Introduction

Antimicrobial pressure, prescribing practices, and cultural factors, such as drug availability and cost, produce significant regional susceptibility differences among certain bacterial pathogens [1–6]. Individual countries first-line treatment options differ and may not be appropriate due to variable resistance patterns in local environments [2].

Country-specific surveillance data are available in most regions, but can be confounding based on the variation of results obtained from different investigations and differences in the susceptibility breakpoints applied in these studies. Regardless, surveillance studies provide useful information to primary care physicians who need to prescribe rational empiric therapy.

Community-acquired urinary tract infections (CA-UTI) and upper respiratory tract infections (CA-URTI) are the leading causes of outpatient infections and the most empirically treated infections worldwide [1, 6, 7]. Increased resistance to antibiotics has complicated the management of both of these outpatient infections. The spread of Enterobacteriaceae that carry a chromosomally mediated AmpC β-lactamase, an extended-spectrum β-lactamase (ESBL), or a carbapenemase is becoming a significant concern in the community [8–11]. Multidrug-resistant Streptococcus pneumoniae and β-hemolytic streptococci with tolerance to penicillin and resistance to macrolides and fluoroquinolones are increasing in Japan and extreme drug-resistant S. pneumoniae have been observed among multidrug-resistant isolates in Canada [12, 13].

Regional and country variations in pathogen resistance and the susceptibility breakpoint differences that are applied must be considered when determining empiric treatment options for both CA-UTI and CA-URTI [14, 15]. Many countries’ adopt or design their own recommendations of treatment guidelines based on local surveillance data which can be scarce in some countries [1–6].

The objective of this study was to investigate the contemporary pattern of antimicrobial susceptibility among pathogens causing CA-UTI and CA-URTI in Argentina, Mexico, Venezuela, Russia, and the Philippines. These five countries have limited local surveillance data available for understanding susceptibility patterns among orally prescribed antimicrobial agents for common outpatient infections.

Methods

Microbiology laboratories in five countries which have scare surveillance data were recruited to collect isolates from community-acquired infections. Only the first isolate collected from a patient who was in a clinic, a physician’s office, a hospital emergency room, or in a community hospital for <48 h at the time of collection was included. Patients were to have had no prior antibiotic exposure within the previous 90 days. Bacterial species from patients with CA-UTI included Escherichia coli, Klebsiella spp., Proteus mirabilis, and other less common Enterobacteriaceae. Species from CA-URTI included S. pneumoniae, Streptococcus pyogenes, H. influenzae, and M. catarrhalis, and a limited number of Enterobacteriaceae. A total of 1501 strains were collected during 2012 (29%) and 2013 (71%) from 12 medical centers, including two each in Argentina, Mexico, Venezuela, and the Philippines and four in Russia.

Identification of the bacterial isolates at the study site was performed using routine laboratory procedures and confirmed by matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry (MALDI Biotyper, Microflex, Bruker Daltonik GmbH, Bremen, Germany) at a reference laboratory (International Health Management Associates [IHMA], Schaumburg, IL, USA). Minimal inhibitory concentrations (MIC) were determined using the Clinical Laboratory Standards Institute (CLSI) broth microdilution procedure and panels prepared by IHMA using CLSI guidelines [15]. Susceptibility breakpoints used were those according to the CLSI and European Committee for Antimicrobial Susceptibility Testing (EUCAST) guidelines [16, 17]. Applied EUCAST breakpoints were those utilized for uncomplicated urinary tract infections which included amoxicillin-clavulanic acid, cefuroxime (oral dosing), cefixime, cefpodoxime, ceftibuten, fosfomycin (oral dosing), and nitrofurantoin. The oral CLSI cefuroxime interpretive criteria were used for Enterobacteriaceae. E. coli ATCC 25922, Pseudomonas aeruginosa ATCC 27853, Staphylococcus aureus ATCC 29213, H. influenzae ATCC 49247, and S. pneumoniae ATCC 49619 were used as quality control (QC) strains. The QC MIC ranges utilized were those of the CLSI [18]. Isolates with an ESBL phenotype were confirmatory tested using cefotaxime ± clavulanic acid and ceftazidime ± clavulanic acid. The confirmatory test was performed on all E. coli, Klebsiella pneumoniae, Klebsiella oxytoca, and P. mirabilis with resistance to cefpodoxime (MICs > 1 mg/L) [18].

This article does not contain any new studies with human or animal subjects performed by any of the authors.

Results

Among the 960 isolates collected from all infection sources with a patient age provided, 331 were from pediatric patients and 629 were from adults (≥18 years old). With respect to patient gender, among the 957 patients with this information recorded, 601 were female and 356 were male.

Enterobacteriaceae from CA-UTI included 345 isolates of E. coli, 87 K. pneumoniae, 68 P. mirabilis, 27 Enterobacter cloacae, and 40 isolates of other species. Isolates were obtained from patients aged 0–17 (n = 76), 18–39 (n = 186), 40–59 (n = 110), 60–79 years (n = 150), and ≥80 years (n = 45). Further, approximately 75% of the isolates were obtained from women, and 23% of these were from women aged ≥65 years. Among the 407 isolates collected from CA-URTI, 64.4% were collected from pediatric patient infections.

Ceftibuten (MIC50, ≤0.25 mg/L) was the most potent oral cephalosporin against Enterobacteriaceae from CA-UTI in each country (Table 1). The difference between the susceptibility rates of ceftibuten between countries ranged from 75.2% in the Philippines to 93.1% in Russia using CLSI criteria and from 71.7% (the Philippines) to 91.3% (Russia) using EUCAST criteria. Cefuroxime, cefixime, cefpodoxime, and cefaclor were less active with a rank order of potency (MIC50) as follows; cefixime (0.5 mg/L) > cefpodoxime (0.5–1 mg/L) > cefuroxime (4–8 mg/L) > cefaclor (4–16 mg/L) and susceptibility rates lower than ceftibuten in all countries. Susceptibility to amoxicillin-clavulanic acid was <60% in all countries with the exception of Russia (75.6%). Susceptibility to norfloxacin and ciprofloxacin was similar and varied from approximately 50% in Argentina and Mexico, 60% in the Philippines, 70% in Venezuela, and 80% in Russia. Trimethoprim-sulfamethoxazole susceptibility was highest in Russia (75.6%) and lowest in Mexico (30.6%). Susceptibility among E. coli to nitrofurantoin was high (≥95%) using EUCAST breakpoints. However, susceptibility among combined Enterobacteriaceae species ranged from 51.5% (Argentina) to 83.3% (Russia) using CLSI breakpoint criteria. All Enterobacteriaceae isolated from CA-URTI were susceptible to fluoroquinolones using CLSI but <80% using EUCAST breakpoints while susceptibility percentages ranged from 64% to 79% for the other agents tested using either CLSI or EUCAST breakpoints (data on file, IHMA Inc.).

Table 1 Susceptibility rates and MIC values for Enterobacteriaceae collected from CA-UTI
Full size table

ESBL percentages for E. coli ranged from 9.2% in Venezuela to 40.7% in Mexico. Similarly, ESBL percentages for K. pneumoniae were lowest (18.8%) in Venezuela and highest (46.4%) in Mexico. ESBL-producing P. mirabilis ranged from 10% to 33% with the highest percentage observed in Argentina. Fosfomycin and nitrofurantoin retained >90% susceptibility against ESBL-positive E. coli in each country (data on file, IHMA Inc.). The activity of fosfomycin and nitrofurantoin diminished against ESBL-positive K. pneumoniae. Among the β-lactams tested, ceftibuten and amoxicillin-clavulanic acid provided the highest susceptibility percentages.

All H. influenzae isolates were susceptible to ceftibuten, cefixime, cefpodoxime, and cefuroxime using CLSI breakpoint criteria (Table 2). Intermediate and resistant MIC values were observed for each of these oral cephalosporins utilizing EUCAST breakpoint criteria. Reduced activity of amoxicillin/clavulanic acid was observed among H. influenzae isolates collected in Argentina and Mexico. One of three Haemophilus parainfluenzae from the Philippines had high-level resistant MIC values for all tested fluoroquinolones (data on file, IHMA Inc.). Isolates of M. catarrhalis were only collected in Argentina and all demonstrated a common broad spectrum susceptible pattern to the agents tested.

Table 2 Susceptibility and MIC values for fastidious respiratory tract pathogens from CA-URTI
Full size table

Streptococcus pneumoniae collected from Argentina, Mexico, and Russia were more susceptible to amoxicillin with and without clavulanic acid compared to other β-lactams (Table 2). Susceptibility oral cephalosporins were higher in Argentina compared to Russia and Mexico. Macrolide susceptibility was highest in Argentina. Levofloxacin susceptibility was high in all countries, though fluoroquinolone-resistant S. pneumoniae isolates were observed in all countries. Trimethoprim-sulfamethoxazole susceptibility was considerably lower in Russia (28–38%) compared to Mexico (43–57%) and Argentina (74–85%) dependent upon the applied breakpoint criteria. All agents had activity against S. pyogenes. Azithromycin- and clarithromycin-resistant S. pyogenes was only observed in the Philippines and fluoroquinolone non-susceptible isolates were observed in all three countries that submitted S. pyogenes.

Among the 47 Enterobacteriaceae collected from CA-URTI using MIC50 values, ceftibuten, cefixime, and fluoroquinolones had similar potency (0.03–0.12 µg/mL).

Discussion

This study determined the activity of oral antimicrobial agents from countries with limited information on the susceptibility patterns for pathogens which cause CA-UTI and CA-URTI. Oral cephalosporins tested in this study had activity against the majority of Enterobacteriaceae. Susceptibility rates were dependent on the breakpoints that were applied. Ceftibuten provided the highest and similar susceptibility rates using either CLSI or EUCAST breakpoint criteria against CA-UTI pathogens when compared to the oral cephalosporins tested. Amoxicillin-clavulanic acid susceptibility was much lower when using CLSI breakpoints compared to that observed using EUCAST uncomplicated UTI breakpoint criteria for this agent. Country variability in susceptibility was also observed, even within the same region with a 20% difference observed for fluoroquinolone susceptibility among the Latin American countries examined.

Sites which routinely do not collect specimens for culture from outpatients with CA-URTI limited the number of pathogens collected for this indication and not all countries provided a representative sample for this investigation. Also the number of countries in the three regions studied would not represent the entire region. Broad spectrum antimicrobial agents including all of those reported in this study should be used selectively. Cephalosporins and fluoroquinolones may not always be the first choice of therapy; due to the antibiotic pressure, these and other classes have on increasing ESBL rates. However, depending on the source of the infection and the pharmacokinetic and pharmacodynamics of the drug class, the loss of activity among other commonly used oral agents warrants the need to continue to monitor the viability of oral cephalosporins and other drugs used for treating outpatient infections.

It is important that local epidemiology efforts continue to determine the rate of emerging or epidemic clones and the resistance rates among several class agents which vary based on the susceptibility breakpoints applied [19–21]. It is equally important to follow the variation of prescribing practices and to reduce antibiotic consumption in the ambulatory setting as this affects antimicrobial resistance in the hospital setting [22, 23].

Conclusion

Information related to the local and regional susceptibility patterns of pathogens causing CA-UTI and CA-URTI is essential for physicians treating patients with these very common infections and that was the intent of this study. Continued surveillance of the pathogens causing CA-UTI and CA-URTI is necessary in this era of increasing antimicrobial resistance. Antimicrobial stewardship should remain a high priority across all countries to promote the best treatment practices to diminish the problem of antimicrobial resistance.