Bacteria Causing Community-Acquired Urinary Tract Infections and Their Antibiotic Susceptibility Patterns in Outpatients Attending at a State Hospital in Turkey

Introduction Clinicians should know the frequency and resistance patterns of bacteria that cause urinary tract infections (UTI) to provide patients with appropriate treatment and antibiotic management. However, the frequency of culture reproducing organisms and resistance patterns change in each community. Therefore, these data must be determined locally to make better treatment decisions. Herein, we aimed to determine the frequency of UTI-causing agents and current antimicrobial resistance profiles in outpatients attending our hospital. Methods This retrospective descriptive study included three hundred eight outpatients attending under the diagnosis of UTI between March and October 2020 who had a positive urine culture for bacterial growth. Age, sex, laboratory tests, urinalysis results, microorganisms grown in urine culture, and antibiograms were evaluated from the patients' medical records. Data were analyzed using SPSS version 23.0 (IBM Corp., Armonk, NY) for Windows. Results In urine culture results, Escherichia coli (E. coli) and Klebsiella species are the most commonly detected agents. The growth in 71 (23%) of the 308 cultures was extended-spectrum beta-lactamase (ESBL) positive. In the E. coli growths, the susceptibility rates to fosfomycin, gentamicin, nitrofurantoin, trimethoprim-sulfamethoxazole, and ampicillin were 95.2%, 90.3%, 95.3%, 76.8%, and 49.3%, respectively. The susceptibility of Klebsiella species to gentamicin was as high as 93.7%, similar to that of E. coli, whereas its susceptibility rates to fosfomycin, trimethoprim-sulfamethoxazole, and nitrofurantoin were lower than those of E. coli (76.1%, 48.4%, and 68.4%, respectively). Of the 71 ESBL-positive growths, 52 were E. coli (17.3% of all UTIs), and 14 were Klebsiella species (4.6% of all UTIs). Of the ESBL-positive strains, 88.7%, 81%, and 76.1% were susceptible to fosfomycin and nitrofurantoin, respectively, and 64.9% and 45.7% were sensitive to cefoxitin and trimethoprim-sulfamethoxazole. Conclusion UTIs are among the most common causes of hospital admission and infections for which empirical antibiotic administration is initiated. The increasing rates of ESBL positivity and resistance to antibiotics such as ampicillin, cephalosporins, trimethoprim-sulfamethoxazole, and quinolones, especially in E. coli and Klebsiella strains, which are the most common pathological agents of UTI in our region, have limited the use of these treatments. However, the high susceptibility of E. Coli growths to fosfomycin and nitrofurantoin and susceptibility of Klebsiella growths to gentamicin may make these antibiotics stand out as suitable options for the empirical treatment of UTI in our setting.


Introduction
Urinary tract infections (UTIs) are among the most common bacterial infections encountered in the community and hospitals [1,2]. UTI is defined as the coexistence of urinary tract symptoms and bacteriuria [3,4]. However, in almost all cases of community-acquired UTI, antibiotic therapy is prescribed empirically before final urine culture or other laboratory results are obtained [2]. Over the years, the frequency of fluoroquinolone-resistant pathogens has increased owing to the increased empirical use of fluoroquinolones, frequency of gram-negative bacteria such as Escherichia coli and Klebsiella that produce extended-spectrum beta-lactamase (ESBL), and frequency of Enterobacteriaceae with multiple resistance mechanisms, including carbapenemase, which influences decisions regarding the empirical treatment of UTI [2,5,6]. Therefore, clinicians should know the frequency and resistance patterns of bacteria that cause UTI to provide patients with appropriate treatment and antibiotic management. The reported data in the literature are conflicting regarding the antimicrobial susceptibility patterns of UTI-causing organisms [7]. Therefore, the clinicians must determine the culture results and antibiotic resistance patterns of the reproducing agents in UTI cases locally to make better antibiotic therapy decisions and prevent the development of antibiotic resistance in the community. This study aimed to determine the frequency of UTI-causing agents and current antimicrobial resistance profiles in outpatients attending our hospital.

Materials And Methods
This retrospective descriptive study was conducted per the Declaration of Helsinki by obtaining data use permission from the Ankara Polatlı Duatepe State Hospital Administration. The researchers were provided with fully anonymized data for the study by the hospital. Ankara Research and Training Hospital Ethics Committee provided ethical approval for this study (Approval number: E-21-672, Approval date: 18.08.2021).
Three hundred eight outpatients with the diagnosis of UTI between March and October 2020 who had a positive urine culture for bacterial growth were included in our study. Age, sex, white blood cell (WBC) count, and C-reactive protein (CRP) levels from laboratory tests, urinalysis results, microorganisms grown in urine culture, and antibiograms were evaluated from the medical records of the patients included in the study. WBC count was measured using the automated cell counter ABX Pentra DF 120 (Horiba Medical, Japan). CRP levels were analyzed with the AU5800 autoanalyzer (Beckman Coulter, Brea, CA) using the immunoturbidimetric method. Chemical examinations for urinalysis were routinely performed with BT Uricell 1600 (BT Products, Turkey), and microscopic examinations were performed with BT Uricell 1280 (BT Products, Turkey). pH, red blood cell (RBC) count per high power field (HPF), and nitrite, leukocyte esterase, protein, and glucose levels were evaluated from the urinalysis. Microscopic findings of >10 WBC per HPF in urine samples were considered significant for UTI [8].
For urine culture, midstream urine samples were routinely collected from the patients after appropriate disinfection practices. After midstream urine samples were collected, 10 mL of urine was taken within 1 hour, inoculated with a round loop to a blood and Eosin Methylene Blue (EMB) agar medium, and incubated at 37°C for 24 hours in an oxygen-stable and suitable humid environment. The culture was checked for bacterial growth the next day. Bacterial identification and an antibiogram test using the cultures were performed with VITEK 2 Compact (bioMérieux SA, Marcy l'Etoile, France). Bacterial growth of ≥105 colonyforming units (CFU)/mL in the culture was considered significant. Antibiograms have not been studied for smaller growths. Contamination is considered when more than two different types of growth are present.
Data were analyzed using SPSS version 23.0 (IBM Corp., Armonk, NY) for Windows. Continuous variables were expressed as median (interquartile [IQR] range), and categorical data as values and percentages. In the comparative analysis, chi-square tests were used for categorical variables, and the Mann-Whitney U test was used for continuous variables. For all the statistical tests, p values < 0.05 were accepted as the statistical significance limit.

Results
Of the 308 outpatients included in the study, 220 (71.4%) were female, and 88 (28.6%) were male. The median age of the patients was 41 years (IQR, 19-69 years). We did not include eight cultures (2.6%) in the antibiogram because the bacterial growths were considered contaminated owing to the growth of a low amount of microorganisms or >2 kinds of microorganisms. In urine culture results, E. coli and Klebsiella species are the most commonly detected agents. The frequency rates of all the agents are shown in Table 1. The serum CRP and WBC measurements and complete urinalysis results are shown in Table 2.    We evaluated gram-negative growths comparatively in three groups, namely E. coli, Klebsiella species, and other gram-negative bacteria. We did not perform subgroup analysis on the gram-positive growths because of their small number. When we compared the demographic and laboratory data according to pathogens, the median ages of the patients with E. coli, Klebsiella species, and other gram-negative growths were 40 years (IQR, 20-65 years), 58 years (IQR, 40-77 years), and 69 years (IQR, 8-74 years), respectively. The median ages of the patients with Klebsiella and E. coli overgrowths were significantly different (Mann-Whitney U test, p = 0.005). The proportions of female patients were 77.8% (n = 172), 51.5% (n = 17), and 50% (n = 9) among the patients with E. coli, Klebsiella, and other gram-negative growths, respectively (chi-square test, p = 0.0001). No significant differences in serum white blood cell count, CRP level, urine pH, RBC count, nitrite, bacteria, leukocyte esterase, protein, and glucose measurements were found between patient groups divided according to the bacterial growths in their urine cultures.
The antibiotic susceptibility of the microorganisms grown in the urine culture is shown in Table 3. In the E. coli growths, the susceptibility rates to fosfomycin, gentamicin, nitrofurantoin, trimethoprimsulfamethoxazole, and ampicillin were 95.2%, 90.3%, 95.3%, 76.8%, and 49.3%, respectively. The susceptibility of Klebsiella species to gentamicin was as high as 93.7%, similar to that of E. coli, whereas its susceptibility rates to fosfomycin, trimethoprim-sulfamethoxazole, and nitrofurantoin were lower than those of E. coli (76.1%, 48.4%, and 68.4%, respectively). The rates of sensitivity of the Klebsiella and Proteus species to ampicillin were 11.1% and 50%, respectively. In the antibiograms of Staphylococcus species, the third most common growth, the sensitivity to trimethoprim-sulfamethoxazole was 92.8%, and to vancomycin and tigecycline was 100%. The sensitivity to nitrofurantoin and ampicillin was also 100%; however, the number of antibiograms was low.  (88.8%) 1/2 (50%) - A gram-negative antibiogram panel susceptibility comparison between E. coli, Klebsiella species, and other gram-negative strains is presented in Table 4. The susceptibility rates to fosfomycin, cefixime, cefuroxime, ampicillin, and trimethoprim-sulfamethoxazole were similar between other gram-negative growths and E. coli growths, but were higher than those of Klebsiella species. The rates of susceptibility of the E. coli growths to ciprofloxacin and nitrofurantoin were higher than those of other gram-negative growths and Klebsiella species growths. A significant difference in cefoxitin susceptibility was only present between E. coli and the other gram-negative growths. The susceptibility to gentamicin was similar between the three groups.

ESBL-positive gram-negative strains
Data were presented as values and percentages. Chi-square tests were used for categorical variables was used for analysis.

Discussion
In cases of community-acquired UTI, antibiotic therapy is often prescribed before culture and susceptibility studies. To prescribe the appropriate antibiotic therapy for patients and reduce the development of antibiotic resistance, clinicians must determine the culture results and antibiotic resistance patterns of the agents grown in cultures locally. In this study, we aimed to determine the frequency and current antimicrobial resistance profiles of agents causing community-acquired UTI in outpatients attending our hospital, a 300-bed secondary care hospital in Turkey.
In our study, E. coli and Klebsiella species were the most common growths detected. Although E. coli is reported to be the most common cause of UTI in the literature, similar to our study, the second most common pathological agent differed between studies. Ağca reported that the second most common urine growths after E. coli were Pseudomonas aeruginosa (6%), Klebsiella species (5%), Enterococcus species (5%), and Staphylococcus aureus (4%) [9]. In another study conducted in Kosovo, the second most commonly isolated pathological agent was the Proteus species [10]. In two other studies from Turkey, Klebsiella species growths were reported as the second most common, similar to our finding [11,12]. Kidwai et al. reported S. aureus and Klebsiella species as the second most common growths after E. coli in patients in low socioeconomic strata [13].
Female sex has been reported as a risk factor of UTI in the literature. UTI occurs twice more frequently in women than in men [1,13]. The short urethra and proximity of the urethra to the anus have been reported to be among the factors that increase the risk of UTI [14]. Although we did not aim to determine the prevalence in our study, we found that 71.4% of the patients with urine culture growth were female. Moreover, the microorganisms reproduced in the culture of patients with UTI may also be related to sex. Similar to our results, previous studies reported that the proportion of females was higher among patients with E. coli growths, and the male-to-female ratio was close to 1 among patients with Klebsiella species growths [12].
Although studies on the antibiotic resistance pattern of all gram-negative growths are limited in the literature, studies on the growth of E. coli and Klebsiella species have been conducted in different centers. The susceptibility of E. coli strains to ampicillin has been reported to range from 11.6% to 28% [12,15], and its susceptibility to fosfomycin ranged from 60% to 98% [7,13,16]. On the other hand, the sensitivity to nitrofurantoin was found to be high in many studies (86.45%-94%) but low (59%) in the study of Kidwai et al. [1,7,8,13,15,16]. While the sensitivity to gentamicin was reported to be as low as 37%-45.65% in some studies [8,13], it was reported to be higher by Daoud et al. and Dash et al., similar to our results (91.3% and 94.1%, respectively) [1,16]. The sensitivity to trimethoprim-sulfamethoxazole ranged from 42.99% to 71.9% [1,7,8,15,16], while that to ciprofloxacin ranged from 68% to 91% [2]. In a study conducted in Turkey, high resistance to ciprofloxacin (22.1%) and cephalosporins (cefepime, 23.5%; ceftazidime, 22.5%; and ceftriaxone, 26.3%) was found in outpatients with E. coli growth in cultures, whereas lower resistance to amoxicillin-clavulanate (16.4%) and nitrofurantoin (4.7%) was observed [11]. In our study, the sensitivity of E. coli strains to ampicillin was 49.3%, higher than those reported in the literature. High susceptibility to fosfomycin, nitrofurantoin, and gentamicin was observed (95.2%, 95.3%, and 90.3%, respectively). While the susceptibility rate to trimethoprim-sulfamethoxazole was lower (76.8%), it was higher than those reported in other studies. The rates of sensitivity to ciprofloxacin and cephalosporins were similar to those reported in the literature.
In studies that evaluated the antibiotic susceptibility of Klebsiella strains, Shaifali et al. reported an ampicillin susceptibility rate of 54.54% [8]. In the literature, the rates of susceptibility to fosfomycin, gentamicin, nitrofurantoin, trimethoprim-sulfamethoxazole, and ciprofloxacin were 53%, 50%-54%, 25%-90%, 47%-81%, 53%-100%, respectively [2,8,13,15]. Rizvi et al. reported a 100% rate of susceptibility to fosfomycin [7]. In our study, the susceptibility rate of the Klebsiella species growth to ampicillin was much lower than those reported in the literature. Similar to the susceptibility of E. coli, the susceptibility of the Klebsiella species to gentamicin was higher (93.7%). The rates of susceptibility of the Klebsiella species to fosfomycin, nitrofurantoin, trimethoprim-sulfamethoxazole, and ciprofloxacin were lower than those of E. coli (76.1%, 68.4%, 48.4%, and 46.4%, respectively) and similar to those reported in other studies.
As the ESBL positivity rates vary among hospitals and regions, hospitals must conduct surveillance studies to determine the ESBL positivity rates and resistance patterns. In different studies from Turkey, the ESBL positivity rates ranged from 7.2% to 53% for E. coli strains and from 32% to 54% for Klebsiella species growths [17][18][19][20]. In our study, the ESBL positivity rate was 23.3% for the E. coli strains, 42.4% for the Klebsiella species, and 10.9% for other gram-negative growths. Carbapenemase production was detected in one E. coli strain. These findings are similar to those reported in the literature, and our ESBL positivity rate may be similar to that in large-center hospitals, as our hospital is a 300-bed secondary care hospital with 30 intensive care beds and a center serving patients from different regions.
The antibiotic susceptibility rates of ESBL-positive microorganisms also differed in the literature. In a study conducted in Tunisia, the rates of susceptibility of ESBL-positive E. coli strains to fosfomycin, nitrofurantoin, and trimethoprim-sulfamethoxazole were 100%, 96.4%, and 36.4%, respectively, and the rate of sensitivity of ESBL-positive E. coli strains to ciprofloxacin was 38.1% [16]. In our study, the rate of susceptibility of ESBL-positive strains was 88.7% for fosfomycin, 81% for nitrofurantoin, 45.7% for trimethoprim-sulfamethoxazole, and 77.1% for ciprofloxacin. This difference may be due to the differences in ESBL genes and antibiotics used by the patient populations in the previous studies.
Our results should be interpreted with consideration of the limitations of this study. Owing to the study's retrospective design and limited data available in the electronic medical records in our hospital, we could not obtain information on individual patient history regarding risk factors such as urinary stones, urinary catheterization, or other instrumentations. In addition, other known risk factors of UTI (e.g., diabetes) were not considered; however, investigating these factors is beyond the scope of our study.

Conclusions
UTIs are among the most common causes of hospital admission and infections for which empirical antibiotic administration is initiated. The increasing rates of ESBL positivity and resistance to antibiotics such as ampicillin, cephalosporins, trimethoprim-sulfamethoxazole, and quinolones, especially in E. coli and Klebsiella strains, which are the most common pathological agents of UTI in our region, have limited the use of these treatments. However, high susceptibility of E. Coli growths to fosfomycin and nitrofurantoin and susceptibility of Klebsiella growths to gentamicin may make these antibiotics stand out as suitable options for the empirical treatment of UTI. Ensuring that hospitals apply the optimum empirical antibiotic treatment by identifying infectious agents and resistance patterns will provide the most effective treatment to patients and prevent the development of antibiotic resistance.
submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.