Interaction between Antibiotic Resistance, Resistance Genes, and Treatment Response for Urinary Tract Infections in Primary Care

Hanley J. Ho, Mei Xuan Tan, Mark Chen, Thean Yen Tan, Seok Hwee Koo, Agnes Koong, Lok Pui Ng, Pei Hu, Kee Tung Tan, Peter Kirm Seng Moey, Yi Ling Eileen Koh, Chia Siong Wong, David Chien Lye, Ngiap Chuan TanJournal of Clinical Microbiology 2019 · DOI

TL;DR

This study investigated antibiotic resistance patterns and their impact on treatment outcomes for urinary tract infections (UTIs) in primary care in Singapore. It found that while empirical antibiotic treatment generally led to better symptom resolution than no treatment, the correlation between in vitro susceptibility and actual clinical response was imperfect, highlighting the need for nuanced approaches to UTI management.

What Problem Does This Paper Address?

Urinary tract infections (UTIs) are incredibly common, especially among women, and are frequently treated empirically in primary care without waiting for urine culture and susceptibility results. This approach is often justified by the predictability of common uropathogens and their resistance profiles. However, rising antimicrobial resistance (AMR) threatens this paradigm, potentially leading to poorer outcomes like delayed symptom resolution and increased medical consults. This paper addresses the critical need to understand current AMR patterns in primary care settings, identify factors associated with resistance to commonly prescribed antibiotics, and evaluate how antibiotic susceptibility and resistance genes influence early symptom resolution in patients with UTIs. This knowledge is crucial for guiding appropriate empirical treatment strategies and potentially refining when urine cultures are necessary.

Methods

The researchers conducted a prospective cohort study in three public primary care clinics in Singapore between 2015 and 2016. They recruited adult patients presenting with UTI symptoms and collected urine samples for culture. Uropathogens were identified using standard phenotypic methods, and antibiotic susceptibility testing was performed using automated systems and supplemental disc diffusion tests.

A key aspect of their methodology was the investigation of specific resistance genes. For isolates showing reduced susceptibility to certain antibiotics (like ceftriaxone, ciprofloxacin, and fosfomycin), they performed molecular testing to detect common plasmid-mediated resistance genes, including bla<sub>CTX-M</sub>, ampC, qnr, and fos.

Patient follow-up was conducted via telephone questionnaires to assess early symptom resolution, defined as resolution within 3 days of the initial consultation. Statistical analyses included univariate and multivariable logistic regression to identify factors associated with antibiotic resistance and early symptom resolution.

Datasets

The study utilized data collected from 743 patients screened for UTI symptoms, with 695 eligible and 299 ultimately included in the analysis due to positive urine cultures. The dataset comprised demographic information, medical history, antibiotic susceptibility profiles of isolated uropathogens, and patient-reported symptom resolution outcomes. No new benchmarks were introduced; the study focused on analyzing existing clinical data within a specific primary care context.

Key Insights and Ideas

  • Imperfect Correlation: The most striking insight is that in vitro antibiotic susceptibility doesn't perfectly predict clinical outcomes. While patients receiving antibiotics active against their uropathogens generally fared better, the difference between susceptible and non-susceptible isolates wasn't always as pronounced as expected, particularly for ciprofloxacin.
  • Resistance Genes in the Community: The study detected common resistance genes (bla<sub>CTX-M</sub>, ampC, qnr, fos) in a significant proportion of isolates, even in primary care. Crucially, a majority of these resistant isolates were from patients without recent hospitalization, suggesting that AMR acquisition might be occurring more frequently in the community than previously assumed.
  • Risk Factors for Resistance: Specific patient characteristics were linked to resistance against certain antibiotics. Indian ethnicity and diabetes mellitus were associated with amoxicillin-clavulanate resistance, while genitourinary abnormalities, recent UTIs, and recent hospitalizations were linked to ciprofloxacin and co-trimoxazole resistance. These factors could help clinicians decide when to order a urine culture.
  • Empirical Therapy Still Matters: Despite the imperfect correlation, receiving any empirical antibiotic was significantly better for early symptom resolution than receiving no antibiotics. This underscores the continued importance of empirical treatment in primary care for uncomplicated UTIs.

Key Results

  • Out of 299 patients with culture-confirmed UTIs, 259 (87%) were female. Escherichia coli was the most common uropathogen (76%).
  • Enterobacteriaceae isolates showed high susceptibility to amoxicillin-clavulanate (86%), nitrofurantoin (87%), and fosfomycin (98%). However, resistance exceeded 20% for ciprofloxacin and co-trimoxazole.
  • Among isolates resistant to indicator antibiotics, resistance genes were detected: bla<sub>CTX-M</sub> (54%), plasmid-mediated ampC (50%), qnr (10%), and fos (17%).
  • Symptom Resolution: 67% of patients treated with antibiotics to which their isolates were susceptible reported early symptom resolution, compared to 45% of those treated with antibiotics to which their isolates were non-susceptible (P = 0.001). Patients not treated with antibiotics had only 27% early resolution (P = 0.018).
  • Amoxicillin-clavulanate: Patients with susceptible isolates had a 70% early resolution rate, versus 44% with non-susceptible isolates (P = 0.036).
  • Ciprofloxacin: Early resolution rates were similar for patients with susceptible (48%) and non-susceptible (54%) isolates (P = 0.762), suggesting a potential issue with this antibiotic's effectiveness or dosing in this setting.
  • Risk Factors: Indian ethnicity and diabetes mellitus were independently associated with amoxicillin-clavulanate resistance. Recent UTI, hospitalization, and genitourinary abnormalities were associated with ciprofloxacin and co-trimoxazole resistance.

Limitations and Future Work

The authors acknowledge several limitations:

  • Catchment Population: The study focused on polyclinic patients, who are generally elderly and may not represent the broader primary care population.
  • UTI Subtypes: The study did not differentiate between various UTI subtypes (e.g., uncomplicated cystitis vs. pyelonephritis), which could influence treatment response.
  • Short Follow-up: The 3-day endpoint for early symptom resolution might have been too short to capture all complications or the full impact of treatment.
  • Small Sample Sizes: Some subgroup analyses were limited by the number of participants.
  • Culture-Negative UTIs: The study only included culture-positive UTIs, excluding a significant proportion of symptomatic women who may still have UTIs.

Future work could involve longer follow-up periods, inclusion of culture-negative UTI patients, and further investigation into optimal dosing strategies, particularly for antibiotics like ciprofloxacin in elderly populations. The authors also suggest the need for strategies beyond traditional surveillance, potentially incorporating routine monitoring of symptom resolution and genotypic resistance markers in primary care.

Summary updated

Abstract

Given increasing antimicrobial resistance, we aimed to determine antibiotic susceptibility and presence of resistance genes in uropathogens in primary care, factors associated with resistance to commonly prescribed antibiotics, and effect of treatment on early symptom resolution. We conducted a prospective study of primary care patients with urinary tract infection (UTI) symptoms and culture-confirmed UTI in Singapore from 2015 to 2016. Cohort characteristics and antimicrobial susceptibility of cultured isolates were analyzed. Among <i>Enterobacteriaceae</i> isolates, early symptom resolution (within 3 days) according to antibiotic prescribed and isolate susceptibility and factors associated with antibiotic resistance were evaluated. Of 695 symptomatic patients, 299 were urine culture positive; of these 299 patients, 259 (87%) were female. <i>Escherichia coli</i> was the most common uropathogen (76%). <i>Enterobacteriaceae</i> isolates (<i>n</i> = 283) were highly susceptible to amoxicillin-clavulanate (86%), nitrofurantoin (87%), and fosfomycin (98%), but >20% were resistant to ciprofloxacin and co-trimoxazole. Isolates resistant to appropriate indicator antibiotics were further tested to determine proportions positive for <i>bla</i><sub>CTX-M</sub> (14/26, 54%), plasmid-mediated <i>ampC</i> (12/24, 50%), <i>qnr</i> (7/69, 10%), and <i>fos</i> (1/6, 17%) resistance genes. A total of 67% of patients given antibiotics with susceptible isolates reported early resolution versus 45% given antibiotics with nonsusceptible isolates (<i>P</i> = 0.001) and 27% not treated (<i>P</i> = 0.018). On multivariable analysis, Indian ethnicity and diabetes mellitus were associated with amoxicillin-clavulanate resistance. Genitourinary abnormalities, UTI in the past 12 months, and hospitalization in the past 6 months were associated with ciprofloxacin and co-trimoxazole resistance. Patients given active empirical antibiotics were most likely to report early symptom resolution, but correlation with <i>in vitro</i> susceptibility was imperfect. Factors associated with resistance may guide the decision to obtain initial urine culture.
BibTeX
@article{ho2019interaction,
  author = {Hanley J. Ho and Mei Xuan Tan and Mark Chen and Thean Yen Tan and Seok Hwee Koo and Agnes Koong and Lok Pui Ng and Pei Hu and Kee Tung Tan and Peter Kirm Seng Moey and Yi Ling Eileen Koh and Chia Siong Wong and David Chien Lye and Ngiap Chuan Tan},
  title = {{Interaction between Antibiotic Resistance, Resistance Genes, and Treatment Response for Urinary Tract Infections in Primary Care}},
  journal = {Journal of Clinical Microbiology},
  year = {2019},
  doi = {10.1128/jcm.00143-19},
}
Figure