Prevalence, Antimicrobial Susceptibility Patterns, and Risk Factors Associated with Enterococci among Pediatric Patients at Dessie Referral Hospital, Northeastern Ethiopia

Background Enterococcus species, which is previously considered as medically not important, now becomes one of the leading causes of nosocomial infections. Even though it becomes the most public health concern and emerging multidrug-resistant pathogen, there is no enough data in the study area. Objective To determine the prevalence, antimicrobial resistance pattern, and associated risk factors of enterococci infection in pediatric patients. Methods A hospital-based cross-sectional study was conducted from February to May 2019 at Dessie Referral Hospital, Northeastern Ethiopia. A total of 403 pediatric patients were included in the study, and a pretested structured questionnaire was used to collect sociodemographic and risk factor-related data. Clinical samples such as urine, blood, wound swabs, discharges, and other body fluids were collected aseptically and inoculated on to Bile Esculin Azide Agar, and colony characteristics, Gram stain, catalase, salt, and temperature tolerance tests were employed for bacterial identification. Antimicrobial sensitivity tests were performed using the modified Kirby-Bauer disk diffusion method. Data was entered into SPSS software version 25 and descriptive statistics; bivariate and multivariate logistic regression analyses were performed. In all cases, a P value ≤ 0.05 with corresponding 95% confidence interval were considered as statistically significant. Result The overall prevalence of enterococci was 2.7% (11/403). Of which, the highest number of enterococci infection was recovered from urine sample (54.5%) followed by blood (27.3%), wound swab (9.1%), and other body fluids (9%). The overall multidrug resistance rate was 54.5%. Higher drug resistance pattern was observed against tetracycline, chloramphenicol, and amoxicillin/clavulanate. Having history of invasive procedure (P < 0.001), chronic illness (P < 0.001) and previous admission history of the children (P < 0.001) were statistically significant associated risk factors for pediatrics enterococci infection. Conclusion The prevalence of enterococci from pediatric patients in this study was relatively low compared to other studies. Significant rates of MDR and VRE were identified, and the risk of infection became high when children had a history of different chronic illnesses and history of admission and underwent invasive treatment procedures. Therefore, efforts should be made to prevent enterococci infections and spread of multidrug-resistant enterococci.


Introduction
Enterococci are members of Gram-positive Enterococcaceae family and are nonspore forming, facultatively anaerobic, oxidase, and catalase negative bacteria that occur singly, in pairs or short chains [1]. Moreover, enterococci are known to survive in a range of hostile environments, including growth in the presence of 6.5% NaCl, temperature of 5 to 65°C, pH 4.5 to 10, and hydrolyze esculin in the presence of 40% bile [1,2]. Enterococci are common indigenous flora of the gastrointestinal tracts and can also normally present in the oral cavity, vagina, oropharynx, and urethra of humans [3].
In fact, enterococci species have low virulence factors. However, under conditions in which the host encountered immune suppression or the integrity of the gastrointestinal or genitourinary tract has been disrupted, they spread to normally sterile sites and cause various infections. These infections include urinary tract infections, wound infection, sepsis, endocarditis, intra-abdominal abscesses, and biliary tract infections [4].
According to Centers for Disease Control and Prevention (CDC) nationwide report, enterococci are one of the leading causes of nosocomial infections worldwide, with healthcareassociated infections occurring more frequently in resourcelimited settings than in developed countries [5]. Moreover, the pathogen distribution and antimicrobial resistance patterns in healthcare-associated infections (HAIs) from pediatric groups also indicated 8% of enterococci species [6].
Different risk factors have been reported to be associated with the spread of enterococci infections including concurrent infections, history antibiotic use, surgery, catheterization, longer duration of hospitalization, and underlying immunosuppressing diseases such as HIV, diabetics, and cancer [7].
The emergence and spread of antimicrobial resistance among enterococcus species pose enormous challenges for clinicians, especially in the management of severe infections. The increased prevalence and dissemination of multidrugresistant enterococcus has narrowed the therapeutic options globally, as the majority of enterococcus isolates exhibit a high level of resistance to ampicillin, penicillin, and vancomycin, which are indeed the most historically useful antienterococci antibiotics [8].
In 2018, the national survey for antimicrobial resistance in Europe report indicated that vancomycin-resistant enterococcus species increased significantly from 10.4% in 2014 to 14.9% in 2017 in different countries [9]. Moreover, according to worldwide surveillance summary of drug potency of Gram-positive pathogens, the overall prevalence of vancomycin-resistant enterococci (VRE) in Europe, Asia and Pacific, and Latin and North America ranges between 1% and 9.8% [10].
The prevalence and drug resistance figures are significantly high in developing countries. In Africa and Ethiopia, some reports indicated the overall prevalence of enterococci and antibiotic resistance ranged from 2.2% to 76% and 6.3% to 95.5%, respectively [11][12][13][14][15][16]. Therefore, efforts should be made to prevent enterococci infections and emergence of multidrug-resistant enterococci. Hence, studying the prevalence, dispersion, and correlation of possible associated risk factors and examining the drug resistance pattern of enterococcus species, particularly in vulnerable target groups such as pediatrics, are important for effective prevention, control, and management of infections. Therefore, the study was aimed at determining the prevalence, antimicrobial susceptibility patterns, and associated risk factors of enterococci among pediatric patients in Dessie Referral Hospital (DRH), Northeastern Ethiopia.

Method and Materials
2.1. Study Design, Area, and Period. A hospital-based crosssectional study was conducted from February 2019 to May 2019 at Dessie Referral Hospital, South Wollo Zone of Amhara regional state, Northeast Ethiopia. Dessie Referral Hospital is located in Dessie city, 400 km from the capital, Addis Ababa, and 471 km far from Bahir Dar, the capital of the Amhara regional state. The city has one referral hospital, one general hospital, three private general hospitals, five health centers, eight private higher clinics, and one Public Health Research Institute. Dessie Referral Hospital provides emergency, antiretroviral therapy (ART) services, chronic care, surgical, dental, medical, pediatric, gynecologic, obstetric, and other services. The hospital serves patients from all parts of the region which comprises a population of more than 4 million people. Dessie Referral Hospital pediatric services consist of inpatient and outpatient departments. The outpatient pediatric department consists of 15 beds and five rooms, while the inpatient pediatric ward consists of 13 rooms and 49 beds, with an average of about 7 children admitted per day.

Inclusion and Exclusion
Criteria. All children younger than 15 years old, attending DRH and who were requested for laboratory investigation during the study period, were included in this study. On the other hand, pediatric patients who received antibiotics within the past 2 weeks were excluded. Moreover, as enterococci are normal flora of some specific sites including respiratory, genital, and gastrointestinal tracts, samples such as sputum, throat swab, stool, and vaginal swabs were excluded from the study.

Sample Size Determination and Sampling
Technique. The sample size was determined using a single population proportion formula considering 50% prevalence, marginal error of 5%, and 95%confidence interval = 1:96 by using the following sample size determination formula: where n is the minimum sample size required, Za/2 is the significant value for 95% confidence interval, P is the expected prevalence of enterococci infection, and d is the margin of error.By including 5% nonresponse rate, a total of 403 children were included in this study using systematic random sampling.

Data and Specimen
Collection. Data were collected using a short interview guided by pretested structured questionnaire consisting of the client's sociodemographic, clinical, and risk factor data. Clinical samples were collected from each study participant aseptically. About 5 ml of the blood sample was collected from children and dispensed into blood culture bottle prepared with 25 ml of Tryptic Soya Broth (FL Medical, Italy) aseptically. Ten ml of freshly voided midstream urine specimen was collected using wide mouth, leak-proof, sterile, plastic container under the supervision of the principal investigator and processed within 2 hours of collection. Approximately 5 ml of cerebrospinal fluid (CSF) sample was collected aseptically into sterile tube by lumbar or ventricular puncture performed by a physician and processed within one hour of collection. Wound swab, pus, eye, and ear discharges were obtained using sterile cotton tip applicator stick aseptically. The blood sample was transported with the blood culture broth while all swabs were transported within BHI broth (HiMedia™  [18]. From a pure culture, 3-5 selected colonies of bacteria were taken and transferred to a tube containing 5 ml sterile normal saline and mixed gently to make homogenous suspension and the turbidity of the suspension was adjusted comparably to a McFarland 0.5 turbidity standard. A sterile cotton swab was used to streak the plates, and the excess suspension was removed by gentle pressing and rotation of the swab against the inside wall surface of the tube. The swab was then used to distribute the bacteria evenly over the entire surface of Mueller Hinton agar (MHA) (Conda Ltd., USA). The inoculated plates were then left at room temperature to dry for 3-5 minutes, and a set of 6 antibiotic discs were applied on the MHA. Based on availability, CLSI recommendation, and prescription practice of DRH, the test was carried out for the following drugs: erythromycin (E, 15 μg), chloramphenicol (C, 30 μg), tetracycline (TE, 30 μg), ampicillin (AMP, 10 μg), nitrofurantoin (NIT, 30 μg), ciprofloxacin (CIP, 5 μg), penicillin (P, 10 IU) and vancomycin (VA, 30 μg), and amoxicillin-clavulanic acid (AMC, 20/10 μg). All antibiotic discs were from Oxoid Ltd., UK. The plates were then incubated at 37°C for 24 hours. Diameters of the zone of inhibition around the discs were measured using a digital caliper. The interpretation of the results of the antimi-crobial susceptibility tests was based on the standardized  (Table 2). Moreover, the peak positivity rates were obtained from pus sample (20%) followed by wound swab (6.2%) and urine specimens (2.9%) (Figure 1).

Association of Risk Factors for the Acquisition of
Enterococci Infection. In this study, a total of 14 independent variables were considered during the bivariate analysis of risk factors for enterococci infection. In the multivariate analysis, the presence of enterococci infection was significantly associated with having a history of invasive procedure (AOR = 26:91, 95% CI: 4.96, 148), chronic illness (AOR = 16:91, 95% CI: 2.98, 88.08), and history of admission (AOR = 13:73, 95% CI: 3.01, 62.59) with P < 0:001 for all (Table 3).
3.5. Multiple Drug Resistance Pattern of the Isolates. All detected enterococcus isolates (11) were resistant to at least one antimicrobial agent, whereas 9 isolates were resistant to ≥2 antimicrobials. Multidrug resistance (defined as nonsusceptibility to at least one agent in three or more antimicrobial categories) was seen in 6/11 of the enterococci isolates (Table 5).

Discussion
In the past 1970s and 1980s, enterococci have been changed from being intestinal normal flora of little clinical significance to becoming one of the most common nosocomial pathogens associated with significant morbidity and mortality worldwide [19]. The present study was conducted to determine the prevalence of enterococci infection, evaluates the pattern of drug susceptibility and identify possible risk factors associated with enterococci infection among pediatric patients. The overall prevalence of enterococci in the present study was 2.7% (11/403). This finding was consistent with the previous report of European Center for Disease Control report 1.8% [6], Nepal 3.4% [20], and India 4% [21]. Similar findings were also reported in Ethiopia of Jimma (2.23%) [14], Addis Ababa (2.23% and 1.82%) [22,23] and Gondar (2.13%) [24]. However, it was lower than the previous reports from Turkey 6.5% [25], India 6% [26], and Gondar 8.9% [16]. The differences with the findings of this study may be attributed to the variation in geographical area, number of samples examined, the extent of time, and method used for sampling.
In this study, enterococci species were most predominantly isolated from urine specimens followed by blood and wound swabs. This was similar to previous reports conducted in Addis Ababa Ethiopia [27], India [28], and Iran [20]. This indicated enterococci infection mostly causes UTI, bacteremia, and wound infections. The most probable reason for the high isolation rate of enterococci from UTI cases might be due to the close proximity of anal opening to the urethra as enterococci reside as commensals in gastrointestinal tract.  With this, uropathogens originate from the fecal flora, spread across the perineum, and invade the bladder through the urethral opening. Moreover, urinary catheterization also has a role in some cases and contributed to higher isolation of enterococci from urine specimens. The present study showed that enterococci isolates had an overall multidrug resistance rate of 54.5%. This finding was in line with a study report from Ethiopia 60% [29]. However, considerably higher in comparison to other study findings from Tunisia, 7.1% [30], and India, 10% [21]. This discrepancy might be due the socioeconomic difference, type of clinical sample used, and age difference of the target study groups. On the contrary, the current finding was lower than a study report from Spain, 80% [31], and India, 82% [32]. Similarly, lower than all studies performed in Ethiopia: Jimma, 89.5%, 80.8% [11,14], Southern Ethiopia, 82.5% [15] and Gondar, 75% [16]. The variation might be due to differences in clinical sample used and age difference of the target study groups. Moreover, these MDR pathogens call for establishment of targeted antimicrobial resistance (AMR) surveillance system, focused infection prevention and control (IPC) mitigation strategies and introduction of cost-effective diagnostic tools for their routine screening in the hospital settings.
Bacterial drug resistance for the commonly used antibiotics is currently spreading globally. In our study, 5/11 (45%) of the enterococci isolates were resistant to ampicillin, penicillin, vancomycin, ciprofloxacin, and nitrofurantoin. This was in line with the report from India, 52.6% [28], and Jimma Ethiopia, 54.5% [14]. However, our finding was higher than study reports from Ethiopia: Addis Ababa, 6.7% [23], Gondar, 5.5% [12], and Canada, 19% [33]. Besides, our finding was lower than study reports from southern Ethiopia, 100% [34], and Gondar 72% [16]. Moreover, a study conducted in two primary hospitals of southwest Nigeria showed 100% resistant to ampicillin [35]. The highest rate of resistance to ampicillin might be the sign of prolonged irrational usage of the drug in the area.
The rate of 5/11 (45.5%) vancomycin-resistant enterococci (VRE) in the current study has made the scenario worse as it is the preferred choice of drug in the case where this bacterium became resistant to other antibiotics in the area. Even though the result was comparable with the study report undertaken in all age groups in Gondar, 41.5% [16], and Nigeria, 42.9% [35], while it was by far greater than reports of other studies conducted among HIV patients in Ethiopia: Gondar, 5.5% [12], southern Ethiopia, 7.5% [15], Addis Ababa, 6.7% [23], Jimma, 22.7% [11], 6.3% [14], west Iran, 24% [36], Canada, 4% [33], and reports of a nationwide study from Europe, Asia and Pacific, Latin and North America, 1% to 9.8% [10]. The possible reason might be due to the gradual change in the MDR strains, antibiotics selective pressure, differences in drug prescription and usage habits, and socioeconomic status.

BioMed Research International
for such upsurge in resistance pattern of enterococci species might be the gradual spread of drug resistant strains, especially in developing countries where there is high usage habit, these antibiotics mainly due to their easy availability and low cost. Moreover, these antibacterial drugs are commonly prescribed irrationally by medical practitioners as an empirical treatment options, and self-prescribing of these drugs is also common [38], which in turn causes emergence and spread of drug resistance.
The multivariate analysis of the present study showed that having a history of invasive treatment procedure was significantly associated with pediatrics enterococci infection (P < 0:001). This report was in line with a study done in primary hospital of South Korea [39]. The main reason for the survival of enterococci infection in patients undertaking invasive medical procedure/treatment was the fact that foreign devices could be used as a portal of access by disruption of the normal integrity of the body. Moreover, invasive  The present study also showed that having a history of chronic illness (P < 0:001) and admission or hospitalization (P < 0:001) was significantly associated with pediatrics enterococci infection. This report was comparable with a study done among hospitalized patients in Germany [40]. This association is mainly attributed to the effect of chronic illness and hospitalization in diminishing immune status of patients which allows enterococci infection to flourish and cause a clinical illness [41]. Moreover, hospitalization further expose patients, particularly children, to acquire enterococcal infection readily through contamination from the healthcare environment, in which high densities of contaminants found on medical devices (such as blood pressure cuffs, intravenous fluid pumps, or stethoscopes), gowns, bed rails, bedside tables, bed linens, urinals, and bedpans [3].
Species investigation was not performed due to financial issue and shortage of reagents and materials.

Conclusions
The prevalence of enterococci from pediatric patients in this study was relatively low compared to other studies. The risk of infection became high when children have a history of different chronic illness and history of admission and undergoes invasive treatment procedures. A significant rate of multidrugresistant enterococci including VRE was identified from clinical samples in the study area. Such occurrence of MDR and VRE showed limited antibiotic treatment options for enterococci infections. Therefore, efforts should be made to prevent nosocomial enterococci infections and spread of multidrugresistant enterococci. Periodic evaluation of drug susceptibility pattern is also essential for rational and appropriate use of antibiotics. Moreover, appropriate prescription and use of antibiotics and large-scale species identification and genotypic studies at national level are necessary.

Data Availability
Data supporting the conclusions of this article are within the manuscript.

Ethical Approval
Ethical clearance was obtained from the School of Biomedical and Laboratory Sciences, University of Gondar Ethical Review Committee, and permission was obtained from the Dessie Referral Hospital and Amhara Public Health Institute, Dessie Branch. Confidentiality was kept throughout the study, and positive findings for enterococci infection were communicated to the respective physicians with their drug susceptibility patterns for treatments.

Consent
Prior to commencing the study, written parental/guardian consent and/or patient accent was obtained for each study participant before data collection.

Conflicts of Interest
The authors declare that they have no conflicts of interest in this research work.

Authors' Contributions
AA, MT, and MAB were involved in proposal writing and designing the study and participated in analysis and interpretation of data. AA, MT, SG, and MAB were involved in data collection and drafting of the manuscript. AA, MT, SG, and MAB finalized the write up of the manuscript. All authors critically revised the manuscript and read and approved the final manuscript for publication.