Bacterial profile, antimicrobial susceptibility, and associated factors of urinary tract infection among people living with HIV at Debre Markos Comprehensive Specialized Hospital, Ethiopia

Abstract

Urinary tract infection (UTI) is a common bacterial infection among people living with HIV (PLWHIV) and can cause complications, particularly due to antimicrobial resistance. Data on UTIs among PLWHIV in Ethiopia are limited. This study assessed the bacterial profile, antimicrobial susceptibility, and associated factors of UTIs among 255 PLWHIV attending the ART outpatient clinic at Debre Markos Comprehensive Specialized Hospital from March to June 2025. Clean-catch midstream urine samples were inoculated on MacConkey and blood agar to identify pathogens, and antimicrobial susceptibility tested by Kirby–Bauer disk diffusion method. Statistical significance was set at p ≤ 0.05 with 95% confidence intervals. UTIs were identified in 28 participants (11%), with E. coli (46.4%), S. aureus (21.4%), and P. aeruginosa (14.3%) predominating. Female sex (AOR = 2.48; 95% CI 1.45–6.27; p = 0.010), HIV RNA > 1000 copies/mL (AOR = 6.30; 95% CI 4.52–14.61; p = 0.001), CD4 +  < 200 cells/mm³ (AOR = 3.17; 95% CI 1.14–9.52; p = 0.028), and being symptomatic (AOR = 3.80; 95% CI 1.63–8.74; p = 0.024) were significantly associated with UTIs. The multidrug-resistant prevalence was 17.9%, highlighting the need for strengthened diagnostics, susceptibility-guided therapy, routine screening, and PLWHIV education on perineal hygiene to improve management and limit antimicrobial resistance.

Introduction

Urinary tract infections (UTIs) are among the most common bacterial infections globally, affecting both the general population and people living with HIV (PLWHIV), who are at higher risk due to immunosuppression¹. HIV infection leads to depletion of CD4+ T lymphocytes, increasing vulnerability to infections such as UTIs. In developing countries, limited healthcare infrastructure, poor sanitation, and weakened immunity contribute to the burden of UTIs, which can lead to complications including pyelonephritis, renal failure, prostatitis, and epididymitis if untreated^2,3.

UTIs involve infection of the urinary tract, including the urethra, bladder, and kidneys.In Ethiopia, UTIs are the second most prevalent infectious disease after respiratory infections⁴. Significant bacteriuria is defined as ≥ 10⁵ colony-forming units per milliliter (CFU/mL) in urine⁵. HIV compromises immunity by reducing CD4+ T lymphocytes⁶. Studies have demonstrated that the prevalence of UTIs is inversely correlated with CD4 + counts, with a higher risk among individuals with CD4+  < 200 cells/mm³⁷. Factors such as viral load, gender, age, diabetes, urinary obstruction, and immunosuppression can influence UTI occurrence among PLWHIV⁸.

Globally, UTIs affect approximately 150 million people annually and incur costs exceeding $6 billion^9,10. Among HIV-positive populations, prevalence ranges from 6.3 to 41%^10,11), with higher rates reported in African settings such as South Africa (48.7%) and sub-Saharan Africa overall (32.12%)^9,12. In Ethiopia, 1.5% of adults are HIV positive, with a higher regional prevalence of 2.2% in Amhara¹³. Although prophylactic cotrimoxazole is widely used to prevent opportunistic infections in PLWHIV, it does not effectively reduce UTI risk in individuals with advanced immunosuppression, and alternative agents such as nitrofurantoin have been adopted in some settings due to high resistance rates^14,15.

Bacterial pathogens causing UTIs are predominantly Gram-negative organisms, with E. coli being the most frequent isolate, followed by other Enterobacteriaceae, while Gram-positive bacteria such as S. aureus also contribute significantly to disease burden^16,17. The rise of multidrug-resistant (MDR) uropathogens has complicated UTI management, leading to treatment failures, recurrent infections, prolonged hospital stays, and increased healthcare costs¹⁰. Geographic variation in bacterial profiles and antimicrobial susceptibility highlights the need for localized data on UTI etiology and resistance patterns¹⁸.

Despite the documented burden of UTIs among PLWHIV, data on the prevalence, bacterial profile, antimicrobial susceptibility patterns, and associated risk factors are limited in Ethiopia, especially in Ethiopia. Therefore, this study aimed to assess the prevalence, bacterial profile, antimicrobial susceptibility pattern, and associated factors of UTIs among PLWHIV attending Debre Markos Comprehensive Specialized Hospital, Ethiopia.

Result

Participant characteristics

Of the 255 study participants living with HIV, 133/255 (52.2%) were females. The mean (± SD) age was 42.4 ± 14.7. Most participants had a secondary level of education (97, 38%), were married (138, 54.1%), urban residents (173, 67.8%), and employees (110, 43.1%) (Table 1).

Table 1.

Socio-demographic characteristics of people living with HIV (n = 255), Debre Markos Comprehensive Specialized Hospital, 2025.

Variable	Character	Frequency (%)
Age	< 15 years	9 (3.5)
	16–45	143 (56.1)
	> 46 years	101 (42.7)
Sex	Male	122 (47.8)
	Female	133 (52.2)
Residence	Urban	173 (67.8)
	Rural	82 (32.2)
Occupation	Employee	110 (43.1)
	Farmers	38 (14.9)
	Daily labors	47 (18.4)
	House wife	33 (12.9)
	Merchants	10 (3.9)
	Others	17 (6.7)
Education	Illiterates	45 (17.7)
	Primary (1–6)	67 (26.3)
	Secondary	97 (38)
	Tertiary (12+)	46 (18)
Marital status	Unmarried	56 (22)
	Married	138 (54.1)
	Widowed	34 (13.3)
	Divorced	26 (10.2)

Clinical characteristics indicated that 83 (32.5%) participants had CD4⁺  < 200 cells/mm³, whereas 215 (84.3%) had no detectable viral load, despite a considerable proportion having low CD4 counts. Further more, 236 (92.5%) had previous catheterization, 241 (94.5%) had previous surgery, 126 (49.4%) did not use water after toilet, and 231 (90.6%) had a previous history of UTIs. Most participants (224, 87.8%) were asymptomatic for UTIs. Storage of urine for an extended period was reported in 100 (39.2%), ART drug use in 232 (91.4%), and duration of ART > 2 years in 231 (90.6%) (Table 4).

Table 4.

Factors associated with bacteriuria among people living with HIV (n = 255), Debre Markos, 2025.

Variable	Character	Growth (n = 28)%	Nogrowth (n = 227)%	COR (95% CI)	P value	AOR (95% CI)	P value
Age	< 15	0 (0)	9 (4.0)	1.68 (0.18, 15.39)	0.647
	15–45	20 (71.4)	125 (55.1)	2.15 (0.87, 5.29)	0.096
	> 46	8 (28.6)	93 (40.9)	1
Sex	Female	20 (71.4)	113 (49.8)	2.48 (1.45, 6.27)	0.040	1.83 (3.20, 10.92	0.010
	Male	8 (28.6)	114 (50.2)	1
Residence	Rural	17 (60.7)	65 (28.6)	3.85 (2.64, 5.06)	0.001	8.29 (0.67, 52.99	0.098
	Urban	11 (39.3)	162 (71.4)	1
Occupation	Employees	6 (21.4)	104 (45.8)	1
	Farmer	2 (7.1)	36 (15.9)	0.63 (0.24, 1.61)	0.599
	Daily labors	8 (28.6)	39 (17.2)	2.31 (0.42, 3.61)	0.198
	Housewife	6 (21.4)	27 (11.9)	2.50 (0.84, 3.52)	0.184
	Merchants	3 (10.7)	7 (3.1)	4.82 (1.61, 9.24)	0.069
	Others	3 (10.7)	14 (6.2)	2.41 (0.84, 7.21)	0.285
Education	Illitrate	11 (3.6)	34 (15)	8.35 (4.35, 18.2)	0.652
	Primary (1–6)	9 (32.1)	58 (25.6)	3.41 (2.31, 5.65)	0.421
	Secondary	6 (21.4)	91 (40.1)	1.46 (0.57, 6.34)	0.128
	Tertiary + 12	2 (7.1)	44 (19.4)	1
Marital status	Unmarried	13 (47.3)	128 (56.3)	1
	Married	2 (7.1)	32 (14.1)	15.93 (0.9, 18.1)	0.47
	Widowed	7 N	19 (8.4)	0.48 (3.10, 7.21)	0.45
	Divorced	6 (21.4)	47 (20.7)	2.83 (1.64, 10.6)	0.39
History of Surgery	Yes	4 (14.3)	10 (4.4)	3.62 (2.14, 5.79)	0.041	0.08 (0.03, 2.01)	0.123
	No	24 (85.7)	217 (95.5)	1
Symptoms of UTI	Yes	9 (32.1 )	22 ( 9.8)	4.41 (2.71, 9.37)	0.001	3.80 (1.63, 8.74)	0.024
	No	19 (67.9)	205 (90.2)	1
Underline condition	Yes	8 (28.6)	7 (3.1)	1.48 (1.05, 4.37)	0.702
	No	20 (71.4)	230 (96.9)	1
History of catheterization	Yes	5 (17.9)	14 (6.2)	3.31 (0.63, 26.1)	0.034	0.15 (0.04, 6.40)	0.323
	No	23 (82.1)	213 (93.8)	1
History of UTI	Yes	8 (28.6)	16 (7.0)	5.28 (2.61, 8.21)	0.001	0.21 (0.01, 3.39)	0.292
	No	20 (71.4)	211 (92.9)	1
Use of water after toilet	Yes	5 (17.9)	124 (54.6)	1
	No	23 (82.1)	103 (45.4)	3.46 (0.02, 8.13)	0.001	7.77 (2.17, 16.30	0.071
Storage of urine for expended time	Yes	15 (53.6)	85 (37.4)	1.93 (1.35, 6.21)	0.36
	No	13 (46.4)	142 (62.6)	1
Clinical stage of HIV	I	16 (57.1)	168 (74.0)	1
	II	10 (42.9)	47 (20.7)	4.21 (1.34, 8.36)	1.072
	III	1 (3.6)	9 (3.9)	2.38 (0.37, 10.9)	0.631
	IV	1 (3.6)	3 (1.3)	0.08 (0.02, 3.61)	9.270
CD4 Level	> 500	2 (7.1)	75 (33.0)	1
	200–500	9 (32.2)	86 (37.9)	3.8 (1.91, 6.24)	0.042
	< 200	17 (60.7)	66 (29.0)	2.64 (4.38, 9.82)	0.003	3.17 (1.13, 9.5)	0.028
Viral load	Not detected	3 (10.7)	212 (93.3)	1
	21–1000	13 (46.4)	9 (3.9)	0.72 (0.14, 0.61)	0.065
	> 1000	12 (42.9)	6 (2.6)	4.17 (2.75, 7.28)	0.001	6.30 (4.5214.61	0.001
Use of ART drug	Yes	25 (89.3)	207 (91.2)	1
	No	3 (10.7)	20 (8.8)	1.24 (0.52, 9.34)	0.740
Duration ART drug receives	< 2 Years	5 (17.9)	19 (8.4)	2.38 (1.62, 6.21)	0.514
	> 2 Years	23 (82.1)	208 (91.6)	1

Prevalence of UTI among HIV-positive participants

Out of 255 participants, 28 (11%) (95% CI 7.4–15.5) had bacteria causing UTIs. Among these, 22/28 (78.6%) were Gram-negative, and 6/28 (21.4%) were Gram-positive. Escherichia coli was the most frequent isolate (13/28, 46.4%), followed by S. aureus (6/28, 21.4%) and P. aeruginosa (4/28, 14.3%) (Table 2).

Table 2.

Prevalence of bacterial isolates causing UTIs among people living with HIV (n = 255), Debre Markos, 2025.

Organism isolated	Frequency (%)
Escherichia coli	13 (46.4%)
Proteus mirabilis	2 (7.1%)
Pseudomonas aeruginosa	4 (14.3%)
Staphylococcus aureus	6 (21. 4%)
Klebsiella pneumonia	2 (7.1%)
Klebsiella ozanie	1 (3.6%)

Antimicrobial susceptibility pattern

Among Gram-positive isolates, only S. aureus was identified. The isolates showed high susceptibility to nitrofurantoin (100%), meropenem (100%), and cefoxitin (100%), followed by levofloxacin (83.3%), penicillin (66.7%), and gentamicin (66.7%). However, a high level of resistance was observed against norfloxacin (83.3%), tetracycline (66.7%), and trimethoprim–sulfamethoxazole (66.7%) (Table 3). Cefoxitin disk testing, used as a surrogate marker for oxacillin susceptibility, showed that all S. aureus isolates were susceptible, indicating no methicillin-resistant S. aureus (MRSA) in this study¹⁹.

Table 3.

Antimicrobial susceptibility patterns of UTI bacterial isolates (n = 28) among people living with HIV, Debre Markos, 2025.

Organism isolate	Pattern	NIT	SXT	PIP	AMC	CRO	MEM	AMK	GEN	TOB	LEV	NOR	TZP	PEN	FOX	TCY
E. coli	R	0	53.8	NA	31	61.5	0	61.5	38.5	23.1	0	15.4	0	NA	NA	NA
	S	100	46.2		69	38.5	100	38.5	61.5	76.9	100	84.6	100
P . aurogensa	R	0	NA	25	NA	NA	0	75	NA	75	25	75	0	NA	NA	NA
	S	100		75			100	25		25	75	25	100
P. mirabilis	R	0	50	NA	50	0	0	100	0	0	0	100	100	NA	NA	NA
	S	100	50		50	100	100	0	100	100	100	0	0
K.pneumonia	R	0	0	NA	0	100	0	100	0	0	0	50	100	NA	NA	NA
	S	100	100		100	0	100	0	100	100	100	50	0
K. ozanie	R	0	0	NA	100	0	0	100	0	0	0	100	0	NA	NA	NA
	S	100	100		0	100	100	0	100	100	100	0	100
Total	R	0	44.4	25	33.3	55.6	0	72.7	27.7	31.8	4.5	40.9	9.1
	S	100%	55.6	75	66.7	44.4	100	27.3	72.2	68.2	95.5	59.1	90.9
S. auerus	R	0	66.7	NA	NA	NA	0	NA	33.3	NA	16.7	83.3	NA	33.3	0	66.7
	S	100	33.3				100		66.7		83.3	16.7		66.7	100	33.3

NA not applicable, GEN gentamicin, AMC amoxicillin-clavulinic acid, AMK amikacin, TZP piperacillin-tazobactam, PIP piperacillin, CRO ceftriaxone, TOB tobramycin, MEM meropenem, NIT nitrofurantoin, NOR norfloxacin, LEV levofloxacilin, SXT trimethoprim-sulfamethoxazole, TCY Tetracycline, FOX cefoxitin, PEN penicillin.

As showen in table-3, all Gram-negative isolates were 100% susceptible to nitrofurantoin and meropenem and showed high resistance to amikacin (72.7%). Specifically, E. coli was 100% susceptible to nitrofurantoin, levofloxacin, and meropenem, 84.6% to norfloxacin, 76.9% to tobramycin, 69% to amoxicillin-clavulinic acid, and 61.5% to gentamicin. Its highest resistance was observed against trimethoprim-sulfamethoxazole (53.8%), amikacin (61.5%), ceftriaxone (61.5%), and piperacillin (53.8%).

P. aeruginosa was fully susceptible (100%) to meropenem, piperacillin-tazobactam, and nitrofurantoin, 75% susceptible to levofloxacin and piperacillin, and resistant to tobramycin (75%), amikacin (75%), and norfloxacin (75%) (Table 3).

Most Klebsiella spp. were 100% sensitive to nitrofurantoin, meropenem, levofloxacin, trimethoprim-sulfamethoxazole, tobramycin, and gentamicin, while they were 100% resistant to amikacin. Specifically, K. pneumoniae was 100% sensitive to amoxicillin-clavulinic acid and 100% resistant to ceftriaxone, whereas K. ozanie was 100% resistant to norfloxacin and amoxicillin-clavulinic acid but 100% sensitive to ceftriaxone. P. mirabilis showed 100% sensitivity to nitrofurantoin, meropenem, levofloxacin, norfloxacin, ceftriaxone, tobramycin, and gentamicin but was 100% resistant to amikacin (Table 3).

Multi-drug resistance pattern of the UTIs causing bacterial isolates

The multidrug resistance (MDR) pattern, defined as nonsusceptibility to at least one agent in three or more antimicrobial categories²⁰, showed that among Gram-negative bacterial isolates, E. coli had 4/13 (18.2%) MDR. Other Gram-negative isolates (P.aeruginosa, P. mirabilis, K. pneumoniae, and K. ozanie) exhibited variable MDR patterns. The overall MDR prevalence among Gram-negative isolates was 4/22 (18.2%). Among Gram-positive isolates, S. aureus showed MDR in 1/6 (16.7%) of isolates. Overall, the MDR prevalence among all bacterial isolates was 5/28 (17.9%) (Table 3).

Associated factors related to UTIs causing bacterial isolates’

The association between socio-demographic characteristics, clinical features, and the presence of bacterial growth in urine was assessed. In bivariate logistic regression analysis, variables such as sex, residence, daily labor occupation, previous history of catheterization, history of UTI, use of water after toilet, CD4 count < 200 cells/mm³, viral load level, duration of ART, history of surgery, and symptoms of UTI were statistically associated with bacteriuria (P ≤ 0.25) and were considered candidates for multivariable logistic regression (Table 4).

In the multivariable analysis (Table 4), female participants had higher odds of developing UTIs compared to males (AOR = 2.48; 95% CI 1.45–6.27, p = 0.010), indicating that women were about 2.5 times more likely to develop bacteriuria than men. Participants with viral load greater than 1000 copies/mL were significantly more likely to have bacteriuria (AOR = 6.30; 95% CI 4.52–14.61, p = 0.001), suggesting a strong positive association between high viral load and UTI risk. Individuals with CD4 counts below 200 cells/mm³ had approximately three times higher odds of bacteriuria compared to those with CD4 counts above 500 (AOR = 3.17; 95% CI 1.14–9.52, p = 0.028), highlighting the role of immunosuppression in susceptibility to infection. Additionally, symptomatic patients had nearly four times higher odds of bacteriuria than asymptomatic individuals (AOR = 3.80; 95% CI 1.63–8.74, p = 0.024), indicating that the presence of UTI symptoms strongly predicts infection.

Discussion

In the present study, the overall prevalence of bacteriuria among people living with HIV (PLWHIV) was 11% (95% CI 7.4–15.5). The result of this finding is consistent with other Ethiopian studies and elsewhere that reported a prevalence of UTI ranging from 10.3 to 13.7%^7,14,21–23. However, a higher prevalence of UTI was recorded elsewhere, ranging from 18 to 70%^3,24–27.These findings should be interpreted in the context of an outpatient ART population, which may differ from inpatient populations with more advanced disease.

In this study, approximately three-fourths (78.6%) of isolates were Gram-negative organisms, similar to different studies from Ethiopia: Hawassa (91.3%)¹⁴, Tigray (85%)²¹, Addis Ababa (60.8%)²², Hiwot Fana Specialized University Hospital (77.8%)²⁴, and Adama (84.2%)³. However, studies in Cameroon (52.8%)²⁶, and Northern Tanzania (67.6%)²³ reported Gram-positive bacteria as the most prevalent UTI-causing bacteria among PLWHIV. This variation may be due to regional climate, socioeconomic status, sanitation, and healthcare infrastructure variations²¹.

Among Gram-negative bacteria, E. coli (46.4%) was the predominant pathogen isolated in this study, aligns with findings in Hawassa (69.6%)¹⁴, Tigray (57%)²¹, Wolaita Sodo (45.7%)⁷, Addis Ababa and Adama (45.1%)³, Dessie²⁸, Addis Ababa (49%)²², and Uganda (72%)²⁹. However, contrary to our study, findings from Cameroon (42.9%)²⁶, and Harar, (45.3%)²⁴ revealed that the most frequently isolated bacterial species causing UTIs was S. aureus. The predominance of E. coli is attributed to its abundance in the intestinal flora, so improper cleaning of the genital area can easily allow the bacteria to enter the urinary tract³⁰ and the ability of uropathogenic E. coli (UPEC) to adhere to, invade, and persist in the urinary tract via fimbriae, pili, toxins, and iron-acquisition systems. This adherence is crucial as it allows UPEC to resist being washed away by urine flow and to form biofilms and intracellular bacterial communities that protect them from the host immune response and antibiotics³¹.

Other Gram-negative bacteria isolated in this study, P. aeruginosa (14.3%) was the the second causative agent for UTI, comparable to Dessie (12%)²⁸ and Wolaita Sodo (11.4%)⁷, but higher than Hawassa (4.8%)¹⁴ and Harar (6.4%)²³.. K. pneumoniae (7.4%) was consistent with Hawassa (8.7%)¹⁴ but lower than Harar (23.8%)²⁴. P. mirabilis (7.1%) agreed with Wolaita Sodo (8.5%)⁷ and Harar (9.5%)²³.

In our study. S. aureus (21.4%) was the second predominant bacterium for the occurrence of UTI among PLWHIV, higher than reports from Harar (11.1%)²⁴, Hawassa (8.7%)¹⁴, Tigray (11%)²¹, and Wolaita Sodo (14.3%)⁷. Our result was lower than that reported from Tanzania (29.7%)²³ and Cameroon (42.9%)²⁶. S.aureus causes UTIs mainly in complicated cases such as in diabetic, immunosuppressed, or catheterized patients. It is facilitated by urease production, an enzyme that hydrolyzes urea to ammonia and raises the PH. This alkaline environment decreases the expression of certain virulence factors while enhancing biofilm formation and adhesion to urinary tract surfaces, promoting persistent infection³². The variation in prevalence is influenced healthcare infrastructure, antibiotic usage patterns, hygienic practices, population demographics, and underlying health conditions in those regions³³.

Antimicrobial resistance is a major clinical challenge in treating infections caused by bacterial pathogens and has increased over the years. In the present study, E.coli was sensitive to nitrofurantoin (100%), levofloxacilin (100%),meropenem (100%), norfloxacin (84.6%), tobramycin (76.9%), and gentamicin (61.5%),but resistant to amikacin, ceftriaxone, piperacillin-tazobactam, and Cotrimoxazole, similar to reports Hawassa, Ethiopia¹⁴, Cameroon²⁶, Harrar²⁴, Dessie²⁸,Tigray²¹ and in Uganda²⁹. The high resistance against these antibiotics might be related to widespread antibiotic use, including prophylaxis, immune suppression allowing persistent infections, overuse, or misuse in the treatment of UTIs²⁹.

Among Gram-positive isolated bacteria, S. aureus showed a high level of susceptibility to nitrofurantoin, meropenem,cefoxitin, levofloxacin, penicillin, and gentamicin. But was resistance to norfloxacin, tetracycline, and trimethoprim-sulfamethoxazole. Our finding was aligned with a study conducted in Northern Tanzania, in which S. aureus was susceptible to nitrofurantoin (88.2%) and gentamycin (69.2%)²³, whereas resistant to cotrimoxazole 100% and tetracycline (75%)²¹.

In our study, both Gram-negative and Gram-positive bacteria were sensitive to gentamycin and nitrofurantoin. This finding is comparable with a study in Cameroon, which found that it were sensitive to gentamycin and nitrofurantoin²⁶. In our findings, Gram-negative isolates showed high susceptibility to nitrofurantoin and meropenem, and moderate susceptibility to gentamicin, levofloxacin, and tobramycin.However, most of the Gram-negative bacterial isolates showed resistance to amikacin, piperacillin-tazobactam, norfloxacin, and ceftriaxone. The finding was similar to those reported in studies from other areas^14,26,28. Higher resistance in Gram-negative bacteria is related to outer membrane structure, hydrophobicity changes, and porin mutations³⁴, whereas Gram-positive bacteria lack this barrier. Reduced outer membrane permeability is a key factor in resistance. High resistance rates emphasize the need for surveillance and control of antimicrobial use³⁵. Antimicrobial resistance is a major clinical challenge and has increased dramatically in recent years.

Multidrug resistance (MDR), which has significant implications for the health outcomes of PLWHIV, was observed in 18.2% of E. coli and 16.7% of S. aureus isolates in this study. These findings are lower than reports from Tanzania, where MDR was 28.0% for S. aureus⁷, and from Hawassa and Addis Ababa, where higher MDR rates were reported for E. coli (57.1% and 88%) and S. aureus (100%)²². The MDR prevalence of E. coli (18.2%) in this study is comparable with findings from Tanzania⁷. The occurrence of MDR in Gram-negative bacteria may be attributed to their intrinsic biological mechanisms, such as modification of antimicrobial targets and efflux pumps, as well as external factors including inappropriate antimicrobial use and complex medical practices.

In our study, women had a higher prevalence of UTI than men, with female PLWHIV had at least two times the chance of developing UTIs compared to males.The finding of this study is in line with reports from Ethiopian studies: Dessie²⁸, Hawassa¹⁴, Tigray²¹, Harar²⁴, Addis Ababa²², and Cameroon (64.4%)²⁶, Uganda²⁹. High prevalence of UTI among female participants may be due to females having shorter and wider urethra, a lack of prostatic fluid, and having a moist urethra, mechanical introduction of pathogens into the bladder and trauma increase the risk of UTI among females irrespective of their HIV status³⁶. However, this contrasts with Nigeria study reporting a higher prevalence of UTI in females than in males¹⁴.Storage of urine in the bladder for an extended period may promote bacterial proliferation, potentially increasing the risk of urinary tract infections, although this association was not statistically significant in our study. Prolonged urine retention provides a favorable environment for bacterial growth, which has been documented as a risk factor for UTIs in previous studies³⁷.

UTIs were strongly associated with immunosuppression in this study; participants with CD4+ counts < 200/mm³ were at least three times more likely to develop UTIs.This finding was supported by studies, Hawassa (31.6%)¹⁴, Tigray (36%)²¹, Harar (56.6%)²⁴, Wolaita Sodo (29.4%)⁷, and Dessie (39.2%)²⁸, Ajmer City (46.5%)²⁷, Cameroon (68.5%)²⁶, Osogabo (37.7%)³⁸, and Northern Tanzania (22.2%)%)²³. This may be due to the impaired immunity at a declining CD4+ count, which makes it easier for bacterial pathogens to adhere to the urinary epithelium^7,39.

Although a high proportion of participants had undetectable viral load, a considerable number still had low CD4 counts. This may be explained by delayed immune reconstitution despite effective virological suppression, particularly among individuals who initiated antiretroviral therapy (ART) late or had long-standing immunosuppression. In such cases, CD4 recovery may lag behind viral suppression, and some individuals may remain immunologically non-responsive despite achieving virological control.

Furthermore, a viral load greater than 1000 copies/mL had the highest significant association with the occurrence of bacteriuria compared to a non-detectable levels. This finding aligns with other studies in Wolaita Sodo (57.3%)⁷, Tigray (38%)²¹, and Northern Tanzania (49%)²³. The link between higher levels of HIV RNA copies and the occurrence of UTIs highlights higher viral replication, progressive CD4+ depletion, and increased susceptibility to opportunistic infections, influenced by co-infections, genetics factors, geographic location, lifestyle, nutrition, stress, and pregnancy²⁴, although differing from a Tanzanian study reporting high UTI prevalence in viral load < 20 copies/mL²³.

In our study, being symptomatic for UTIs was at least four times more likely to occur than asymptomatic UTIs. This finding is comparable to findings in Addis Ababa (42%)²², Hawassa (10.7%)¹⁴, Tigray (27.8%)²¹, Addis Ababa (69.2%)³, Dessie (27.6%)²⁸. Furthermore, the link between clinical symptoms of UTIs and the occurrence of UTIs may be attributed to the irritation of the bladder trigone or urethra. This irritation causes the bladder to contract, resulting in more frequent and painful urination³.

In conclusion, the study showed a high prevalence of UTIs among PLWHIV, with significant associations observed with female sex, higher viral load, low CD4 count (< 200/mm³), and presence of symptoms. Although use of water after toilet was not significant in multivariable analysis, it may reflect broader hygiene practices, which were not fully assessed in this study. The predominance of multidrug-resistant uropathogens highlights the need for routine culture and antimicrobial susceptibility testing. However, the findings should be interpreted considering limitations including the single-center design, use of convenience sampling, and absence of molecular resistance characterization.

Materials and methods

Study design, duration, site, and participants

A hospital-based prospective cross-sectional study was conducted from March to June 2025 among 255 people living with HIV (PLWHIV) attending the ART outpatient clinic at Debre Markos Comprehensive Specialized Hospital, located in Debre Markos town. Debre Markos is situated approximately 295 km northwest of Addis Ababa and 265 km southeast of Bahir Dar, Ethiopia.The hospital provides health services for inpatients and outpatients in pediatrics, internal medicine, ophthalmology, gynecology, orthopedics, and surgery. The ART clinic of the hospital serves 3534 HIV patients, and the hospital provides services to more than 5 million people in Debre Markos town and its surrounding catchment areas.

All participants were outpatients and were recruited consecutively during routine ART clinic visits. The inclusion criteria included all PLWHIV attending the ART outpatient clinic during the study period who voluntarily agreed to participate after providing written informed consent or assent. Participants who had used antibiotics within two weeks prior to data collection were excluded from the study.

Data collection

Data were collected by trained nurses using a structured questionnaire to obtain sociodemographic characteristics (age, sex, marital status, residence, occupation, education), clinical characteristics (CD4 count, viral load, previous history of UTIs, clinical stage of HIV, symptoms of UTIs, catheterization, surgery, underlying conditions, ART use, duration of ART), and behavioral characteristics (use of water after toilet, prolonged urine storage in the bladder). Data were collected from participants who fulfilled the inclusion criteria after obtaining written informed consent or assent.

Sample collection and transportation

Participants were instructed on how to collect a clean-catch midstream urine sample. Approximately 30–60 mL of urine was collected in a sterile, wide-mouthed container after cleansing the genital area. Participants details (name, age, collection time) were recorded on the bottles. Samples were analyzed within one hour in the microbiology laboratory of DMCSH using standard methods. Specimens not processed within 2 h were refrigerated at 4 °C until analysis.

Isolation and identification of bacteria

A calibrated sterile platinum wire loop (4.0 mm diameter) was used to deliver 0.001 mL of urine onto MacConkey (HIMEDIA) and Blood Agar (HIMEDIA) plates. Plates were incubated aerobically at 37 °C for 24 h. Colony counts were multiplied by 1000 to estimate bacterial load. Specimens producing ≥ 10⁵ CFU/mL were considered significant for UTI, while < 10⁵ CFU/mL was considered insignificant or contamination⁴⁰. Standard bacteriological procedures were followed for identification. Colony morphology, hemolytic pattern, Gram reaction, and microscopic characteristics were observed. Gram-negative bacteria were further identified by biochemical tests including indole production, lactose fermentation, urea hydrolysis, citrate utilization, lysine decarboxylation, motility, oxidase, and H₂S production. Gram-positive bacteria were identified using mannitol fermentation, catalase, and coagulase tests⁴¹.

Antimicrobial susceptibility testing

Antimicrobial susceptibility was tested by Kirby–Bauer disc diffusion according to CLSI guidelines¹⁹.Cefoxitin disk test was used as a surrogate marker for oxacillin susceptibility among S.aureus isolates¹⁹. Pure culture suspensions were prepared in sterile physiological saline and adjusted to 0.5 McFarland standard. Mueller–Hinton agar (HIMEDIA) plates were swabbed with the suspension, antimicrobial discs placed, and incubated at 37 °C for 18–24 h. Zones of inhibition were measured, and results interpreted as susceptible, intermediate, or resistant based on CLSI breakpoints¹⁹.

The following antimicrobials were selected based on CLSI recommendations and local prescriptions for UTIs among HIV patients, and all were obtained from Oxoid: for gram-positive bacteria, Penicillin (PEP, 30 μg), Cefoxitin (CXT, 30 μg), Tetracycline (TE, 30 μg), Nitrofurantoin (F, 300 μg), Gentamicin (GEN, 10 μg), Meropenem (MEM, 10 μg), Norfloxacin (NOR, 15 μg), Levofloxacin (LEV, 30 μg), and Trimethoprim-Sulfamethoxazole (SXT, 1.25/23.75 μg); for gram-negative bacteria, Gentamicin (GEN, 10 μg), Piperacillin/Tazobactam (PIP/TA, 100/10 μg), Ceftriaxone (CRO, 30 μg), Amikacin (AMK, 30 μg), Trimethoprim-Sulfamethoxazole (SXT, 1.25/23.75 μg), Meropenem (MEM, 10 μg), Nitrofurantoin (F, 300 μg), Amoxicillin-clavulinic acid (AMC, 10 μg), Piperacillin (PI, 10 μg), Tobramycin (TOB, 30 μg), Norfloxacin (NOR, 15 μg), and Levofloxacin (LEV, 30 μg) were used.

Quality control

Quality control was performed by the reference strain E. coli ATCC 25922, K. pneumoniae ATCC 700603, and S. aureus ATCC 25923 to check the quality of culture media and antibiotics disks.

Statistical analysis

The data were entered into Epidata Version 4.6 and transferred to and analyzed using Statistical Package for the Social Sciences (SPSS) version 27.0 for analysis. Descriptive statistics were calculated for demographic and health characteristics. Logistic regression was employed to asses associations between variables. All significant variables in bivariate analysis with a p value < 0.25 were entered into a multivariate logistic regression model to avoid possible confounders. A p value of < 0.05 was considered statistically significant in the regression models with a 95% confidence interval to test the level of significance. Hosmer–Lemeshow Goodness-Of-Fit statistic was used to check model fitness.

Author contributions

LS: Conceived the idea for this study, and participated in conception and design, data collection, and data analysis. AA: Conceived the idea for this study, Participated in conception and design, and conducted data analysis. AM: Conceived the idea for this study, participated in conception and design, conducted data analysis, and drafted and finalized the manuscript for publication.

Funding

No funding was received for this research.

Data availability

The data used to support the findings of this study are included in the article.

Declarations

Competing interests

The authors declare that they have no competing interests.

Ethical approval and consent to participate

This research was conducted in accordance with the Declaration of Helsinki. Ethical clearance was obtained from the Ethical Review Committee (ERC) of Debre Markos University (No: Hsc/R/C/Ser/PG/478/01/17), College of Medicine and Health Sciences, along with a support letter from the Amhara Public Health Institution and the Debre Markos Comprehensive Specialized Hospital. Participation was voluntary, and written informed consent was obtained from all adult PLWHIV. For minors, assent was obtained along with written consent from their parents or legal guardians. For illiterate participants, data collectors read the informed consent for each respondent and confirmed their willingness by signing the informed consent sheet. Moreover, the confidentiality of the patient’s information was safeguarded.