adjunctive or alternative measures to
UTI therapy, and may want guidance
on their use. Advise patients that in-
creased fluid intake, consumption of
cranberry juice, acupuncture, and other
nondrug approaches to the manage-
ment of cystitis are of no known bene-
fit. Patients should be counseled that
relying on nondrug therapies to treat
acute cystitis is essentially similar to
treating the condition with a placebo.
A meta-analysis of 5 trials of placebo
vs. antibiotic therapy for uncompli-
cated cystitis showed that antibiotic
therapy was clearly superior for
achieving symptom resolution (OR,
4.8; CI, 2.59.2) and microbio-logical
cure (OR, 10.7; CI, 2.9638.4) (30).
A randomized, controlled trial of 5 ap-
proaches to management of UTI in non-
pregnant adult women found that all
approaches achieved similar symptom con-
trol, but women who delayed the start of an-
tibiotics had a longer duration of symptoms.
The 5 strategies tested were immediate an-
tibiotics, delayed empirical antibiotics for
persistent symptoms, antibiotics based on a
symptom score, antibiotics based on dip-
stick results, and antibiotics based on a pos-
itive urine culture result (31). This finding is in
accord with the meta-analysis of Bent and
colleagues (16), which found that women
who present with symptoms of UTI have a
high pretest probability of UTI and will prob-
ably require treatment for UTI unless anoth-
er explanation for their symptoms is found.
The patient may raise the issue of
cranberry products to prevent or to
treat UTI. Cranberries contain sub-
stances that inhibit certain adhesins
of uropathogenic E. coli, thereby
preventing the bacteria from adher-
ing to human cells. However, these
in vitro findings have not yet been
proven to have clinical relevance.
A recent randomized trial attempted to
address this issue and found a potentially
protective effect on recurrent UTI with cran-
berry products compared with placebo as
6 March 2012 Annals of Internal Medicine
Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013
well as a reduction in urinary P-fimbriated
E. coli; however, the study was under-
powered to reach statistical significance
for either outcome (32). Another recent,
randomized, controlled trial of twice-daily
cranberry juice consumption versus place-
bo did not find a difference in the rate of
UTI recurrence between the groups (33). An-
other randomized study compared cran-
berry capsules with TMP-SMX and showed
that the antibiotic was more effective at
preventing recurrent UTI, albeit at the ex-
pense of promoting antibiotic resistance
(34). At this point there is no definitive
evidence that commercially available cran-
berry products can be used for treating or
preventing UTI.
When should patients be
hospitalized for UTI?
Uncomplicated cystitis by definition
does not require hospitalization, but
complicated cystitis or pyelonephritis
may require inpatient management.
tem or other anatomical abnormali-
Factors for which hospitalization
ties, or vesicoureteral reflux. The
may be necessary include serious co-
morbid conditions, including preg-
nancy. Patients with a high fever,
dehydration, high leukocyte count, or
other evidence of sepsis may require
inpatient supervision, as do those
who are vomiting and thus unable to
take oral therapy. Another considera-
tion is whether the patient has an
upper urinary tract condition that
would require drainage or surgical in-
tervention, including abscesses, em-
physematous pyelonephritis, papillary
necrosis, or xanthogranulomatous
pyelonephritis. On rare occasions, a
patient will have a multidrug-resistant
urine organism that is susceptible
only to parenterally administered an-
timicrobial agents or is intolerant of
agents that can be given orally. How-
ever, many women with pyelo-
nephritis can be managed successfully
at home, as treatment trials have
shown the efficacy of oral therapy
with a fluoroquinolone for mild, acute
pyelonephritis in this group (35).
Clinical judgment is required to
determine which patients are appro-
priate for outpatient management,
and a period of observation in
the emergency department may be
warranted.
In the Clinic ITC3-11
What are the usual reasons for
failure of UTI therapy?
Underlying medical conditions that
increase the risk for delayed or
incomplete response to therapy, re-
lapse, or infectious complications in-
clude pregnancy, poorly controlled
diabetes mellitus, and immunosup-
pression. Advanced age, recent
UTIs, recent antibiotic use, previous
infection with resistant organism,
prolonged duration of symptoms
before presentation (>3 d), and re-
cent travel to an area with a high
prevalence of antibiotic-resistant
uropathogens all increase the risk for
resistant infection. Urologic compli-
cations that can put the patient at
risk for treatment failure include uri-
nary tract stones, voiding disorders,
indwelling catheters, stents, urinary
obstruction, duplicated collecting sys-
combination of obstruction and in-
fection puts the patient at increased
risk for upper UTI (pyelonephritis),
sepsis, and perinephric or intrarenal
abscess. In particular, patients with
CAUTI who do not promptly re-
spond to appropriate antibiotic therapy
may have obstruction or stones in the
upper urinary tract. Effective urine
drainage reduces intraluminal pres-
sure and restores the flow of urine.
When should clinicians consider
consultation with a specialist?
Consultation is rarely needed for
acute, uncomplicated cystitis in am-
bulatory women but can be helpful
in complicated UTI, particularly in
patients with indwelling bladder
catheters, those hospitalized because
of upper tract disease, and men with
UTI. Consider consulting an in-
fectious disease specialist when the
organisms isolated in the urine are
resistant to standard antibiotics.
Consultation with an infectious dis-
eases specialist or a urologist may also
be necessary in patients with possible
upper urinary tract involvement who
do not respond to appropriate antibi-
otic therapy within 72 hours. A
surgically correctable lesion may be
2012 American College of Physicians
 present in men who report voiding
difficulties or acute urine retention
or who have early recurrent UTI or
persistent microscopic hematuria.
Stones, strictures, and occult cancer
are among the differential diagnoses
for men with these symptoms.
How should patients treated for
UTI be followed?
In uncomplicated cystitis, no specific
follow-up is required as long as
symptoms resolve. In pregnant
women, obtain urine cultures after
treatment to confirm eradication of
bacteriuria, and obtain repeated uri-
nalyses or urine cultures at intervals
to confirm sterility of the urine
through the time of delivery. The ra-
tionale for repeated screening is that
pregnant women who have already
had bacteriuria during their pregnan-
cy are at increased risk for recurrence
and need closer monitoring than do
women who had sterile urine on ini-
tial screening. However, more evi-
dence is needed to define the appro-
priate frequency of follow-up cultures
and retreatment strategies (10).
It seems prudent to monitor patients
with complicated UTI during and
after therapy, because failure of ther-
apy for cystitis in the presence of un-
derlying complicating conditions is
both more probable and more likely
to produce significant morbidity
than in patients without complicat-
ing conditions. Because complicated
UTI is an extremely heterogenous
disorder, the approach to monitoring
during and after therapy must be in-
dividualized. Patients who are more
fragile, are infected with more diffi-
cult-to-treat organisms, have more
severe symptoms at the outset,
multiple predisposing conditions,
or multiple medication allergies
need closer follow-up than do pa-
tients with only mild symptoms, an
easily-treated organism, a single
minor predisposing condition, and
otherwise good health status. Pa-
tients with complicated UTI should
be followed clinically for sympto-
matic resolution and be reevaluated
if symptoms do not improve within
24 to 48 hours, worsen, or recur
quickly after therapy. In CAUTI, it
is important to monitor response to
therapy by the patients symptoms
rather than by repeated urine cul-
tures, as recurrent bacteriuria is the
norm in patients whose indwelling
catheter remains in place.
What is the correct approach to
secondary prevention in patients
who have a history of UTI?
Recurrent UTI is very common, es-
timated to occur in up to 50% of
women within 1 year of an initial
UTI (3). Other than the antimicro-
bial prophylaxis recommended in
the Prevention section above, there
are few measures that women can
take to prevent recurrent UTI.
As for the often-cited behavioral
risk factors for recurrent UTI, a
large casecontrol study in pre-
menopausal women did not find
any associations between risk for
recurrent UTI and pre- and post-
coital voiding, frequency of urina-
tion, wiping patterns, douching,
use of hot tubs, frequent use of panty-
hose or tights, or daily fluid con-
sumption (2). Women with recurrent
UTI should be counseled about the
true risk factors for UTI so they do
not blame themselves or make un-
necessary behavioral changes.

Treatment and Management... The IDSA has recently released new standard-of-
care guidelines for treatment of acute, uncomplicated cystitis; acute uncompli-
cated pyelonephritis; and catheter-associated UTI. Nonpharmacologic therapies
for acute cystitis do not have proven benefits and may lead to adverse outcomes.
Posttreatment follow-up should include monitoring the response to therapy
rather than repeated urine cultures, except in pregnant women.

CLINICAL BOTTOM LINE

2012 American College of Physicians ITC3-12 In the Clinic Annals of Internal Medicine 6 March 2012

Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013


What measures do U.S. stakeholders
use to evaluate the quality of care
for patients with UTI?
The recent guidelines on acute un-
complicated cystitis and
pyelonephritis published by the
IDSA in 2011 identified several
performance measures as appropri-
ate indicators for management of
acute uncomplicated UTI in
women. Use an available recom-
mended antimicrobial for treatment
of uncomplicated cystitis and un-
complicated pyelonephritis if the
patient is not allergic. Use fluoro-
quinolones for acute uncomplicated
cystitis only when a recommended
antimicrobial cannot be used. Ob-
tain a pretherapy urine culture in
cases of acute uncomplicated
pyelonephritis; start empirical
therapy and modify according to
culture results. In other words, flu-
oroquinolone antibiotics should not
be the first choice agents for acute,
uncomplicated cystitis, and urine
culture results should be used to
guide therapy in pyelonephritis
(27). CAUTI has been in the spot-
light recently because of several
important social changes. On
1 October 2008, the Centers for
Medicare & Medicaid Services
stopped reimbursing U.S. hospitals
for several complications of hospi-
talization, including CAUTI that
develops during hospitalization.
Although the rule has provoked
controversy over whether CAUTI
is truly preventable, the increased
focus on prevention of this disor-
der, particularly through decreasing
unnecessary use of urinary
catheters, is likely to be beneficial
(36). Around the same time, the
focus of infection control agencies
shifted from control of hospital-
acquired infections to prevention,
as summarized in the mission
statement of the Targeting Zero
campaign of the Association for
Professionals in Infection Control
and Epidemiology (37). The other
6 March 2012 Annals of Internal Medicine
Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013
Practice
important social change affecting
Improvement
CAUTI research and prevention
efforts has been the upsurge in
public reporting of hospital quality-
of-care data. Although CAUTI is
not among the standard outcome of
care measures, prevention of
CAUTI has been announced as a
2012 National Patient Safety Goals
by the Joint Commission. Therefore,
monitoring for appropriate urinary
catheter care and use is rapidly be-
coming standard hospital policy.
What do professional organizations
recommend regarding care of
patients with UTI?
Professional organizations in both
the United States and internation-
ally have released practice guide-
lines concerning UTI that address
3 categories of UTI: acute cystitis,
CAUTI, and asymptomatic bac-
teriuria. The IDSA published the
Guidelines for Antimicrobial
Treatment of Acute Uncomplicated
Cystitis and Pyelonephritis in
Women in 2011. These guidelines
focus on treatment of women with
acute uncomplicated cystitis and
pyelonephritis, and their recom-
mendations are limited to pre-
menopausal, nonpregnant women
with no known urologic abnormali-
ties or comorbid conditions. An
important and unifying theme in
these guidelines is that the issues of
in vitro resistance prevalence and
the potential for collateral damage
by various antimicrobial agents are
taken into account in the recom-
mended treatment choices.
CAUTI has received considerable
press lately. Guidelines concerning
management and prevention of
this condition have been released
by the Centers for Diseases Con-
trol Healthcare Infection Control
Practices Advisory Committee, by
the Association for Professionals in
Infection Control and Epidemiolo-
gy in its Compendium of Strategies
to Prevent Healthcare-Associated
In the Clinic ITC3-13 2012 American College of Physicians
 Infections in Acute Care Hospitals,
and also by the IDSA. This topic
is also addressed in the very com-
prehensive European Association
of Urologists Guidelines on Uro-
logical Infections released in 2009.
Asymptomatic bacteriuria guide-
lines were published by the IDSA
in 2005. Abundant evidence sup-
ports nontreatment of asympto-
matic bacteriuria (with the
exceptions of pregnant women
and patients undergoing urologic
procedures). These IDSA guide-
lines recommendations were
reviewed and endorsed by the
U.S. Preventive Services Task Force
in 2008. Specifically, the Task
Force recommended against screen-
ing for asymptomatic bacteriuria
in men and nonpregnant women
and recommended for screening
for asymptomatic bacteriuria
with urine culture for pregnant
women at 12 to 16 weeks gestation
or at the first prenatal visit, if
later.

PIER Module
In the Clinic

In the Clinic
http://pier.acponline.org/physicians/diseases/d162/d162.html

Tool Kit
PIER module on UTI from the American College of Physicians
(ACP). PIER modules provide evidence-based, updated
information on current diagnosis and treatment in an electronic
format designed for rapid access at the point of care.
Patient Information
www.annals.org/intheclinic/toolkit-uti.html
Urinary Tract Patient information that appears on the next page for duplication
and distribution to patients.
Infection www.nlm.nih.gov/medlineplus/ency/article/000483.htm
www.nlm.nih.gov/medlineplus/spanish/ency/article/000483.htm
Information on catheter-associated UTIs from the National
Institutes of Healths MedlinePLUS, in English and Spanish.
http://familydoctor.org/online/famdocen/home/women/
gen-health/190.html
Frequently asked questions on UTIs in women from the American
Academy of Family Physicians.
http://kidney.niddk.nih.gov/kudiseases/pubs/uti_ez/ (English)
http://kidney.niddk.nih.gov/spanish/pubs/uti_ez/ (Spanish)
A handout titled What I need to know about Urinary Tract
Infection from the National Institute of Diabetes and Digestive
and Kidney Diseases (NIDDK), in English and Spanish.
Clinical Guidelines
www.annals.org/content/149/1/W-20.full
Recommendation statement on screening for asymptomatic
bacteriuria in adults from the U.S. Preventive Services Task Force.
http://cid.oxfordjournals.org/content/50/5/625.full
International clinical practice guidelines from the Infectious Diseases
Society of America on the diagnosis, prevention, and treatment of
catheter-associated UTI in adults, released in 2009.
Diagnostic Tests and Criteria
http://pier.acponline.org/physicians/diseases/d162/tables/d162-tlab.html
Table listing laboratory and other studies for acute cystitis in women.
Quality of Care Guidelines
http://qualitymeasures.ahrq.gov/browse/by-topic-detail.aspx?
id=13254&ct=1&term=urinary
AHRQ quality indicators for UTI, including measures on the hospital
admission rate of UTI and the percentage of nursing home
residents with UTI.
www.annals.org/content/144/2/116.full
Systematic review on antimicrobial urinary catheters to prevent
catheter-associated UTI in hospitalized patients published in
Annals of Internal Medicine in 2006.

2012 American College of Physicians ITC3-14 In the Clinic Annals of Internal Medicine 6 March 2012

Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013

 THINGS YOU SHOULD In the Clinic
Annals of Internal Medicine
KNOW ABOUT URINARY
TRACT INFECTION

What is a urinary tract infection (UTI)?

A UTI is a common infection of the urinary system
(i.e., urethra, bladder, ureters, and kidneys).
It occurs when bacteria at the opening of the urethra
ascend to the bladder.
Infection limited to the urethra is called urethritis.
Infection that moves from the urethra to the bladder
is called cystitis.
Infection that moves further up the urinary tract to
the kidneys is called pyelonephritis.

What factors increase the risk for UTI?

Being a sexually active adult woman.
Using a diaphragm or spermicide.
Having an abnormality of the urinary tract that
obstructs the flow of urine (such as a kidney stone
or enlarged prostate gland).
Having a catheter (a tube inserted in the urethra
and bladder of people who cannot pass urine or who
lack bladder control).
Having poorly controlled diabetes in postmenopause.

What are common symptoms?

Painful urination and feeling an urgent need to
urinate frequently.
Urine that contains blood or looks cloudy.
Discomfort in the lower abdomen and pain in the
back or pelvic area.
Fever may develop.
If you have recurrent or resistant infection, your doc-
tor may order tests to determine if your urinary sys-
tem is normal.
How is it treated?

How is it diagnosed?
Your doctor may diagnose UTI based on your
symptoms or may do additional tests.
Your urine is checked for evidence of infection with
a test called a urinalysis or dipstick.
A urine culture for bacteria if you are pregnant or
are likely to have a kidney infection or resistant
bacteria.
Patient Information
Your doctor will prescribe an oral antibacterial drug.
It is important to take the full course of treatment
even if your symptoms disappear.
Patients with severe kidney infections may need to
be hospitalized and receive IV treatments.
If you have recurring infection, you may be advised
to take low doses of an antibiotic daily for many
months or to take a single dose of an antibiotic after
sexual intercourse to prevent infection.

For More Information

www.urologyhealth.org/urology/index.cfm?article=47
Information on urinary tract infections in adults from the
American Urological Association.

http://familydoctor.org/online/famdocen/home/women/
gen-health/284.html
Answers to questions about painful urination, from the American
Academy of Family Physicians.

http://womenshealth.gov/publications/our-publications/fact-sheet/
urinary-tract-infection.cfm
Urinary tract infection fact sheet from the U.S. Department of
Health and Human Services Office on Womens Health.

Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013

 CME Questions

1. A 64-year-old woman is evaluated for
symptoms of a urinary tract infection
(UTI). She has had 3 UTIs in the past
2 years. She is not sexually active. She
has no other medical problems and takes
no medications. A pelvic examination
reveals pale, dry vaginal epithelium that
is smooth and shiny with loss of most
rugation. Urinalysis reveals 2+ leukocyte
esterase, leukocytes too numerous to
count, and 10 to 20 erythrocytes/hpf.
Urine culture grows escherichia coli.
In addition to treating the current UTI,
which of the following is the most
reasonable management option for this
patient?
A. Continuous antibiotic prophylaxis
B. CT imaging of the abdomen and
pelvis
C. Topical estrogen therapy
D. Vaginal lubricants
2. A 65-year-old woman is evaluated
because a screening urine culture for
an insurance policy grows greater than
105 colony-forming units/mL of E. coli.
She does not have fever, dysuria, urinary
frequency, or other symptoms. Medical
history is unremarkable. She has no
allergies and takes no medications.
Physical examination findings are normal.
Which of the following is the most
appropriate treatment?
A. Amoxicillin
B. Ciprofloxacin
C. Trimethoprim-sulfamethoxazole
D. No treatment
3. An otherwise-healthy 28-year-old
woman has had 2 episodes of acute
cystitis within the past 6 months. The
patient is sexually active and has
intercourse with her husband on average
2 times per week and says her cystitis
does not seem to be intercourse-related.
Each time, symptoms remit after a
single course of trimethoprim-
sulfamethoxazole. The patient is
currently asymptomatic but will be
traveling abroad for the next 2 months
and is concerned about recurrent
infections. Her only medication is an oral
contraceptive for birth control. She
reports no allergies.
Which of the following is the most
appropriate management?
A. Ciprofloxacin after intercourse
B. Ciprofloxacin for 10 days when
symptoms develop
C. Trimethoprim chronic suppressive
therapy
D. Trimethoprim-sulfamethoxazole for
3 days when symptoms develop
4. A 32-year-old sexually active woman
with type 1 diabetes mellitus is evaluated
because of recurrent UTIs. She has had
3 episodes this year. The most recent
episode occurred 2 weeks ago. Physical
examination, including vital signs, is
normal. Urinalysis is normal except for
the microscopic examination, which
shows 4+ bacteria.
Which of the following management
strategies is most appropriate at this
time?
A. Patient-initiated empiric antibiotic
therapy
B. Continuous standard-dose antibiotic
therapy
C. Urinalysis and culture at the onset of
dysuria
D. Postcoital empiric antibiotic therapy

Questions are largely from the ACPs Medical Knowledge Self-Assessment Program (MKSAP, accessed at
http://www.acponline.org/products_services/mksap/15/?pr31). Go to www.annals.org/intheclinic/
to complete the quiz and earn up to 1.5 CME credits, or to purchase the complete MKSAP program.

2012 American College of Physicians ITC3-16 In the Clinic Annals of Internal Medicine 6 March 2012

Downloaded From: http://annals.org/ by a Pontificia Universidad Ca User on 01/14/2013