Introduction
Currently, urinary tract infections (UTIS) are the second most frequent reason for healthcare visits, and approximately 150 million cases per year are diagnosed worldwide 1. The most common causative pathogen is Escherichia coli, which is found in 75%-90% of the urine cultures of community-acquired UTIS 2, 3.
Indiscriminate use of antibiotics in this or other bacterial infections has been associated with high bacterial resistance 4. The development of antibiotic resistance is not new. Within the last two decades, antibiotic resistance among gram-negative bacteria of the Enterobacteriaceae family has increased significantly worldwide, with the emergence of organisms producing extended-spectrum beta-lactamase (ESBL) being especially noteworthy 5.
There are several reports on the prevalence of UTIS due to ESBL-producing microorganisms in Latin America. Colombia faces values between 3.1% and 12.5% 6,7. In Brazil, values of 11% were observed 8; however, these values are higher in Mexico, where a study reports that 50% of UTIS due to E. coli were caused by ESBL-producing bacteria 9. In Peru, the frequency of UTI due to ESBL-producing microorganisms were high, reaching 40.1% 10; in special populations such as that of pregnant women, a prevalence of 11.8% was found 11. These results show a diverse variability, which may be explained by the various epidemiological factors involved and the level of development of the programs for rational antibiotic use implemented in each health system.
Identifying the sources of transmission of highly-resistant microorganisms is crucial for their prevention. In addition, the importance of identifying risk factors to predict which patients present a higher risk is related with being able to limit the spread of antibiotic-resistant pathogens 3.
Male sex, advanced age, comorbidities (most frequently diabetes), international travel, previous UTIS, recent hospitalizations, and previous use of antibiotics have been described as factors associated with the development of antibiotic resistant and ESBL-producing bacteria 3. A study on the fecal carriage of ESBL-producing E. coli in the Dutch population showed an association between the use of antacids and high loads of that microorganism 12. Furthermore, it has been found that the administration of nitrofurantoin in patients with recurrent UTIS is associated with a twofold increased risk of being a carrier of an ESBL-producing pathogen, and this was also observed in a recent study carried out in the United Kingdom 3, 13.
In this context, the current investigation is aimed at assessing those factors associated with the presence of extended-spectrum beta-lactamase-producing pathogens in urinary infections.
Materials and Methods
retrospective analytical cross-sectional study was conducted on patients with UTI treated at Clínica Jesús del Norte, Lima, Peru. The study included positive urine cultures from patients treated in the outpatient, emergency, or hospitalization units from July 2016 to July 2017. Data were collated and organized with Excel software. Samples comprised 1542 positive cultures, 137 of which were discarded due to registration errors or insufficient information.
The main study variables were demographic aspects including: Patient age and sex, healthcare unit in which they were treated, previous hospitalizations, and number of UTIS before the appearance of a positive culture for ESBL-producing pathogens.
Urine culture collection: Urine samples were collected from mid-stream urine after urogenital cleaning with tap water and soap. Calibrated loops were used to perform semiquantitative urine cultures. Isolation of ESBL-producing Enterobacteriaceae was carried out by disk diffusion, and the minimal inhibitory concentration was determined by broth microdilution.
Statistical analysis was performed with STATA 14.0 software. Univariate analysis was conducted using relative and absolute frequencies as well as means and standard deviation; for the bivariate analysis, Poisson regression was used to find the prevalence ratio (PR) of the different variables in relation to the presence of UTI caused by ESBL-producing microorganisms. A 95% confidence interval (CI) and a p < 0.05 significance level were considered for this study. Permission was requested from the clinic to access the clinical laboratory database. Because it was a secondary data source analysis, no informed consent was obtained; however, confidentiality of information was kept using a correlative number to identify the cases.
Results
total of 1405 patients with positive urine cultures met the study inclusion criteria. The mean age was 39.9 ±24.5 years, 85.48% belonged to women, and 24.13% were over 60 years. Of these, 55.56% were treated on an outpatient basis. Furthermore, 3.42% had presented previous UTIS, and only 10.25% had been hospitalized before their current visit. Of the total positive urine cultures, 49.18% tested positive for ESBL UTI (table 1).
Features | n | % | |
---|---|---|---|
Sex | Female | 1201 | 85.48 |
Male | 204 | 14.52 | |
Age | Under10 | 222 | 15.80 |
Between 11 and 20 | 43 | 3.06 | |
Between 21 and 40 | 473 | 33.67 | |
Between 41 and 60 | 328 | 23.35 | |
Over 60 | 339 | 24.13 | |
Healthcare unit | Ambulatory | 780 | 55.56 |
Emergency | 546 | 38.89 | |
Hospitalization | 64 | 4.56 | |
ICU | 14 | 1.00 | |
UTI caused by ESBL | No | 714 | 50.82 |
Yes | 691 | 49.18 | |
Previous UTIS | No | 1357 | 96.58 |
Yes | 48 | 3.42 | |
Previous hospitalization | No | 1261 | 89.75 |
Yes | 144 | 10.25 |
The most frequently isolated microorganism was E. coli (85.41%), followed by Klebsiella pneumoniae (4.48%) and Staphylococcus saprophyticus (3.13%) (table 2 ). Of the 714 patients that tested positive for ESBL UTI, 92.1% presented ESBL-producing E. coli, 5.7% presented ESBL-producing K. pneumoniae, and 2.6% were positive for ESBL-producing Proteus mirabilis (table 3).
Germ | n | % |
---|---|---|
Escherichia COÜ-ESBL | 644 | 92.13 |
Klebsiella pneumoniae-ESBL | 40 | 5.72 |
Proteus mirabilis-ESBL | 18 | 2.58 |
Klebsiella oxytoca-ESBL | 3 | 0.43 |
Proteus vulgaris-ESBL | 2 | 0.29 |
A statistically significant association was found between the male sex and the development of UTI caused by ESBL-producing microorganisms (PR: 1.224; 95% CI: 1.035-1.448; p = 0.007). In addition, age, number of previous hospitalizations, and specific care unit (i.e., coming from hospitalization or ICU), also showed a statistically significant association (p < 0.05) with ESBL-associated UTI. However, no significant association was found between ESBL-associated UTI and previous UTIS, having been treated by the emergency unit, or having been an outpatient (p > 0.05) (table 4).
Discussion
his study, in addition to other studies, showed that UTI episodes occur more frequently in women 14-16. Whereas some literature indicates that the average age of the most frequently affected patients is 65 years, other studies report ages from 35 to 40 years 17, 18. This study, as well as various international studies, found that the age group with the highest incidence of UTIS was that over 60 years, with the mean age of cases of UTI patients being 37 years.
Among publications that were reviewed herein, the most frequently isolated UTI pathogen was E. coli3, 19-22. This represents approximately 50%-80% of uncomplicated UTIS in women 23. The present study identified E. coli in 8 out of 10 positive cultures, being more frequently found in women, which is consistent with findings from other research studies 2, 3, 6.
ESBL is an emerging health concern worldwide, and Peru is no exception 3-5, 16, 20. This investigation found that more than half of the cultures tested positive for ESBL UTI, a figure that exceeds previous studies in this field 5, 21. In addition, 92.1% of all ESBL pathogen-positive cultures corresponded to ESBL-producing E. coli, followed by ESBL-producing K. pneumoniae.
The concern about resistance to multiple drugs was first detected in enteric bacteria such as E. coli, Shigella, and Salmonella between 1950 and 1960 4. There are variable patterns of antibiotic resistance; genes responsible for it can be transferred among different bacteria through elements such as bacteriophages, plasmids, DNA, or transposons. These genes often show resistance to a specific antibiotic, but several genes may accumulate in a single bacterial cell, with each gene presenting a different type of resistance 4. With bacterial genetic endowment, in the case of ESBL, these bacteria develop enzymes that are capable of hydrolyzing latest-generation cephalosporins and aztreonam.
UTI recurrence has been proposed as a risk factor in previous studies 7, 19; however, the present study found that more than 90% of patients with a positive urine culture for ESBL UTI had no previous UTIS. Although the results of this investigation differ from those of previous studies, it is important to emphasize that this could be due to data under-reporting or to patients having been previously treated in other health institutions.
As in other research studies, this study found a significant association between the presence of UTIS due to ESBL-producing pathogens and male sex, age, number of previous hospitalizations, and healthcare unit 22, 23. An explanation for this could be that adult male patients, probably with prostate pathology, require more invasive procedures in the urinary tract and may also suffer from alterations in the bladder function and anatomy. As a result, they present a greater predisposition to UTIS caused by ESBL-producing pathogens. Furthermore, patients with previous hospitalizations are exposed to the hostile environment of health centers, where there is a greater presence of resistant pathogen strains. Likewise, older patients tend to visit hospital centers more often and are more vulnerable to infections due to their physiological conditions 24-26.
Finally, it should be mentioned that this study had certain limitations such as the lack of assessment of diagnoses from previous hospitalizations or the use of antibiotics during hospitalization. Although the information was obtained from a single-center database and the cases found were numerous, extrapolation of the results to the general population of the country may correspond to certain socioeconomic strata with a higher purchasing power, and, therefore, large-scale results could vary.
In conclusion, although the presence of UTI is associated with female sex and patient age, the development of UTIS caused by an ESBL-producing pathogen is usually significantly associated with the male sex, number of previous hospitalizations, and the healthcare unit in which the patient was attended