Antimicrobial susceptibility patterns and associated risk factors of bacterial isolate among urinary tract infection suspected patients at University of Gondar Comprehensive Specialized Hospital North west Ethiopia

Abstract

Background

Urinary tract infections (UTIs) are the most common infections affecting millions of people worldwide. There is a shortage of comprehensive data addressing the current prevalence, antimicrobial susceptibility pattern, and associated risk factors in clinically suspected UTI patients.

Objective

To assess the prevalence, bacterial profile, antimicrobial susceptibility pattern and associated risk factors for urinary tract infection among clinically suspected patients.

Method

A hospital-based cross-sectional study was conducted from May 1, 2023, to November 30, 2023, in patients clinically suspected of having a UTI. Sociodemographic and clinical data were collected using structured questionnaires. A total of 320 patients were enrolled, and a systematic random sampling technique was used to select the studyparticipantsApproximately 5 to 10 ml of morning midstream urine was collected and inoculated primarily on the CLED agar. Thereafter, the isolates were subculture on Blood agar and MacConkey media for identification. The Kirby-Bauer disc diffusion method was used to determine the antibiotic susceptibility pattern of the isolates. Extended-spectrum beta-lactamase and carbapenemase-producing strains were detected using combined disk and modified carbapenem inhibition test methods, respectively. The data were entered and analyzed using SPSS version 27 software. A p value < 0.05 was considered to indicate statistical significance.

Results

Of the total participants, 212 (66.25%) were females. The mean age of the participants was 35.99 ± 17.17 years. The prevalence of UTIs was 21.9%. A total of 70 different bacterial isolates were recovered. Of these, 58 (82.9%) and 12 (17.1%) were gram-negative and gram-positive bacterial isolates, respectively. Among the gram-negative bacteria,E.coli isoltes was 34 (58.6%) predominant, followed by K. pneumoniae 22 (37.9%) and P. mirabilis 2 (3.5%), while S. aureus 9 (75%) was prevalent among the gram-positive isolates. Meropenem and nitrofurantoin are effective antibiotics against gram-negative bacteria.The prevalence of multidrug resistance was 71.4%. Approximately 27 (46.6%) and 4 (6.9%) gram-negative isolates were found to be extended-spectrum beta-lactamases and carbapenemase producers, respectively. A history of UTI, catheterization, HIV, and diabetes were independent risk factors associated with UTI (p < 0.05).

Conclusion

The prevalence of UTIs among clinically suspected patients was high, with increased positivity rates for MDR, ESBL and CP-producing strains. Patients with UTI should have a critical need for appropriate antimicrobial administration and the practice of providing patients with suitable antibioticsAntibiotic should be prescribed based on results obtained from the laboratory diagnosis of AST and complying with local guidelines by considering the most often domestically isolated organisms.

Supplementary information

The online version contains supplementary material available at 10.1186/s12879-025-12125-5.

Introduction

Urinary tract infections (UTIs) are among the most common bacterial infections acquired in the community or in health institutions [1]. Globally, an estimated 150 million people are affected by UTIs each year, and approximately 236,786 people die from the infection [2]. The problem is more frequent in women, and approximately 50% of women are affected by UTIs in their lifetime [3, 4]. The increased frequency of UTIs in women is because of the shorter and wider urethral opening, which makes it easier for bacterial colonizers to reach the bladder [5]. In addition, UTIs are more complex in persons at either extreme of the age spectrum: infants, young children and elderly people [6]. The most common signs and symptoms of UTIs include urgency, odorous urine, painful urination (dysuria), supra pubic pain, fever, vomiting, septicaemia, and hematuria [7].

In Ethiopia, urinary tract infections are among the common reasons for seeking medical attention in the community [8]. Different studies have revealed that members of Enterobacteriaceae and gram-positive cocci are common etiologic agents of UTIs [9–12]. The leading pathogens associated with UTIs are Escherichia coli, Klebsiella, Proteus spp., Staphylococcus aureus, and Enterococcus species [13]. Among these bacterial species, Escherichia coli is responsible for more than 80% of infections, followed by Klebsiella pneumoniae and Pseudomonas aeruginosa [14]. However, the relative prevalence of the bacterial species detected in patients with UTIs varies significantly because of differences in age, sex, hospitalization, catheterization, and prior exposure to antibiotics [15]. Moreover, increasing the risk of urinary tract infection in females,womens who have been engaging in sexual activity, vaginal infections, diabetes, pregnancy, a history of catheterization, and obesity might increase the risk of UTIs [16]. The diagnosis of urinary tract infection requires the detection and identification of the etiological agents by performing a urine culture. Infection of the urinary tract is confirmed when a significant number of colony-forming units (CFU) are detected in the culture of morning-midstream urine, which is 10⁵ CFU/ml [17].

Antimicrobial resistance is one of the most serious global public health threats in the treatment of bacterial infections, including UTIs [18]. UTI is the second most frequent reason for using Antibiotics in the community, which can contribute to the emergence of MDR bacteria [19]. The prevalence of MDR bacteria, defined as bacteria with non-susceptibility to at least one agent in ≥ 3 antimicrobial categories associated with UTI has increased worldwide, thus limiting the therapeutic options for the treatment of infections caused by those microorganisms [20, 21]. According to the World Health Organization (WHO), antibiotic resistance is increasing worldwide, resulting in increased morbidity and mortality [22]. Hospitalization costs for UTIs caused by drug-resistant uropathogens are 29% greater than those for UTIs associated with susceptible strains [23]. Urinary tract infections caused by ESBL-producing Enterobacteriaceae have changed drastically and pose a significant burden [24]. Recently, carbapenem-resistant Enterobacteriaceae (CRE) have also emerged as a major concern in the treatment of patients with UTIs [25]. The increase in carbapenem resistance compromises the efficacy of these last-resort antibiotics, leading to increased morbidity, mortality, and healthcare costs [26].

In previous studies, studies on the prevalence of UTIs and antimicrobial resistance patterns in specific segments of populations, such as pregnant women and diabetic patients, were conducted [27]. However, comprehensive data addressing the burden, antimicrobial resistance, including CP and ESBL, and associated risk factors for UTIs in clinically suspected patients with different demographic and clinical backgrounds are lacking. Moreover, in Ethiopia, antibiotics are widely used to treat clinically suspected UTIs empirically, and there could be an increasing incidence of microbial resistance to antibiotics used to treat such infections [28]. Understanding the extent of this problem is paramount for guiding clinical decision-making, implementing effective treatment strategies, and formulating preventive measures. Therefore, this study aimed to assess the prevalence, bacterial profile, antimicrobial susceptibility pattern and associated risk factors for UTIs and the detection of ESBL- and CP-producing uropathogens in clinically suspected patients attending the University of Gondar Comprehensive Specialized Hospital, Northwest Ethiopia.

Materials and methods

Study area, design and period

A hospital-based cross-sectional study was conducted from May 1, 2023, to November 30, 2023. The study was performed at the University of Gondar Comprehensive Specialized Hospital (UoGCSH), Northwest Ethiopia. Gondar is located 727 km from Addis Ababa, the capital city of Ethiopia, and 180 km from Bahir Dar, which is the capital city of the Amhara National Region State [29]. Gondar has an estimated total population of 465,973 [30]. The hospital is one of the largest hospitals in the country [31]. It has 1500 beds and 32 wards and provides services for more than five million people in the catchment area. Annually, approximately 12,000 patients visit this hospital for the diagnosis of urinary tract infections. Urine culture was performed in the medical microbiology laboratory section. As a teaching hospital, it plays an important role in teaching, research and community services [32]. Urine culture was performed on the medical microbiology laboratory section.

Study population

The study population included all selected patients suspected of having UTIs who were admitted to the outpatient and inpatient departments of the UoGCSH for the diagnosis and treatment of UTIs during the study period.

Inclusion and exclusion criteria

All patients clinically suspected of having UTIs (i.e., the presence of at least two of the following clinical symptoms; frequency, urgency, dysuria, and hematuria) presenting at inpatient and outpatient departments of the UoGCSH during the data collection period were included. Patients who were treated with antibiotics for UTIs for the last two weeks of the data collection were excluded from the study.

Sample size determination and sampling technique

The sample size was calculated using a single population proportion formula, n= (Z a/2)²*P (1-P)/d² [33], where p = the prevalence of urinary tract infection (23.7%) from a previous study [34], Z = the 95% level of significance (Zα/2 = 1.96) and d = the 5% margin of error. The sample size was calculated to be 278. By considering a 15% contingency (n = 42) for the nonresponse rates, the final sample size was 320. The data of the study participants were obtained from the inpatient and outpatient departments. The number of clinically suspected patients with UTIs admitted to the outpatient and inpatient departments of the UGCSH every three months was 2100 and 900, respectively.

The total sample size was proportionally allocated to the outpatient and inpatient departments using the formula ni = Ni*(n/N), where N = total population, Ni = initial population, ni = initial sample population, and n = final sample size. A systematic random sampling technique was used to recruit the study participants. This sampling technique uses the formula K = N/n, where K is the sampling interval, N is the total population, and n is the final sample size. Then, 2100/320 = 7 and 900/320 = 3. Therefore, the inpatient and outpatient study subjects were selected at 3rd and 7th intervals, respectively.

Data collection process

Socio-demographic and clinical data were collected using a structured questionnaire through face-to-face interviews. The questionnaire was prepared first in the English language, then translated into the local language (Amharic), and finally back into English to ensure consistency and accuracy. The questionnaire included socio-demographic information such as age, sex, residence, level of education, occupation and marital status. It also covered associated risk factors, including previous UTIs, catheterization status, prior use of antibiotics, comorbidities, male circumcision, and clinical symptoms such as dysuria and urgency. The data collection was carried out after obtaining written informed consent from adult participants and assent from the children’s parents or guardians. Socio demographic, clinical and data related to associated factors were collected via face-to-face interviews.

Urine sample collection

A morning mid-stream freshly voided urine sample (5 to 10 mL) was collected aseptically after the participants were instructed. The female participants were informed to clean their pre urethral area with water. Participants were given sterile, clean, dry, leak-proof and wide-mouthed bottles with tight caps. For the children, the sample collection container and instructions were given to the parents or their guardians, and the importance of contamination-free specimens was explained. Then, the collected midstream urine specimen was transported to the UoGCSH microbiology laboratory within 30 minutes of collection. In case of a delay in processing within 2 hours of collection, the specimens were refrigerated at 4 °C for up to 8 hours.

Culture and identification of isolates

A well-mixed urine sample was inoculated onto cysteine lactose electrolyte deficient medium (CLED) (Himedia, Ltd., Mumbai, India) by using a standard calibrated loop (0.001 mL). The inoculated plates were incubated for 18–24 hours at 37 °C. After overnight incubation, the CLED media was inspected for the growth of bacteria. The incubation period was extended to 48 hours at the optimal temperature for slow-growing pathogens. Then, the colonies were counted on the plate to check for significant bacteriuria (≥100 CFU per 0.001 mL of urine) [35]. In other words, urine culture was considered positive if it contained ≥10⁵ cfu/mL. Thereafter, the colonies were inoculated on mannitol salt agar (Oxiod, Hampisher, UK), blood agar (Oxiod, Hampshier, UK) and MacConkey agar (Oxiod, Hampshire, UK) for identification. Gram reactions, colony characteristics and biochemical tests were used to identify the bacterial isolates.

Gram-negative isolates were identified by using a variety of biochemical tests, including urease, motility, oxidase, H₂S production, and citrate and carbohydrate utilization tests in the TSI [36]. On the other hand, the Gram-positive isolates were further characterized using catalase, novobiocin, mannitol salt agar and coagulase tests. The bile-esculin test was conducted to identify Enterococci species. Using an inoculating wire loop, approximately 2–3 morphologically similar enterococci colonies were picked, inoculated, and streaked onto the surface of a bile-esculin tube for isolation. The inoculated tubes were then incubated at 37 °C for 24 hours, and the results were subsequently determined. Black halos observed around isolated colonies were considered positive for enterococci in the bile esculin test. For further confirmation, two drops of 3% freshly prepared hydrogen peroxide (H₂0₂) were placed on a grease-free slide using a sterile glass rod. Twenty-four-hour-old test organisms were transferred onto slides and immediately observed for the presence of gas bubbles. Gas bubbles indicate a positive reaction, while the absence of gas bubbles indicates a negative reaction [37, 38].

Antimicrobial susceptibility testing

The antimicrobial susceptibility pattern of the bacterial isolate was determined by the Kirby-Bauer disk diffusion technique using commercial disks (Oxoid) according to Clinical and Laboratory Standards Institute CLSI-2022 on Muller Hinton agar (MHA) [39]. From a pure culture, 3–5 freshly grown colonies were removed, transferred to a tube containing 5 ml of sterile normal saline solution and mixed gently until a homogeneous suspension formed. Subsequently, the turbidity of the suspension was adjusted to match the density of the McFarland 0.5 standard [40] to obtain a standardized inoculum size. A sterile cotton-tipped applicator stick was dipped into the suspension, and the excess was removed by gently pressing and rotating the swab against the surface of the tube. Then, the swab was used to distribute the bacteria evenly over the entire surface of MHA (Oxoid). The inoculated plates were left at room temperature to dry for 3–5 minutes; then, the discs were evenly distributed on the inoculated plates using sterile forceps and incubated in an aerobic atmosphere at 37 °C for 18–24 hrs. The diameter of the zone of inhibition around the disc was measured using a ruler. The results were interpreted as sensitive, intermediate, or resistant based on the CLSI 2022 guidelines [40]. Antibiotic disks were placed on the surface of MHA (Oxoid) using sterile forceps as follows: ampicillin (10 μg), amoxicillin clavulanic acid (30 μg), cefoxitin (30 μg), cefotaxime (30 μg), ceftriaxone (30 μg), ceftazidime (30 μg), gentamycin (10 µg), nitrofurantoin (300 µg), nalidixic acid (30 μg), chloramphenicol (30 μg), ciprofloxacin (5 μg), cotrimoxazole (25 μg), tetracycline (30 μg), meropenem (10 µg), clindamycin (2 μg), erythromycin (15 μg), vancomycin (30 µg) and rifampicin (5 μg) [31]. MDR was defined as the resistance of an isolate to three or more antimicrobial classes tested [40].

Extended spectrum beta-lactamase detection

Initial screening for ESBL was performed by determining the diameters of the zones of inhibition produced by ceftriaxone (30 μg), ceftazidime (30 μg) and cefotaxime (30 μg) from the antimicrobial susceptibility test on Mueller-Hinton agar (Oxoid) according to the CLSI screening criteria.

Based on CLSI-2022, bacterial isolates showing zones of inhibition with diameters ≤22 mm in response to ceftazidime (30 μg) or ≤ 27 mm in response to cefotaxime (30 μg) were subjected to an ESBL production test. Phenotypic confirmation of ESBL production was performed by using the combined disk method with ceftazidime (30 μg) and ceftazidime clavulanic acid (30/10 μg). The bacterial suspension was prepared by taking 2–3 fresh colonies and adjusting to the 0.5 McFarland standards. Lawn culture was performed on MHA plates. The ceftazidime and ceftazidime-clavulanic acid discs were placed 20 mm apart on the agar surface and incubated overnight at 37 °C for 24 hrs. After overnight incubation, a ≥ 5 mm increase in zone diameter for either cefotaxime or ceftazidime tested in combination with clavulanic acid was considered indicative of ESBL production. E. coli ATCC 25922 was used as an ESBL-negative strain, whereas K. pneumoniae ATCC 700,603 was used as an ESBL-positive reference strain [40].

Carbapenemase detection

Initially, bacterial isolates resistant to meropenem (≤19) were suspected to produce CP, and CP-producing strains were detected using the modified carbapenem inhibition method (MCIM); approximately one loop (1 µL) of a fresh colony of the tested strain was transferred to a tube containing 2 mL of tryptic soy broth (TSB). Meropenem disk (Oxoid) (10 mg) was incubated in a suspension of the tested strain for 4 h at 35 °C. Then, the disk was removed from the suspension and placed onto an MHA plate seeded with a suspension of the E. coli ATCC 25922 indicator strains. After overnight incubation, the zone of inhibition was measured to determine whether the meropenem had been hydrolyzed. The MCIM was considered positive when the inhibition zone diameter was approximately 6–15 mm with small colonies in the inhibitory zone and negative if there was a large zone of inhibition around the disk (≥18 mm) [40].

Data quality assurance

The questionnaire was pretested on 5% of clinically suspected UTI patients at Kolladiba District Hospital before the actual data collection to ensure its accuracy and consistency. Socio demographic and clinical data were collected by the principal investigator. The completion, accuracy, and clarity of the collected data were regularly checked. The quality control of the laboratory was assessed using laboratory manuals and standard operating procedures (SOPs) of the microbiology laboratory. Culture media preparation was carried out based on the manufacturer’s instructions, and the sterility of the culture media was verified by incubating the media overnight at 37 °C without specimen inoculation. The culture medium was regularly inspected for the presence or absence of any physical changes. To evaluate the performance of the media (for quality control of cultures), standard strains of E. coli (ATCC 25922), Pseudomonas aeruginosa (ATCC 27853), and S. aureus (ATCC 25923) were used throughout the isolation procedure during the study period. The quality control for the antimicrobial susceptibility test was guided by CLSI-2022. The functions of the weighing balance, incubator and refrigerator were regularly checked and calibrated.

Data analysis and interpretation

The data were entered into Epi Data version 4.6 to assess data completeness and clearance. The data were transferred, summarized, tabulated and analyzed using Statistical Package for Social Science (SPSS) version 27 software. The results are presented in words, tables and figures. Bivariate and multivariate logistic regression analyses were conducted to identify potential factors associated with UTIs in clinically suspected patients. A p value < 0.2 was considered to indicate statistical significance in the bivariate analysis. The results were summarized using descriptive statistics. A p value ≤0.05 at the 95% confidence interval was considered to indicate statistical significance.

Results

Socio-demographic characteristics

A total of 320 study participants suspected of having UTIs were enrolled in the present study. Of the total participants, 212 (66.25%) were females. The age of the participants ranged from 5 to 80 years. The mean age of the patients was 35.9 years ±17.17 years. Most of the study participantslived in an urban area, 190 (59.4%),, and the remaining participants were rural. Approximately 134 (41.9%) participants had completed primary school (Table 1).

Table 1.

Socio-demographic characteristics of study participants attending the University of Gondar Comprehensive Specialized Hospital outpatient and inpatient departments from May 1 to November 30, 2023, Northwest Ethiopia

Socio-demographic characteristics		Number	Percentage (%)
Sex	Female	212	66.3
	Male	108	33.7
Age in years	5–17	53	16.6
	18–65	247	77.1
	> 65	20	6.3
Residence	Rural	130	40.6
	Urban	190	59.4
Level of Education	Unable to write and read	127	39.7
	Primary	134	41.9
	Secondary	43	13.4
	College and above	16	5.0
Occupation	Unemployed	143	44.7
	Farmer	42	13.1
	Government employee	108	33.8
	Private employee	27	8.4
Marital status	Single	105	32.8
	Married	215	67.2

Clinical characteristics of the study participants

Of the total 320 study subjects, 110 (34.4%) participants had a previous history of UTI. Most (262, 81.9%) experienced pain, burning, or dysuria during urination. One-fourth, 82 (25.6%) patients had a history of catheterization, 32 (10%) had diabetes, 41 (12.8%) had HIV-positive patients, and 18 (5.6%) were pregnant. Nearly three-fourths of the patients (224, 70%) were outpatients, while 96 (30%) were inpatients (Table 2).

Table 2.

Clinical characteristics of patients suspected of having a UTI at the UOGCSH from May 1 to November 30, 2023, Northwest Ethiopia

Clinical and medical history		Number	Percentage
Previous history of UTI	Yes	110	34.4
	No	210	65.6
Vomiting	Yes	117	36.6
	No	203	63.4
Pain, burning or dysuria	Yes	262	81.9
	No	58	18.1
Urinary frequency	Yes	253	79.1
	No	67	20.9
Abdominal pain	Yes	77	24.1
	No	243	75.9
Hematuria	Yes	104	32.5
	No	216	67.5
Convulsion	Yes	72	22.5
	No	248	77.5
Indwelling catheterization	Yes	82	25.6
	No	238	74.4
Patient type	Outpatient	224	70.0
	Inpatient	96	30.0
Ward type	Surgical	47	14.7
	Pediatric	21	6.6
	Medical	28	8.8
Diabetes mellitus	Known	32	10.0
	Unknown	288	90.0
HIV status	Known	41	12.8
	Unknown	279	87.2
Cancer	Known	19	5.9
	Unknown	301	94.1
Pregnancy	Yes	18	8.5
	No	194	91.5

Prevalence of bacterial UTI

The overall prevalence of urinary tract infection among the study participants was 70 (21.9%) (95% CI, 17.5–26.8). Of the total culture-positive participants, 55/70 (78.6%) were females. The prevalence of UTIs in females and males was 55/212 (25.9%) and 15/108 (13.9%), respectively. Among participants aged over 65 years, there was a 30% prevalence of UTIs. The rate of UTI positivity among children aged 5–17 years was 12 (22.6%). The prevalence of UTIs among married and urban patients was 44 (20.5%) and 46 (24.2%), respectively. Among the female participants, 18 were pregnant, and the prevalence of UTIs among them was 4 (22.2%). The prevalence of UTIs in outpatients and inpatients was 43/224 (19.2%) and 27/96 (28.1%), respectively (Table 3).

Table 3.

Prevalence and distribution of urinary tract infection among study participants attending the UOGCSH from May 1 to November 30, 2023, Northwest Ethiopia

Characteristics		UTI		Total (%)
		Positive (%)	Negative (%)
Sex	Female	55 (25.9)	157 (74.1)	212 (100)
	Male	15 (13.9)	93 (86.1)	108 (100)
Age (year)	5–17	12 (22.6)	41 (77.4)	53 (100)
	18–65	52 (21.1)	195 (78.9)	247 (100)
	> 65	6 (30)	14 (70)	20 (100)
Residence	Rural	24 (18.5)	106 (81.5)	130 (100)
	Urban	46 (24.2)	144 (75.8)	190 (100)
Level of education	Unable to write and read	31 (24.4)	96 (75.6)	127 (100)
	Primary	28 (20.9)	106 (79.1)	134 (100)
	Secondary	8 (18.6)	35 (81.4)	43 (100)
	College and above	3 (18.8)	13 (81.2)	16 (100)
Occupation	Unemployed	29 (20.3)	114 (79.7)	143 (100)
	Farmer	10 (23.8)	32 (76.2)	42 (100)
	Government employ	26 (24.1)	82 (75.9)	108 (100)
	Private employee	5 (18.5)	22 (81.5)	27 (100)
Marital status	Single	26 (24.8)	79 (75.2)	105 (100)
	Married	44 (20.5)	171 (79.5)	215 (100)

Type of bacterial isolates

Both gram-negative and gram-positive isolates were recovered (58/70 [82.9%] and 12/70 [17.1%], respectively). Among the gram-negative bacterial isolates, E. coli was the predominant species, followed by K. pneumoniae and P. mirabilis 34/58 (58.6%), 22/58 (37.9%) and 2/58 (3.5%), respectively. Among the gram-positive isolates, S. aureus was the most common, followed by Enterococcus species and S. saprophyticus 9/12 (75.0%), 2/12 (16.7%) and 1/12 (8.3%), respectively. Among the gram-negative isolates, 23/58 (39.7%) were recovered from inpatients, and 35/58 (60.3%) were from outpatients (Table 4).

Table 4.

Distribution of uropathogens isolated from patients suspected of having UTIs at the UOGCSH from May 1 to November 30, 2023, Northwest Ethiopia

Type of isolates	Frequency	Percent (%)
gram-negative	58	82.9
E. coli	34	58.6
K. pneumonia	22	37.9
P. mirabilis	2	3.5
gram-positive	12	17.1
S. aureus	9	75.0
Enterococcus spp	2	16.7
S. saprophyticus	1	8.3
Total	70	100.0

Associated risk factors for UTI

Based on the multivariate logistic regression, previous history of UTI (AOR = 4.03, 95% CI = 1.66–9.79), catheterization (AOR = 2.58, 95% CI = 1.03–6.43), HIV status (AOR = 4.50, 95% CI = 1.56–12.97) and diabetes mellitus (AOR = 7.83, 95% CI = 2.24–27.35) were found to be independent risk factors for UTI in clinically suspected patients. Therefore, those who were previously infected with UTIs were four times more likely to develop UTIs than those who had no history of infection. Individuals who were catheterized were two times more likely to develop UTIs than those who were not catheterized. On the other hand, patients with HIV are five times more likely to be culture-positive than those with an unknown HIV status. There were nearly eight times more diabetic patients who contracted UTIs than in those with an unknown status (Table 5).

Table 5.

Factors associated with urinary tract infection among patients attending at the UOGCSH from May 1 to November 30, 2023, Northwest Ethiopia

Variables		UTI status (%)		COR (95%CI)	P value	AOR (95% CI)	P value
		Positive	Negative
Sex	Female	55 (25.9)	157 (74.1)	0.46 (0.25–0.86)	0.015	1.76 (0.85–3.69)	0.101
	Male	15 (13.9)	93 (86.1)	1
Age (years)	5–17	12 (22.6)	41 (77.4)	1
	18–65	52 (21.1)	195 (78.9)	1.65 (0.79–3.45)	0.185
	> 65	6 (30)	14 (70)	0.97 (0.30–3.11)	0.953
Residency	Rural	24 (18.5)	106 (81.5)	1
	Urban	46 (24.2)	144 (75.8)	0.71 (0.41–1.23)	0.223
Level of Education	Illiterate	31 (24.4)	96 (75.6)	0.72 (1.93–2.7)	0.628
	Primary	28 (20.9)	106 (79.1)	0.86 (0.23–3.24)	0.83
	Secondary	8 (18.6)	35 (81.4)	1.01 (0.23–4.39)	0.99
	Higher	3 (18.8)	13 (81.2)	1
Occupation	Unemployed	29 (20.3)	114 (79.7)	0.89 (0.31–2.56)	0.834
	Farmer	10 (23.8)	32 (76.2)	0.72 (0.21–2.42)	0.604
	Government	26 (24.1)	82 (75.9)	0.71 (0.24–2.08)	0.541
	Private	5 (18.5)	22 (81.5)	1
Marital status	Married	44 (20.5)	171(79.5)	1.27 (0.76–2.22)	0.383
	Single	26 (24.8)	79 (75.2)	1
History of UTI	Yes	45 (40.9)	65 (59.1)	0.19 (0.11–0.34)	0.001	4.03 (1.66–9.79)	0.002*
	No	25 (11.9)	185 (88.1)	1
Vomiting	Yes	27 (23.1)	90 (76.9)	0.9 (0.52–1.55)	0.693
	No	43 (21.2)	160 (78.8)	1
Pain and dysuria	Yes	68 (26)	194 (74)	0.10 (0.24–0.43)	0.002
	No	2 (3.4)	56 (96.6)	1
Urinary frequency	Yes	65 (25.7)	188 (74.3)	0.23 (0.09–0.61)	0.003	1.12 (0.30–4.10)	0.869
	No	5 (7.5)	62 (92.5)	1
Abdominal pain	Yes	21 (27.3)	56 (72.7)	0.67 (0.37–1.22)	0.190
	No	49 (20.2)	194 (79.8)	1
Hematuria	Yes	55 (52.9)	49 (47.1)	0.66 (0.35–0.13)	0.00	5.92 (2.49–14.06)	0.001*
	No	15 (6.9)	201 (93.1)	1
Convulsion	Yes	22 (31)	50 (69)	0.54 (0.3–0.98)	0.045	2.89 (0.88–9.47)	0.08
	No	48 (19)	200 (81)	1
History of catheterization	Yes	47 (57.3)	35 (42.7)	0.80 (0.04–0.15)	0.002	2.57 (1.03–6.43)	0.043*
	No	23 (9.7)	215 (90.3)	1
HIV status	Known	19 (46.3)	22 (53.7)	0.26 (0.13–0.51)	0.001	4.49 (1.56–12.97)	0.005*
	Unknown	51 (18.3)	228 (81.7)	1
Diabetes mellitus	Known	16 (50)	16 (50)	0.23 (0.11–0.49)	0.001	7.83 (2.24–27.35)	0.003*
	Unknown	54 (18.8)	234 (81.2)	1
Cancer	Yes	16 (84.2)	3 (15.8)	0.04 (0.01–0.14)	0.05	2.28 (0.34–8.27)	0.651
	No	54 (17.9)	247 (82.1)	1
Pregnancy	Yes	4 (22.2)	14 (77.8)	0.98 (0.31–3.07)	0.971
	No	51 (26.3)	143 (73.7)	1

COR = Crude odds ratio, AOR = Adjusted odds ratio. * = Significantly associated

NM = Circumcision did not meet the assumptions of the chi-square test and was not subjected to bivariate analysis

Antimicrobial susceptibility pattern of gram-negative bacterial isolates

Among the tested gram-negative bacterial isolates, approximately 54 (93.1%) were susceptible to meropenem, followed by nitrofurantoin (38; 65.52%), chloramphenicol (38; 65.5%), gentamycin (33; 56.9%) and ciprofloxacin (26; 44.8%). However, the resistance rate was 56% for ampicillin (96.6%), followed by 51% for cotrimoxazole (87.9%) and 51% for tetracycline (87.9%). Approximately 97.1% of the E. coli isolates were sensitive to meropenem followed by nitrofurantoin (85.3%). K. pneumoniae was another isolate that showed susceptibility to meropenem 19 (86.4%), followed by chloramphenicol 15 (68.2%) (Table 6).

Table 6.

Antimicrobial susceptibility patterns of gram-negative isolates from urine samples collected at the University of Gondar Comprehensive Specialized Hospital, Northwest Ethiopia, between May 1, 2023, and November 30, 2023

Antimicrobials	Susceptibility	Types of isolates N (%)
		E. coli (n = 34)	K. pneumonae (n = 22)	P. mirabilis (n = 2)
Ampicillin	Sensitive	2 (5.9)	-	-
	Resistant	32 (94.1)	22 (100)	2 (100)
Amox+Clavulnic	Sensitive	8 (23.5)	9 (41)	2 (100)
	Intermediate	1 (2.9)	-
	Resistant	25 (73.5)	13 (59)
Cefoxitin	Sensitive	10 (29.4)	11 (50)	1 (50)
	Intermediate	2 (5.9)	0	-
	Resistant	22 (64.7)	11(50)	1 (50)
Cefotaxime	Sensitive	12 (35.3)	8 (36.5)	-
	Intermediate	2 (5.9)	-	-
	Resistant	20 (58.8)	14 (63.6)	2 (100)
Ceftriaxone	Sensitive	10 (29.4)	9 (41)	1 (50)
	Resistant	24 (70.6)	13 (59)	1 (50)
Ceftazidime	Sensitive	5 (14.7)	10 (45.5)	1 (50)
	Resistant	29 (85.3)	12 (54.5)	1 (50)
Gentamycin	Sensitive	20 (58.8)	12 (54.6)	1 (50)
	Resistant	14 (41.2)	10 (45.5)	1 (50)
Nitrofurantoin	Sensitive	29 (85.3)	13 (59.1)	2 (100)
	Resistant	5 (14.7)	9 (40.9)
Nalidixic acid	Sensitive	8 (23.5)	6 (27.3)
	Resistant	26 (76.5)	16 (72.7)	2 (100)
Chloramphenicol	Sensitive	22 (64.7)	15 (68.2)	1 (50)
	Resistant	12 (35.3)	7 (31.8)	1 (50)
Ciprofloxacin	Sensitive	17 (50)	9 (40.1)
	Resistant	17 (50)	13 (59.1)	2(100)
Cotrimoxazole	Sensitive	5 (14.7)	2 (9.1)
	Resistant	29 (85.3)	20 (20.9)	2 (100)
Meropenem	Sensitive	33 (97.1)	19 (86.4)	2 (100)
	Resistant	1 (2.9)	3 (13.6)
Tetracycline	Sensitive	4 (11.76)	3 (13.6)	1 (50)
	Resistant	30 (88.2)	19 (86.4)	1 (50)

Multidrug resistance pattern of gram-negative bacterial isolates

The overall multidrug resistance rate (MDR) was 71.4% (95% CI: 59.4–81.6, n = 50). Of these, 39/50 (78%) were gram-negative, whereas the remaining 11/50 (22%) were gram-positive. The majority of the gram-negative isolates (39/58, 67.2%) were MDR. More than half of the K. pneumoniae 15/22 (68.2%) and E. coli 22/34 (64.7%) isolates were MDR (Table 7). The distributions of MDR isolates in the inpatient and outpatient participants were 20/27 (74.1%) and 30/43 (69.8%), respectively.

Table 7.

Multidrug resistance profile of gram-negative bacterial isolates

Antimicrobial resistance pattern	Bacterial isolates			Total
	E. coli	K. pneumoniae	P. mirabilis
AMP,CTX,CTR,CAZ,GEN,CHL,NAL,CIP,TET	3	-	1	4
AMP*,AMC,CTX,CTR,CAZ,GEN,CHL,NIT,COT,CIP,TET	2	3		5
AMP*,AMC,CTX,CTR,CAZ,GEN,CHL,NAL,COT,TET	2	1		3
AMP*,AMC,CTR,CAZ,GEN,NAL,CIP,TET	1	1		2
AMP*, AMC, CTX,CTR,CAZ,GEN, CHL,NAL,CIP,TET	2	2		4
AMP, AMC, CTX,CTR,CAZ,GEN,CHL,NIT,NAL,CIP,TET	2	-		2
AMP*, AMC,CTX,CTR,CAZ, CIP,NAL,TET	2	3		5
AMP*,AMC,CTX,CTR,CAZ,GEN, CIP,NAL,CIP,COT	2	2	1	5
AMP, AMC, CTX,CTR,CAZ,TET, NAL	3	-		3
AMP, AMC,CTX,CTR,CAZ,TET,CIP,NAL, COT	2	-		2
AMP*,AMC, CTR,CAZ, NAL, CIP, NIT,NAL,COT, MER	1	1		2
AMP*, AMC,CTR,CAZ,NAL, CIP, COT, TER,MER	-	1		1
AMP*, AMC.CTX. GEN,CHL,CIP, COT, MER	-	1		1
Total	22	15	2	39

Note: Klebsella spp. are intrinsically resistant to ampicillin and are no longer resistant to ampicillin according to the current CLSI-2022. Abbreviations: AMP, ampicillin; AMC, Amox+Clavulnic; CXT, cefoxitin; CTX, cefotaxime; CTR, ceftriaxone; CAZ, ceftazidime; GEN, gentamycin; TET, tetracycline; NIT, nitrofurantoin; NAL, nalidixic acid; CHL, chloramphenicol; CIP, ciprofloxacin; COT, cotrimoxazole; MER, meropenem. AMP* = Ampicillin is not applicable for K. pneumonae

Prevalence of extended-spectrum beta-lactamase-producing gram-negative bacteria

Thirty-four out of the 58 gram-negative bacterial isolates were Cefotaxime resistant and became candidates for the ESBL-production test. Of these, 20 and 14 were E. coli and K. pneumonae, respectively. The combined disk method revealed that 18 (52.9%) of the E. coli strains and 9 (40.9%) of the K. pneumoniae strains were phenotypically confirmed for the production of ESBL. The overall prevalence of ESBL-producing Enterobacteriaceae was 27/58 (46.6%, 95% CI, 33.3–60.1), with a clear dominance of 52.9% for E. coli, followed by 40.9% for K. pneumoniae. The prevalence of ESBL-producing Enterobacteriaceae among isolates from inpatients and outpatients was 12/23 (52.2%) and 15/35 (42.9%), respectively.

Prevalence of carbapenemase-producing gram-negative bacterial isolates

One E. coli and three K. pneumoniae isolates were meropenem resistant (≤19 mm). These bacterial isolates were candidates for carbapenemase production, and a modified carbapenem inactivation test was performed to confirm carbapenemase production. The overall prevalence of carbapenemase-producing gram-negative bacterial isolates was 4/58 (6.9%, 95% CI, 19–16.7). K. pneumoniae isolates were the most prevalent carbapenemase producers, at 3/22 (13.5%), followed by E. coli isolates at 1/22 (4.5%). All carbapenemase-producing bacterial isolates were recovered from inpatients, and all of them were ESBL producers.

Antimicrobial susceptibility pattern of gram-positive bacterial isolates

Approximately 5/12 (41.6%) of the gram-positive isolates were susceptible to chloramphenicol, followed by ciprofloxacin (5/12, 41.7%) and gentamycin (2/12, 16.7%). Among the isolates tested, tetracycline (12, 100%) had the highest resistance, followed by cotrimoxazole (10, 83.3%). Enterococcus isolates were resistant to all antibiotics 2 (100%) (Table 8).

Table 8.

Antimicrobial susceptibility pattern of gram-positive bacterial isolates from patients with UTIs between May 1, 2023, and November 30, 2023

Antimicrobials	Susceptibility	Type of isolate, N (%)
		S. aureus (n = 11)	S. saprophyticus (n = 1)	Enterococcus (n = 2)
Ampicillin	Sensitive	1 (11.9)
	Resistant	8 (88.1)	1 (100)	2 (100)
Cefoxitin	Sensitive	1 (4.54)		*
	Resistant	8 (95.46)	1 (100)
Gentamycin	Sensitive	2 (22.22)	1 (100)	*
	Resistant	7 (77.78)
Chloramphenicol	Sensitive	4 (44.55)	1 (100)	*
	Resistant	5 (55.45)
Ciprofloxacin	Sensitive	3 (33.33)	1 (100)
	Resistant	6 (66.67)		2 (100)
Cotrimoxazole	Sensitive			*
	Resistant	9 (100)	1 (100)
Clindamycin	Sensitive	1 (11.1)	1 (100)	*
	Resistant	8 (88.9)
Erythromycin	Sensitive	1 (11.1)
	Resistant	8 (88.9)	1 (100)	2 (100)
Vancomycin	Sensitive	*	*
	Resistant			2 (100)
Rifampicin	Sensitive	*	*
	Resistant			2 (100)
Tetracycline	Sensitive
	Resistant	9 (75)	1 (100)	2 (100)

NB: * = Antibiotics are not applicable for the isolate

The overall prevalence of MDR in gram-positive isolates was 11/12 (91.7%). Among the gram-positivepositive strains, Enterococcus spp. were resistant to all the antibiotics tested. Approximately 8 (88.9%) S. aureus isolates were MDR (Table 9).

Table 9.

Multidrug resistance profile of gram-positive bacterial isolates

Antimicrobial resistance pattern	Bacterial isolates			Total
	S. aureus	S. saprophyticus	Enterococcus
AMP,CXT,GEN,CHL,CIP,	1			2
AMP, CXT, GEN,CHL,TET	1			1
AMP,CTX,GEN, CIP,COT	1			1
AMP, CXT,COT,ERY,TET	1	1		2
AMP,CXT,CIP, CLN, ERY, TET	1			1
AMP,CIP, COT,CLN,ERY,TET	1			1
AMP,CIP, COT,CLN,ERY,RIF,TET	1		1	2
AMP,CIP,COT,CLN,ERY,RIF,TET	1		1	1
Total	8	1	2	11

Abbreviations: CLN = Clindamycin, ERY = Erythromycin, RIF = Rifampicin

Discussion

Urinary tract infection is one of the most common bacterial infections encountered in daily clinical practice [41]. Uropathogens are becoming a public health threat at an alarming rate across the globe; perhaps aggravated in resource-limited settings [42]. This study samples the patterns of antimicrobial susceptibility patterns in bacteria associated with UTIs in symptomatic patients at university of Gondar comprehensive specialized hospital. Our main finding is that rates of AMR of the main uropathogens isolated from UTIs (E. coli, Klebsiella spp., P.mirablies, S.aurus, Enterococcus spp. and S. saprophyticus) are severely high. Further, approximately half of the bacterial pathogens isolated from UTIs have MDR. The prevalence of UTIs in the present study was 21.9% (95% CI, 17.5–26.8; N = 70). This finding was in line with the findings of studies performed in Dessie (23.7%) [43], Pawe Hospital (20.6%) [15] and Ayder Mekelle (21.54%) [44]. However, this percentage was higher than that reported in studies in Bahir Dar (15.8%) [45], Kenya (3.9%) and Nepal (16%) [46, 47]. This difference might be due to differences in the characteristics, socioeconomic status of the populations, study population, duration of the study and distribution of uropathogens among the various regions and populations.

In previous studies, gram-negative bacteria were shown to be responsible for more than 90% of UTIs [48]. In agreement with previous findings, the present study also showed that 82.9% of the bacterial isolates were gram-negative. Of these, E. coli was the most frequently isolated uropathogen, accounting for 58.6% of the gram-negative isolates, followed by K. pneumoniae (37.9%). Similarly, a predominance of E. coli was observed in similar studies in Pawe and Arsho, accounting for 42.6%, 55.3% and 44.4%, respectively, of the gram-negative isolates [49, 50]. The predominance of E. coli may be attributed to its unique virulent structures, such as flagella and pili, which aid in its attachment to the uroepithelium and increase the risk of infection [51, 52]. In addition, E. coli is one of the most common normal flora of the lower digestive tract and can access the urinary tract and cause ascending infections [53]. The most important surface virulence factors involved in the colonization process inside the urinary tract are type 1 fimbriae and P fimbriae, and the expression of these virulence factors converts a commensal strain to a uropathogen [54, 55].

Gram-positive bacterial isolates, which accounted for 17.1% of the infections, were also detected in the present study. Of these, S. aureus was the predominant isolate responsible for 75% of the Gram-positive bacterial infections in the present study. This occurs because S. aureus can compete more easily with other members of the gut microbiota, thereby favoring changes in its composition and inhibiting the immune system [56].

In this study, among the assessed risk factors, a previous history of UTI, a history of catheterization, diabetes mellitus, and HIV were the independent risk factors for the occurrence of a UTI. For instance, those who were previously infected with UTIs were four times more likely to develop UTIs than those who did not have a history of infection. This result was in agreement with studies reported in Hiwot Fana Hospital (AOR = 5.64, p = 0.001) and Tigray hospitals (AOR = 4.40, p ≤ 0.001) [57, 58]. In the present study, patients who had a history of catheterization were twice more likely to develop UTIs than those who did not have a history of catheterization. This finding aligned with reports in Arbaminch Hospital (AOR = 3.65, p ≤ 0.001) [59] but differed from the findings documented at Mekelle University (AOR = 18.2, p = 0.001) [8]. The increased risk of UTIs among study participants who had a history of cauterization might be because catheter insertion predisposes individuals to the development of catheter-associated UTIs by creating a new entry point for bacterial invasion and by pushing bacteria into the bladder [8]. Similarly, diabetic patients were nearly eight times more likely to contract UTIs than were those without diabetes mellitus in this study. This finding was comparable to those of studies performed by Hiwot Fana (AOR = 3.69, p = 0.014) and Hawassa (AOR = 3.71, p = 0.033) [58, 60]. Because of poor metabolic control, various impairments in the immune system, and incomplete bladder emptying caused by autonomic neuropathy, diabetic patients might experience frequent UTIs [61]).

Approximately 93.1% of the gram-negative isolates were sensitive to meropenem. This finding was comparable with previous reports from Gondar (97.3%) [25]. This significant level of sensitivity could be attributed to the low availability of the drug in the study area. According to previous studies in Uganda (87.1%) [62] and Somali Hospital (90.6%) [63], approximately 85.3% of the E. coli in this study were sensitive to nitrofurantoin. In the present study, 96.7% of the gram-negative bacterial isolates were resistant to ampicillin, followed by 87.9% to cotrimoxazole. Similar findings were observed in Gondar (97.5%) [25] and Dessie (100%) [43]. Additionally, the highest resistance rate of the isolates was observed for third-generation cephalosporins such as cefoxitin, cefotaxime, ceftriaxone and ceftazidime. All gram-positive bacterial isolates in this study were resistant to tetracycline. Similar findings were reported in a previous study from Addis Ababa [64]. This high resistance rate could stem from the indiscriminate empirical prescription of these drugs for various infections without laboratory confirmation, the continuous use of antibiotics for many years, the easy availability of these antibiotics and the tendency of patients to use less expensive antibiotics.

In the present study, the overall prevalence of MDR bacteria was 71.4% (95% CI, 59.4–81.6; N = 50). This result was consistent with the findings reported at Felege-Hiwot Hospital (66%) [65]. Notably, 91.7% of the gram-positive bacterial isolates exhibited significant MDR. This result was consistent with the findings reported at Yekatit 12 Hospital, Addis Ababa (91.7%) [66]. Relatively, a higher rate of MDR was observed among bacterial isolates recovered from inpatients (74.1%) than among isolates from outpatients (69.8%). This could be attributed to the fact that inpatients were extensively exposed to multiple antibiotics during their ward stays, leading to the acquisition of drug-resistant strains in the hospital environment [42, 67, 68].

In this study, the prevalence of ESBL-producing bacterial strains was 46.6% (95% CI, 33.3–60.1). The findings of this study were greater than previously reported findings of 18.2% in Gondar [69] and 25.5% in Mekelle [8]. This increased prevalence of ESBL isolates might be due to the spread of drug-resistant isolates through natural selection, the transfer of resistance genes through various horizontal gene transfer mechanisms and the lack of early treatment for cases [70]. The present study also revealed that the prevalence of carbapenemase-producing gram-negative uropathogens was 6.9% (95% CI, 1.9–16.7). The prevalence of carbapenemase-producing uropathogens in this study was comparable to that reported in previous studies in Gondar (2.73%) [25]. However, the findings of this study were high compared to previous findings in southern Ethiopia (1.65%) [71] and Nepal (1.7%) [72]. This relatively high prevalence could be attributed to the improper use of antibiotics as empirical therapy for treating multidrug-resistant pathogens [73]. In this study, K. pneumoniae was the predominant carbapenemase producer and accounted for three-fourths of the carbapenem-resistant isolates. This predominance might stem from the fact that the deficiency of the porins OmpK35 and OmpK36 in K. pneumoniae is important for the development of carbapenem-resistant strains of K. pneumoniae [74].

Limitations of the study

In the present study, HIV and diabetes mellitus were indicated as associated risk factors for UTIs; however, laboratory methods were not used to confirm the presence or absence of these diseases in participants with unknown status. ESBL- and CP-producing bacterial strains were not detected via genotypic methods.

Conclusion

The prevalence of bacterial UTIs among clinically suspected patients was 21.9%. The gram-negative bacterial isolates were the largest contributors to the UTIs, with Escherichia coli being more prevalent, followed by Klebsiella pneumoniae. A history of previous UTIs, catheterization, diabetes mellitus, and HIV were identified as risk factors for UTIs. Antimicrobials such as meropenem, nitrofurantoin, chloramphenicol, and gentamycin were relatively effective against gram-negative isolates. Nearly all the gram-positive and two-thirds of the gram-negative isolates were MDR. The detection of extended-spectrum beta-lactamases and carbapenemase-producing uropathogens indicates an ongoing increasing trend. Patients with UTI should have a critical need for appropriate antimicrobial administration and the practice of providing patients with suitable antibiotics based on results obtained from the laboratory diagnosis of AST and complying with local guidelines by considering the most often domestically isolated organisms.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

AMR

Antimicrobial Resistance

ATCC

American Type Culture Collection

AST

Antibiotic susceptibility testing

CLED

Cystine Lactose Electrolyte Deficient

CLSI

Clinical Laboratory Standards Institute

CP

Carbapenemase

CPE

Carbapenem-resistant Enterobacteriaceae

ESBL

Extended Spectrum Beta Lactamase

MDR

Multi-Drug Resistance

MHA

Muller Hinton Agar

MSU

Mid-stream Urine

OECD

Organization for Economic Cooperation Development

SB

Significant bacteriuria

SPSS

Statistical Package for the Social Sciences

UPEC

Uropathogenic Escherichia coli

UTI

Urinary tract infection

Author contributions

Conceptualization: Aschalew Gelaw, Kumlgn Tesfa,Debaka Belete. Data curation: Kumlgn Tesfa. Formal data analysis: Kumlgn Tesfa. Investigation: Kumlgn Tesfa, Aschalew Gelaw, Meseret Mulu,Debaka Belete. Resources: Kumlgn Tesfa, Aschalew Gelaw, Meseret Mulu, Debaka Belete. Software: Kumlgn Tesfa, Aschalew Gelaw,Debaka Belete. Supervision: Aschalew Gelaw, Debaka Belete. Writing the original draft: Kumlgn Tesfa. Revised and edited the manuscript: Kumlgn Tesfa, Aschalew Gelaw,Meseret Mulu, Debaka Belete. Approval for publication: All the authors.

Funding

The authors declare that no funds or grants were received.

Data availability

All the data have been incorporated into the manuscript and are available from the corresponding author upon request.

Declarations

Ethics approval and informed consent

The study was conducted according to the guidelines of the Declaration of Helsinki. The ethical clearance was obtained from the Research and Ethical Review Committee of the School of Biomedical and Laboratory Sciences, College of Medicine and Health Sciences, University of Gondar (Ref no SBMLS/483/23). Written informed consent was obtained from the participants or their guardians. Study participants who were not willing to participate in the study were not forced to participate.

Competing interests

The authors declare no competing interests.