addressing antibiotic resistance

Addressing Antibiotic Resistance

Kalpana Gupta, MD, MPH

Management of uncomplicated urinary tract infec-tions (UTIs) has traditionally been based on 2 impor-tant principles: the spectrum of organisms causingacute UTI is highly predictable (Escherichia coli ac-counts for 75% to 90% and Staphylococcus sapro-phyticus accounts for 5% to 15% of isolates), and thesusceptibility patterns of these organisms have alsobeen relatively predictable. As a result, empiric ther-apy with short-course trimethoprim-sulfamethox-azole (TMP-SMX) has been a standard managementapproach for uncomplicated cystitis.

However, antibiotic resistance is now becoming amajor factor not only in nosocomial complicatedUTIs, but also in uncomplicated community-acquiredUTIs. Resistance to TMP-SMX now approaches 18%to 22% in some regions of the United States, andnearly 1 in 3 bacterial strains causing cystitis or py-elonephritis demonstrate resistance to amoxicillin.Fortunately, resistance to other agents, such as ni-trofurantoin and the fluoroquinolones, has remainedlow, at approximately 2%. Preliminary data suggestthat the increase in TMP-SMX resistance is associ-ated with poorer bacteriologic and clinical outcomeswhen TMP-SMX is used for therapy. As a result,these trends have necessitated a change in the man-agement approach to community-acquired UTI. Theuse of TMP-SMX as a first-line agent for empirictherapy of uncomplicated cystitis is only appropriatein areas where TMP-SMX resistance prevalence is<10% to 20%. In areas where resistance to TMP-SMX exceeds this rate, alternative agents need to beconsidered. Am J Med. 2002;113(1A):29S–34S.© 2002 by Excerpta Medica, Inc.

Management of uncomplicated urinary tract in-fections (UTIs) has traditionally been based on2 important principles. First, the spectrum of

organisms causing acute UTI is highly predictable, withEscherichia coli accounting for 75% to 90% of isolates,Staphylococcus saprophyticus accounting for 5% to 15% ofisolates, and other organisms accounting for 5% to 10%of isolates. Second, the susceptibility patterns of these or-ganisms have also been relatively predictable. Based onthese principles, empiric therapy with short-course tri-methoprim-sulfamethoxazole (TMP-SMX) has been astandard management approach for uncomplicated cys-titis.1,2 In addition, these principles have guided initialtherapeutic choices while awaiting urine culture and sus-ceptibility results for uncomplicated pyelonephritis andcomplicated UTIs.

However, antibiotic resistance, which has traditionallybeen a problem only in nosocomial complicated UTIs, isnow also becoming a major factor in uncomplicatedcommunity-acquired UTIs. Resistance among cystitisisolates to ampicillin and cephalothin has increased fromapproximately 20% to 30% to 40% within the last de-cade.3– 8 Resistance to TMP-SMX, the current drug ofchoice in the United States for empiric therapy of uncom-plicated UTI in women, now approaches 18% to 22% insome regions of the United States.5,8 Fortunately, resis-tance to other agents, such as nitrofurantoin and the fluo-roquinolones, has remained low.4,5,8 These trends havenecessitated a change in the management approach tocommunity-acquired UTI. Specifically, the use of TMP-SMX as a first-line agent for empiric therapy of uncom-plicated cystitis is recommended only in areas whereTMP-SMX resistance prevalence is �10% to 20%.2 Inareas where resistance to TMP-SMX exceeds this rate,alternative agents need to be considered. In addition togeographic variation in the prevalence of resistance, con-sideration of individual patient risk factors for resistanceneeds to be incorporated into the management strategyfor uncomplicated UTI. An understanding of the trendsin resistance in uncomplicated UTI is essential for suc-cessfully managing UTI in an era of evolving antibioticresistance.

ACUTE UNCOMPLICATED CYSTITIS

Susceptibility PatternsSeveral studies have examined the resistance profiles ofuropathogens causing community-acquired UTI. A mul-ticenter trial of fluoroquinolones versus TMP-SMX fortherapy of acute uncomplicated cystitis that was con-

From the Department of Medicine, Division of Allergy and InfectiousDiseases, University of Washington School of Medicine, Seattle, Wash-ington, USA.

Requests for reprints should be addressed to Kalpana Gupta, MD,MPH, Department of Medicine, Division of Allergy and Infectious Dis-eases (356523), University of Washington School of Medicine, 1959 NEPacific Street, 881221, Seattle, Washington 98195.

© 2002 by Excerpta Medica, Inc. 0002-9343/02/$20.00 29SAll rights reserved. PII S0002-9343(02)01057-4

ducted in the early 1990s reported a TMP-SMX resistancerate of 7% among E coli isolates.9 Resistance to the fluo-roquinolones studied, ofloxacin and ciprofloxacin, was0%. A population-based survey of urine isolates fromoutpatient women with acute uncomplicated cystitis alsofound that the prevalence of TMP-SMX resistanceamong E coli was relatively low during the early 1990s(9% in 1992) but showed a significantly increasing trendover time, reaching 18% by 1996.4 Resistance to the�-lactams studied, ampicillin and cephalothin, was al-ready high in 1992 at approximately 20% to 30% andincreased to 28% to 34% by 1996. Resistance to nitro-furantoin and ciprofloxacin was �1% in each study year(Figure 1). Larger, nationwide laboratory surveys havedemonstrated the same rates of resistance among UTIisolates collected from 1998 through 2000 (Table 1).5,8

Another important trend in in vitro resistance profilesamong community-acquired uropathogens is that ofmultidrug resistance. A recent nationwide survey of UTIisolates collected from January 1 through September 30,

2000 from various patient groups found that 7% of E coliisolates were resistant to �3 of the drugs tested.8 Themost common phenotype of the multidrug-resistant Ecoli was resistance to ampicillin, cephalothin, and TMP-SMX. This survey included isolates from both upper andlower urinary tract sources, inpatients and outpatients, aswell as men and women of all age groups. Thus, the prev-alence of multidrug resistance among uncomplicatedcystitis isolates is still unknown. Stratified analyses didreveal that isolates from patients with complicated UTIs(patients who were male, �65 years, or inpatients) weremore likely to be multidrug resistant. The prevalence ofmultidrug-resistant isolates also varied geographically(see below).

Geographic VariationBecause there is no systematic surveillance system formonitoring susceptibility profiles of community-ac-quired UTI isolates, few data are available to practitionersregarding resistance rates in their specific practice area.

Figure 1. Prevalence of resistance among Escherichia coli causing uncomplicated cystitis in women. TMP-SMX � trimethoprim-sulfamethoxazole. (Adapted from JAMA.4)

Table 1. Urinary Tract Infection Isolates: 1999 National Outpatient Data

Antimicrobial

% Susceptible

Escherichia coliStaphylococcussaprophyticus

Ampicillin 60.6 29.3Ceftazidime 99.1 No dataCeftriaxone 99.7 74.5Ciprofloxacin 97.4 99.1Levofloxacin 97 98.2Ofloxacin 97 100TMP-SMX 81.7 93.9

TMP-SMX � trimethoprim-sulfamethoxazole.Adapted from The Surveillance Network (TSN™) Databases-MRL.

A Symposium: Addressing Antibiotic Resistance/Gupta

30S July 8, 2002 THE AMERICAN JOURNAL OF MEDICINE� Volume 113 (1A)

Most studies assess resistance rates from 1 institution orreport national averages. There are now commercial or-ganizations that attempt to collect susceptibility datafrom laboratories throughout the United States. In addi-tion to culture and susceptibility data, information onsuch factors as sex, age, culture source (urine versusblood), and patient location (inpatient versus outpatient)is also collected, allowing for better epidemiologically de-fined studies. Two recent studies have been publishedbased on such data from The Surveillance Network (TSN;MRL, Herndon, VA) Database–USA, a national surveil-lance system that collects antimicrobial susceptibility re-sults from clinical microbiology laboratories distributedin 9 geographic regions of the United States.5,8 One of thestudies examined data specifically from community-ac-quired UTI isolates from females 15 to 50 years of age and�50 years. In this study, significant geographic variationin TMP-SMX resistance among E coli isolates was re-ported, with rates as high as 22% in the western andsouthern regions of the United States and as low as 10%in the northeast. There was no significant geographicvariation in rates of resistance to nitrofurantoin or thefluoroquinolones studied. However, there was a differ-ence by age group. Women �50 years tended to havehigher resistance rates to both of these classes of drugs,mainly because of an increased percentage of non–E coliorganisms in this age group. In the study of multidrugresistance that was mentioned earlier, rates of multidrug-resistant E coli were higher in the Pacific and West SouthCentral US Bureau of Census regions of the UnitedStates.8

Thus, these studies verify that there is significant vari-ation in resistance rates by geographic location within theUnited States. However, there are still limitations inher-ent in laboratory-based surveillance studies. Many prac-titioners do not obtain a urine culture before treatingepisodes of truly uncomplicated cystitis. Thus, data fromthese episodes is not included in a laboratory-based anal-ysis. Urine cultures may be more likely to be collectedfrom cystitis episodes that are associated with some com-plicating factor, thus potentially skewing the susceptibil-ity results in these studies. In addition, although thesestudies provide important regional information, there

still is little information available for individual practiceareas; that is, the regions encompass many small areaswhere resistance rates may be higher or lower than theregional average.

Risk Factors for ResistanceIn addition to the resistance prevalence in the local com-munity, specific host factors likely influence whether agiven individual will have a TMP-SMX–resistant UTI.There are few data available regarding this issue. One ofthe best studies to date is that by Wright et al.10 Theyconducted a case-control study of patients arriving at anemergency room with cystitis symptoms. All patients hadurine cultures performed. Potential factors contributingto resistance were ascertained from chart review on eachpatient and were compared between women with a TMP-SMX–resistant uropathogen (cases) and women with aTMP-SMX–susceptible infection (controls). Age, historyof UTI, the presence of diabetes, cancer, chronic neuro-logic or urologic disorder, residence in a long-term carefacility, recent hospitalization, and recent antimicrobialuse were some of the factors considered. Four variableswere significantly and independently associated with riskof having a TMP-SMX–resistant UTI in the multivariateanalysis: diabetes, recent hospitalization, current use ofany antibiotic, and current or recent use of TMP-SMX(Table 2). Women who were using TMP-SMX at the timeof presentation or who had used it in the past 3 monthshad a 5-fold increased risk of having a TMP-SMX-resis-tant uropathogen. More surprising was the fact that useof any antimicrobial within the past 3 months was asso-ciated with a 4.5-fold increased risk of having a TMP-SMX-resistant organism. Diabetes was not an indepen-dent risk factor for antibiotic resistance after recent hos-pitalization was excluded. Thus, it may be thathospitalization is a mechanism by which diabetic patientsacquire resistant organisms. Another smaller study con-ducted in the United Kingdom also found that use oftrimethoprim in the previous 6 months was associatedwith an increased risk of having a UTI with a tri-methoprim-resistant organism and that diabetes was notindependently associated with resistance.11

In summary, it is likely that current or recent use of

Table 2. Risk Factors for Trimethoprim-Sulfamethoxazole (TMP-SMX) Resistance*

Independent Factors Odds Ratio (95% CI)

Diabetes 3.1 (1.2–8.4)Recent hospitalization 2.5 (1.1–5.7)Current use of antimicrobials 4.5 (2.0–10.2)Current/recent use of TMP-SMX 5.1 (2.2–11.5)

CI � confidence interval.* Of 448 emergency department patients with Escherichia coli UTI, 67 (15%) had TMP-SMX–

resistant strains.Adapted from J Gen Intern Med.10


July 8, 2002 THE AMERICAN JOURNAL OF MEDICINE� Volume 113 (1A) 31S

TMP-SMX or recent use of other antimicrobials increasesthe risk that a patient with acute cystitis has an organismresistant to TMP-SMX. However, the contribution of ad-ditional factors, such as comorbid diseases, travel to areaswith high rates of TMP-SMX resistance, and exposure tofamily members who are on antibiotics or children indaycare, all need further study.12–14

ACUTE UNCOMPLICATEDPYELONEPHRITISSusceptibility PatternsManagement of pyelonephritis should always includeurine culture and susceptibility testing. However, therapyis usually chosen and initiated before this information isavailable. An understanding of susceptibility patterns incommunity-acquired pyelonephritis is therefore essen-tial, especially in settings where the patient will be treatedas an outpatient. Compilation of data from 3 studies re-veals that the trends in resistance among cystitis strainsare also being seen among strains from episodes of acuteuncomplicated pyelonephritis (Table 3).3,15,16 In the1980s, susceptibility of community-acquired pyelone-phritis isolates approached 75% for ampicillin and 100%for TMP-SMX.3,15 Susceptibility rates for fluoroquinolo-nes were not reported in this study. By the mid-1990s,susceptibility of community-acquired pyelonephritis iso-lates decreased to approximately 50% for ampicillin and70% to 82% for TMP-SMX. On the other hand, suscep-tibility of these isolates to the fluoroquinolones was ex-cellent (�98%) during this time period. The study byTalan et al,16 which included pyelonephritis isolates from1994 to 1997, found that �1% of 255 isolates were resis-tant to ciprofloxacin as compared with an 18% resistancerate to TMP-SMX. As seen with cystitis isolates, the rateof TMP-SMX resistance among E coli strains ranged con-siderably by geographic region, with only 7% resistanceamong isolates from study sites in the eastern UnitedStates as compared with 32% resistance among isolatesfrom study sites in the western United States.

Aside from current resistance profiles of pyelonephri-tis isolates, this study also provides data with importanttherapeutic implications for uncomplicated pyelonephri-tis in women. The main aim was to compare the efficacyof ciprofloxacin 500 mg twice a day for 7 days versusTMP-SMX 160/800 mg twice a day for 14 days for outpa-tient oral therapy of acute uncomplicated pyelonephritisin women. Patients randomized to ciprofloxacin were el-igible to receive 1 dose of intravenous ciprofloxacin, andpatients randomized to TMP-SMX were eligible to re-ceive 1 dose of intravenous ceftriaxone if deemed neces-sary by the treating physician. Patients were continued ontheir study drug even if the infecting organism was resis-tant to the drug unless they had a clinical failure. In anintent-to-treat analysis, patients randomized to 7 days ofciprofloxacin had better bacteriologic and clinical cure

rates than patients in the TMP-SMX treatment group, atleast in the early follow-up period of 4 to 11 days aftertherapy. In a safety and cost analysis, adverse drug eventswere not significantly different between the 2 treatmentgroups. However, the mean cost of treatment was lowerin the ciprofloxacin group by $155 as compared with theTMP-SMX group. This study also provided importantinformation regarding the clinical implications of in vitroresistance in treatment of pyelonephritis (see below).

IMPLICATIONS OF TRIMETHOPRIM-SULFAMETHOXAZOLE RESISTANCEAcute Uncomplicated CystitisWith increasing evidence for significant in vitro changesin TMP-SMX resistance among community-acquiredcystitis isolates, a major research area of interest has beenunderstanding the clinical and bacteriologic ramifica-tions of this resistance. Because it is known that mostantimicrobials used to treat UTIs achieve urinary con-centrations in excess of serum concentrations (on whichNational Committee for Clinical Laboratory Standardssusceptibility breakpoints for most drugs are based), it isreasonable to expect that not all cases of in vitro resistancewill translate into therapeutic failure. Until recently, therehas been relatively little study of this issue. However, datafrom a few older studies as well as a recent trial specificallydesigned to address this issue suggest that laboratory-de-fined resistance does have a significant impact on micro-biological and clinical outcomes (Table 4).

In a treatment trial comparing a high-dose, single-sa-chet formulation of amoxicillin-clavulanate versus 7-daytherapy with TMP-SMX for treatment of acute uncom-plicated cystitis, 12% of 135 women randomized to TMP-SMX had a TMP-SMX–resistant uropathogen.17 Bacte-rial eradication at day 14 was achieved in 50% of women(7 of 14) with a TMP-SMX–resistant uropathogen ascompared with 86% of all women (106 of 123) treatedwith TMP-SMX. Clinical cure rates in the resistant groupwere not reported. In another treatment trial, McCarty etal9 found a 50% bacterial eradication rate and 60% clin-ical cure rate among 10 women with a TMP-SMX–resis-tant uropathogen who had been randomized to receiveTMP-SMX. All of the women with clinical failure alsohad bacteriologic failure. Definitive conclusions cannotbe drawn from either of these studies because of the smallsample sizes and the fact that they were not designed tospecifically address this issue.

More recently, we and others have used modeling toestimate the outcomes and costs associated with in vitroresistance.18,19 In a simple mathematical model, we esti-mated that a 10% TMP-SMX resistance prevalencewould result in a clinical cure rate of 92% and bacterio-logic cure rate of 89%. At a resistance prevalence of 20%,the estimated cure rates would be 88% and 84%, respec-tively. Using a decision tree model, Le and Miller found



that the cost of empirical treatment of uncomplicatedcystitis with TMP-SMX increased only slightly with in-creasing rates of TMP-SMX resistance among E coli, withmean costs of $92, $99, $106, and $113 with TMP-SMXresistance rates of 0%, 10%, 20% and 30%, respectively.19

They calculated the mean cost of treatment with a fluo-roquinolone at $107 and concluded that only at a TMP-SMX resistance prevalence of 22% does it become lessexpensive to use a fluoroquinolone rather than TMP-SMX for empiric UTI therapy.

Although these models are interesting and potentiallyuseful for clinicians trying to determine when to use analternative agent for acute cystitis, they are both based onvery limited data regarding actual outcomes in the settingof in vitro resistance. Fortunately, the results of a trialspecifically designed to address this issue recently becameavailable.20 This study was conducted in Israel, where theprevalence of TMP-SMX resistance among community-acquired uropathogens is estimated at 30% to 40%.Healthy women with acute cystitis symptoms had aurine culture performed and were treated with TMP-SMX 160/800 mg twice daily for 5 days. The overall TMP-SMX resistance rate was 29%. Bacteriologic cure wasachieved in 86% of women with a susceptible uropatho-gen as compared with 42% of women with a resistanturopathogen. The corresponding clinical cure rates were

88% and 54%, respectively. The authors conclude thatTMP-SMX should not be used in a setting of such highresistance.

It does appear, based on older treatment trials, model-ing systems, and this most recent study, that the clinicalfailure rate in the setting of TMP-SMX resistance is sig-nificant. The correlation between in vitro resistance andclinical failure is not 100% but is probably in the range of40% to 50%. Bacterial failure is even more closely corre-lated with in vitro resistance, occurring in 50% to 58% ofcases. Thus, the Infectious Diseases Society of Americaguidelines are quite reasonable in recommending an al-ternative agent for therapy of uncomplicated cystitiswhen the TMP-SMX resistance prevalence is about 20%.2

If the threshold for clinical failure is low because of otherpatient factors, then the use of a fluoroquinolone couldbe considered even at the level of 10% TMP-SMX resis-tance.

Acute Uncomplicated PyelonephritisFrom the Talan study,16 it is also clear that in vitro resis-tance correlates highly with clinical and bacteriologicoutcomes in pyelonephritis. In the TMP-SMX treatmentarm, the bacteriologic cure rate at 4 to 11 days after ther-apy was 96% among the 76 women with a TMP-SMX–susceptible uropathogen versus 50% among the 14women with a TMP-SMX–resistant uropathogen(P �0.001). The clinical cure rates in the TMP-SMX–treated group were 92% in women with a TMP-SMX–susceptible uropathogen compared with 35% inwomen with a TMP-SMX–resistant uropathogen (P�0.001). Interestingly, the use of an initial dose of in-travenous ceftriaxone resulted in improved bacterio-logic outcomes in the setting of TMP-SMX resistance.Among women with a TMP-SMX–resistant uropatho-gen, 5 of 5 who received an initial intravenous dose ofceftriaxone had achieved bacteriologic cure comparedwith the 2 of 9 patients with a resistant uropathogenwho received only oral TMP-SMX. The clinical curerates were also better when ceftriaxone was given, al-though the sample size was too small to reach statisticalsignificance. Thus, this study clearly demonstrateslower bacteriologic and clinical efficacy among pa-tients with pyelonephritis treated with TMP-SMX inthe setting of a TMP-SMX–resistant uropathogen.

CONCLUSIONS

Antimicrobial resistance rates among bacteria causingcommunity-acquired cystitis and pyelonephritis are in-creasing. The resistance rates for TMP-SMX exceed 20%in some regions of the United States, and nearly 1 in 3bacterial strains causing cystitis or pyelonephritis dem-onstrate resistance to amoxicillin. However, resistance tothe fluoroquinolones and nitrofurantoin remains low, atapproximately 2%. Studies demonstrate that the increase

Table 3. Antimicrobial In Vitro Resistance: Pyelonephritis

Antimicrobial

Year of Study/% Susceptible

1985–1987* 1995* 1994–1997†

Ampicillin 72 57 NATMP-SMX 100 70 82Fluoroquinolones NA 98 �99

NA � not available; TMP-SMX � trimethoprim-sulfamethoxazole.* Data from Infect Dis Clin North Am.3† Data from JAMA.16

Table 4. Success Rates with Use of Trimethoprim-Sulfame-thoxazole (TMP-SMX) for Treatment of TMP-SMX–ResistantUrinary Tract Infections (UTIs)

Study Population

Success Rates inTMP-SMX–resistant UTIsTreated with TMP-SMX

n (%)

BacterialEradication

ClinicalCure

Outpatient women (n � 17) 7/14 (50) NAOutpatient women, 18–93

yr of age (n � 9)5/10 (50) 6/10 (60)

Healthy premenopausalwomen (n � 20)

64/151 (42) 81/151 (54)

NA � not available.


July 8, 2002 THE AMERICAN JOURNAL OF MEDICINE� Volume 113 (1A) 33S

in TMP-SMX resistance does correlate with poorer bac-teriologic and clinical outcomes when TMP-SMX is usedfor therapy. Thus, the empiric use of TMP-SMX for acuteuncomplicated cystitis is recommended only in those re-gions where E coli resistance to TMP-SMX is �20%. IfE coli resistance to TMP-SMX is �20%, the empiric useof fluoroquinolones, including ciprofloxacin, gatifloxa-cin, and levofloxacin, is recommended for equal efficacyin a 3-day regimen. Alternatively, nitrofurantoin may beused in a 7-day regimen or fosfomycin may be consideredas a single-dose regimen. A fluoroquinolone is also therecommended drug for outpatient oral therapy of acuteuncomplicated pyelonephritis unless the infecting patho-gen is known to be susceptible to TMP-SMX. Furtherresearch on which factors best predict resistance and con-tinued surveillance of resistance rates in the communityare needed to enable clinicians to choose appropriate em-piric therapy for uncomplicated UTI in this era of evolv-ing resistance.

REFERENCES1. Hooton TM, Stamm WE. Management of acute uncompli-

cated urinary tract infection in adults. Med Clin North Am.1991;75:339–357.

2. Warren JW, Abrutyn E, Hebel JR, Johnson JR, SchaefferAJ, Stamm WE. Guidelines for antimicrobial treatment ofuncomplicated acute bacterial cystitis and acute pyelone-phritis in women. Infectious Diseases Society of America(IDSA). Clin Infect Dis. 1999;29:745–758.

3. Hooton TM, Stamm WE. Diagnosis and treatment of un-complicated urinary tract infection. Infect Dis Clin NorthAm. 1997;11:551–581.

4. Gupta K, Scholes D, Stamm WE. Increasing prevalence ofantimicrobial resistance among uropathogens causingacute uncomplicated cystitis in women. JAMA. 1999;281:736–738.

5. Gupta K, Sahm DF, Mayfield D, Stamm WE. Antimicrobialresistance among uropathogens that cause community-acquired urinary tract infections in women: a nationwideanalysis. Clin Infect Dis. 2001;33:89–94.

6. Dyer IE, Sankary TM, Dawson JA. Antibiotic resistance inbacterial urinary tract infections, 1991 to 1997. West J Med.1998;169:265–268.

7. Karlowsky JA, Jones ME, Thornsberry C, Critchley I, KellyLJ, Sahm DF. Prevalence of antimicrobial resistance amongurinary tract pathogens isolated from female outpatientsacross the US in 1999. Int J Antimicrob Agents. 2001;18:121–127.

8. Sahm DF, Thornsberry C, Mayfield DC, Jones ME, Kar-

lowsky JA. Multidrug-resistant urinary tract isolates ofEscherichia coli: prevalence and patient demographics inthe United States in 2000. Antimicrob Agents Chemother.2001;45:1402–1406.

9. McCarty JM, Richard G, Huck W, et al. A randomized trial ofshort-course ciprofloxacin, ofloxacin, or trimethoprim/sul-famethoxazole for the treatment of acute urinary tract in-fection in women. Ciprofloxacin Urinary Tract InfectionGroup. Am J Med. 1999;106:292–299.

10. Wright SW, Wrenn KD, Haynes ML. Trimethoprim-sulfame-thoxazole resistance among urinary coliform isolates. J GenIntern Med. 1999;14:606–609.

11. Steinke DT, Seaton RA, Phillips G, MacDonald TM, DaveyPG. Factors associated with trimethoprim-resistant bacte-ria isolated from urine samples. J Antimicrob Chemother.1999;43:841–843.

12. Murray BE, Mathewson JJ, DuPont HL, Ericsson CD, RevesRR. Emergence of resistant fecal Escherichia coli in travel-ers not taking prophylactic antimicrobial agents. Antimi-crob Agents Chemother. 1990;34:515–518.

13. Reves RR, Fong M, Pickering LK, Bartlett A III, Alvarez M,Murray BE. Risk factors for fecal colonization with tri-methoprim-resistant and multiresistant Escherichia coliamong children in day-care centers in Houston, Texas.Antimicrob Agents Chemother. 1990;34:1429–1434.

14. Rydberg J, Cederberg A. Intrafamilial spreading of Esche-richia coli resistant to trimethoprim. Scand J Infect Dis.1986;18:457–460.

15. Johnson JR, Stamm WE. Diagnosis and treatment of acuteurinary tract infections. Infect Dis Clin North Am. 1987;1:773–791.

16. Talan DA, Stamm WE, Hooton TM, et al. Comparison ofciprofloxacin (7 days) and trimethoprim-sulfamethoxazole(14 days) for acute uncomplicated pyelonephritis in women:a randomized trial. JAMA. 2000;283:1583–1590.

17. Masterton RG, Bochsler JA. High-dosage co-amoxiclav ina single dose versus 7 days of co-trimoxazole as treatmentof uncomplicated lower urinary tract infection in women. JAntimicrob Chemother. 1995;35:129–137.

18. Gupta K, Hooton TM, Stamm WE. Increasing antimicrobialresistance and the management of uncomplicated commu-nity-acquired urinary tract infections. Ann Intern Med. 2001;135:41–50.

19. Le TP, Miller LG. Empirical therapy for uncomplicated uri-nary tract infections in an era of increasing antimicrobialresistance: a decision and cost analysis. Clin Infect Dis.2001;33:615–621.

20. Raz R, Chazan B, Kennes Y, et al. Empiric use of tri-methoprim-sulfamethoxazole (TMP-SMX) in the treatmentof women with uncomplicated urinary tract infections, in ageographical area with a high prevalence of TMP-SMX–resistant uropathogens. Clin Infect Dis. 2002;34:1165–1169.



addressing antibiotic resistance

Documents