Abstract
Background
Uropathogen resistance, fluoroquinolone‐resistance (FQR), and extended spectrum beta‐lactamase (ESBL), has been observed to be emerging worldwide with prevalences above recommended thresholds for routine empirical treatment. The primary aim of our study was to determine the prevalence of FQR from a geographically diverse sample of United States emergency departments (EDs).
Methods
We conducted a multi‐center, observational cohort study using a network of 15 geographically diverse US EDs. All patients ≥18 years of age with the primary or secondary diagnosis of urinary tract infection (UTI) in the ED identified using International Classification of Diseases (ICD‐10) diagnosis code of cystitis, pyelonephritis, or UTI from 2018 to 2020 were included. We calculated descriptive statistics for uropathogens and susceptibilities. Logistic regression analysis was used to identify antimicrobial resistance risk factors associated with FQR Escherichia coli.
Results
Among 3779 patients who met inclusion criteria, median age was 62.9 years (interquartile range [IQR]: 41–77.6) and 76.3% were female. The most common diagnoses were complicated (41.2%) and uncomplicated cystitis (40.3%). E. coli was the most common pathogen (63.2%), followed by Klebsiella pneumoniae (13.2%) and Enterococcus species (5.8%). Across all sites, overall E. coli FQ‐resistance prevalence was 22.1%, ranging from 10.5 to 29.7% by site. The prevalence of ESBL‐producing uropathogen was 7.4%, ranging from 3.6% to 11.6% by site. Previous IV or oral antimicrobial use in the past 90‐days and history of a multi‐drug resistant pathogen were associated with FQ‐resistant E. coli (odds ratio [OR] 2.68, 95% confidence interval [CI]: 2.04–3.51, and OR 6.93, 95% CI: 4.95–9.70, respectively). Of the patients who had FQ‐resistant E. coli or an ESBL‐producing uropathogen isolated, 116 (37.1%) and 61 (36.7%) did not have any documented risk factors for resistance.
Conclusion
FQ‐resistant E. coli is widely prevalent across US sites highlighting the need for ongoing monitoring of antimicrobial resistance and, at some locations, modification of empirical treatments.
INTRODUCTION
Urinary tract infections (UTI) are a commonly treated infection in the emergency department (ED), accounting for approximately 3 million visits annually, and 15% of all outpatient antibiotics in the United States (US). 1 , 2 , 3 , 4 Enterobacterales remain the most common cause of UTIs and are associated with increased rates of in vitro resistance to commonly prescribed antibioitics. 5 , 6 In particular, the prevalence of Escherichia coli resistance to commonly prescribed antibiotics such as trimethoprim/sulfamethoxazole (TMP/SMX), fluroquinolones (FQ; e.g., ciprofloxacin and levofloxacin), and beta‐lactams has continued to increase in most regions of the United States, and underscores the importance of using local antibiograms for selecting empiric treatment in patients diagnosed with a UTI in the ED. 5 , 7
Many geographic regions in the United States are reporting prevalences of FQ‐resistant (FQR) and TMP/SMX‐resistant Enterobacterales of >10%, with some areas with rates >20%, exceeding threshold rates (>10%) recommended by the Infectious Diseases Society of America (IDSA) to change from one antibiotic class to another for empirical treatment. 5 , 6 Additionally, the prevalence of extended spectrum β‐lactamase (ESBL)‐producing Enterobacterales continues to increase, now exceeding 20% in some US locations. 6
Given a shrinking armamentarium of effective antibiotic treatment for UTIs, the IDSA and Centers for Disease Control and Prevention (CDC) have identified the need for continued surveillance of resistance patterns at the local and national health care system level as a top priority to best inform antibiotic treatment decisions. 7 , 8 In 2019, the CDC classified ESBL‐producing Enterobacterales as a serious threat due to the increase in community infections. 9 ESBL‐producing Enterobacterales are oftentimes also resistant to FQ as well as other agents commonly used to treat UTIs. As antimicrobial resistance can change rapidly, surveillance and reporting of the prevalence of E.coli resistant and ESBL‐producing uropathogens is necessary to guide empiric antimicrobial treatment. The objective of our study was to determine the prevalence of FQ resistance among patients presenting to the ED and diagnosed with a UTI among a geographically diverse group of US sites. Secondary objectives were to identify geographic variation in E. coli resistance and risk factors associated with antimicrobial resistance.
METHODS
We conducted a multi‐center, retrospective observational cohort study at 15 geographically diverse, US hospital EDs, that participate in the Emergency Medicine PHARMacotherapy Research NETwork (EMPHARM‐NET) (see Figure 1 for full site listings). Patients were identified based on primary diagnosis of uncomplicated or complicated cystitis and uncomplicated or complicated pyelonephritis using International Classification of Diseases (ICD)‐10 codes: ICD‐10, N30.00, N30.01, and N39.0 between January 1st, 2018 – December 31st, 2020. All patients ≥18 years of age who had a primary or secondary diagnosis of UTI in the ED, reported symptoms of a UTI, and had a urine culture obtained in the ED were included. Exclusion criteria were as follows: pregnancy; suspected or confirmed acute bacterial prostatitis; orchitis, epididymitis; or chronic bacterial prostatitis as determined by history and/or physical examination; gross hematuria requiring intervention other than administration of antibiotics for UTI or removal or exchange of a urinary catheter; and urinary tract surgery within 7 days prior to ED presentation. Patients could only have one ED visit included in the study with subsequent visits being excluded in the study population.
FIGURE 1.
Prevalence of fluoroquinolone‐resistant Escherichia coli based on EMPHARM‐NET site location, 2018–2020
Data Collection
All data variables were defined a priori and were available for abstraction from the electronic medical record (EMR). Data collected included patient‐specific characteristics (e.g., age, sex, past medical history, laboratory results), signs and symptoms (e.g., fever, dysuria, flank pain, frequency/urgency) of UTI, risk factors for antimicrobial resistance, ED disposition, and urine microbiological results and susceptibilities. Data were abstracted at each site by the principal investigator (PI) or by a trained data abstractor with an audit of a random sample of charts (~10%) completed to ensure that abstractor entries were accurate and complete. All site institutional review boards approved the study. The study is reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement. 10 We used REDCap© electronic‐data capture tool (Version 6.18.1, 2019; Vanderbilt University, Nashville, TN) hosted by the University of Iowa to manage all study data. 11
Definitions
Cystitis was defined as patients reporting dysuria, urinary frequency or urgency, suprapubic pain, or hematuria. 12 , 13 Pyelonephritis was defined as patients reporting urinary symptoms in addition to a fever (temperature > 38°C), chills, flank pain, costovertebral‐angle tenderness, and nausea or vomiting. 12 , 13 Patients were further classified as having complicated urinary tract infections (cystitis or pyelonephritis) if they were male or had a pre‐existing anatomical condition or current immunocompromising condition that may increase their risk for treatment failure. 6 Pre‐existing urinary tract anatomical conditions were: history of kidney stones, urinary obstruction, neurogenic bladder, renal insufficiency, long‐term urinary catheter [indwelling foley or suprapubic catheter], renal transplant, and nephrostomy tubes). 6 Immunocompromising conditions included active cancer, chronic systemic corticosteroid use, current use of immunosuppressants (e.g., renal transplant), and human immunodeficiency virus (HIV) disease. We classified patients without criteria for complicated cystitis or pyelonephritis as uncomplicated infections.
A positive urine culture was defined as a specimen with ≥104 CFU/mL bacteria isolated in the urine culture with ≤2 organism isolated in the urine culture. 14 , 15 , 16 , 17 If a urine culture grew >2 organisms, it was considered to be contaminated; if no organisms were isolated, it was considered no growth. 5 , 14 The following pathogens were considered to be contaminants: Lactobacillus, non‐saprophyticus coagulase‐negative Staphylococcus, Corynebacterium species; or β‐hemolytic streptococci. 14 Contaminated cultures and no growth cultures were reported as a negative culture in the final analysis. To test for the presence of ESBL‐Enterobacterales or carbapenemase‐producing Enterobacterales, the majority of the sites used VITEK® 2 (bioMerieux) to test for ESBL presence and Xpert® Carba‐R (Cepheid) test for the presence of carbapenemases. If the site did not have capability to test specifically for ESBL‐producing Enterobacterales, we considered isolates that were non‐susceptible to ceftriaxone (i.e., minimum inhibitory concentration [MIC] > 1 μ/ml) to be ESBL producing. 6
We evaluated risk factors for antimicrobial resistance, which included the following: previous intravenous (IV) or oral antibiotic use in the prior 90 days, hemodialysis dependence, pre‐existing urinary tract anatomical conditions as defined above, residence in a long‐term care facility, and history (within the past 12 months) of multi‐drug resistant (MDR) pathogen in the past 12 months. 18 , 19 In all cases, the EMR was reviewed for the following MDR pathogens: ESBL‐producing Enterobacterales; carbapenem‐resistant Enterobacterales (CRE); methicillin‐resistant Staphylococcus aureus (MRSA); vancomycin resistant Enterococci (VRE); Stenotrophomonas maltophilia; Pseudomonas aeruginosa; Acinetobacter spp.; AmpC‐ beta‐lactamase‐producing bacteria; or resistance to antibiotics from three different classes within the past 12 months.
Statistical analysis
Descriptive characteristics were used to characterize overall patient demographics for the total cohort, as well as for culture‐positive and culture‐negative patients. Among those with positive cultures, we characterized the distribution of uropathogens by UTI type including complicated and uncomplicated cystitis and pyelonephritis. We quantified the prevalence of FQ‐resistant E. coli by participating centers to describe geographical variability. Similarly, we quantified antimicrobial resistant prevalences for E. coli isolates. We assessed bivariate associations using logistic regression between select demographic and clinical factors on the presence of ESBL and FQ‐resistant E. coli and present odds ratios and 95% confidence intervals (CI). A multivariable analysis was conducted to analyze potential variables associated with the presence of ESBL or FQ‐resistant E. coli. Variables identified for regression analysis came from previously identified risk factors (defined above) or were considered clinically significant variables that may be associated with risk of MDR (e.g., kidney disease). All analyses were conducted using SAS v9.4 (Cary, NC).
RESULTS
Over the 3‐year period (January 2018 – December 2020), 3714 patients met inclusion criteria for our study. Of the included patients, 2242 patients (60%) had a urine culture that grew ≤2 uropathogens at ≥104 CFU/ml and 1472 had a urine culture that was negative (e.g., no growth or contaminated). Patient median age was 62.9 years (Interquartile range [IQR] 41–77.6 years), 76.1% were female, and 36.7% and 28.8% of patients reported having dysuria and frequent/urgent urinary symptoms, respectively (Table 1). The most common UTI diagnosis was complicated cystitis (41.2%). Most patients (66.2%) were discharged home from the ED. Antimicrobial resistance risk factors were identified in 44.2% of patients, with 15.1% having at least two risk factors; the most common (29.1%) risk factor was IV or oral antibiotic use in the prior 90 days.
TABLE 1.
Characteristics of patients diagnosed with a urinary tract infection from 15 US EDs
| Demographic and clinical characteristics | Overall (N = 3,714) | Culture positive (N = 2,242) | Culture negative (N = 1,472) | |||
|---|---|---|---|---|---|---|
| Demographics | ||||||
| Age in years, median (IQR) | 62.9 | (41.0–77.6) | 66.0 | (45.5–79.4) | 58.1 | (36.3–74.0) |
| Sex, n (%) | ||||||
| Male | 885 | (23.8) | 582 | (26.0) | 303 | (20.6) |
| Female | 2,827 | (76.1) | 1,659 | (74.0) | 1,168 | (79.3) |
| Disposition, n (%) | ||||||
| Discharged from ED | 2,457 | (66.2) | 1,329 | (59.3) | 1,128 | (76.6) |
| Admitted ‐ Non‐ICU | 1,151 | (31.0) | 829 | (37.0) | 322 | (21.9) |
| Admitted ‐ ICU | 104 | (2.8) | 83 | (3.7) | 21 | (1.4) |
| Clinical history, n (%) | ||||||
| Kidney disease | 506 | (13.6) | 350 | (15.6) | 156 | (10.6) |
| Advanced liver disease (cirrhosis/ ESLD) | 67 | (1.8) | 47 | (2.1) | 20 | (1.4) |
| Diabetes | 989 | (26.6) | 624 | (27.8) | 365 | (24.8) |
| UTI characteristics, n (%) | ||||||
| Chief complaint UTI | 1,875 | (50.5) | 1,158 | (51.7) | 717 | (48.7) |
| UTI Type | ||||||
| Pyelonephritis, uncomplicated | 294 | (7.9) | 187 | (8.3) | 107 | (7.3) |
| Pyelonephritis, complicated | 393 | (10.6) | 269 | (12.0) | 124 | (8.4) |
| Cystitis, uncomplicated | 1,495 | (40.3) | 823 | (36.7) | 672 | (45.7) |
| Cystitis, complicated | 1,532 | (41.2) | 963 | (43.0) | 569 | (38.7) |
| UTI symptoms | ||||||
| Altered mental status | 548 | (14.8) | 381 | (17.0) | 167 | (11.3) |
| Fever | 483 | (13.0) | 344 | (15.3) | 139 | (9.4) |
| Dysuria | 1,364 | (36.7) | 795 | (35.5) | 569 | (38.7) |
| Flank pain | 514 | (13.8) | 304 | (13.6) | 210 | (14.3) |
| Frequent/urgent urinary symptoms | 1,071 | (28.8) | 642 | (28.6) | 429 | (29.1) |
| Suprapubic pain | 645 | (17.4) | 406 | (18.1) | 239 | (16.2) |
| Characteristics that may contribute to antimicrobial resistance, n (%) | ||||||
| Previous IV or oral antibiotic use in the past 90 days | 1,080 | (29.1) | 639 | (28.5) | 441 | (30.0) |
| Hemodialysis dependence | 54 | (1.50 | 34 | (1.5) | 20 | (1.4) |
| Urinary tract abnormality (e.g., catheter) | 583 | (15.7) | 385 | (17.2) | 198 | (13.5) |
| Long‐term or intermittent urinary catheter | 366 | (9.9) | 262 | (11.7) | 104 | (7.1) |
| Nephrolithiasis | 107 | (2.9) | 62 | (2.8) | 45 | (3.1) |
| Renal transplant | 42 | (1.1) | 21 | (0.9) | 21 | (1.4) |
| Neurogenic bladder | 135 | (3.6) | 107 | (4.8) | 28 | (1.9) |
| Nephrostomy tubes | 51 | (1.4) | 30 | (1.3) | 21 | (1.4) |
| Residence in a long‐term care facility | 280 | (7.5) | 205 | (9.1) | 75 | (5.1) |
| History of multi‐drug resistant pathogen | 385 | (10.4) | 296 | (13.2) | 89 | (6.0) |
| Extended spectrum beta‐lactamase | 128 | (3.4) | 109 | (4.9) | 19 | (1.3) |
| Carbapenem‐resistant Enterobacterales | 17 | (0.5) | 10 | (0.4) | 7 | (0.5) |
| Methicillin‐resistant Staphylococcus aureus | 103 | (2.8) | 68 | (3.0) | 35 | (2.4) |
| Vancomycin resistant Enterococci | 44 | (1.2) | 31 | (1.4) | 13 | (0.9) |
| Stenotrophomonas maltophilia | 6 | (0.2) | 5 | (0.2) | 1 | (0.1) |
| Pseudomonas aeruginosa | 69 | (1.9) | 56 | (2.5) | 13 | (0.9) |
| Acinetobacter spp. | 14 | (0.4) | 11 | (0.5) | 3 | (0.2) |
| ≥3 different classes | 160 | (4.3) | 128 | (5.7) | 32 | (2.2) |
Of the patients with a positive urine culture, 84.7% had cultures that grew Enterobacterales with E. coli (63.2%) being the most common pathogen isolated (Table 2). Overall prevalences of ESBL‐producing and carbapenemase‐producing uropathogens were 7.4% and 0.3%, respectively.
TABLE 2.
Pathogens identified based on type of urinary tract infection
| Uropathogen | Overall (n = 3714) | Pyelonephritis | Cystitis | ||
|---|---|---|---|---|---|
| complicated (n = 393) | Uncomplicated (n = 294) | complicated (n = 1532) | Uncomplicated (n = 1495) | ||
| n | n | n | n | n | |
| Enterbacterales | |||||
| E. coli | 1417 (38.1) | 144 (36.6) | 139 (47.3) | 521 (34.0) | 613 (41.0) |
| E. cloacae | 49 (2.2) | 6 (1.5) | 1 (0.3) | 26 (1.7) | 16 (1.1) |
| E. aerogenes | 24 (1.1) | 3 (0.8) | 1 (0.3) | 17 (1.1) | 3 (0.2) |
| K. oxytoca | 35 (1.6) | 3 (0.8) | 3 (1.0) | 23 (1.5) | 6 (0.4) |
| K. pneumoniae | 296 (13.2) | 44 (11.2) | 21 (7.1) | 144 (9.4) | 87 (5.8) |
| Citrobacter spp. | 55 (2.5) | 8 (2.0) | 3 (1.0) | 30 (2.0) | 14 (0.9) |
| Proteus spp. | 122 (5.4) | 17 (4.3) | 9 (3.1) | 74 (4.8) | 22 (1.5) |
| Non‐enterbacterales | |||||
| Enterococcus species | 130 (5.8) | 25 (6.4) | 8 (2.7) | 70 (4.6) | 27 (1.8) |
| P. aeruginosa | 82 (3.7) | 18 (4.6) | 3 (1.0) | 45 (2.9) | 16 (1.1) |
| Staphylococcus aureus | 44 (2.0) | 8 (2.0) | 1 (0.3) | 27 (1.8) | 8 (0.5) |
| Staphylococcus saprophyticus | 22 (1.0) | 3 (0.8) | 5 (1.7) | 2 (0.1) | 12 (0.8) |
| Multi‐drug resistant | |||||
| ESBL‐producing pathogen a | 166 (7.4) | 32 (8.1) | 14 (4.8) | 83 (5.4) | 37 (2.5) |
| CRE‐producing pathogen a | 6 (0.3) | 1 (0.3) | 0 (0.0) | 3 (0.2) | 2 (0.1) |
| Other b | 146 (6.5) | 16 (4.1) | 6 (2.0) | 78 (5.1) | 46 (3.1) |
The multi‐drug resistant pathogens presented consist of a subset of the Enterbacterales pathogens.
Other = pathogens not commonly isolated in the urinary tract but are considered pathogenic (e.g., Aerococcus urinae, Stenotrophomonas maltophilia, etc.).
For the primary outcome, 22.1% of E. coli isolates showed resistance to FQs (Table 3). Overall, E. coli resistance was most prevalent to ampicillin (39.2%) and TMP/SMX (24.3%). Of patients with E. coli, resistance to carbapenems (e.g., meropenem, imipenem, ertapenem) was reported in 0.3% of patients. FQ‐resistant E. coli occurred in complicated pyelonephritis (33.3%), followed by complicated cystitis (28.2%), uncomplicated pyelonephritis (17.3%), and uncomplicated cystitis (15.3%). (Table 3). There was geographic variation in the prevalence of FQ‐resistant E. coli ranging from 10.5–29.7%. Fourteen of the enrolling institutions reported FQ‐resistant E. coli rates >15%, with seven reporting rates >20% (Figure 1).
TABLE 3.
Antimicrobial drug resistance prevalence for Escherichia coli isolates (2018‐2020) a
| Antimicrobial drug | Overall (N = 1,471) | Pyelonephritis | Cystitis | ||
|---|---|---|---|---|---|
| complicated (n = 144) | Uncomplicated (n = 139) | complicated (n = 521) | Uncomplicated (n = 613) | ||
| n | n | n | n | n | |
| Beta lactams | |||||
| Ampicillin/sulbactam | 245 (17.3) | 28 (19.4) | 37 (26.6) | 90 (17.3) | 90 (14.7) |
| Ampicillin | 556 (39.2) | 61 (42.4) | 66 (47.5) | 208 (39.9) | 221 (36.1) |
| Cefazolin | 180 (12.7) | 31 (21.5) | 19 (13.7) | 85 (16.3) | 45 (7.3) |
| Cefepime | 83 (5.9) | 15 (10.4) | 6 (4.3) | 38 (7.3) | 24 (3.69) |
| Ceftazidime | 45 (3.2) | 11 (7.6) | 5 (3.6) | 20 (3.8) | 9 (1.5) |
| Ceftriaxone | 124 (8.8) | 22 (15.3) | 10 (7.2) | 62 (11.9) | 30 (4.9) |
| Ertapenem | 2 (0.1) | 1 (0.7) | 0 (0.0) | 1 (0.2) | 0 (0.0) |
| Imipenem | 1 (0.1) | 0 (0.0) | 0 (0.0) | 1 (0.2) | 0 (0.0) |
| Meropenem | 2 (0.1) | 2 (1.4) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
| Piperacillin/tazobactam | 20 (1.4) | 7 (4.9) | 1 (0.7) | 6 (1.2) | 6 (1.0) |
| Aminoglycosides | |||||
| Amikacin | 5 (0.4) | 1 (0.7) | 0 (0.0) | 2 (0.4) | 2 (0.3) |
| Gentamicin | 99 (7.0) | 15 (10.4) | 8 (5.8) | 39 (7.5) | 37 (6.0) |
| Tobramycin | 40 (2.8) | 14 (9.7) | 3 (2.2) | 15 (2.9) | 8 (1.3) |
| Fluoroquinolones | |||||
| Ciprofloxacin | 269 (19.0) | 41 (28.5) | 22 (15.8) | 122 (23.4) | 84 (13.7) |
| Levofloxacin | 157 (11.1) | 31 (21.5) | 14 (10.1) | 65 (12.5) | 47 (7.7) |
| Fluoroquinolone resistance b | 313 (22.1) | 48 (33.3) | 24 (17.3) | 147 (28.2) | 94 (15.3) |
| Other | |||||
| Aztreonam | 41 (2.9) | 10 (6.9) | 4 (2.9) | 20 (3.8) | 7 (1.1) |
| Nitrofurantoin | 28 (2.0) | 4 (2.8) | 1 (0.7) | 17 (3.3) | 6 (1.0) |
| Trimethoprim/sulfamethoxazole | 344 (24.3) | 44 (30.6) | 46 (33.1) | 123 (23.6) | 131 (21.4) |
These results are from testing conducted at site hospital microbiology laboratories. Not all antimicrobials were tested at each site. Denominators with total tested for each antimicrobial are presented.
Resistant to ciprofloxacin or levofloxacin.
Use of IV or oral antibiotics in the prior 90 days and a history of having a MDR pathogen were significantly associated with isolating FQ‐resistant E. coli (aOR, 1.87 [95% CI 1.38–2.53]; aOR, 5.25 [95% CI 3.67–7.51], respectively) (Table 4). History of having a MDR pathogen was significantly associated with isolating an ESBL‐producing uropathogen (aOR, 4.71 [95% CI 3.26–6.81) (Table 5). Of the patients who had FQ‐resistant E. coli or an ESBL‐producing uropathogen isolated, 116 (37.1%) and 61 (36.7%) did not have any documented risk factors for resistance.
TABLE 4.
Risk factors associated with isolating FQ‐resistant E. coli
| Characteristic | FQ Res E coli | |||
|---|---|---|---|---|
| uOR a | 95% CI | aOR b | 95% CI | |
| Sex (ref = female) | 1.54 | 1.13–2.09 | 1.11 | 0.78–1.60 |
| Kidney disease | 1.91 | 1.36–2.68 | 1.37 | 0.92–2.03 |
| Advanced liver disease (cirrhosis/ ESLD) | 0.74 | 0.25–2.19 | 0.41 | 0.13–1.33 |
| Previous IV or oral antibiotic use in the past 90 days | 2.68 | 2.04–3.51 | 1.87 | 1.38–2.53 |
| Hemodialysis dependence | 2.97 | 0.90–9.80 | 0.96 | 0.21–4.29 |
| Urinary tract abnormality (e.g., catheter) | 2.74 | 1.92–3.91 | 1.60 | 0.51–5.00 |
| Long‐term or intermittent urinary catheter | 3.45 | 2.25–5.28 | 1.26 | 0.42–3.80 |
| Nephrolithiasis | 1.13 | 0.51–2.54 | 0.64 | 0.17–2.34 |
| Renal transplant | 2.54 | 0.80–8.07 | 0.78 | 0.19–3.22 |
| Neurogenic bladder | 3.69 | 1.93–7.06 | 1.12 | 0.46–2.83 |
| Nephrostomy tubes | 1.77 | 0.32–9.71 | 0.56 | 0.08–4.04 |
| Residence in a long‐term care facility | 2.17 | 1.43–3.31 | 1.93 | 1.22–3.07 |
| History of multi‐drug resistant pathogen | 6.93 | 4.95–9.70 | 5.25 | 3.67–7.51 |
Represents the unadjusted odds ratios of each risk factor for FQ‐resistance among E coli isolates (n = 1471).
Among patients with positive E coli culture, adjusted for all variables listed.
TABLE 5.
Risk factors associated with isolating an ESBL‐producing uropathogen
| Characteristic | ESBL | |||
|---|---|---|---|---|
| uOR a | 95% CI | aOR b | 95% CI | |
| Sex (ref = female) | 1.29 | 0.91–1.82 | 1.16 | 0.79–1.70 |
| Kidney disease | 1.55 | 1.05–2.29 | 1.17 | 0.76–1.80 |
| Advanced liver disease (cirrhosis/ ESLD) | 1.17 | 0.41–3.29 | 0.82 | 0.28–2.42 |
| Previous IV or oral antibiotic use in the past 90 days | 2.04 | 1.48–2.81 | 1.37 | 0.96–1.95 |
| Hemodialysis dependence | 1.68 | 0.59–4.84 | 0.78 | 0.23–2.67 |
| Urinary tract abnormality | 1.71 | 1.18–2.47 | 1.33 | 0.52–3.40 |
| Long‐term or intermittent urinary catheter | 1.60 | 1.04–2.45 | 0.79 | 0.32–1.95 |
| Nephrolithiasis | 1.10 | 0.44–2.78 | 0.72 | 0.22–2.31 |
| Renal transplant | 4.00 | 1.45–11.05 | 2.17 | 0.64–7.40 |
| Neurogenic bladder | 1.79 | 0.98–3.27 | 0.78 | 0.36–1.67 |
| Nephrostomy tubes | 1.95 | 0.67–5.65 | 0.88 | 0.24–3.27 |
| Residence in a long‐term care facility | 1.58 | 0.98–2.54 | 1.34 | 0.81–2.19 |
| History of multi‐drug resistant pathogen | 5.30 | 3.77–7.44 | 4.71 | 3.26–6.81 |
Represents the unadjusted odds ratios of each risk factor for ESBL‐producing uropathogens among culture positive cohort (n = 2242).
Among culture positive patients, adjusted for all variables listed.
DISCUSSION
Our study of contemporary patients from 15 centers across the US found that the prevalence of FQ‐resistant E. coli exceeded 15% at 14 of the 15 sites and was >20% at 7 of the 15 sites signifying that geographic variation exists in E. coli FQ‐resistance. Patients presenting with a complicated UTI had E. coli FQ resistance prevalences that exceeded 50% at many sites, whereas those with uncomplicated pyelonephritis and cystitis had resistance prevalences being 29.3% and 29.9% for pyelonephritis and cystitis, respectively. Previous studies have shown that prevalence of E. coli resistance to FQs has been increasing over the past two decades. 20 , 21 In an analysis of over 12 million urine specimens, E. coli resistance to FQs was reported to be 3% in 2000 but increased to 17.1% in 2010. 21 A study enrolling participants from 2013 to 2014 evaluated the prevalence of FQ‐resistant E. coli in patients presenting to the ED with acute uncomplicated pyelonephritis and found that 19.9% of isolates were resistant. 5 The combined E. coli FQ‐resistance rate at all sites in this study was 22.1% (range 10.5%–29.7% by site) demonstrating that resistance continues to increase across the US.
The most recent IDSA guidelines (published >10 years ago) for the treatment of uncomplicated pyelonephritis recommend empiric FQ treatment unless the local prevalence of E coli resistance is greater than 10%, in which case, an initial parenteral antibiotic (e.g., ceftriaxone, consolidated 24‐hour dose of an aminoglycoside) dose is recommended to be administered before the patient is discharged home. 7 All 15 sites in our study reported E. coli FQ‐resistance >10%, suggesting that empiric antimicrobial regimens should be modified for patients with pyelonephritis. For uncomplicated cystitis, FQ‐resistant was reported in 29.9% of all of E. coli isolates. Fortunately, current guidelines recommend nitrofurantoin as first‐line treatment for uncomplicated cystitis as resistance rates have remained low. 7 In the total cohort, we found E. coli resistance to nitrofurantoin was 2%, with it being slightly lower at 1.9% in patients with uncomplicated cystitis suggesting that nitrofurantoin remains a viable treatment option for uncomplicated cystitis.
As FQ‐resistant E. coli continues to impact empiric treatment options in the ED, of growing concern is resistance emerging to alternative treatments mediated through bacterial production of ESBLs. 22 , 23 A recent study evaluating the emergence of ESBL UTI infections among ED patients who were hospitalized reported the prevalence of ESBL‐producing Enterobacterales to be 17.2% (range: 4.6%–45.4%). 6 The prevalence of ESBL‐producing Enterobacterales in our cohort was significantly lower at 7.4% (range: 3.6–11.6), but is consistent with previous reports evaluating community‐associated ESBL infections. 24 The difference in findings can likely be explained by the differing enrollment criteria and the predominance of outpatient treatment. Our study sought to determine resistance rates among all ED patients who were diagnosed with a UTI, whereas Talan et al. evaluated only patients hospitalized for treatment of UTI. 6 Additional differences could be related to the prevalence of risk factors for ESBL‐producing Enterobacterales. In the previously mentioned paper, over 75% of the patients had received antibiotic treatment within the prior 90‐days, and over 30% had a history of FQ‐ or ceftriaxone‐resistant isolate in the prior 90 days. 6 In contrast, 29.1% of our cohort had received antibiotic treatment within the prior 90‐days, and only 10.4% had a history of a MDR pathogen. It is evident that the prevalence of ESBL‐producing uropathogens is increasing, but our results suggest they may be less prevalent in patients who are less ill, being discharged home from the ED, and without any risk factors for MDR pathogens.
As the prevalence of antimicrobial resistance continues to rise among commonly prescribed antibiotics, it is evident that certain patients are at much higher risk for having a UTI caused by a resistant uropathogen. Multiple studies have evaluated the impact of risk factors on the likelihood of having a resistant uropathogen with the most common risks factors associated with isolating a resistant uropathogen being recent exposure (e.g., 60–90 days) to antimicrobial therapy, history of MDR pathogen, and complicating genitourinary factors (e.g., chronic indwelling catheters). 5 , 25 , 26 , 27 Our research is consistent with the existing literature in demonstrating the importance of identifying those patients with increased risk for resistant organisms, notable those who received an antibiotic in the prior 90 days or had a history of MDR pathogen. Importantly, however, we found that over 1/3 of patients with resistant organisms did not have identifiable risk factors. This has been described in patients hospitalized with UTI and for patients with ESBL, and we also demonstrate that this is also the case for patients with FQ resistance. 6 , 26 Our results taken together with those of Frazee et al 26 and Talan et al, 5 , 6 highlight serious concerns with increasing resistance rates to commonly used antibiotic treatments and that resistance impacts patients with and without identifiable risk factors.
STRENGTHS AND LIMITATIONS
Strengths of our study include being a multi‐center study conducted at 15 US sites with geographic variation using routinely collected and relatively current (2018–2020) health care data. We used broad inclusion criteria that should be representative of the ED patient population and is more representative of community prevalences. Additionally, because we stratified our patient population, our reported resistance prevalances are more granular based on UTI type (e.g., complicated vs. uncomplicated).
However, our study has several important limitations that must be addressed. First, our study was retrospective and observational, which introduces the prospect of inaccurate documentation. We intentionally selected variables for analysis that are likely to be accurately reported and are easily abstractable from the EMR; however, there may have been slight variation between centers in recording of data. Second, UTIs are one of the most common infectious related diagnoses in the ED and the diagnostic error for UTI in the ED has been reported to be high, which could have impacted our included population. 28 We created inclusion criteria so that patients not only had to have an ICD‐10 diagnosis code for UTI during their ED visit, but they also had to have reported symptoms of UTI and urine culture obtained increasing the likelihood of only enrolling patients with a UTI. Third, there is some difficulty in the ability to accurately capture some of the proposed antimicrobial risk factors (e.g., previous antibiotic exposure in past 90‐days, history of MDR) because of the nature of the health care system (not a closed system). Fourth, while we have 15 sites included in our study representing various geographic areas across the US, our network lacks sites in the southern and pacific regions decreasing the generalizability of our findings to those regions. Finally, we only included patients presenting to the ED so the generalizability of our resistance findings may be limited.
CONCLUSIONS
FQ‐resistant E. coli is widely prevalent and ESBL‐mediated resistance appears to be increasing across US sites highlighting the need for ongoing monitoring of antimicrobial resistance and, at some locations, modification of empirical treatments.
AUTHOR CONTRIBUTION
Study Concept and Design: B.F., M.R., D.T., A.G. Acquisition of Data: A.G., C.B., A.Z., S.R., J.D., L.N., B.P., P.S., A.M., A.B., G.K., B.F., M.R., S.H., G.S., D.Z., G.H., M.C., C.T., M.J., T.F. Analysis and interpretation of data: B.F., M.R., P.V., D.T., A.G. Drafting of manuscript: A.G., C.B., A.Z., S.R., J.D., L.N., B.P., P.S., A.M., A.B., G.K., B.F., M.R., S.H., G.S., D.Z., G.H., M.C., C.T., M.J., T.F., P.V., and D.T. Critical revision of manuscript for important intellectual content, statistical expertise, and acquisition of funding: A.G., C.B., A.Z., S.R., J.D., L.N., B.P., P.S., A.M., A.B., G.K., B.F., M.R., P.V., S.H., G.S., D.Z., G.H., M.C., C.T., M.J., T.F., and D.T.
FUNDING INFORMATION
Some of the work was supported by Spero Therapeutics Investigator Initiated Research Program.
CONFLICT OF INTEREST
The following authors report no conflict of interest: A.G., C.B., A.Z., S.R., J.D., L.N., B.P., P.S., A.M., A.B., and G.K. The following authors received funding by Spero Therapeutics provided unrestricted research grant via its Investigator initiated research program, B.F., M.R., P.V., S.H., G.S., D.Z., G.H., M.C., C.T., M.J., T.F., and D.T. REDCap database was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR002537. CB was partially funded by a research grant from the Mayo Midwest Pharmacy Research Committee.
PRESENTATION
This work was accepted and presented at IDWeek Virtual Conference: 2021.
ACKNOWLEDGMENTS
Keith Burrell, University of Iowa. Devin Spolsdoff, MS, University of Iowa. Joann Huynh, PharmD Candidate (2022), University of Illinois at Chicago College of Pharmacy. Kevin Johns, PharmD Candidate (2022), University of Illinois at Chicago College of Pharmacy. Danielle Garza, PharmD. Monica Frauhiger, PharmD. Lyudmila Garbovsky, PharmD, BCPS, BCCCP, Hospital of the University of Pennsylvania, Philadelphia, PA. David Gajdosik, PharmD, BCPS, Hospital of the University of Pennsylvania, Philadelphia, PA. Rachel Lam, Stritch School of Medicine, Loyola University Chicago. Mollie Shutter, Stritch School of Medicine, Loyola University Chicago.
Faine BA, Rech MA, Vakkalanka P, et al. High prevalence of fluoroquinolone‐resistant UTI among US emergency department patients diagnosed with urinary tract infection, 2018–2020. Acad Emerg Med. 2022;29:1096‐1105. doi: 10.1111/acem.14545
Supervising Editor: Dr. Brigitte Baumann
REFERENCES
- 1. Takhar SS, Moran GJ. Diagnosis and management of urinary tract infection in the emergency department and outpatient settings. Infect Dis Clin North Am. 2014;28:33‐48. [DOI] [PubMed] [Google Scholar]
- 2. Gupta K, Grigoryan L, Trautner B. Urinary tract infection. Ann Intern Med. 2017;167:ITC49‐ITC64. [DOI] [PubMed] [Google Scholar]
- 3. Foxman B. The epidemiology of urinary tract infection. Nat Rev Urol. 2010;7:653‐660. [DOI] [PubMed] [Google Scholar]
- 4. Foxman B. Urinary tract infection syndromes: occurrence, recurrence, bacteriology, risk factors, and disease burden. Infect Dis Clin North Am. 2014;28:1‐13. [DOI] [PubMed] [Google Scholar]
- 5. Talan DA, Takhar SS, Krishnadasan A, et al. Fluoroquinolone‐resistant and extended‐Spectrum β‐lactamase‐producing Escherichia coli infections in patients with pyelonephritis, United States(1). Emerg Infect Dis. 2016;22:1594‐1603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Talan DA, Takhar SS, Krishnadasan A, et al. Emergence of extended‐Spectrum β‐lactamase urinary tract infections among hospitalized emergency department patients in the United States. Ann Emerg Med. 2021;77:32‐43. [DOI] [PubMed] [Google Scholar]
- 7. Gupta K, Hooton TM, Naber KG, et al. International clinical practice guidelines for the treatment of acute uncomplicated cystitis and pyelonephritis in women: a 2010 update by the Infectious Diseases Society of America and the European Society for Microbiology and Infectious Diseases. Clin Infect Dis. 2011;52:e103‐e120. [DOI] [PubMed] [Google Scholar]
- 8. Core Elements of Hospital Antibiotic Stewardship Programs . Centers for Disease Control and Prevention. 2019. 2021. https://www.cdc.gov/antibiotic‐use/healthcare/pdfs/hospital‐core‐elements‐H.pdf [DOI] [PMC free article] [PubMed]
- 9. Antibiotic Resistance Threats in the United States . 2019. 2021. https://www.cdc.gov/drugresistance/biggest‐threats.html
- 10. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Int J Surg. 2014;12:1495‐1499. [DOI] [PubMed] [Google Scholar]
- 11. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata‐driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377‐381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Hooton TM. Uncomplicated urinary tract infection. New England Journal of Medicine. 2012;366:1028‐1037. [DOI] [PubMed] [Google Scholar]
- 13. Abrahamian FM, Moran GJ, Talan DA. Urinary tract infections in the emergency department. Infect Dis Clin North Am. 2008;22(73–87):vi‐87. [DOI] [PubMed] [Google Scholar]
- 14. Jackson GG, Grieble HG, Knudsen KB. URINARY FINDINGS DIAGNOSTIC OF PYELONEPHRITIS. Jama. 1958;166:14‐17. [DOI] [PubMed] [Google Scholar]
- 15. McCarter Y, Burd EM, Hall GS, Zervos M. Cumitech 2c: Laboratory Diagnosis of Urinary Tract Infections. AMS Press; 2009. [Google Scholar]
- 16. Burd EM, Kehl KS. A critical appraisal of the role of the clinical microbiology Laboratory in the Diagnosis of urinary tract infections. J Clin Microbiol. 2011;49:S34‐S38. [Google Scholar]
- 17. Wilson C. Clinical Microbiology Procedures Handbook. 4th ed. AMS Press; 2016. [Google Scholar]
- 18. Jorgensen S, Zurayk M, Yeung S, et al. Emergency department urinary antibiograms differ by specific patient group. J Clin Microbiol. 2017;55:2629‐2636. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Tenney J, Hudson N, Alnifaidy H, Li JTC, Fung KH. Risk factors for aquiring multidrug‐resistant organisms in urinary tract infections: a systematic literature review. Saudi Pharm J. 2018;26:678‐684. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Zhanel GG, Hisanaga TL, Laing NM, et al. Antibiotic resistance in Escherichia coli outpatient urinary isolates: final results from the north American urinary tract infection collaborative Alliance (NAUTICA). Int J Antimicrob Agents. 2006;27:468‐475. [DOI] [PubMed] [Google Scholar]
- 21. Sanchez Guillermo V, Master Ronald N, Karlowsky James A, Bordon JM. In Vitro Antimicrobial Resistance of Urinary Escherichia coli Isolates among U.S. Outpatients from 2000 to 2010. Antimicrob Agents Chemother. 2012;56:2181‐2183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Pitout JDD, Laupland KB. Extended‐spectrum β‐lactamase‐producing Enterobacteriaceae: an emerging public‐health concern. Lancet Infect Dis. 2008;8:159‐166. [DOI] [PubMed] [Google Scholar]
- 23. Kaye KS, Gupta V, Mulgirigama A, et al. Antimicrobial Resistance trends in urine Escherichia coli isolates from adult and adolescent females in the United States from 2011 to 2019: rising ESBL strains and impact on patient management. Clin Infect Dis. 2021;73:1992‐1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Doi Y, Park YS, Rivera JI, et al. Community‐associated extended‐spectrum β‐lactamase‐producing Escherichia coli infection in the United States. Clin Infect Dis. 2013;56:641‐648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Faine BA, Harland KK, Porter B, Liang SY, Mohr N. A clinical decision rule identifies risk factors associated with antimicrobial‐resistant urinary pathogens in the emergency department: a retrospective validation study. Ann Pharmacother. 2015;49:649‐655. [DOI] [PubMed] [Google Scholar]
- 26. Frazee BW, Trivedi T, Montgomery M, Petrovic D‐F, Yamaji R, Riley L. Emergency department urinary tract infections caused by extended‐Spectrum β‐lactamase–producing Enterobacteriaceae: many patients have no identifiable risk factor and discordant empiric therapy is common. Ann Emerg Med. 2018;72:449‐456. [DOI] [PubMed] [Google Scholar]
- 27. Khawcharoenporn T, Vasoo S, Singh K. Urinary tract infections due to multidrug‐resistant Enterobacteriaceae: prevalence and risk factors in a Chicago emergency department. Emerg Med Int. 2013;2013:258517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Taylor RA, Moore CL, Cheung K‐H, Brandt C. Predicting urinary tract infections in the emergency department with machine learning. PLoS One. 2018;13:e0194085. [DOI] [PMC free article] [PubMed] [Google Scholar]