Antimicrobial Resistance in Uropathogenic Escherichia coli Strains Isolated from Relapses from Recurrent Urinary Tract Infections

Abstract

Little is known about the characteristics of uropathogenic Escherichia coli (UPEC) associated with recurrent urinary tract infections (RUTIs). The present study aimed to analyze the phenotypic antimicrobial resistance of recurrent UPEC isolates attributable to either relapse or reinfection. A total of 140 E. coli strains were isolated from 70 outpatients with RUTIs. All isolates were analyzed by random amplified polymorphic DNA-polymerase chain reaction to evaluate genetic similarity between the first and second isolates. We found that 64.2% (45/70) of outpatients had a relapse with the primary infecting E. coli strain and 35.7% (25/70) had reinfection with a new E. coli strain. Compared with reinfecting strains, relapse UPEC isolates exhibited much higher antimicrobial resistance; 89% of these isolates were multidrug-resistant and 46.6% were extended-spectrum β-lactamase producers. Our study provides evidence that RUTIs are mainly driven by the persistence of the original strain in the host (relapses) despite appropriate antibiotic treatments, and only RUTIs attributed to relapses seem to favor multidrug resistance in UPEC isolates.

Introduction

Urinary tract infections (UTIs) are characterized by a high frequency of recurrence, even after appropriate antibiotic therapy,¹ and they are mostly caused by uropathogenic Escherichia coli (UPEC).² Recurrent urinary tract infections (RUTIs) are defined as “relapse” when they are caused by the resurgence of the same strain that caused the original UTI, or as “reinfection” when they are caused by the reintroduction of a new strain into the urinary tract.^3,4 The ability of UPEC to persist at subclinical levels as intracellular reservoirs that are not susceptible to most antibiotic treatments likely contributes to the recurrence of UTIs after cessation of antibiotic administration.^5–7

The emergence of antimicrobial resistance is a major medical concern (particularly in low- and middle-income countries)⁸ that limits therapeutic alternatives and increases morbidity and mortality.⁹ Of particular concern is the increase in extended-spectrum β-lactamase (ESBL)-producing E. coli, causing UTIs in recent years.^10,11 Moreover, UPEC is becoming increasingly resistant to many of the antimicrobials used to treat UTIs. In many areas of the world (e.g., Mexico City), resistance to ciprofloxacin and TMP/SMX has a prevalence of over 55%.^12–14 The impact of multiple antimicrobial treatments on the increase in antimicrobial resistance in patients with RUTIs is still unclear. We believe that recurrent infections, especially those attributable to failure of the initial treatment (i.e., relapse), may lead to antibiotic resistance since persisting bacteria are exposed to multiple antibiotics. Here, we analyze phenotypic antimicrobial resistance of recurrent UPEC isolates attributable to either relapse or reinfection RUTIs and compare them with resistance profiles of isolates from nonrecurrent UTIs.

Materials and Methods

Source of bacterial strains and selection criteria

Urinary samples of outpatients aged 1–70, suffering from RUTI, were collected between 2016 and 2019 from General Hospital “Dr. Manuel Gea González” (SSA, Ministry of Health). Informed consent was obtained from patients who donated urine samples. Sample collection, quantification, and identification were performed as previously described.¹⁴ Isolates that were biochemically confirmed as E. coli were maintained in Luria Bertani (LB) broth with 20% glycerol at –20°C. Seemingly healthy individuals with uncomplicated acute cystitis were selected as urine donors. Samples were collected during the diagnosis of UTIs. Patients were excluded if they met one of the following criteria: if they were pregnant; had diabetes, obstruction, or functional abnormalities of the urinary tract; had been treated with antibiotics for UTI within the last 30 days; were immunocompromised; or if other sources of infection were present. A total of 140 E. coli clinical isolates were obtained from urine samples from 70 outpatients with symptoms of lower UTI: 2 isolates per patient from repeated UTI episodes occurring within an interval between 2 weeks and 6 months. The first E. coli sample was isolated before antimicrobial treatment and the second sample was isolated when the patient returned to the hospital with clinical symptoms of UTI despite conducting the appropriate antimicrobial treatment (after an antibiotic sensitivity test, the patients were treated with an appropriate antimicrobial drug following the clinical practice guidelines established by The Secretary of Health of Mexico). Additionally, 85 strains were isolated from patients with uncomplicated and sporadic UTI (nonrecurrent), without a history of RUTIs; urine samples were collected in the same year and with the same selection criteria as the RUTIs.

Definitions

A case of acute E. coli UTI was defined according to the presence of the following symptoms: significant bacteriuria with an E. coli count of ≥10⁴ CFU/mL of midstream urine, pyuria with ≥5 leukocytes/mm³, and clinical signs or symptoms of UTI (urgency for urination, dysuria, suprapubic pain, or low lumbar/flank pain). RUTI refers to at least two episodes of repeatedly occurring UTIs, documented by urine culture within more than 2 weeks. We classified recurrence as a relapse when a strain from the second episode had an identical random amplified polymorphic DNA-polymerase chain reaction (RAPD-PCR) pattern to that of the initial episode. Recurrence was considered a reinfection when the RAPD-PCR pattern was different between the initial and subsequent episodes.

DNA extraction

We prepared templates for RAPD-PCR by boiling;¹⁵ single colonies were inoculated in 1 mL of LB in a 1.5-mL microcentrifuge tube and were incubated as standing cultures at 37°C for 16 hours. The cells were centrifuged (in a bench centrifuge) at maximum speed for 2 minutes. The pellet was washed twice in sterile distilled water; cells were then suspended in 300 µL of distilled water and boiled for 10 minutes. The supernatant was used immediately as a template source for PCR. The DNA concentration and purity in the supernatant were determined using a Nanodrop One spectrophotometer (Thermo Scientific).

RAPD-PCR analysis

To determine whether two isolates from each patient (70 patients) had identical E. coli clones, a RAPD-PCR typing assay was developed for this study. A PCR was applied to each sample, containing one primer; 1254 (5′-CCGCAGCCAA-3′).^16,17 RAPD analysis of E. coli isolates was performed according to procedures described previously by Bart et al.¹⁸ with a few modifications. Briefly, the mixture used for PCR of RAPD contained: 100 ng of template DNA, 5 pmol/µL (primer 1254), each nucleotide (dATP, dCTP, dGTP, dTTP) at 0.2 mM, 3.0 mM MgCl₂, 1× of buffer 10×, and 1U (1254 primers) of Taq polymerase (Vivantis) in a final volume of 25 µL. The following cycling conditions were used: 94°C for 1 minute, 4 cycles of 94°C for 4 minutes, 36°C for 4 minutes, and 72°C for 4 minutes, and 35 cycles of 94°C for 1 minute, 36°C for 1 minute, and 72°C for 2 minutes with a final 10 minutes elongation step at 72°C. The PCR reaction products were separated on 1.5% (wt/vol) agarose gels in 0.5× TBE buffer or polyacrylamide gels and stained with ethidium bromide (BIO-RAD) for 45 minutes. The BIONUMERICS software (Applied Maths NV), was used for analyzing DNA fingerprint gel images.¹⁹ The Jaccard coefficient was used to determine the percentage of similarity between the samples. The dendrograms were prepared applying the algorithm UPGMA (unweighted pair-group method with arithmetical averages) for the formation of clusters for graphical representation.

Antimicrobial susceptibility testing and detection of ESBL phenotype

The antimicrobials drugs were amikacin (AMK, 30 μg), amoxicillin/clavulanic acid (AMC, 20/10 µg), ampicillin (AMP, 10 μg), cephazolin (KZ, 30 μg), cefepime (FEP, 30 μg), ceftriaxone (CRO, 30 μg), ciprofloxacin (CIP, 5 μg), fosfomycin (FOS, 200 μg), gentamicin (CN, 10 μg), meropenem (MEM, 10 μg), nitrofurantoin (F, 300 μg), and trimethoprim-sulfamethoxazole (SXT, 25 μg). Antibiotic disks were obtained from Oxoid Ltd. UK. The United States National Committee for Clinical and Laboratory Standards (CLSI) guidelines were used to produce the antibiograms (disc diffusion on Mueller-Hinton agar, Oxoid; with incubation for 24 hours at 37°C) and to interpret the inhibition halos.²⁰ The inhibition zones were measured and then classified as sensitive or resistant; reduced susceptibility was also classified as resistance. E. coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853 were used as reference strains for antibiotic disk controls. Multidrug-resistant (MDR) was defined as resistance to three or more different classes of antibiotics. ESBL production was determined by the double-disk synergy test according to Jarlier et al.²¹ This test was performed by placing disks of ceftazidime, cefotaxime, and ceftriaxone (30 μg each) at a distance of 20 mm (from center to center) from a disk containing amoxicillin with clavulanic acid (20 and 10 μg, respectively). The strains that showed synergy between oxyimino-β-lactams and clavulanic acid were classified as ESBL-producing strains. Klebsiella pneumoniae ATCC 700603 was used as the positive control and E. coli ATCC 25922 as the negative control.

Statistical analyses

For statistical analysis, a χ² test was used to assess the significance of differences between groups. For the tests, a p value of <0.05 was considered statistically significant.

Results

Inclusion of the study participants

A total of 140 E. coli strains were isolated from 70 outpatients with RUTIs (two isolates per patient; men, women, or children) with a variety of ages (1–70) from 2016 to 2019. Strains consisted of one isolate before the beginning of antimicrobial treatment and one isolate at the first bacteriological recurrence, after treatment of patients with significant E. coli bacteriuria. All patients had recurrences within 5 months of the first UTI episode. For comparative purposes, the data of patients from nonrecurrent cases are included in Table 1. Patients were 70% female and 30% male; 48.6% of them were 30–59 years old (79.4% females and 20.5% males) (Table 1). Notably, Table 1 shows that 14.2% of the recurrence cases occurred in girls under 14 years of age, while in nonrecurrent UTI cases, the percentage was 4.7; this difference was statistically significant. RUTI cases increased by 4.4% in men older than 60 years compared with nonrecurrent UTI cases.

Table 1.

Age, Gender, and Number of Patients with Recurrent Urinary Tract Infections and Patients with Nonrecurrent Urinary Tract Infections

	RUTI patients		Nonrecurrent UTI patients
	n = 70 (%)		n = 85 (%)
Age (years)	Females	Males	Females	Males	p value
≤14	10 (14.2)	0 (0.0)	4 (4.7)	0 (0.0)	0.0384
15–29	4 (5.7)	1 (1.4)	11 (13)	2 (2.3)	0.1149
30–59	27 (38.5)	7 (10)	34 (40)	11 (13)	0.5881
60–70	8 (11.4)	13 (18.5)	11 (13)	12 (14.1)	0.6861
Total	49 (70)	21 (30)	60 (70.6)	25 (29.4)

A p value <0.05 (bold) was considered statistically significant.

RUTI, recurrent urinary tract infection; UTI, urinary tract infection.

Genetic relatedness by RAPD-PCR

All isolates were analyzed by RAPD-PCR (using primer 1254) to evaluate genetic similarity between the first and second isolates and to discriminate between relapses (i.e., the same strain) or reinfections (a new strain), and based on this RAPD analysis, a dendrogram was constructed (Fig. 1), (Fig. 2). Identical electrophoretic profiles in the first and second isolates evidenced relapses in 45 patients (64.2%), while differences between the first and second isolates evidenced reinfections in 35.7% of them.

FIG. 1.

Dendrogram derived from UPGMA cluster analysis using Jaccard’s similarity indices based on RAPD profiles for 16 strains: 2 strains isolated from the same patient in 2 episodes of infections. Primer 1254 was used in the PCR.

FIG. 2.

Dendrogram derived from UPGMA cluster analysis using Jaccard’s similarity indices based on RAPD profiles for 74 strains: two strains isolated from the same patient in two episodes of infections. Primer 1254 was used in the PCR.

Antimicrobial susceptibility and ESBL phenotypes

The antibiotics tested were those commonly prescribed to treat UTIs. The antimicrobial resistance profiles of E. coli isolates from relapse or reinfection RUTIs cases are compared in Table 2. Antibiotic resistance rates were higher in relapse isolates; there was a statistically significant difference in resistance to 6 out of 12 antibiotics (ampicillin, amoxicillin-clavulanic acid, cefazolin, ciprofloxacin, trimethoprim-sulfamethoxazole, and gentamicin) (Table 2). Interestingly, the resistance profiles of isolates from reinfections and nonrecurrent UTIs were very similar. However, the prevalence of MDR isolates was 89% (40/45) in relapse cases, 52% (26/50) in reinfection cases, and 45.8% (39/85) in nonrecurrent UTI cases.

Table 2.

Percentage of Resistance to Different Antimicrobial Agents in Escherichia coli Isolated from Recurrent Urinary Tract Infections (Relapse or Reinfection) or Nonrecurrent Urinary Tract Infections

Antimicrobials	Number of isolates from RUTIs attributable to:			Number of isolates:
Antimicrobials	Relapse: persistence of the primary infecting strain, n total = 45 (100%)	Reinfection: with a new strain, n total = 50 (100%)	p values (reinfection vs. relapse)	Isolates from nonrecurrent UTIs, n total = 85 (100%)	p values (nonrecurring UTIs vs. relapse)
Ampicillin	41 (91.1)	31 (62.0)	0.0009	57 (67.0)	0.0020
Amoxicillin-clavulanic acid	36 (80.0)	24 (48.0)	0.0012	45 (52.9)	0.0025
Cefazolin	32 (71.1)	25 (50.0)	0.0360	45 (52.9)	0.0449
Ceftriaxone	21 (46.6)	16 (32.0)	0.1433	26 (30.5)	0.0695
Cefepime	20 (44.4)	16 (32.0)	0.2119	25 (29.4)	0.0865
Meropenem	0.0 (0.0)	1 (2.0)	0.3402	0.0 (0.0)	—
Ciprofloxacin	39 (86.6)	34 (68.0)	0.0313	57 (67.0)	0.0155
Trimethoprim-sulfamethoxazole	31 (68.8)	24 (48.0)	0.0395	42 (49.4)	0.0597
Amikacin	2 (4.4)	0.0 (0.0)	0.1319	1 (1.1)	0.2378
Gentamicin	25 (55.5)	16 (32.0)	0.0206	24 (28.2)	0.0022
Fosfomycin	3 (6.6)	2 (4.0)	0.5611	6 (7.0)	0.9332
Nitrofurantoin	9 (20.0)	6 (12.0)	0.2857	5 (5.8)	0.0135

A p value of <0.05 (bold) was considered statistically significant.

The first and second isolates of 40 relapse cases presented identical antimicrobial resistance profile. Only in five relapse cases some differences in resistance were found between the first and second isolates (Table 3). These changes occurred in the isolates; key 12, where the 1st isolate was resistant to SXT and CN and in the 2nd isolate, they changed to susceptible; isolate key 45, the 1st isolate was susceptible to SXT and in the 2nd isolate changed to resistant; isolate key 47, the 1st was resistant to CRO and in the 2nd, changed to susceptible; isolate key 51, the 1st was resistant to CRO and FEP and in the 2nd, changed to susceptible; finally, isolate key 52, the 1st was susceptible to AMK and CN and in the 2nd, changed to resistant (Table 3).

Table 3.

Antimicrobial Resistance to Antimicrobial Agents in 90 Escherichia coli Causing Urinary Tract Infection Relapses, First and Second Isolates from Two Infectious Episodes per Patient

	Antimicrobials
Isolates key	AMP	AMC	KZ	CRO	FEP	MEM	CIP	SXT	AMK	CN	FOS	F
1A 1B
1st	R	R					R	R		R
2nd	R	R					R	R		R
2A 2B
1st							R	R
2nd							R	R
4A 4B
1st	R	R	R	R	R		R			R		R
2nd	R	R	R	R	R		R			R		R
5A 5B
1st	R	R	R	R	R		R	R		R
2nd	R	R	R	R	R		R	R		R
6A 6B
1st	R		R				R	R
2nd	R		R				R	R
11A 11B
1st	R	R								R
2nd	R	R								R
14A 14B
1st	R	R	R	R			R	R				R
2nd	R	R	R	R			R	R				R
17A 17B
1st	R	R	R				R	R
2nd	R	R	R				R	R
12A 12B
1st	R	R					R	R		R
2nd	R	R					R	C-S		C-S
18A 18B
1st	R	R	R	R	R		R	R		R
2nd	R	R	R	R	R		R	R		R
19A 19B
1st	R	R	R	R	R		R			R
2nd	R	R	R	R	R		R			R
20B 20B
1st	R	R	R				R	R
2nd	R	R	R				R	R
21A 21B
1st	R	R	R	R			R	R		R		R
2nd	R	R	R	R			R	R		R		R
22A 22B
1st	R	R	R	R	R		R	R
2nd	R	R	R	R	R		R	R
23A 23B
1st	R	R	R				R	R		R
2nd	R	R	R				R	R		R
24A 24B
1st	R	R	R					R				R
2nd	R	R	R					R				R
25A 25B
1st	R	R	R					R
2nd	R	R	R					R
26A 26B
1st		R					R	R				R
2nd		R					R	R				R
27A 27B
1st	R	R						R
2nd	R	R						R
28A 28B
1st	R						R	R
2nd	R						R	R
29A 29B
1st	R						R	R		R
2nd	R						R	R		R
30A 30B
1st	R	R	R	R	R		R			R
2nd	R	R	R	R	R		R			R
31A 31B
1st	R	R	R	R	R		R	R		R
2nd	R	R	R	R	R		R	R		R
32A 32B
1st	R	R	R	R	R		R			R
2nd	R	R	R	R	R		R			R
33A 33B
1st	R	R	R	R			R
2nd	R	R	R	R			R
34A 34B
1st	R	R	R	R	R		R	R		R
2nd	R	R	R	R	R		R	R		R
35A 35B
1st		R					R	R		R
2nd		R					R	R		R
36A 36B
1st	R						R	R		R
2nd	R						R	R		R
37A 37B
1st	R		R
2nd	R		R
38A 38B
1st	R	R					R	R		R	R
2nd	R	R					R	R		R	R
39A 39B
1st	R	R	R				R	R		R
2nd	R	R	R				R	R		R
40A 40B
1st	R		R	R	R		R	R			R	R
2nd	R		R	R	R		R	R			R	R
41A 41B
1st	R	R	R				R	R
2nd	R	R	R				R	R
42A 42B
1st	R	R						R
2nd	R	R						R
43A 43B
1st	R	R	R	R	R		R		R			R
2nd	R	R	R	R	R		R		R			R
44A 44B
1st							R	R
2nd							R	R
45A 45B
1st	R	R	R	R	R		R	S		R
2nd	R	R	R	R	R		R	C-R		R
46A 46B
1st	R		R		R		R	R		R
2nd	R		R		R		R	R		R
47A 47B
1st	R	R	R	R	R		R
2nd	R	R	R	C-S	R		R
48A 48B
1st	R	R	R	R	R		R		R	R
2nd	R	R	R	R	R		R		R	R
2nd	R	R	R	R	R		R			R
50A 50B
1st	R	R	R		R		R			R	R	R
2nd	R	R	R		R		R			R	R	R
51A 51B
1st	R	R	R	R	R		R	R		R
2nd	R	R	R	C-S	C-S		R	R		C-S
52A 52B
1st	R	R	R	R	R		R		S	S		R
2nd	R	R	R	R	R		R		C-R	C-R		R
53A 53B
1st	R	R	R	R	R		R	R		R
2nd	R	R	R	R	R		R	R		R
No. of resistantstrains of the1st isolates	41	36	32	21	20	0.0	39	31	2	25	3	9

Two isolates per patient (1st and 2nd) from two different infectious episodes of 45 patients, in total 90 isolates.

R, resistant; C-S, change to susceptible; C-R, change to resistant; white box means susceptible.

AMP, ampicillin; AMC, amoxicillin-clavulanic acid; KZ, cefazolin; CRO, ceftriaxone; FEP, cefepime; MEM, meropenem; CIP, ciprofloxacin; SXT, trimethoprim-sulfamethoxazole; AMK, amikacin; CN, gentamicin; FOS, fosfomycin; F, nitrofurantoin.

In 95 isolates from RUTIs, ESBL-producing E. coli were phenotypically detected in 46.6% (21/45) of relapses and 26% (13/50) of reinfections. All ESBL-producing E. coli were MDR in relapses and 11/13 were MDR in reinfections.

Discussion

We sought to determine whether recurrent infections, especially those attributable to the failure of the initial treatment (i.e., relapse), may lead to antibiotic resistance. Here, we analyzed the antimicrobial resistance of E. coli strains from relapse or reinfection RUTIs and compared them with resistance profiles of isolates from nonrecurrent urinary infections; we show that RUTIs are mainly relapses despite appropriate antibiotic treatment, which could lead to the emergence of antimicrobial resistant strains.

We found that 64.2% of the RUTIs were owing to relapses caused by the same UPEC strain, which is consistent with previous studies also using molecular typing methods; relapse rates up to 68% have been reported.^4,22,23 Although our study involves a few cases from a single area of Mexico City, it is worth noting that RUTIs in girls under 14 years of age and in men over 60 years of age were more common than previously known. In our study, relapses occurred even after appropriate antimicrobial treatment indicating that the bacteria were not eliminated from the urinary tract, thus, possibly re-emerging from quiescent intracellular reservoirs that can persist in the bladder,²⁴ or from reinfection with the same bacterial strain from the gut or the vagina.²⁵

We found a significant difference in antimicrobial resistance between isolates from relapses and reinfections (Table 2): a much higher percentage of relapse isolates (18–32% higher than in reinfection isolates) were resistant to antibiotics (e.g., gentamicin, trimethoprim/sulfamethoxazole, ciprofloxacin, cefazolin, amoxicillin/clavulanic ac., and ampicillin). Moreover, 89% of relapse isolates were MDR, while only 52% of reinfection isolates were MDR. Additionally, the number of relapse isolates containing ESBL producers was nearly twice that of reinfection isolates. This big difference in antimicrobial resistance is particularly striking since previous studies reported no significant differences in antimicrobial susceptibility between relapse or reinfection isolates.^26,27

In a similar study, Ahn et al.²⁸ showed that among patients who experienced UTI recurrence caused by ESBL-producing E. coli, 71% had recurrences with ESBL-producing E. coli. In another study conducted in Mexico, recurrent symptomatic UTIs were significantly associated with development of ESBL-producing E. coli causing UTIs.²⁹

We believe that the observed increase in bacterial resistance is owing, in part, to constant exposure of bacteria to antibiotics after numerous treatments selecting resistant bacteria, as resistance patterns are strongly associated with previous antibiotic exposure.³⁰ It is well known that continuous ingestion of antibiotics promotes the emergence and selection for bacterial resistance.³¹ It is noteworthy that the most empirically prescribed antibiotics in Mexico are trimethoprim/sulfamethoxazole, ciprofloxacin, and cephalosporins (cefepime and ceftriaxone),³² which were among the antibiotics with the highest levels of resistance (particularly in relapse isolates).

Bacterial cells called “persisters” are a subpopulation of slow-growing or growth-arrested bacterial cells with decreased susceptibility to antibiotics that coexist with a susceptible clonal bacterial population.³³ We still do not know whether some of our isolates are of the “persister” type; however, accumulating evidence indicates that “persisters” have a significant role in the relapse and recalcitrance of infections.³³

Conclusions

Our study provides evidence that UPEC isolates from RUTIs, especially those that cause relapses (strains that persist in the host for long periods), tend to be MDR. We find that isolates from relapses were more resistant to the antimicrobials tested than isolates from reinfections. However, the biological basis of MDR in these isolates remains to be explored.

Footnotes

Acknowledgment

We thank the Faculty of Medicine, UNAM, for the facilities and donations of reagents.

Authors’ Contributions

S.G.-P.: Conceptualization, methodology, validation, investigation, project administration, writing original draft. P.C.-G.: Methodology, formal analysis, investigation. S.V.-R.: Resourses, validation. R.H.-C.: Investigation, Resourses, writing—review and editing. A.M.-H.: Conceptualization, formal analysis, resourses, visualization, supervision, funding acquisition, writing—review and editing.

Disclosure Statement

The authors declare no conflict of interest.

Funding Information

This work has not been supported by any funding agency.

References

Foxman

. Epidemiology of urinary tract infections: Incidence, morbidity, and economic costs. Am J Med, 2002; 113 Suppl 1A:5S–13S.

Flores-Mireles

, Walker

, Caparon

, Hultgren

. Urinary tract infections: Epidemiology, mechanisms of infection and treatment options. Nat Rev Microbiol, 2015; 13(5):269–284.

Hooton

. Recurrent urinary tract infection in women. Int J Antimicrob Agents, 2001; 17(4):259–268.

Ejrnaes

, Sandvang

, Lundgren

, et al. Pulsed-field gel electrophoresis typing of Escherichia coli strains from samples collected before and after pivmecillinam or placebo treatment of uncomplicated community-acquired urinary tract infection in women. J Clin Microbiol, 2006; 44(5):1776–1781.

Blango

, Mulvey

. Persistence of uropathogenic Escherichia coli in the face of multiple antibiotics. Antimicrob. Agents Chemother, 2010; 54:1855–1863.

Kerrn

, Struve

, Blom

, et al. Intracellular persistence of Escherichia coli in urinary bladders from mecillinam-treated mice. J Antimicrob Chemother, 2005; 55(3):383–386.

Schilling

, Hultgren

. Recent advances into the pathogenesis of recurrent urinary tract infections: The bladder as a reservoir for uropathogenic Escherichia coli. Int J Antimicrob Agents, 2002; 19(6):457–460.

World Health Organization. 2014. Antimicrobial resistance: Global report on surveillance. Available from: http://apps.who.int/iris/bitstream/10665/112642/1/9789241564748_eng.pdf?ua=1. [Last Accessed December 9, 2015]

Shamsrizi

, Gladstone

, Carrara

, et al. Variation of effect estimates in the analysis of mortality and length of hospital stay in patients with infections caused by bacteria-producing extended-spectrum beta-lactamases: A systematic review and meta-analysis. BMJ Open, 2020; 10(1):e030266.

10.

Pitout

, Laupland

. Extended-spectrum beta-lactamase-producing enterobacteriaceae: An emerging public-health concern. Lancet Infect. Dis, 2008; 8:159–166.

11.

Chen

, Ding

, Apisarnthanarak

, et al. The higher prevalence of extended spectrum beta-lactamases among Escherichia coli ST131 in Southeast Asia is driven by expansion of a single, locally prevalent subclone. Sci Rep, 2019; 9(1):13245.

12.

Schito

, Naber

, Botto

, et al. The ARESC study: An international survey on the antimicrobial resistance of pathogens involved in uncomplicated urinary tract infections. Int J Antimicrob Agents, 2009; 34(5):407–413.

13.

Bonkat

, Müller

, Braissant

, et al. Increasing prevalence of ciprofloxacin resistance in extended-spectrum-β-lactamase-producing Escherichia coli urinary isolates. World J. Urol, 2013; 31:1427–1432.

14.

Molina-López

, Aparicio-Ozores

, Ribas-Aparicio

, et al. Drug resistance, serotypes, and phylogenetic groups among uropathogenic Escherichia coli including O25-ST131 in Mexico City. J Infect Dev Ctries, 2011; 5(12):840–849.

15.

Le Bouguenec

, Archambaud

, Labigne

. Rapid and specific detection of the pap, afa, and sfa adhesin-encoding operons in uropathogenic Escherichia coli strains by polymerase chain reaction. J Clin Microbiol, 1992; 30(5):1189–1193.

16.

Venieri

, Vantarakis

, Komninou

, Papapetropoulou

. Differentiation of faecal Escherichia coli from human and animal sources by random amplified polymorphic DNA-PCR (RAPD-PCR). Water Sci. Technol, 2004; 50(1):193–198.

17.

Aslam

, Nattress

, Greer

, et al. Origin of contamination and genetic diversity of Escherichia coli in beef cattle. Appl Environ Microbiol, 2003; 69(5):2794–2799.

18.

Bart

, Schuurman

, Achtman

, et al. Randomly amplified polymorphic DNA genotyping of serogroup A meningococci yields results similar to those obtained by multilocus enzyme electrophoresis and reveals new genotypes. J Clin Microbiol, 1998; 36(6):1746–1749.

19.

Benacer

, Mohd Zain

, Sim

, et al. Determination of Leptospira borgpetersenii serovar Javanica and Leptospira interrogans serovar Bataviae as the persistent Leptospira serovars circulating in the urban rat populations in Peninsular Malaysia. Parasit Vectors, 2016; 1(9):117.

20.

Clinical and Laboratory Standards Institute. 2019. Performance standards for antimicrobial susceptibility testing. 29th ed. CLSI supplement M100: Clinical and Laboratory Standards Institute. Wayne, PA.

21.

Jarlier

, Nicolas

, Fournier

, Philippon

. Extended broad-spectrum β-lactamases conferring transferable resistance to newer β-lactam agents in Enterobacteriaceae: Hospital prevalence and susceptibility patterns. Rev Infect Dis, 1988; 10(4):867–878.

22.

Russo

, Stapleton

, Wenderoth

, et al. Chromosomal restriction fragment length polymorphism analysis of Escherichia coli strains causing recurrent urinary tract infections in young women. J Infect Dis, 1995; 172(2):440–445.

23.

Kõljalg

, Truusalu

, Vainumäe

, et al. Persistence of Escherichia coli clones and phenotypic and genotypic antibiotic resistance in recurrent urinary tract infections in childhood. J Clin Microbiol, 2009; 47(1):99–105.

24.

Mysorekar

, Hultgren

. Mechanisms of uropathogenic Escherichia coli persistence and eradication from the urinary tract. Proc Natl Acad Sci U S A, 2006; 103(38):14170–14175.

25.

Hooton

. Uncomplicated urinary tract infection. N Engl J Med, 2012; 366(11):1028–1037.

26.

Ejrnæs

, Stegger

, Reisner

, et al. Characteristics of Escherichia coli causing persistence or relapse of urinary tract infections: Phylogenetic groups, virulence factors and biofilm formation. Virulence, 2011; 2(6):528–537.

27.

Luo

, Ma

, Zhao

, et al. Similarity and divergence of phylogenies, antimicrobial susceptibilities, and virulence factor profiles of Escherichia coli isolates causing recurrent urinary tract infections that persist or result from reinfection. J Clin Microbiol, 2012; 50(12):4002–4007.

28.

Ahn

, Kim

, et al. Does urinary tract infection caused by extended-spectrum β-lactamase-producing Escherichia coli show same antibiotic resistance when it recurs? J Infect Chemother, 2019; 25(7):498–502.

29.

Alcántar-Curiel

, Alpuche-Aranda

, Varona-Bobadilla

, et al. Risk factors for extended-spectrum b-lactamases-producing Escherichia coli urinary tract infections in a tertiary hospital. Salud Publica Mex, 2015; 57(5):412–418.

30.

Holmes

, Moore

LSP

, Sundsfjord

, et al. Understanding the mechanisms and drivers of antimicrobial resistance. Lancet, 2016; 387(10014):176–187.

31.

Ben

, Fu

, Hu

, et al. Human health risk assessment of antibiotic resistance with antibiotic residues in the environment: A review. Environ Res, 2019; 169:483–493.

32.

Miranda-Novales

, Flores-Moreno

, López-Vidal

, et al. Limited therapeutic options in mexico for the treatment of urinary tract infections. Antibiotics, 2022; 11(11):1656.

33.

Fisher

, Gollan

, Helaine

. Persistent bacterial infections and persister cells. Nat Rev Microbiol, 2017; 15(8):453–464.