Bacterial profile, their antibiotic susceptibility pattern, and associated factors of urinary tract infections in children at Hiwot Fana Specialized University Hospital, Eastern Ethiopia

Background Urinary tract infections (UTIs) are common pediatric infections and contribute to high morbidity and mortality. At present, the antimicrobial resistance emergency has quadrupled worldwide and poses a serious threat to the treatment of patients. However, there have been few studies on UTIs in children in Ethiopia, particularly in the east. Objective This study aimed to assess the bacterial profile of urinary tract infections, their susceptibility to antimicrobial agents, and associated factors in under-five children at Hiwot Fana Specialized University Hospital, eastern Ethiopia. Method We conducted hospital-based quantitative study on 332 consecutively selected under-five children from March 20 to June 10, 2021. Parents and guardians were interviewed to collect data using a structured questionnaire. Random urine samples were collected aseptically, and standard microbiological techniques were used to identify the bacteria and test for susceptibility to various antibiotics. Data were entered into Epi Info version 7 and exported to Statistical Package for the Social Sciences (SPSS) version 25 for analysis. Data were analyzed using descriptive analysis, bivariate, and multivariable logistic regression analysis. The crude odds ratio (COR) and adjusted odds ratio (AOR) with their respective 95% confidence intervals (CI) were used to determine the significance of the predictors. A p-value at a 95% confidence interval of less than 0.05 was considered statistically significant. Results The overall prevalence of bacterial urinary tract infections was 80 (24.1%) 95% CI:19.40–29.00%). Most of the bacterial isolates 55 (68.75%) were gram-negative bacteria, predominantly E. coli 23 (28.75%) and K. pneumoniae 10 (12.50%). Being a rural resident (AOR: 4.10, 95%CI: 1.45 11.54), uncircumcised male (AOR: 3.52, 95%CI: 1.33, 9.39), previous history of antibiotic usage (AOR: 7.32, 95%CI: 2.11, 25.37), indwelling catheterization (AOR: 10.35, 95%CI: 3.74, 28.63), previous history of urinary tract infections (AOR: 5.64, 95% CI: 1.36, 23.38), and urinary frequency (AOR: 5.56, 95%CI: 2.03, 15.25) had higher odds of culture positive result. The majority of the isolates have shown high levels of antibiotic resistance. Meropenem, ciprofloxacin, and amoxicillin-clavulanic acid were effective against gram-negative uropathogens, whereas rifampin and ciprofloxacin were the most sensitive drugs for gram-positive isolates. From the tested bacterial isolates, 53/86 (61.6%), 11/86 (11.6%), and 2/86 (2.3%) were found to have multidrug resistance (MDR), extreme drug resistance (XDR), and pan drug resistance (PDR), respectively. Conclusions About one-fourth of the children were culture-positive for many types of bacterial uropathogens; this is higher compared with most of the previous studies in Africa. Rural dwellers, uncircumcised males, indwelling catheterization, a history of antibiotic use and urinary tract infection, and frequent urination all had a higher risk of bacterial infections. Many isolates were resistant to multiple drugs, primarily beta-lactams. Urinary tract infections as well as the growth and spread of resistant bacterial pathogens should be monitor regularly.


Introduction
Infections of the urinary tract (UTIs) are some of the most common causes of acute morbidity and chronic medical conditions like hypertension, failure to thrive, and end-stage renal disease in children [1]. UTIs, which are caused by bacteria, are a serious public health problem and a significant cause of morbidity in infant boys, older men, women of all ages, and about 5% of girls and 2% of boys by the age of seven [1]. Virulence factors used by the main uropathogens are adherence, production of the toxin, immune evasion, and iron acquisition [2]. UTIs begin when bacterial uropathogens reside anywhere in the urinary tract: the kidney, ureter, bladder, and urethra. Expression of pili and adhesins results in colonization and invasion of the superficial umbrella cells. Host inflammatory responses begin to clear extracellular bacteria, though some bacteria evade the immune system through host cell invasion, morphological changes, multiplication, and biofilm formation, as well as by producing toxins and proteases that cause host cell damage, releasing essential nutrients that promote bacterial survival and ascent to the kidneys, and progressing to bacteremia [2,3]. The clinical manifestations of UTIs depend on the causative agent, the severity of the infection, the parts of the urinary tract involved, and the immune response of patients [4]. UTIs can be lower (cystitis) or upper (pyelonephritis) UTIs based on the patient's clinical manifestations [5]. Serious sequelae include frequent recurrences, Urosepsis, renal scarring, progressive kidney damage in young children, preterm birth, and high-level antibiotic resistance, which can lead to high health risks, mortality, and a considerable financial burden on society [2,6]. Globally, UTIs are common bacterial infections in humans that affect more than 150 million people and require enormous antibiotic expenditure yearly, with a global cost of more than 6 billion US dollars [7,8]. UTIs are a common disease in developing countries, with an estimated incidence of at least 8.3 million doctor visits annually [8].
The etiologic agents of UTIs are different and usually depend on the time, geographical area, and age of patients [9]. The most prevalent bacteria causing UTIs are Escherichia coli, followed by Klebsiella spp., Staphylococcus spp., Proteus spp., Pseudomonas aeruginosa, Enterobacter spp., Serratia spp., Citrobacter spp., Enterococcus spp., and Streptococci agalactiae, with variations in their sequence of prevalence in pediatric patients [10,11].
Several associated factors have been identified in the literature; however, the studies are not reproducible. Age, gender, premature children, and young children with severe constipation are risk factors more commonly associated with bacterial UTIs and result in a high financial burden. Other risk factors include prolonged hospitalization, congenital urinary tract obstruction, males who have not undergone circumcision, malnutrition, a previous history of UTI, urinary catheterization, and malnutrition [12][13][14][15]. The antimicrobial resistance (AMR) emergence of bacterial uropathogenic has challenged the current therapies to treat and control the spread of infections [16], and treatment has not improved and does not prevent reinfections [17]. Especially, UTIs caused by resistant bacteria are an important global medical cause of severe infections with increasing rates of morbidity and mortality [18], and they can also result in prolonged hospital stays and poverty for pediatric patients [19,20]. In third-world countries, the problem is immense, as most healthcare facilities lack culture-based diagnostic facilities [20,21]. Recently, multidrug-resistant (MDR) and carbapenem-resistant Enterobacterales (CREs) have emerged and gradually become a significant public health challenge [22]. Several recent studies reported the emergence of MDR bacterial uropathogens [23][24][25] that increase the need for the routine application of antimicrobial susceptibility testing to detect the antibiotic of choice as well as the screening of the emerging MDR strains [14,15]. MDR, extensively drug-resistant (XDR), and pan drug-resistant (PDR) bacteria have been clearly defined according to standardized international terminology developed by the European Centre for Disease Control (ECDC) and the Center for Disease Control and Prevention (CDC), Atlanta [26]. To qualify as MDR, an agent must have gained nonsusceptibility to at least one antimicrobial agent from three or more categories. Being resistant to at least one agent in all but two or fewer antimicrobial categories is XDR (i.e., bacterial isolates remain susceptible to only one or two antimicrobial categories). Non-susceptibility to any agent in any antimicrobial category is PDR. In Ethiopia, there are limited studies available on the prevalence, antibiotic susceptibility profiles, and associated factors of UTIs in children under the age of five, and this subject has never been addressed in the study area.

Study area and period
We conducted the study in Harar town at Hiwot Fana Specialized University Hospital (HFSUH), eastern Ethiopia, from March 20 to June 10, 2021. HFSUH is the referral teaching hospital for Health and Medical Sciences students at Haramaya University that provides healthcare services to more than 5 million people around Harar and neighboring regions.

Study design and population
We conducted hospital-based quantitative study to assess the data and know the status of bacterial UTIs and associated factors in children hospitalized with suspected cases. All under-five children who were clinically suspected of UTIs and their parents or guardians who agreed to give consent to participate in the study were included from March 20 to June 10, 2021. Children under the age of five who had taken antibiotics within the previous 10 days of the datacollecting period were excluded. This study's source population included all patients who visited HFSUH for a specific clinical condition diagnosis. However, during the study period, children who went to the Mother and Child Health (MCH) clinic for UTI diagnoses were regarded as the target population.

Sample size determination and sampling technique
The sample size of 332 under-five children was determined by taking the 26.8% prevalence [27] at a 95% confidence interval (CI), a 5% margin of error, and a 10% non-respondent rate. Study participants were recruited by convenient sampling technique until we got the required sample size.

Data and sample collection
Data on sociodemographic characteristics, clinical future, medical history, and associated factors were collected using a pretested, structured questionnaire adopted after reviewing various literature [12,15,28,29] through a face-to-face interview of parents or guardians in their local languages and collecting urine samples from under-five children at the same time. The interviewer nurses performed an appropriate physical examination and card review of the child during medical registration. For toilet-trained children, a freshly cleaned random urine sample was collected in a wide-mouthed, sterile, leak-proof bottle after instructing the parents or guardians of the enrolled children to clean their genitals with soap and water. Urine samples were obtained from children who were not toilet trained by urethral catheterization, suprapubic aspiration (SPA), use of a pediatric urine collection bag, or leaving the child with the diaper off and obtaining a clean catch of urine when the child voids. Contamination was managed by giving proper instructions on how to handle the specimen correctly. The samples were delivered to the microbiology laboratory at Haramaya University College of Health and Medical Sciences within 30 minutes of collection [30], in a cold box.

Bacterial isolation and identification
We used Chesbrough's recommended culture and biochemical procedures for bacterial isolation and phenotypic characterizations [31]. Well-mixed urine specimens were inoculated on 5% blood agar and Cysteine Lactose Electrolyte Deficient Medium (CLED) (HiMedia, Mumbai, India) plates [5,32], using a 0.001 ml calibrated loop. After overnight aerobic incubation at 37˚C, colony morphology, gram staining, and biochemical tests were used for phenotypic characterizations [31] (supplementary material).

Antimicrobial susceptibility testing (AST)
Based on recommendations from the Clinical and Laboratory Standards Institute (CLSI) [34], modified Kirby-Bauer disk diffusion techniques were used for susceptibility testing. The suspension of bacterial inoculum, equivalent to 0.5 McFarland standards, was uniformly spread on Mueller-Hinton agar (HiMedia, Mumbai, India) plates using a sterile applicator cotton swab.

Data quality control
To ensure the validity of the questionnaire, the English version was translated into local languages (Amharic and Afan Oromo), and vice versa, by separate language experts and pretested at Jugol Hospital. Data collectors (laboratory personnel and nurses) were trained regarding all stages of the data collection process. To prepare the culture media, manufacturers' directions were followed. The sterility of culture media was checked by incubating overnight at 35-37˚C without specimen inoculation. The performance of the culture medium and drug disks was checked using reference strains such as E. coli (American Type Culture Collection, ATCC 25922), P. aeruginosa (ATCC 27853), and Staphylococcus aureus (ATCC 25923) [34]. To perform a drug susceptibility test, the turbidity of the bacterial suspension was adjusted to the 0.5 McFarland standard.

Method of data analysis
The data were double-entered into Epi Info version 7 and exported to Statistical Package for Social Science (SPSS) version 25 for analysis. A descriptive statistical tool was used to summarize the findings. The results were presented in words, graphs, and tables. Bivariate and multivariable logistic regression models were used to predict the relationship between dependent and independent variables. In the bivariate logistic regression model, independent variables with a P-value of less than 0.25 and clinical relevance of the variables were considered candidates for the multivariable logistic regression model [35]. The crude odds ratio (COR) and adjusted odds ratio (AOR) with their respective 95% CI were used to determine the significance of predictors. Finally, every variable with P-values less than 0.05 at a 95% confidence interval was considered statistically significant. Hosmer and Lemeshow goodness-of-fit-tests were used for adherence to multivariate logistic regression model assumptions, and a pvalue > 0.05 was considered a good fit.

Ethical considerations
An ethical review was obtained from the Haramaya University Institutional Health Research Ethics Review Committee (IHRERC) of the College of Health and Medical Sciences before conducting research. The reference number of the ethical letter was "Ref No IHRERC /027/ 2021)". An official letter of support was written to HFSUH. Each parent or guardian of a child received information on the study, including its goals, methods, potential risks, and benefits. The study participants were informed of their right to refuse or withdraw from the study at any time. Each study was unaffected by participants' refusal to take part. Informed, voluntary, written, and signed parental consent was obtained from all parents or guardians of children after explaining the purpose and objective of the study. Participants' confidentiality was strictly assured. Moreover, the positive case was reported to the attending physician or health professional. Besides, both study participants and data collectors used sanitizer and wore face masks to protect them from COVID-19.

Socio-demographic characteristics
A total of 332 study participants were enrolled in this study. Out of these participants, 64.2% were males. The median age of the children was 2 years. Most of them, 88.3%, were older than 12 months. About 69% of children were rural dwellers. Around 41.9% of study participants had parents or guardians who were unable to read and write, and 54.8% were farmers (Table 1).

Clinical future and medical history
About 139/213 (65.3%) of the males were circumcised. More than half of the participants were from inpatient wards. Most of the children (87.7%) did not have a previous history of antibiotic usage, 90.1% did not have a previous history of UTI, and 78.3% did not have a history or diagnosis of diabetes mellitus. About 29.5% of children have indwelling catheters. The majority of study participants did not complain about hematuria (78%), convulsion (74.4%), or urinary frequency (69.6%) during diagnosis. One-third of the children had a fever (Table 2).

Associated factors
In the bivariate logistic regression analysis, 12 independent variables with a p-value of less than 0.25 were candidates for the multivariable logistic regression analysis (Table 3).   (Table 4). Gram-positive bacteria. Among the tested antibiotics, rifampin (81.8%), ciprofloxacin (81.25%), and gentamycin (68.75%) were effective against gram-positive isolates. On the other hand, they showed resistance to tetracycline (56.2%), tobramycin (54.5%), and trimethoprimsulphamethoxazole (54%). S. aureus was the most predominant isolate, which revealed 83.3% sensitivity to each of rifampin and ciprofloxacin. However, there is 83.3% resistance to each of the antibiotics, ampicillin and tetracycline (Table 5).

Discussion
Although UTIs can cause serious illness in children, the definitive diagnosis of UTIs is often overlooked in most developing countries due to a lack of facilities, high laboratory costs, and the challenges of obtaining urine from children, especially those who would not void voluntarily [14]. Presently, the effectiveness of commonly prescribed drugs is decreasing globally due to increasing resistant strains. As a result, an infection caused by those strains is challenging; it becomes difficult to treat [5,36]. Colony counts below 10 5 CFU/mL in random, clean urine samples for toilet trained children and below 5 x 10 4 CFU/mL for children under 2 years old and catheterized or SPA urine are not considered significant for defining a UTI [37]; however, this cannot be ignored in patients with immunosuppressed.
In this study, the overall prevalence of bacterial UTIs in children was 24.1%, which is in agreement with other studies in Hawassa, Ethiopia (27.5%) [14], Uganda (26.8%) [27], and

PLOS ONE
Bacterial profile, antimicrobial susceptibility pattern, and associated factors of urinary tract infections  This study's most common isolate was Escherichia coli (28.75%), which is comparable to some studies of Ethiopia [14,15]. E. coli is the most common organism identified in nearly all UTI studies worldwide: 71.7% [13], 31.8% [38], 35.7% [42], and 41.2% [43]. The high rate of E. coli might be due to the high abundance of E. coli in the fecal flora, which ascends through the genitalia to cause UTI, in addition to having a unique structure such as P-fimbriae or pili adherence factors, which promote E. coli attachment to the uroepithelial cells, allowing for multiplication and tissue invasion [44]. But, this finding contradicts a study done in Abakaliki, Nigeria in which Klebsiella spp. (24.5%) were most commonly isolated [39], and in Uganda, Proteus spp. (39.5%) were most frequently isolated [27]. This could be due to variations in specimen collection techniques and the existence of many virulence factors [44][45][46].
On the analysis, being a rural resident was significantly associated with bacterial UTIs (p = 0.008) because the majority of study participants were from rural settings and host fecal flora might be the source of the infections [47]. Boys who were not circumcised had a nearly four-fold higher risk of UTIs. This could be because men with uncircumcision and obstructive uropathies have a higher risk of UTIs. As a result, a tight foreskin may interfere with the normal passage of urine and hinder the full emptying of the bladder [48]. patients who self-medicated without prescription were 7 times more likely to acquire UTIs; this finding contrast with another study conducted in Hawassa, Ethiopia [14]. These variations could be attributed to the low rate of self-medication in Hawassa (5.9%) when compared to this study's (12%). Table 5. Antimicrobial susceptibility pattern of gram-positive bacteria isolated from the urine of under-five children at HFSUH, eastern Ethiopia, 2021.

Bacterial isolate
Pattern Antimicrobial

PLOS ONE
Bacterial profile, antimicrobial susceptibility pattern, and associated factors of urinary tract infections    The history of previous UTIs shows six times more susceptibility to culture-positive results than the correlative one, similar to earlier studies in Bahirdar, Ethiopia [15]. It may be caused by the persistence of resistant strains from previous uropathogenic infections or by the recurrence of those illnesses [49]. Patients with indwelling catheterization have an eleven times higher risk of developing catheter associated UTIs than the non-indwelling catheterized group, which is contrary to the study in Gondar, Ethiopia [47]. Due to the presence of an indwelling catheter device and potentially pathogenic multidrug-resistant organisms in the hospital environment, catheter insertion predisposes to the development of catheter-associated UTIs by creating a new entry point for bacterial invasion and pushing bacteria into the bladder [50]. On the analysis, diabetic patients were an independent risk factor for the acquisition of bacterial UTIs when compared with non-diabetic patients (p = 0.014). This might be due to frequent urination and a high blood sugar level because the high sugar level gives a favorable growth environment to the pathogens [51]. Study participants with a body temperature of � 38.5˚C had 4 times more risk for bacterial UTI compared to their counterparts; this result is consistent with the study done in Tanzania [12]. Though it was contradicted by the study in Nigeria [52]. It could be due to geographical variation, climatic change, and differences in the number of study participants. Under-five children with symptoms of urinary frequency were six times more likely to have a culture-positive result. This finding was a disparity from the study in Nigeria [13]. Frequently urinating, infections, injuries, or irritation of the bladder, as well as changes in the muscles and nerves that regulate bladder function, could all be to blame for these variances [53].  naturally resistant to penicillin); through the making of broad spectrum β-lactamases; through the yielding of ESBLs or penicillinases; and through non-carbapenemase mechanisms of resistance such as the turn of outer membrane proteins and the occurrence of efflux pumps, which can also act harmoniously with the overexpression of β-lactamases [63]. S. aureus has been able to develop resistance mechanisms to most antibiotics used against it through modification of penicillin-binding protein-2a (PBP2a), penicillinase, ribosomal methylation of binding sites, and efflux pumps. Similar to all β-lactam medications, S. aureus gained methicillin resistance through the development of PBP2a, which is resistant to all β-lactam antibiotics [65,66].

Limitations of the study
One major limitation is the inability to identify some bacterial isolates into species and serotypes. In additions, that the possibility of contamination (especially of gram-negative organisms) exists in the bag or open diaper collection methods. Furthermore, the study could not indicate the molecular characteristics, ESBL and CRE of isolates due to a lack of resources.

Conclusion and recommendation
In comparison to the majority of other investigations conducted in Africa, more than onefourth of the children had cultures that were positive for various bacterial uropathogens. UTIs were made more common by indwelling catheterization, a prior history of the condition, males who had not undergone circumcision, a history of using antibiotics without a prescription, rural dwellers, diabetes mellitus, children's body temperatures, and urine frequency. Both gram-positive species and a variety of gram-negative species were documented, which yielded 24%. The most common isolates that cause UTIs are E. coli, K. pnemoniae, and S. aureus. These strains showed different percentages of susceptibility to the tested antibiotics. The current study points out that meropenem, ciprofloxacin, and amoxicillin-clavulanic acid are choices of drugs for UTI in the study area. Significantly high proportions of MDR strains were observed, especially against commonly used drugs. As patients are diagnosed and treated empirically in the study setting, hospitals should establish antimicrobial stewardship programs. Patients should be managed with microbiological evidence to avoid antimicrobial prescriptions for patients without bacterial infection and better control the spread of drug resistance. The careful prescribing of antibiotics is needed to halt the progress of drug resistance. Hence Working on the identified associated factors is vital to minimize the observed high prevalence of UTIs.