Setting, sample collection and processing

The Laboratory of Clinical Microbiology and Infectious Diseases of Isala Hospital in Zwolle provides infectious disease diagnostics for over 250 general practitioners (GPs) in the northeast of the Netherlands. The study was conducted from October 2022 to June 2024, involving patients from two general practices in Zwolle. All patients suspected of acute UTI according to the Dutch College of general Practitioners (NHG) UTI guideline were included [2]. An acute urinary tract infection is an infection of the urinary tract characterized by the sudden onset of symptoms such as painful or burning urination, frequent urge to urinate, urgency without actual urine production, and possibly blood in the urine [2]. Patients with an indwelling catheter were excluded as well as urine samples sent for treatment control purposes. Patients were included more than once if they presented with a new UTI episode at least 28 days after their previous visit during the study period.

The general practitioners in the two participating practices follow this guideline: when a UTI is suspected, a medical history is taken and, if necessary, a physical examination is performed. Preferably, the first morning void of (midstream) urine is tested for nitrate and leukocytes. If the urine test strip results are inconclusive, a dip slide test is performed to confirm bacterial growth. Treatment is then prescribed blindly according to the guideline. In most cases, no initial urine sample is sent to the laboratory for culture.The guideline recommends culturing only after one or more therapy failures, unless the patient belongs to a special category: elderly, child, pregnant, immunocompromised, etc., or if tissue invasion is suspected. For our study, between October 11^th, 2022, up to and including June 11^th, 2024, of each patient, the first urine sample was sent to our laboratory and cultured according to standard laboratory procedures. All samples were processed as follows: urine was cultured on a set of selective agar plates and a Gram stain is made. Bacterial growth is evaluated based on potential uropathogens in relation to the quantity of urogenital flora and the number of leucocytes. Isolates were identified with Matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF; Bruker Daltonik, Bremen, Germany), antimicrobial susceptibility testing (AST) was performed using the Vitek 2 system (bioMérieux, Marcy-l’Ėtoile, France) with EUCAST breakpoints. Culture results were reported to the GP as usual.

The local Medical Research Ethics Committee considered the study not subject to the Dutch Medical Research Involving Human Subjects Act (WMO) and declared it to be exempted from further review. No consent from the participants was needed.

Data comparison and follow up

Resistance percentages were calculated for all E. coli strains isolated from the included cultures for the oral antibiotics ampicillin/amoxicillin, amoxicillin-clavulanic acid, cefuroxime axetil, ciprofloxacin, fosfomycin, nitrofurantoin, trimethoprim, trimethoprim-sulfamethoxazole and third generation cephalosporin cefotaxime/ceftriaxone. Of these, the oral agents nitrofurantoin, fosfomycin and trimethoprim are the most commonly used to treat acute UTI. Pivmecilliman is currently being investigated in the Netherlands and is not (yet) used as a first-line treatment for UTI.

The observed local resistance rates were compared with data from 2023 in the Dutch national AMR database Infectious Disease Surveillance Information System for Antibiotic Resistance (ISIS-AR, accessed at www.ISIS-web.nl in January 2025), where national AST data sent in by most Dutch clinical microbiology laboratories are registered by the National Institute for Public Health and the Environment (RIVM) [7]. Querying the database provides data on the first entered isolate per species per patient per calendar year. If two Escherichia coli isolates with different resistance patterns were cultured from one sample, the first isolate is selected.

Comparison was also made with regional data from the same period as from the two participating general practices, i.e., obtained from all other general practices in the region that consistently submit urine samples to our laboratory and of which identification and AST data are stored in our laboratory information system (LIS). The same selection method as was used for the national database was applied here. After data collection and analysis, the data could not be traced back to individual participants.

Demographic data comprising participants’ age and sex were recorded. Differences in resistance rates were calculated using SPSS version 28, with a p-value < 0.05 considered statistically significant.

To assess treatment success or failure we recorded which antibiotic was empirically prescribed after the first urine sample was collected, as well as the frequency and, if possible, the reason of patient return visits within 28 days thereafter. Treatment failure was defined as microbiological, i.e., the cultured uropathogen was found to be resistant against the administered antibiotic, or as clinical: the findings indicated colonization and no true UTI, or the case involved a complicated UTI for which the treatment proved ineffective.

Culture results

In total 366 first urine samples were collected from 272 patients (243 female (89%), 29 male), with a median age of 57 years (range 1–98 years). Cultured uropathogens are shown in Table 1. Mixed flora (≥ 3 species) was reported in 19 cultures (5.2%), in 27 cultures (7.4%) urethral flora only, and in one sample (0.3%) no growth was found. E. coli was the most frequently isolated pathogen (82.2%), followed by Staphylococcus saprophyticus and Klebsiella pneumoniae.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Uropathogens cultured from first urine samples.

https://doi.org/10.1371/journal.pone.0334222.t001

Resistance rates of E. coli

Resistance rates of the 273 E. coli study isolates were compared to regional percentages (10,872 isolates) and to national percentages (approximately 138,000 isolates), Table 2. National ISIS-AR database resistance rates for amoxicillin/ampicillin, cefuroxime, ciprofloxacin, fosfomycin, trimethoprim and trimethoprim-sulfamethoxazole were significantly higher than those found in the study of isolates recovered from first urine samples. Regional rates were higher for amoxicillin/ampicillin and ciprofloxacin compared to the study rates.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. E. coli resistance rates.

https://doi.org/10.1371/journal.pone.0334222.t002

Therapy outcome and follow up

Of the 272 participating patients, for 256 patients no follow-up cultures were submitted within 28 days after the first urine sample. Of these, 207 patients were registered with one UTI episode, others with two (n = 36), three (n = 13), four (n = 3) and five (n = 1) recurring UTIs. This total of 335 episodes were apparently successfully empirically treated with nitrofurantoin (n = 245), fosfomycin (n = 30), ciprofloxacin (n = 24), trimethoprim (n = 8), amoxicillin (n = 5) and amoxicillin/clavulanic acid (n = 4). Fourteen suspected UTIs were not treated and in five cases the treatment was unknown.

In 19 cases (18 patients), a repeat urine sample was cultured within 28 days after the first sample because of persistent symptoms consistent with a UTI. Of the 13 patients empirically treated with nitrofurantoin, one patient had a culture with a nitrofurantoin resistant Klebsiella pneumoniae. For the other twelve patients the reason of treatment failure was unknown. Unexplained treatment failure also occurred after prescription of fosfomycin (n = 3) and ciprofloxacin (n = 1). One patient who was initially treated with ciprofloxacin was admitted immediately with an urosepsis, which was subsequently treated successfully with ceftriaxone. One patient who did not receive treatment at the first visit kept complaints, which were successfully treated with nitrofurantoin prescribed at the second visit.

Successful empirical treatment with nitrofurantoin was achieved in 245 of 258 cases (95%). Success rates for fosfomycin, ciprofloxacin and trimethoprim were 90% (27/30), 92% (24/26), and 100% (8/8) respectively.