INTRODUCTION
Urinary tract infection (UTI) is one of the most common bacterial infections faced by clinicians worldwide and represents a significant part of the clinical microbiology laboratory workload [1, 2]. Moreover, UTIs have become a huge economic burden, with an annual incidence of 150 million cases worldwide and treatment costs of about six billion US dollars per year [3]. The most common clinical manifestations of UTI include cystitis, pyelonephritis, as well as asymptomatic bacteriuria [2]. The prevalence of UTI varies depending on age, sex, catheterization, hospitalization, and previous exposure to antimicrobials [1]. The overall prevalence of UTI worldwide ranges from 8.7% in Iran to 90.1% in Ethiopia [4].
It is estimated that 40% of women and 12% of men experience at least one symptomatic UTI during their lifetime [3]. In the neonatal period, the incidence of UTI is less than 2% in both males and females. The incidence of UTI among men remains relatively low after the neonatal period until age 60 when prostate enlargement begins to interfere with bladder emptying. In the second decade of life, UTIs predominate in females at a ratio of 4:1. The recurrence and persistence of UTIs are common in women of reproductive age due to various factors such as frequent sexual intercourse, use of spermicides for contraception, frequent change of sexual partners, and features of the genitourinary system [5]. Pregnancy, diabetes, and immunosuppression increase the risk of UTIs in women.
UTIs can be caused by fungal or viral pathogens but, in most cases, are caused by bacteria. The overall prevalence of UTIs is due to gram-negative bacteria (90%), with Escherichia coli (E. coli) causing the majority of cases (70-95%) followed by other pathogens such as Klebsiella pneumoniae (K. pneumoniae), Proteus mirabilis, Staphylococcus aureus (S. aureus), S. saprophyticus, and Enterococcus faecalis [1, 3]. To select the most appropriate treatment, it is necessary to understand how modern UTI pathogens function and the spectrum of their antibiotic resistance. The present study had the following goals: to find the most common gram-negative and gram-positive microorganisms in adult hospital patients with UTI and to determine the susceptibility of these uropathogens to antibiotics.
MATERIALS AND METHODS
A hospital-based retrospective study was performed at the Department of Microbiology at Safdarjung Hospital, New Delhi, India. A total of 2099 urine samples was obtained over six months from December 2019 to May 2020. During this period, we collected retrospective data from the microbiology laboratory logbook. The study was performed in accordance with the ethical standards as laid down in the 1964 Helsinki Declaration and its later amendments.
Ethical approval and consent
Since this was a retrospective analysis, the approval of the responsible Department of Bacteriology was obtained. During data collection, every effort was made to ensure the patient’s confidentiality. All measures available were taken to maintain the confidentiality related to patients’ details while compiling the data for the study.
Inclusion and exclusion criteria
The adult patients (>18 years old) admitted to the hospital with UTI infection were included in the study. The patient data records included patient age, sex, urine culture, and antimicrobial susceptibility test results.
Specimen collection
Urine samples were collected in sterile universal containers with a wide neck and hermetic lid. The form for the necessary tests was duly filled out for each patient.
Specimen processing
Samples were processed within 2-4 h of collection for aerobic culture and sensitivity to antibiotics. Isolation of uropathogens was performed on Blood agar and MacConkey agar using a 0.01 mm calibrated loop for a semi-quantitative method. The plates were incubated overnight at 37°C and observed for discrete growth. The grown bacterial colonies were identified and characterized. The number of colonies was expressed in colony-forming units (CFU/ml). Colony counts >105 CFU/ml were considered significant in the patients with no risk factors, and in symptomatic patients (with known health history), ≥103 CFU/ml was considered significant. The grown bacteria were identified using the following methods: gram-staining, production of catalase, oxidase, and coagulase as well as Triple Sugar Iron (TSI) test, indole, citrate utilization, urea hydrolysis, SIM (sulfur indole, motility) medium, Mannitol Salt Agar (MSA) medium, DNase, and bile aesculin hydrolysis 1
Antibiotic Susceptibility Testing
The antibiotic susceptibility test was carried out for each isolated bacteria using the Kirby–Bauer disc diffusion method according to the Clinical and Laboratory Standards Institute guidelines (2019). Bacterial suspensions were prepared by emulsifying 3-5 pure colonies (only the top of the colony was touched) in nutrient broth and adjusted to 0.5 McFarland standards. The surface of the Mueller–Hinton agar plate was swabbed with a sterile cotton swab after dipping it into the suspension. Standard antibiotic discs were placed aseptically, and the inoculated Mueller–Hinton agar plates were incubated at 37°C for 16-18 h. The diameters of the zones of complete inhibition were measured using a ruler, and bacterial species were reported as sensitive, intermediate, or resistant according to the CLSI 2019 guidelines 2
All the antibiotic discs used were obtained from HiMedia Laboratories Pvt. Ltd (Mumbai, India): nitrofurantoin (300 μg), amikacin (30 μg), cotrimoxazole (25 μg), gentamicin (10 μg), ciprofloxacin (5 μg), norfloxacin (10 μg), ampicillin (10 μg), imipenem (10 μg), meropenem (10 μg), cefoxitin (30 μg), piperacillin/tazobactam (100/10 μg), ceftazidime (30 μg), amoxiclav (20/10 μg), cefuroxime (30 μg), colistin (10 μg), nalidixic acid (30 μg), vancomycin (30 μg), linezolid (30 μg), clindamycin(2 μg), erythromycin(15 μg), and penicillin(10 U).
Quality control and quality assurance
The procedures were carried out according to the standard operating procedures (SOPs). Each new sample was checked with the reference strains obtained from the American Type Culture Collection (ATCC) such as E. coli (ATCC 25922), S. aureus (ATCC 25923) and Pseudomonas aeruginosa (P. aeruginosa) (ATCC 27853).
RESULTS
The analysis of 2099 samples received from December 2019 to May 2020 showed that 212 (10.1%) of them were culture-positive. The vast majority – 2049 (97.6%) – of all samples was received from females, while only 50 (2.4%) were from males. Out of 212 culture-positive samples, 16 (7.5%) were recovered from males and 196 (92.5%) from females (Table 1).
| Patient gender | Total number of samples, (%) | Number of culture-positive samples, (%) | Number of culture-negative samples, (%) |
|---|---|---|---|
| Male | 50 (2.4) | 16 (0.8) | 34 (1.6) |
| Female | 2049 (97.6) | 196 (9.3)**** | 1853 (88.3) |
| Total | 2099 (100) | 212 (10.1) | 1887 (89.9) |
indicates a significant difference between the number of positive samples obtained from female patients and that obtained from males, as determined by χ2-test, p < 0.0001.
Three departments of Safdarjung Hospital participated in the study: Obstetrics and Gynecology, Surgery, and Medicine. The maximum number of samples (1339/63.8%) was received from the Department of Obstetrics and Gynecology followed by the Department of Medicine (653/31.1%) and the Department of Surgery (107/5.1%) with 155 (7.4%), 47 (2.2%), and 10 (0.5%) culture positive samples, respectively (Table 2).
| Department | Total number of samples, (%) | Number of culture-positive samples, (%) | Number of culture-negative samples, (%) |
|---|---|---|---|
| Obstetrics and Gynecology | 1339 (63.8) | 155 (7.4)**** | 1184 (56.4) |
| Medicine | 653 (31.1) | 47 (2.2) | 606 (28.9) |
| Surgery | 107 (5.1) | 10 (0.5) | 97 (4.6) |
| Total | 2099 (100) | 212 (10.1) | 1887 (89.9) |
indicates a significant difference between the number of positive samples obtained from patients in the Obstetrics and Gynecology Department and that from Medicine and Surgery departments, as determined by χ2-test, p < 0.0001.
Our results showed a significant prevalence of gram-negative bacteria in the analyzed samples (159/75%) in comparison to gram-positive microorganisms (53/25%). The most frequently isolated uropathogen was E. coli (74/34.9%), followed by K. pneumoniae (41/19.4%), Acinetobacter spp. (14/6.6%), Pseudomonas aeruginosa (10/4.7%), Enterobacter spp. (6/2.8%), Proteus mirabilis (6/2.8%), Klebsiella oxytoca (5/2.4%), and Citrobacter spp. (3/1.4%), as shown in Table 3.
| Isolated pathogens | Number of positive isolates, (%) |
|---|---|
| Gram-negative bacteria −159 (75%) | |
| Escherichia coli | 74 (34.9) |
| Klebsiella pneumoniae | 41(19.4) |
| Acinetobacter spp. | 14 (6.6) |
| Pseudomonas aeruginosa | 10 (4.7) |
| Enterobacter spp. | 6 (2.8) |
| Proteus mirabilis | 6 (2.8) |
| Klebsiella oxytoca | 5 (2.4) |
| Citrobacter spp. | 3 (1.4) |
| Gram-positive bacteria − 53 (25%) | |
| Enterococcus spp. | 32 (15.1) |
| Staphylococcus aureus | 16 (7.5) |
| Coagulase-negative Staphylococcus (CoNS) | 5 (2.4) |
| Total | 212 (100%) |
The results of the antibiotic susceptibility tests for gram-negative bacteria are shown in Table 4. E. coli was found to be the most susceptible to nitrofurantoin (90.5%) and colistin (98.6%) and the least sensitive to cefuroxime (6.7%) and cefotaxime (8.1%). K. pneumoniae showed the highest sensitivity to colistin (92.6%). Acinetobacter spp. was 100% sensitive to colistin, followed by minocycline (50%). Pseudomonas aeruginosa was the most sensitive to colistin (100%) and ceftazidime (50%), while it showed less sensitivity (40%) to other drugs such as piperacillin-tazobactam, ciprofloxacin, netilmicin, amikacin, imipenem, and meropenem. Enterobacter spp. showed the maximum sensitivity to colistin (100%), netilmicin (66.6%), and imipenem (66.6%) while was the least sensitive to piperacillin-tazobactam (16.6%) and cotrimoxazole (16.6%). Proteus mirabilis was the most sensitive to imipenem (66.6%) and piperacillin-tazobactam (50%) while showed the least sensitivity (16.6%) to cefuroxime, cefotaxime, and ciprofloxacin. K. oxytoca demonstrated a high sensitivity to colistin (100%) while being resistant to most of the other tested drugs. Citrobacter spp. was 100% sensitive to amikacin, netilmicin, nitrofurantoin, colistin, and imipenem, although it was only 33.3% sensitive to cotrimoxazole, cefuroxime, cefotaxime, amoxiclav, and ciprofloxacin.
| Antimicrobial agents | Antimicrobial susceptibility of isolated gram-negative bacteria (%) | |||||||
|---|---|---|---|---|---|---|---|---|
| E. coli (n=74) | K. pneumoniae (n=41) | Acinetobacter spp. (n=14) | P. aeruginosa (n=10) | Enterobacter spp. (n=6) | Proteus mirabilis (n=6) | K. oxytoca (n=5) | Citrobacter spp. (n=3) | |
| Amikacin | 71.6 | 31.7 | 14.3 | 40 | 50 | 33.3 | 20 | 100 |
| Netilmicin | 74.3 | 31.7 | NA b | 40 | 66.6 | 0 | 20 | 100 |
| Piperacillin-Tazobactam | 31.1 | 17 | 0 | 40 | 16.6 | 50 | 0 | 66.6 |
| Cotrimoxazole | 37.8 | 17 | 7.1 | IR a | 50 | 0 | 0 | 33.3 |
| Imipenem | 71.6 | 31.7 | 14.3 | 40 | 66.6 | 66.6 | 20 | 100 |
| Meropenem e | - | - | - | 40 | - | - | - | - |
| Nitrofurantoin | 90.5 | 36.5 | NA | NA | 16.6 | IR | NA | 100 |
| Nalidixic Acid | 14.8 | 24.4 | NA | NA | 33.3 | 0 | 0 | 33.3 |
| Amoxiclav | 27 | 17 | IR | IR | 33.3 | 0 | 0 | 33.3 |
| Cefuroxime | 6.7 | 4.8 | NT d | IR | 0 | 16.6 | 0 | 33.3 |
| Cefotaxime | 8.1 | 9.7 | NT | IR | 0 | 16.6 | 0 | 33.3 |
| Ciprofloxacin | 18.9 | 17 | NA | 40 | 0 | 16.6 | 0 | 33.3 |
| Colistin | 98.6 | 92.6 | 100 | 100 | 100 | IR | 100 | 100 |
| Ceftazidime | NT | NAv c | 7.1 | 50 | Nav | NA | NA | NA |
| Minocycline | NT | NT | 50 | NA | NT | NT | NT | NT |
IR – intrinsic resistance (according to CLSI 2019);
NA – not applicable (according to CLSI 2019);
NAv – not available (shortage of antibiotic discs for testing during study period);
NT – not tested (not included in the panel of drugs tested against these bacterial species);
Meropenem – due to shortage of its supply during study period, the drug was included only for P. aeruginosa panel; n – number of isolates tested to each class of antibiotics.
Among the gram-positive bacteria, Enterococcus spp. (15.1%) was isolated most frequently followed by Staphylococcus aureus (7.5%) and CoNS (2.4%) (Table 3). The results of their susceptibility testing are shown in Table 5. Enterococcus spp. was the most susceptible to linezolid (100%) and the least to ciprofloxacin (40.6%). Staphylococcus aureus was 100% sensitive to vancomycin, linezolid, and nitrofurantoin but only 25% sensitive to ciprofloxacin and ampicillin. CoNS was 100% sensitive to vancomycin, linezolid, and nitrofurantoin while only 20% sensitive to ampicillin and ciprofloxacin.
| Antimicrobial agents | Antibiotic susceptibility of isolated gram-positive bacteria (%) | ||
|---|---|---|---|
| Enterococcus spp. (n=32) | Staphylococcus aureus (n=16) | Coagulase- negative Staphylococcus spp. (CоNS) (n=5) | |
| Penicillin | NT a | 0 | 0 |
| Ampicillin | 56.2 | 25 | 20 |
| Cotrimoxazole | IR b | 37.5 | 40 |
| Ciprofloxacin | 40.6 | 25 | 40 |
| Gentamicin (low dose) | IR | 68.7 | 60 |
| Cefoxitin | NA c | 37.5 | 0 |
| Vancomycin | 87.5 | 100 | 100 |
| Linezolid | 100 | 100 | 100 |
| Gentamicin (high dose) | 53.1 | NA | NA |
| Nitrofurantoin | 87.5 | 100 | 100 |
NT – not tested (not included in the panel of drugs tested against these bacterial species);
IR – intrinsic resistance (according to CLSI 2019);
NA – not applicable (according to CLSI 2019); n – number of isolates tested to each class of antibiotics.
DISCUSSION
We examined the prevalence of urinary tract infections among inpatients of Safdarjung Hospital in India. Only 10.1% of the analyzed samples appeared to be positive for urinary pathogens, which corresponds to a low prevalence of infection [4]. Significantly more infected samples were obtained from women (97.6%) than from men, which is consistent with the literature data [1, 6, 7]. Most of these samples (69.3%) were received from the patients of the Department of Obstetrics and Gynecology.
Gram-negative bacteria comprised 75% of the isolated uropathogens, of which E. coli was the most frequently isolated (34.9%), followed by Klebsiella pneumoniae (19.4%), Acinetobacter spp. (6.6%), Pseudomonas aeruginosa (4.7%), Enterobacter spp. (2.8%), Proteus mirabilis (2.8%), Klebsiella oxytoca (2.4%), and Citrobacter spp. (1.4%). These results are in agreement with the data published by other authors [8-11].
Among the gram-positive bacteria, the most isolated one was Enterococcus spp. (15.1%) followed by Staphylococcus aureus (7.5%) and coagulase-negative Staphylococcus spp. (2.4%). Similar results were described by Akhter et al. [7], although Staphylococcus aureus (6.6%) and coagulase-negative Staphylococcus spp. (10.2%) were dominant according to other studies [8, 9]. E. coli was found to be the most susceptible to colistin (98.6%) and nitrofurantoin (90.5%) while being the least sensitive to cefuroxime (6.7%) and cefotaxime (8.1%), which corresponds to the results of Kasew et al. [9]. We also confirmed their data demonstrating that K. pneumoniae was the most sensitive to colistin (98.6%) and the least sensitive to cefuroxime (4.8%) and cefotaxime (9.7%). Acinetobacter spp. was the most sensitive to colistin (100%) and netilmicin (42.8%), as also shown by Uwingabiye et al. [12], while Arshi et al. [13] showed that Acinetobacter spp. is susceptible to carbapenems (70%). Pseudomonas aeruginosa was the most sensitive to colistin (100%) and ceftazidime (50%) in contrast to the data described by Akhter et al. [7], who found that Pseudomonas aeruginosa is the most sensitive to carbapenems. Enterobacter spp. demonstrated the maximum sensitivity to colistin (100%), netilmicin (66.6%), and imipenem (66.6%), which was in concordance with Arshi et al. [13] and Mukherjee et al. [14], while Kasew et al. [9] reported contradicting results. Proteus mirabilis was the most sensitive to meropenem (83.3%) and imipenem (66.6%), which is similar to the findings reported by Akhter et al. [7], Arshi et al. [13], and Mukherjee et al. [14]. In our study, K. oxytoca showed the maximum sensitivity to colistin (100%) and was resistant to most of the other drugs, which contradicts the results of Mukherjee et al. [14] who showed that these bacteria are highly sensitive to carbapenems. Citrobacter spp. was the most sensitive to amikacin (100%), netilmicin (100%), and nitrofurantoin (100%). Similar results were reported by Mukherjee et al. [14] and Kasew et al. [9].
Among the gram-positive organisms, Enterococcus spp. (15.1%) was the most frequently isolated bacteria followed by Staphylococcus aureus (7.5%) and coagulase-negative Staphylococcus spp. (2.4%). Akhter et al. [7] and Naik et al. [6] reported similar results, while Derbie et al. [11] found S. aureus to be the most common gram-positive isolate. Our experiments showed that Enterococcus spp. was the most sensitive to linezolid (100%), which is similar to the results of Mukherjee et al. [14] and Arshi et al. [13]. We found S. aureus to be the most sensitive to vancomycin (100%), linezolid (100%) and nitrofurantoin (100%), which is in accordance with the findings reported by Arshi et al. [13], Mukherjee et al. [14], and Naik et al. [6]. Coagulase-negative Staphylococcus spp. was 100% sensitive to vancomycin, linezolid, and nitrofurantoin. Similar results were observed by Mukherjee et al. [14].
Thus, our study presents valuable regional (New Delhi, India) data on the prevalence and antimicrobial susceptibility spectra of different uropathogens in the adult inpatient population. However, we must mention that the study had several limitations including a small sample size, a short study period, and recruitment of inpatients from a few departments. Since this is a retrospective study, the analysis of the risk factors for UTI or their specific symptoms is out of the scope of this study.