The epidemiology of urinary tract infection

Abstract

Urinary tract infections (UTIs) are among the most common bacterial infections acquired in the community and in hospitals. In individuals without anatomical or functional abnormalities, UTIs are generally self limiting, but have a propensity to recur. Uropathogens have specialized characteristics, such as the production of adhesins, siderophores and toxins that enable them to colonize and invade the urinary tract, and are transmitted between individuals both through person-to-person contact and possibly via food or water. Although generally self limiting, treatment of UTIs with antibiotics leads to a more rapid resolution of symptoms and is more likely to clear bacteriuria, but also selects for resistant uropathogens and commensal bacteria and adversely affects the gut and vaginal microbiota. As uropathogens are increasingly becoming resistant to currently available antibiotics, it may be time to explore alternative strategies for managing UTI.

Full text

Translate

Turn on search term navigation

REVIEWS

The epidemiology of urinary tract infection

Betsy Foxman

Abstract | Urinary tract infections (UTIs) are among the most common bacterial infections acquired in the community and in hospitals. In individuals without anatomical or functional abnormalities, UTIs are generally self limiting, but have a propensity to recur. Uropathogens have specialized characteristics, such as the production of adhesins, siderophores and toxins that enable them to colonize and invade the urinary tract, and are transmitted between individuals both through person-to-person contact and possibly via food or water. Although generally self limiting, treatment of UTIs with antibiotics leads to a more rapid resolution of symptoms and is more likely to clear bacteriuria, but also selects for resistant uropathogens and commensal bacteria and adversely affects the gut and vaginal microbiota. As uropathogens are increasingly becoming resistant to currently available antibiotics, it may be time to explore alternative strategies for managing UTI.

Foxman, B. Nat. Rev. Urol. 7, 653660 (2010); http://www.nature.com/doifinder/10.1038/nrurol.2010.190

Web End =doi:10.1038/nrurol.2010.190

Introduction

The urethra is a portal for the exit of urine, but also allows the entry of microbes, including pathogens, into the urinary tract. Bacteria live around the urethral opening in both men and women and routinely colonize urine in the urethra, but are washed out during micturition. The shorter distance to the bladder in women makes it possible for bacterial colonizers to reach the bladder more easily before they are removed by micturi tion. In addition, the urethral opening in women is proxi mate to the vaginal cavity and rectum, which harbor large bacterial communities. Urogenital manipulations associ ated with activities of daily living and medical interventions facilitate the movement of bacteria from the vaginal cavity, rectal opening and periurethal area into the urethra; however, even if bacteria reach the bladder and multiply to significant numbers, only rarely does bacterial coloniza tion result in symptoms. Given human anatomy, it is not surprising that urinary tract infections (UTIs) are among the most common bacterial infections; indeed, we should perhaps be surprised that, given the constant assault on our urinary system by microbes, UTIs are not more common.

Both host and bacterial factors influence the probability that asymptomatic colonization resolves spontaneously or progresses to symptomatic infection (Table 1). Host factors include the presence of anatomical or functional abnormalities, genetic predisposition, and behaviors that increase exposure to uropathogens or that move bacteria into the bladder (such as sexual intercourse). Bacterial factors include a variety of virulence characteristics that enable the pathogen to move to and colonize the bladder and evade the human immune system.

In this Review, I first describe the difficulties in defining UTI, and provide an overview of the bacteriology of

UTI. I then describe how the majority of UTIs can be attributed to a highly heterogeneous group of Escherichia coli that are specifically adapted to colonizing the urinary tract, and how sexual activity increases the risk of acquiring uropathogens and facilitates their movement into and colonization of the bladder. Given the selflimiting nature of lower UTI, the increasing prevalence of bacterial resistance to antibiotics, and results of placebocontrolled trials, I argue that alternative management strategies seem to be in order for patients with uncompli

cated UTI.

The Review closes with a comment on the contribution of epidemiology to our understanding of the trans

mission

and pathogenesis of UTI.

Definition of UTI

UTIs are classified as either lower (confined to the bladder) or upper (pyelonephritis), and as either uncompli cated or complicated. An uncomplicated UTI is one occurring in a normal host who has no structural or functional abnormali ties, is not pregnant, or who has not been instrumented (for example, with a catheter). All other UTIs are considered complicated. This Review focuses on uncompli

cated UTI and catheterassociated UTI.

UTI is diagnosed using a combination of urinary symptoms and urine culture demonstrating numbers of a known uropathogen above a given threshold (usually defined as >1,000 cfu/ml of urine,1 but thresholds as low as 100 cfu/ml and as high as 100,000 cfu/ml are also used2). However, urinary symptoms and bacteri uria frequently occur independently of each other: ~20% of women presenting with classic UTI symptoms have negative urine cultures.3 Significant numbers of bacteria are often found in the urine of otherwise healthy, asymptomatic individuals.4 The risk of asymptomatic bacteriuria increases following sexual activity and with advancing age. Although asymptomatic bacteriuria increases the risk of symptomatic UTI, it should not be treated except

Department of Epidemiology, University of Michigan School of Public Health, 1415 Washington Heights, 5108 SPH Tower, Ann Arbor,MI 48109, USA. mailto:[email protected]

Web End [email protected]

Competing interestsThe author declares no competing interests.

NATURE REVIEWS | UROLOGY VOLUME 7 | DECEMBER 2010 | 653

REVIEWS

the selflimiting nature of uncomplicated UTI, suggest that UTIs are overdiagnosed and overtreatedespecially as urine cultures are usually not performed for uncomplicated UTI.

UTI secondary to catheterization is the most common nosocomial infection. Indeed, asymptomatic bacteri uria is almost universal in individuals catheterized for any length of time, with an incidence of significant bacteriuria following catheterization of 5% per day.12 In a

prospective study of 1,497 catheterized patients, there were no significant differences in the frequency of pain, urgency, dysuria or fever among those with and without bacteriuria.13 Mean leukocyte counts were signifi

cantly

higher among those with catheterassociated UTI; however, pyuria (defined as a white blood cell count greater than 10 cells/ml , or >5 cells per highpower field in conventional urinalysis), although fairly specific (90%), was not a very sensitive (37%) predictor of signifi

cant bacteriuria.14

Guidelines published by the Infectious Disease Society of America define catheterassociated UTI as the presence of symptoms (new onset or worsening of fever, rigors, altered mental status, malaise, or lethargy with no other identified cause; flank pain; costovertebral angle tenderness; acute hematuria; or pelvic discomfort) and >1,000 cfu/ml of one or more bacterial species.15 The

guidelines recommend against screening for or treating asymptomatic bacteriuria in catheterized patients.

Bacteriology

The bacteriology of UTI is very predictable: although a number of different species can cause UTI, the majority of infections in all populations are caused by the Gramnegative, facultative anaerobic, uropathogenic E. coli (UPEC). Among otherwise healthy women aged 1839 years, 80% of UTI are caused by E. coli.16 UPEC

are different from the E. coli strains that normally inhabit the gastrointestinal tract, in that they are better adapted to living within the urinary tract and evading the hosts immune response. Biofilm formation and urothelial cell invasion17 are two mechanisms that contribute to the success of UPEC in this regard. UPEC are very heterogeneous, and as yet there are no tests that enable us to determine whether an E. coli strain is UPEC if it is not isolated from the urine. Other organisms that cause UTI include members of the family Enterobacteriaceae, strepto cocci and staphylococci species, and yeast.

In a 20042005 laboratorybased surveillance of communityacquired UTI in the Calgary Health Region of Canada, 40,618 UTI episodes were recorded among 30,851 residents (annual incidence of 17.5 episodes per 1,000 people). Most (74%) of the specimens were obtained from ambulatory patients, with 18% from hospitalized patients whose infection occurred within 2 days of hospitalization and the remainder (9%) from nursing home residents. Using culturebased methods, the most common bacterial isolate in all groups was E. coli (74.2% of ambulatory, 65.5% of hospitalized and 46.6% of nursing home patients).18 Among ambulatory patients, only Klebsiella pneumoniae (6.2%),

Key points

Urinary tract infection (UTI) is diagnosed using a combination of urinary

symptoms and urine culture; ~20% of women presenting with symptoms indicative of UTI will have a negative urine culture

Escherichia coli

are the bacteria most frequently implicated in uncomplicated UTI and catheter-associated UTI, and are becoming increasingly resistant to antibiotics

Positive urine culture in the absence of symptoms should not be treated, except

in pregnant women or those undergoing invasive genitourinary procedures

Table 1 | Risk factors for uncomplicated and catheter-associated UTI

Category Uncomplicated UTI Catheter-associated UTI

Transmission of uropathogens

Sexual intercourse Food or water?

Catheter insertionHands of hospital personnel

Host factors Female sexPrior UTIFrequency of sexual intercourse Vaginal infectionCondom useSpermicide useGenetic susceptibility

Female sexPrior UTIGenetic susceptibility Trauma

Bacterial factors Adherence factors Siderophores Bacteriocins ToxinsBiolm formation

Adherence factors Siderophores Bacteriocins ToxinsBiolm formation

Abbreviation: UTI, urinary tract infection.

in those who are pregnant or undergoing invasive genitourinary procedures,5 as symptomatic UTI often follows in these patients. Furthermore, treatment of asymptomatic bacteriuria can potentially cause harm by adversely affecting the individuals microbiota6 and selecting for antibioticresistant organisms (described below). Urinary symptoms, particularly increased frequency and urgency, are also very common, and often occur independently of infection.7 Asymptomatic microhematuria has been observed among otherwise healthy women and men in the absence of infection, with no apparent cause or longterm effects.810

Given the frequency of bacteriuria and urinary symptoms in the population, an individual without UTI may have both urinary symptoms and bacteriuria by chance alone. This can result in overdiagnosis, particularly if treatment is based on the presence of symptoms alone or even symptoms plus results of a dip stick urinalysis. Schmielmann et al.11 reviewed the sensitivity and specificity of various diagnostic algorithms for UTI, and found that algorithms combining symptoms and dip stick urinalysis results vary widely in sensitivity (6582%) and specificity (5395%). Using the most sensitive criterion for urine culture (>100 cfu/ml), an algorithm that included urinary symptoms and detection of leukocytes and nitrate in the urine had a sensitivity of 80.3% and a specificity of 53.7%; in other words, 20% of patients with a positive culture are missed, and almost half of those with negative cultures are unnecessarily treated. The low sensitivity and specificity of symptoms in predicting bacteriuria, and the transient nature of asymptomatic bacteriuria combined with

654 | DECEMBER 2010 | VOLUME 7 www.nature.com/nrurol

REVIEWS

Ambulatory patients

Hospital patients

0.9% 0.8%0.9%

1.4%

2.0%

2.8%

5.3%

6.2%

1.8%

1.6% 1.4%

0.7%

2.2%

1.9%

8.0%

5.5%

8.9%

74.2%

65.5%

Nursing home patients

11.3%

Escherichia coli

Klebsiella pneumoniae

Enterococcus sp.

Streptococcus agalactiae

Proteus mirabilis

Staphylococcus saprophyticus

Viridans streptococci

Klebsiella oxytoca

Pseudomonas aeruginosa

Other

3.2%

0.9%

4.8%

46.6%

10.1%

2.0%

11.4%

9.7%

Figure 1 | The variation in bacteriology of urinary tract infection (UTI) according to population. Results from a 20042005 laboratory-based surveillance study of all community-acquired UTI in the Calgary Health Region of Canada.18 The annual incidence of UTI in this region (based on positive urine cultures) was 1.75%.

Enterococcus species (5.3%), Group B Streptococcus (also known as Streptococcus agalactiae) (2.8%), Proteus mirabilis (2.0%) and Staphylococcus saprophyticus (1.4%) accounted for more than 1% of the remainder. By contrast, these other species also occurred among both hospitalized and nursing home patients but at a higher frequency, and Viridans streptococci, Klebsiella oxytoca, Pseudomonas aeruginosa, Citrobacter freundii, Enterobacter cloacae, and Staphylococcus aureus accounted for 1% or more (Figure 1).

PCR can be used to detect species that are not cultivable using standard clinical microbiology methods. A German study screened consecutive urine samples from 1,449 patients at a university hospital with both culture

and nonculture (PCR) techniques.19 Of these samples, 128 were culturepositive and PCRpositive, and 37 were culturenegative but PCRpositive. While the majority of samples contained E. coli (35%), Enterococcus faecalis (19%) or other Enterobacteriaceae, or Streptococcus or Staphylococcus species (23%), 22% were fastidious bacteria detected only by PCR. The fastidious bacteria were mostly obligate anaerobes. Furthermore, in contrast to results using culture techniques, where ~95% of cultures find only one organism in uncomplicated UTI, 38% of the specimens tested by PCR contained two or more species. The most common combinations were two Gramnegative species, such as E. coli and E. faecalis.

NATURE REVIEWS | UROLOGY VOLUME 7 | DECEMBER 2010 | 655

REVIEWS

Transmission

As E. coli and other Enterobacteriaceae that cause UTI normally colonize the gastrointestinal tract, it had been presumed that UTIs result from movement of normal bowel bacteria into the urinary tract. However, it is now clear that in the otherwise healthy individual, bowel bacteria that make their way into the urinary tract will be rapidly cleared. UPEC, the uropathogens most commonly associated with UTI, are adapted to the urinary tract and have specialized features that enable them to initiate and maintain colonization in this environment, reaching large numbers in the urine (exceeding 100,000 cfu/ml). Presumably, if uropathogens move from the vaginal cavity and gastrointestinal tract into the bladder, they may also move in the opposite direction. Given the large numbers that grow in the urine, it is likely that hands become contaminated during urination, which can directly transfer the uropathogen to a partner during intimate activity or indirectly transmit the uropathogen via food or water. Sexual activity may lead to direct transmission from the vaginal cavity to the urethra of a sex partner (and vice versa), or indirectly during oral sex (Table 1).

UTI has long been associated with sexual activity; hence, uncomplicated UTI has been dubbed honeymoon cystitis. Medical and epidemiologic evidence supports this bit of folklore. Several medical case reports of sexual transmission exist in the literature,20,21 and, at a population level, UTIs occur most frequently among women aged 1829 yearsa time of life when women are most likely to be initiating sexual activity. Recent, frequent vaginal intercourse is a major risk factor in this age group.22 If

a woman has a UTI caused by UPEC, the same UPEC is twice as likely as an E. coli strain isolated from her rectal specimen to be found in a urethral or rectal specimen of her most recent sexual partner.23 Male urethral coloniza tion with the same UPEC for a duration as long as 16 days has been documented.24 Moreover, condom use is associated with a decreased risk of initial UTI;24

however, trauma associated with condom use seems to increase the risk of recurrence. These data, along with the increased risks associated with a higher frequency of sexual intercourse and with a new sex partner, suggest that UPEC is transmitted by sexual contact.

If UPEC is sexually transmitted, then we might expect to observe outbreaks or epidemics (that is, more cases than expected in a given geographic area in the observed time period). UPEC is a very large, hetero geneous group of organisms with a high background incidence, which has made it difficult to identify and track outbreaks. However, unusual antibiotic resistance profiles provide useful markers of epidemics. One outbreak occurred in South London, UK in the late 1980s, and was associated with a multidrugresistant strain of E. coli serotype O15.25 Since 2000, an extended spectrum lactamaseproducing strain of E. coli, sequence type (ST) 131, has been identified as a major source of communityacquired UTI, with resistance conferred by the CTXMtype lactamase enzyme.26 ST is determined using a sequencingbased molecular typing method known as

multi locus sequence typing. In a chartreview study of patients with communityacquired UTI caused by CTXM15producing E. coli in New Zealand between 2004 and 2006, 10 of the 11 patients whose acquisition could not be attributed to hospitalization in the previous 6 months reported recent travel to India.27 India has high prevalences of E. coli strains with CTXMmediated resistance.28 While not definitive, this does suggest possible transmission of UPEC via food or water or through persontoperson contact during travel. Although there have been some studies suggesting food as a possible source for UTI, the data are inconclusive at present.29

Epidemics of nosocomial UTI have also been observed, again identified by unusual antibiotic resistance profiles. Most are caused by the transmission of outbreak strains between patients on the hands of hospital personnel.30 The transmission cycle can be broken by proper hygiene practices that interrupt spread by contact and by diminishing the pathogen reservoirthe urine of patients with indwelling urinary catheters. Sporadic nosocomial UTI is primarily attributable to catheterization, but urogenital manipulation during surgery or medical care also increases the risk of transmission. There are two sources of catheterassociated UTI: the bacteria found in the perineal or rectal microbiota of the patient, and those present on the hands of the healthcare professional inserting or manipulating the instrument; therefore, good hygiene practices and removing catheters as quickly as possible reduce the risk of catheterassociated UTI.15,31

Prevalence and incidence

The prevalence of bacteriuria has a Jshaped distribution, with a higher frequency among the very young and a gradual increase with age in both men and women. Until the age of 60 years and older, the prevalence is significantly higher for women than men.32 The distribution of symptomatic infection has a somewhat different shape: women aged 1529 years have the highest frequency (approaching 20%).33 The annual incidence of UTI in the USA was estimated using selfreported UTI history for the previous year in the National Health and Nutrition Examination Survey (NHANES). Among women aged 18 years and older, the estimated incidence was 12.6%; for men, this incidence was only 3%.32 These percentages are an order of magni tude higher than incidence estimates obtained using laboratory surveil lance data: in a populationbased study conducted using laboratory surveillance of inpatient and outpatient visits in the Calgary Health Region of Canada during 20042005, the overall annual incidence was 1.75%.18 The differences in estimates are most likely attributable to the use of urine culture in the Canadian study rather than selfreported doctor diagnosis, as most lower UTIs in the USA are treated empirically without the benefit of culture findings.

An estimated 1 million cases of nosocomial UTIs occur in the USA annually,13 of which 80% are attributed to catheters. Catheterassociated UTIs account for 40% of all hospitalacquired infections. A 20072008

656 | DECEMBER 2010 | VOLUME 7 www.nature.com/nrurol

REVIEWS

surveillance study conducted at 34% of hospitals in The Netherlands found that 1.2% of patients had a catheterrelated UTI and 1.7% had a symptomatic UTI.34 In a

prospec tive study of 1,497 patients catheterized for at least 24 h and followed daily for 14 days, the incidence of UTI was 15.6%.13 The risk of a symptomatic UTI is also high following removal of the catheter. In a study to assess postoperative complications in 262 women who were randomized to either catheterization or noncatheterization prior to gynecological laparoscopic surgery, the rate of postoperative UTI after 7 days was 10% for the catheterized group compared to 3.8% in the noncatheterized group.35 The rate in the noncatheterized group is more than an order of magnitude higher than the estimated population rates for US women (0.24% per week based on the NHANES data32).

Morbidity

Lower UTIs resolve quickly: in an observational study of 511 nonpregnant adult women (aged 1870 years) presenting with cultureconfirmed UTI treated in primary care, the mean duration of urinary frequency was 3.46 days (SD 2.59), 1.88 days (SD 1.75) for hematuria, and 3.06 days (SD 2.54) for urgency. The mean duration of symptoms was 12 days longer among those with a UTI caused by an antibioticresistant organism or in those who were not treated, but the differences were not statistically significant.36 Furthermore, the risk of cystitis progressing to pyelonephritis is low, and the risks are similar among those who are treated with antibiotics and those who are not: a small placebo controlled trial reported only one case in the placebo group (1 [2.6%] of 38 patients),37 and a larger study reported one case in the placebo group (1 [0.4%] of 227 patients) and one case in the treatment group (1 [0.15%] of 657 patients).3

A metaanalysis of five treatment trials found a similarly low incidence of pyelonephritis (2 [0.34%] of 582 patients). The estimates from the various studies were not statistically different.37,38

Despite the low morbidity, however, the high incidence of UTI results in considerable healthcare costs on a population scale: the estimated annual direct and indirect cost of UTI in the USA in 1995 was $1.6 billion (equivalent to $2.3 billion in 2010, adjusted for inflation).39 Furthermore, in 2000, a nosocomial UTI was found to add on average $676 to a hospital bill (equivalent to $842 in 2010), and necessitates one extra hospital day per patientnearly 1 million extra hospital days each year.40 In addition, the rate of bloodstream infection (bacteremia) associated with shortterm catheter use in prospective studies is ~1%,13 a complication that is estimated to cost $2,832 per episode (equivalent to $3,530 in 2010).40

UTIs have a propensity to recur. Among otherwise healthy college women with a first UTI, the risk of a second episode within 6 months was 24%,22 and in those with a history of one or more UTIs, the risk of a second within 1 year was 70%.41 In the Canadian surveillance study,18 14% of the 30,851 residents with UTI had more than one episode during the 2year study period, and 2%

had six or more episodes. A similar pattern is observed for acute pyelonephritis.42 Some recurrences may be attributable to the ability of UPEC to evade the immune system by creating intracellular bacterial communities43 or by maintaining a reservoir in the vaginal cavity or gastrointestinal tract, despite the fact that UPEC do induce adaptive immune responses.44,45 However, if the host can become immune, why do as many as 5% of women with a first UTI experience at least 1 year with three or more symptomatic episodes? Although recurrences in this highrisk group are still most often caused by UPEC (66%), and many different UPEC strains exist, recurrences are not solely a function of the infecting microbe.46 Human genetic variations, particularly those affecting the immune response, have been associated with an increased risk of recurrent UTI.4750 One

hypothesis is that the initial, index infection leads to a resetting of the host response, so thatlike an oversensitive smoke detectorsymptoms are generated when faced with the same bacteria that would, before the index infection, have caused only a transient and asymptomatic coloniza tion of the urinary tract. Over time, the response drifts back to its preindex UTI level. This hypothesis is consistent with the observation that the longer a woman remains symptom free, the more likely she is to remain so. Anecdotal evidence seems to suggest that recurring UTI is cleared when women undergo a period of sexual abstinence or become pregnant, or when they change their sex partners. An alternative hypothesis suggests that while the host defense mechanism exfoliates bladder cells bound by bacteria,17 the bladder is more susceptible to attack by other pathogens until it is able to fully heal.

Antibiotic resistance

Even short courses of antibiotics can profoundly affect the microbial community of the vaginal cavity51 and gastrointestinal tract.6 Vaginal yeast infections, for example, are a relatively common sequela of antibiotic therapy for UTI.52 After a 5day course of ciprofloxacin, several taxa were lost from the gut microbiota of healthy volunteers who had not taken antibiotics in the previous year; some taxa did not reappear for as long as 6 months.6 As the second most common reason for prescribing antibiotics (following otitis media), UTI prescribing has a considerable effect on the normal microbiota. Even ignoring the effects on those taxa that are present in this microbial community, antibiotic use selects for resistant pathogens and commensal organisms: a major risk factor for an antibioticresistant UTI is prior antibiotic use.52,53

Antibiotic resistance initially imposes a fitness cost on a bacterium (such as a slower growth rate), which suggests that resistant bacteria should revert to sensi tivity once the selecting antibiotic is removed from its environment. Further study has suggested that this reversion rarely occurs: bacteria seem to rapidly evolve mechanisms to compensate for any fitness costs imposed by resistance. For example, in an interventional trial performed in Sweden, restricting the use of trimethoprimcontaining drugs for 24 months did not result in a loss of resistance to this antibiotic in E. coli isolates from patients

NATURE REVIEWS | UROLOGY VOLUME 7 | DECEMBER 2010 | 657

REVIEWS

Levooxacin

Ciprooxacin

Nitrofurantoin

Trimethoprim/sulfamethoxazole

Ampicillin

13.0

12.4

8.6

26.4

48.7

0 20 40 60 80 100

Figure 2 | Geographical variation in the prevalence of uropathogenic E. coli resistant to selected antibiotics. Mean prevalences for each antibiotic are represented by bars, with error bars showing the range of values reported. Data taken from three international surveillance studies performed in North American, South American and European populations.5557

Resistant E. coli isolates (%)

75.0%

60.6%

Bleidorn et al. (2010)63

Ferry et al. (2004)3

Christiaens et al. (2001)37

25.0%

Placebo

Treatment

55.5%

54.0%

88.0%

Symptomatic cure at 7 days

0 10 20 30 40 50

Cure rate (%)

60 70 80 90 100

Bacteriologic cure at 7 days

48.5%

Bleidorn et al. (2010)63

Ferry et al. (2004)3

Christiaens et al. (2001)37

71.9%

34.0%

89.0%

41.0%

74.0%

Figure 3 | Symptomatic and bacteriologic cure rates at 7 days in two placebo-controlled trials (pivmecillinam versus placebo3 and nitrofurantoin versus placebo37) and an equivalency trial (ciprofloxacin versus ibruprofen [placebo]63), all performed in women with uncomplicated lower UTI. As three different treatment regimens were included in the pivmecillinam study, the mean cure rate of all treatment groups combined is shown here. Statistically significant differences between treatment and placebo groups (asterisks) were calculated using an intention-to-treat analysis (P <0.05).

with UTI.54 As many antibiotic resistance mechanisms are obtained via horizontal gene transfer (for example, transmitted on a plasmid), a bacterium may acquire resistance to multiple antibioticseven those of different classesat once. Thus, using one antibiotic can select for resistance to several other antibiotics. Furthermore, some resistance mechanisms provide resistance to multiple classes of antibiotics.

A summary of several international surveillance systems that have reported the frequency of UPEC resistance to selected antibiotics in North and South America and Europe shows that there is considerable local variation in resistance (Figure 2). Ampicillin resistance among uropathogens worldwide ranges from a low of 33% in Quebec, Canada to almost 80% in Mexico.5559 Resistance

to trimethoprim, the firstline antibiotic therapy for uncomplicated UTI, ranges from 9% in Quebec to 61% in Mexico. Resistance to ciprofloxacin remains low in some areas (0% in Nova Scotia, Canada) but is found in 72% of isolates in Mexico; in the USA, 510% of uropathogens are now resistant to ciprofloxacin.57 Although surveillance data such as these tend to overestimate resistance (only recurrent and complicated UTI tend to be cultured and recorded), these trends are troubling. Also troubling is the emergence among patients with community acquired UTI of E. coli possessing the extendedspectrum lactamase enzymes, which confer resistance against penicillins, thirdgeneration cephalo

sporins and monobactams.6062 Alternative UTI therapies

Lower UTI are generally self limiting (Figure 3). In one placebocontrolled trial in 78 nonpregnant women, the intentiontotreat analysis indicated that there was no difference in the rate of symptomatic cure at 7 days between patients treated with nitrofurantoin and those who received placebo, although bacteriologic cure rates were significantly higher in those treated with nitrofurantoin.37 A much larger trial by Ferry et al.3 compared three regimens of pivmecillinam (which differed in terms of dosage and duration) to each other and placebo in 1,143 nonpregnant women. At 47 weeks, bacteriologic cure rates were 8389% in the active treatment groups and 70% in the placebo group; symptomatic cure rates at 57 weeks were 5962% in the pivmecillinam groups and 51% in the placebo group.

Interestingly, in a 2010 equivalency trial that compared ciprofloxacin with ibuprofen in 80 nonpregnant women with communityacquired UTI in Germany, symptomatic cure rates at 7 days were higher in the ibuprofen group, and the proportion of patients with unresolved symptoms who were switched to an alternative antibiotic was similar in the two groups.63 The results of this study are thoughtprovoking: given that anti biotic therapy for UTI is not without adverse effects, that uncomplicated UTI will spontaneously resolve (albeit at a slower rate than with antibiotic therapy), and that resistance to antibiotic therapy is increasing, it may be time to reconsider standard UTI therapy in order to preserve the effectiveness of our current antibiotics. One possible strategy that might be explored is one of watchful waiting, with analgesics prescribed for symptomatic relief, in a manner similar to that suggested for otitis media in children. This would effectively return UTI treatment to protocols followed prior to the institution of empiric therapy, where treatment was not prescribed until results of urine culture were available, which typically takes 23 days.

658 | DECEMBER 2010 | VOLUME 7 www.nature.com/nrurol

REVIEWS

Conclusions

Epidemiology focuses on population distributions and the determinants of that distribution. In the context of UTI, epidemiologic studies have been essential for identi fying behavioral and bacterial risk factors for trans

mission of causative organisms, demon strating trans mission between sex partners and personto person spread within hospitals, and possible trans mission via food or water from human or animal sources. By examining the distribution of microbes that cause UTI, epi demiologic studies have demonstrated that, although UTIs are primarily caused by E. coli, these E. coli are markedly different from commensal E. coli and are highly hetero geneous at the genetic level. Surveillance is another important epidemiologic activity, as monitoring the emergence and spread of anti

biotic resistance among uropathogens informs treatment decisions. Furthermore, as described above, investigating outbreaks of uropathogens with unique antibiotic resistance patterns has identified new risk factors and possible transmission modes.

Review criteria

PubMed was searched for full-length, English-language articles using the following search terms: urinary tract infection, cystitis, uropathogenic Escherichia coli, and catheter-associated urinary tract infection combined with placebo-controlled clinical trial, epidemiology, prevention, surveillance, antibiotic resistance, and extended spectrum beta-lactamase resistance.

1. Rubin, R. H., Shapiro, E. D., Andriole, V. T., Davis, R. J. & Stamm, W. E. Evaluation of new anti-infective drugs for the treatment of urinary tract infection. Infectious Diseases Society of America and the Food and Drug Administration. Clin. Infect. Dis. 15 (Suppl. 1), S216S227 (1992).

2. Warren, J. W. et al. Guidelines for antimicrobial treatment of uncomplicated acute bacterial cystitis and acute pyelonephritis in women. Infectious Diseases Society of America (IDSA). Clin. Infect. Dis. 29, 745758 (1999).

3. Ferry, S. A., Holm, S. E., Stenlund, H., Lundholm, R. & Monsen, T. J. Clinical and bacteriological outcome of different doses and duration of pivmecillinam compared with placebo therapy of uncomplicated lower urinary tract infection in women: the LUTIW project. Scand. J. Prim. Health Care 25, 4957 (2007).

4. Nicolle, L. E. Asymptomatic bacteriuria important or not? N. Engl. J. Med. 343, 10371039 (2000).

5. Nicolle, L. E. Infectious Diseases Society of America guidelines for the diagnosis and treatment of asymptomatic bacteriuria in adults. Clin. Infect. Dis. 40, 643654 (2005).

6. Dethlefsen, L., Huse, S., Sogin, M. L. & Relman, D. A. The pervasive effects of an antibiotic on the human gut microbiota, as revealed by deep 16S rRNA sequencing. PLoS Biol. 6, e280 (2008).

7. Heidler, S. et al. The natural history of lower urinary tract symptoms in females: analysis of a health screening project. Eur. Urol. 52, 17441750 (2007).

8. Wollin, T., Laroche, B. & Psooy, K. Canadian guidelines for the management of asymptomatic microscopic hematuria in adults. Can. Urol. Assoc. J. 3, 7780 (2009).

9. Tomson, C. & Porter, T. Asymptomatic microscopic or dipstick haematuria in adults: which investigations for which patients? A review of the evidence. BJU Int. 90, 185198 (2002).

10. Madeb, R. et al. Long-term outcome of patients with a negative work-up for asymptomatic microhematuria. Urology 75, 2025 (2010).

11. Schmiemann, G., Kniehl, E., Gebhardt, K., Matejczyk, M. M. & Hummers-Pradier, E. The diagnosis of urinary tract infection: a systematic review. Dtsch Arztebl. Int. 107, 361367 (2010).

12. Maki, D. G. & Tambyah, P. A. Engineering out the risk for infection with urinary catheters. Emerg. Infect. Dis. 7, 342347 (2001).

13. Tambyah, P. A. & Maki, D. G. Catheter-associated urinary tract infection is rarely symptomatic: a prospective study of 1,497 catheterized

patients. Arch. Intern. Med. 160, 678682 (2000).14. Tambyah, P. A. & Maki, D. G. The relationship between pyuria and infection in patients with indwelling urinary catheters: a prospective study of 761 patients. Arch. Intern. Med. 160, 673677 (2000).

15. Hooton, T. M. et al. Diagnosis, prevention, and treatment of catheter-associated urinary tract infection in adults: 2009 International Clinical Practice Guidelines from the Infectious Diseases Society of America. Clin. Infect. Dis. 50, 625663 (2010).

16. Stamm, W. E. Scientific and clinical challenges in the management of urinary tract infections. Am. J. Med. 113 (Suppl. 1A), 1S4S (2002).

17. Mulvey, M. A., Schilling, J. D., Martinez, J. J. & Hultgren, S. J. Bad bugs and beleaguered bladders: interplay between uropathogenic Escherichia coli and innate host defenses. Proc. Natl Acad. Sci. USA 97, 88298835 (2000).

18. Laupland, K. B., Ross, T., Pitout, J. D.,Church, D. L. & Gregson, D. B. Community-onset urinary tract infections: a population-based assessment. Infection 35, 150153 (2007).

19. Imirzalioglu, C., Hain, T., Chakraborty, T. & Domann, E. Hidden pathogens uncovered: metagenomic analysis of urinary tract infections. Andrologia 40, 6671 (2008).

20. al-Wali, W., Hamilton-Miller, J. M., Joshi, S. & Brumfitt, W. A case of recurrently sexually transmitted urinary tract infection. Genitourin. Med. 65, 397398 (1989).

21. Bailey, R. R., Peddie, B. A., Swainson, C. P. & Kirkpatrick, D. Sexual acquisition of urinary tract infection in a man. Nephron 44, 217218 (1986).

22. Foxman, B. et al. Risk factors for second urinary tract infection among college women. Am. J. Epidemiol. 151, 11941205 (2000).

23. Foxman, B. et al. Uropathogenic Escherichia coli are more likely than commensal E. coli to be shared between heterosexual sex partners. Am. J. Epidemiol. 156, 11331140 (2002).24. Foxman, B. et al. Condom use and first-time urinary tract infection. Epidemiology 8, 637641 (1997).

25. ONeill, P. M., Talboys, C. A., Roberts, A. P. & Azadian, B. S. The rise and fall of Escherichia coli O15 in a London teaching hospital. J. Med. Microbiol. 33, 2327 (1990).

26. Peirano, G. et al. High prevalence of ST131 isolates producing CTX-M-15 and CTX-M-14 among extended-spectrum--lactamase-producing

Escherichia coli isolates from Canada. Antimicrob. Agents Chemother. 54, 13271330 (2010).

27. Freeman, J. T. et al. Community-onset genitourinary tract infection due to CTX-M-15-producing Escherichia coli among travelers to the Indian subcontinent in New Zealand. Clin. Infect. Dis. 47, 689692 (2008).

28. Pitout, J. D. Infections with extended-spectrum -lactamase-producing enterobacteriaceae: changing epidemiology and drug treatment choices. Drugs 70, 313333 (2010).

29. Vincent, C. et al. Food reservoir for Escherichia coli causing urinary tract infections. Emerg. Infect. Dis. 16, 8895 (2010).

30. Schaberg, D. R., Weinstein, R. A. & Stamm, W. E. Epidemics of nosocomial urinary tract infection caused by multiply resistant Gram-negative bacilli: epidemiology and control. J. Infect. Dis. 133, 363366 (1976).

31. Huang, W. C. et al. Catheter-associated urinary tract infections in intensive care units can be reduced by prompting physicians to remove unnecessary catheters. Infect. Control Hosp. Epidemiol. 25, 974978 (2004).

32. Johnson, C. C. Definitions, classification, and clinical presentation of urinary tract infections. Med. Clin. North Am. 75, 241252 (1991).

33. Foxman, B. & Brown, P. Epidemiology of urinary tract infections: transmission and risk factors, incidence, and costs. Infect. Dis. Clin. North Am. 17, 227241 (2003).

34. van der Kooi, T. I., Mannin, J., Wille, J. C. &van Benthem, B. H. Prevalence of nosocomial infections in The Netherlands, 20072008: results of the first four national studies. J. Hosp. Infect. 75, 168172 (2010).

35. Tang, K. K., Wong, C. K., Lo, S. F. & Ng, T. K. Is it necessary to catheterise the bladder routinely before gynaecological laparoscopic surgery? Aust. N. Z. J. Obstet. Gynaecol. 45, 380383 (2005).36. Little, P. et al. Presentation, pattern, and natural course of severe symptoms, and role of antibiotics and antibiotic resistance among patients presenting with suspected uncomplicated urinary tract infection in primary care: observational study. BMJ 340, b5633 (2010).

37. Christiaens, T. C. et al. Randomised controlled trial of nitrofurantoin versus placebo in the treatment of uncomplicated urinary tract infection in adult women. Br. J. Gen. Pract. 52, 729734 (2002).

38. Katchman, E. A. et al. Three-day vs longer duration of antibiotic treatment for cystitis in women: systematic review and meta-analysis. Am. J. Med. 118, 11961207 (2005).

39. Foxman, B., Barlow, R., DArcy, H., Gillespie, B. & Sobel, J. D. Urinary tract infection: self-reported

NATURE REVIEWS | UROLOGY VOLUME 7 | DECEMBER 2010 | 659

REVIEWS

incidence and associated costs. Ann. Epidemiol. 10, 509515 (2000).40. Saint, S. Clinical and economic consequences of nosocomial catheter-related bacteriuria. Am. J. Infect. Control 28, 6875 (2000).

41. Hooton, T. M. et al. A prospective study of risk factors for symptomatic urinary tract infection in young women. N. Engl. J. Med. 335, 468474 (1996).

42. Ki, M., Park, T., Choi, B. & Foxman, B. The epidemiology of acute pyelonephritis in South Korea, 19971999. Am. J. Epidemiol. 160, 985993 (2004).

43. Rosen, D. A., Hooton, T. M., Stamm, W. E., Humphrey, P. A. & Hultgren, S. J. Detection of intracellular bacterial communities in human urinary tract infection. PLoS Med. 4, e329 (2007).

44. Weichhart, T., Haidinger, M., Hrl, W. H. & Semann, M. D. Current concepts of molecular defence mechanisms operative during urinary tract infection. Eur. J. Clin. Invest. 38 (Suppl. 2), 2938 (2008).

45. Song, J. & Abraham, S. N. Innate and adaptive immune responses in the urinary tract. Eur. J. Clin. Invest. 38 (Suppl. 2), 2128 (2008).

46. Stamm, W. E., McKevitt, M., Roberts, P. L. & White, N. J. Natural history of recurrent urinary tract infections in women. Rev. Infect. Dis. 13, 7784 (1991).

47. Hawn, T. R. et al. Genetic variation of the human urinary tract innate immune response and asymptomatic bacteriuria in women. PLoS ONE 4, e8300 (2009).

48. Lundstedt, A. C. et al. Inherited susceptibility to acute pyelonephritis: a family study of urinary

tract infection. J. Infect. Dis. 195, 12271234 (2007).49. Hawn, T. R. et al. Toll-like receptor polymorphisms and susceptibility to urinary tract infections in adult women. PLoS ONE 4, e5990 (2009).50. Sheinfeld, J., Schaeffer, A. J., Cordon-Cardo, C., Rogatko, A. & Fair, W. R. Association of the Lewis blood-group phenotype with recurrent urinary tract infections in women. N. Engl. J. Med. 320, 773777 (1989).

51. Tempera, G. et al. The impact of prulifloxacin on vaginal lactobacillus microflora: an in vivo study. J. Chemother. 21, 646650 (2009).52. Kurtaran, B. et al. Antibiotic resistance in community-acquired urinary tract infections: prevalence and risk factors. Med. Sci. Monit. 16, CR246CR251 (2010).

53. Colodner, R., Kometiani, I., Chazan, B. & Raz, R. Risk factors for community-acquired urinary tract infection due to quinolone-resistant E. coli. Infection 36, 4145 (2008).

54. Sundqvist, M. et al. Little evidence for reversibility of trimethoprim resistance after a drastic reduction in trimethoprim use. J. Antimicrob. Chemother. 65, 350360 (2010).55. Schito, G. C. et al. The ARESC study: an international survey on the antimicrobial resistance of pathogens involved in uncomplicated urinary tract infections. Int. J. Antimicrob. Agents 34, 407413 (2009).

56. Karlowsky, J. A., Kelly, L. J., Thornsberry, C., Jones, M. E. & Sahm, D. F. Trends in antimicrobial resistance among urinary tract infection isolates of Escherichia coli from female outpatients in the United States. Antimicrob. Agents Chemother. 46, 25402545 (2002).

57. Zhanel, G. G. et al. Antibiotic resistance in outpatient urinary isolates: final results from the North American Urinary Tract Infection Collaborative Alliance (NAUTICA). Int. J. Antimicrob. Agents 26, 380388 (2005).

58. Moreira, E. D. Jr et al. Antimicrobial resistance of Escherichia coli strains causing community-acquired urinary tract infections among insured and uninsured populations in a large urban center. J. Chemother. 18, 255260 (2006).

59. Als, J. I., Serrano, M. G., Gmez-Garcs, J. L. & Perianes, J. Antibiotic resistance of Escherichia coli from community-acquired urinary tract infections in relation to demographic and clinical data. Clin. Microbiol. Infect. 11, 199203 (2005).

60. DeBusscher, J., Zhang, L., Buxton, M., Foxman, B. & Barbosa-Cesnik, C. Persistent extended-spectrum -lactamase urinary tract infection. Emerg. Infect. Dis. 15, 18621864 (2009).

61. Doi, Y. et al. Community-acquired extended-spectrum -lactamase producers, United

States. Emerg. Infect. Dis. 13, 11211123 (2007).62. Azap, O. K. et al. Risk factors for extended-spectrum -lactamase positivity in uropathogenic Escherichia coli isolated from community-acquired urinary tract infections. Clin. Microbiol. Infect. 16, 147151 (2010).

63. Bleidorn, J., Ggyor, I., Kochen, M. M., Wegscheider, K. & Hummers-Pradier, E. Symptomatic treatment (ibuprofen) or antibiotics (ciprofloxacin) for uncomplicated urinary tract infection? Results of a randomized controlled pilot trial. BMC Med. 8, 30 (2010).

660 | DECEMBER 2010 | VOLUME 7 www.nature.com/nrurol

Word count: 6705

Show less

The epidemiology of urinary tract infection

Content area

Abstract

Full text