Introduction

Urinary-tract infections (UTIs) are the second most common cause of community-acquired infections, with Escherichia coli being the most common causative pathogen [1]. The great majority of UTIs are easily manageable, although some patients experience frequent relapses [2]. Development of resistance is a common characteristic of uropathogenic microorganisms in the case of relapse [3].

Cefprozil and cefixime are antimicrobials introduced in the market more than two decades ago. Cefprozil is a cephem antimicrobial and cefixime is an orally available third-generation cephalosporin. Since enterobacteriaceae belong to their antimicrobial spectrum, both these antimicrobials are suggested for the management of UTIs [4, 5]. For many years, cefixime was considered an ideal alternative for patients with acute pyelonephritis switching from intravenous to oral therapy [5]. The application of both these agents in the therapeutic armamentarium against UTIs has been abandoned over the years, whereas their activity on urinary pathogens is not reported in studies published over the last 5 years. The emerging resistance of uropathogenic enterobacteriaceae to commonly prescribed antimicrobials [3] led us to conduct the current study to investigate the activity of cefprozil and cefixime against enterobacteriaceae pathogens from patients with community-acquired UTIs.

Methods

This was a multicenter study that was conducted during the period November 2012 until April 2013 among patients admitted for urine culture into 10 different microbiology laboratories in Greece. The study was approved by the Ethics Committees of the Prefectures that the labs refer to. Inclusion criteria were: (a) written informed consent; (b) female gender; (c) age ≥18 years; (d) at least two of the following symptoms of acute cystitis, i.e., micturition, pain at urination and increased frequency of urination; (e) one Gram-negative isolate grown at quantity ≥105 cfu/ml from midstream urine culture; (f) uncomplicated cystitis; and (g) community-acquired UTI. The episode of cystitis was considered uncomplicated for non-pregnant immunocompetent women without signs of urinary obstruction on ultrasound, without any abnormal findings on gynecologic evaluation and without any urinary catheter. Each patient could be enrolled once in the study. For the UTI to be considered community acquired, the patients should not have any contact with health care systems for more than 90 days (i.e., no hospitalization, no residence in long-term care facilities, no out-patient drug infusions and no patients under chronic hemodilution).

All isolates were transported onto slant agar into the central lab of the 4th Department of Internal Medicine at ATTIKON University Hospital. After re-culture onto MacConkey agar (BBL Becton–Dickinson, Cockeysville, MD, USA), susceptibilities to ampicillin, amoxicillin/clavulanate, cefoxitin, cefotaxime, ceftazidime, cefepime, gentamycin, amikacin, fosfomycin and trimethoprim/sulfamethoxazole were determined by the disk diffusion method using commercially available antimicrobial-impregnated disks (Oxoid Ltd, London, UK). Susceptibilities were interpreted by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) breakpoints [6]. Extended-spectrum β-lactamases (ESBL) production was tested with the double disk approximation test [7].

All ESBL-producing isolates, all non-E. coli isolates and almost half of the non-ESBL-producing E. coli isolates were selected for further study of susceptibilities to cefprozil and cefixime. This was done by selecting half of the non-ESBL-producing E. coli coming from each laboratory. Minimum inhibitory concentrations (MIC) to cefprozil, cefixime and ciprofloxacin were determined by a microdilution technique at a 0.1 ml final volume using one 5 × 105 cfu/ml log-phase inoculum. Commercially available antibiotic powders were purchased (Sigma Co, St. Louis, USA). MIC was considered the lower antimicrobial concentration inhibiting visible bacterial growth after 18 h of incubation at 35 °C. The EUCAST MIC susceptibility breakpoints were used [6]. Minimum bactericidal concentrations (MBC) were determined after plating the content of clear wells onto MacConkey agar performing three times 1:10 serial dilutions. MBC was considered as the lowest concentration which resulted in the killing of 99.9% of the plated inoculum. American Tissue Cell Collection isolate E. coli 25,922 was used as a reference strain.

MIC50 and MBC50 were determined as the MIC and MBC respective values that inhibited/killed 50% of isolates; MIC90 and MCB90 were determined as the respective values inhibiting/killing 90% of isolates. Susceptibilities were compared by the Fischer exact test (SPSS Statistics version 22.0, IBM, NY, USA). Any value of p below 0.05 was considered statistically significant.

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1964, as revised in 2013. Informed consent was obtained from all patients for being included in the study.

Results

A total of 747 isolates were collected belonging to the species of E. coli, Klebsiella pneumoniae and Proteus mirabilis (Fig. 1). Overall prevalence of ESBL production was 6.7%. E. coli was the most common pathogen isolated from 85.7% of patients. Among E. coli isolates, 93.8% were non-ESBL producers and 6.2% were ESBL producers. Among K. pneumoniae isolates, 81.8% were non-ESBL producers and 18.2% were ESBL producers. None of the P. mirabilis isolates were ESBL producers.

Fig. 1
figure 1

Flowchart of the selection of isolates for the study. ESBL extended-spectrum β-lactamase

Full size image

Against non-ESBL-producing E. coli, cefprozil was more active than ciprofloxacin (93.7% vs 80.2% of isolates inhibited, p < 0.0001). This was not the case for cefixime (85.7% vs 80.2% of isolates inhibited, p 0.125) (Table 1).

Table 1 Susceptibility patterns of 400 urinary pathogens to cefprozil, cefixime and ciprofloxacin in relation to their resistance phenotype
Full size table

Although the activity of cefprozil and cefixime was limited against ESBL-producing isolates, they were active against isolates resistant to ciprofloxacin (Table 2).

Table 2 Comparative susceptibility of 13 antimicrobials against urinary-tract pathogens in relation to resistance phenotype
Full size table

Discussion

Current results suggest that in an era of emerging antimicrobial resistance, cefprozil and cefixime retain good activity against enterobacteria from patients with community-acquired UTI. However, both antimicrobials are not active against isolates that produce ESBL. Both tested drugs are available for oral administration. Cefixime was studied instead of other oral third-generation cephalosporins because it is the only drug of this class still in the market both in Greece and in many other European countries. The plasma half-life of cefprozil is 1.3 h and its bioavailability is 100%; it should be administered at a 500-mg twice-daily oral dose for 5–7 days for the treatment of UTIs. The plasma half-life of cefprozil is 4 h and its bioavailability is 50%; it should be administered at a 400-mg once-daily oral dose for 5 days for the treatment of UTIs [8].

Current guidelines suggest oral quinolones as the first-line treatment for community-acquired UTIs [9]. However, resistance to quinolones has emerged and mounts close to 20% as reported in a large survey of 499 E. coli isolates from Germany [10]. This increase in resistance of E. coli urine isolates to ciprofloxacin seems to be a worldwide phenomenon [2]. A retrospective analysis of the yearly susceptibility trends of 1107 community E. coli urine isolates in the Adelaide and Meath Hospital, that is a national tertiary reference center for urology in Ireland, indicated resistance rate of 10.6%. This amounts to 17.8% for hospital-acquired isolates and to 28.6% for isolates coming from patients hospitalized in urology departments [2]. However, the resistance rates to ciprofloxacin of 723 community-acquired urine E. coli isolates from inpatients enrolled in the SMART program of antimicrobial surveillance in USA during the years 2009–2011 was 22.4%. This was 7.4% for the 167 community-acquired urine isolates of K. pneumoniae [11]. The main driver for acquisition of ciprofloxacin resistance in the community is consumption of ciprofloxacin the last 3 months associated with odds ratio 4.20 [3]. In our study, cefprozil inhibited E. coli at a rate significantly greater than ciprofloxacin; both cefprozil and cefixime retained good activity against ciprofloxacin-resistant isolates. As a consequence, they remain a good therapeutic alternative either for empirical treatment, in the case of high suspicion for ciprofloxacin-resistant pathogens, or when one ciprofloxacin-resistant pathogen is isolated from urine.

One huge emerging problem in the management of community-acquired UTIs is enterobacteria pathogens that are β-lactam-resistant through the production of ESBL. Current publications suggest that the prevalence of these isolates ranges between 2 and 6% in E. coli and it is higher in K. pneumoniae [10, 12, 13]. These rates are in accordance with the 6.7% prevalence reported in the present study. Female gender, recurrent UTIs and presence of comorbidities are the most common risk factors for the acquisition of these isolates [12, 13]. Our findings suggest that both cefprozil and cefixime have poor activity against these isolates and that they cannot be suggested for management. Their empirical use should not be considered when there is increased clinical suspicion for UTIs by ESBL-producing pathogens.

The current study did not focus on epidemiological information of patients with UTIs caused by ESBL-producing isolates and ciprofloxacin-resistant isolates. This is because a previous large-scale epidemiological study in Greece has clearly shown the impact of previous antimicrobial consumption in the last 3 months as a risk factor for the emergence of these resistant isolates [3]. Major emphasis was given on the use of cefprozil and cefixime as alternative treatments for UTIs.

Conclusion

The results of the present study indicate that cefprozil and cefixime retain good activity against urine pathogens. According to our findings, they can be empirically used for management. Their use is particularly encouraged in the following cases: (a) UTIs with documented or high suspicion for implication of ciprofloxacin-resistant pathogens; and (b) UTIs by pathogens cross-resistant to other antimicrobials provided that they do not produce ESBL.