Resistance is one of many reasons why antibiotic therapy can be ineffective. Efforts to forestall further development of antimicrobial resistance include judicial prescribing of antibiotics, implementing infection-control measures, and developing institutional stewardship of antimicrobial agents. This article, the third and final in a series on antibiotic resistance, discusses selected common infections that have changing epidemiology and/or for which the recommended evaluation and treatment guidelines have been updated.

Urinary tract infection (UTI) is the most frequently diagnosed bacterial infection among community-living women.1 In the United States, UTIs account for approximately 8 million medical visits per year.2 Acute cystitis is considered uncomplicated when a symptomatic infection occurs in an otherwise healthy, nonpregnant adult female. Escherichia coli is the causative organism in 80% to 85% of cases of acute cystitis; the causative organisms in the remaining cases are Staphylococcus saprophyticus, Klebsiella pneumoniae, and Proteus mirabilis.3

Increasing antimicrobial resistance of E coli urine culture isolates has made empirical treatment of UTIs more difficult. In 1999, the Infectious Diseases Society of America (IDSA) published guidelines for the treatment of UTIs.2 These guidelines recommended a 3-day course of trimethoprimsulfamethoxazole (TMP-SMX) as initial therapy for women with acute uncomplicated cystitis except in communities where resistance exceeds 10% to 20%. Other authors, however, suggested a threshold of 22% to 30%.4

A study published in 2007 evaluated adherence to the IDSA guidelines.5 The major outcomes measure was whether these guidelines influenced antibiotic selection. The study reviewed 2,339 cases of uncomplicated UTI treated between 1996 and 2001. The data showed that the use of TMP-SMX did not change significantly; however, the use of ciprofloxacin increased significantly.5

Concern about increasing antimicrobial resistance to TMPSMX is likely the reason that medical providers did not alter their prescribing habits in response to the IDSA guidelines. Prior to 1990, E coli resistance to TMP-SMX was 0% to 5%.6 In 2001, reported resistance among urine isolates of E coli obtained from female outpatients from across the United States was 16%; even allowing for considerable geographic variation, TMP-SMX resistance was higher than 10% in all nine US Census Bureau regions.7 However, resistance may be overestimated as uncomplicated cases of UTI are often treated empirically; therefore, isolates from laboratory cultures may have been obtained from women in whom previous antimicrobial treatment had failed or who had had underlying risk factors. Local hospital antibiograms often include both inpatient and outpatient isolates and, therefore, may also overstate the prevalence of E coli resistance as the cause of uncomplicated cystitis. Risk factors for E coli resistance to TMP-SMX include current use of antibiotics, use of TMPSMX within the previous 3 months, diabetes, and recent hospitalization.8

An increase in the use of fluoroquinolones for cystitis may increase resistance, thereby limiting the effectiveness of this class of antibiotics for other types of infections. E coli isolates from outpatient urine cultures showed a greater than threefold stepwise increase in resistance from 1995 (0.7%) to 2001 (2.5%).7 Given this concern, what would be the most appropriate treatment for uncomplicated cystitis if TMP-SMX cannot be used because of allergy, recent antibiotic administration, recent hospitalization, or a community-resistance prevalence greater than 20%? Nitrofurantoin should be considered for women with mild to moderate cystitis symptoms. This compound has activity against most uropathogenic E coli and gram-positive cocci; however, it is inactive against most Proteus species and some Enterobacter and Klebsiella strains. Nitrofurantoin is administered for 5 to 7 days for cystitis; a 7-day regimen was recommended in the past, but a recent study demonstrated efficacy with 5 days of treatment.9 Nitrofurantoin should not be administered for complicated UTIs, including pyelonephritis, because it does not attain appreciable serum levels.10 Fluoroquinolones, such as ciprofloxacin, are excellent drugs for treating UTIs; however, because of increasing resistance, these agents should be reserved for women with more severe cystitis symptoms and a risk factor for TMP-SMX resistance. Limited use of fluoroquinolones for UTIs will help to maintain the efficacy of this important drug class.

ASYMPTOMATIC BACTERIURIA IN ADULTS

Asymptomatic bacteriuria is the isolation of bacteria from a urine specimen obtained from a person without symptoms or signs referable to UTI. Pyuria may or may not be present. Asymptomatic bacteriuria is a common occurrence, particularly among older adults because of the physiologic changes related to aging.11

IDSA guidelines for diagnosis and treatment of asymptomatic bacteriuria in adults are based on published evidence.12 Results of analyzed studies show that premenopausal women with asymptomatic bacteriuria are at increased risk for subsequent symptomatic UTI; however, an association between asymptomatic bacteriuria and long-term adverse outcomes, including hypertension, chronic kidney disease, cancer, or increased mortality, has not been found. Interestingly, antimicrobial treatment of asymptomatic bacteriuria neither decreases the frequency of symptomatic infection nor prevents future episodes of asymptomatic bacteriuria. Women with certain host factors appear to have increased susceptibility to both asymptomatic and symptomatic UTI that is not altered by treatment.12

Likewise, screening and treatment of asymptomatic bacteriuria in women with diabetes, older persons who reside in the community, or elderly residents of long-term-care facilities are not beneficial. Symptomatic UTI is defined by the presence of symptoms referable to the GU tract; however, nonurinary tract-specific symptoms may indicate the presence of a UTI in elderly persons. Consensus-based criteria have been developed to define symptomatic UTI in nursing-home residents who do not have an indwelling catheter. Per these criteria, symptomatic UTI is identified if the person exhibits three of these four criteria: (1) fever 100.4ºF (38ºC) or higher; (2) new or increased burning sensation with urination, frequency, or urgency; (3) new flank or suprapubic pain or tenderness; and (4) worsening of mental or functional status.13

Women with asymptomatic bacteriuria during early pregnancy have a 20- to 30-fold increase in the risk of developing pyelonephritis.12 Antimicrobial treatment is proven to decrease the risk of subsequent pyelonephritis, as well as the frequency of low birth weight and preterm delivery. Thus, the IDSA guidelines recommend that pregnant women be screened for bacteriuria by urine culture in early pregnancy and treated if test results are positive. Asymptomatic bacteriuria is also a risk for patients who undergo traumatic urologic interventions that involve mucosal bleeding. These patients should be treated prior to such interventions. For all other adults, the condition has not been shown to be harmful, and screening for, or treatment of, asymptomatic bacteriuria is not appropriate.12