Bacterial etiologies, antimicrobial susceptibility pattern and associated factors among patients suspected sterile body site infections at Debre Markos Comprehensive Specialized Hospital, Northwest Ethiopia

Background Sterile body locations are usually associated with clinical urgency and life-threatening illnesses, and they are typically contaminated with diverse bacterial etiologies. If the bacteria acquire resistance to antimicrobial drugs, the public health crisis will only worsen. In developing countries, drug-resistant bacteria are common because of poor surveillance, diagnostic capacity, and control measures. Early diagnosis, and assessing the drug resistance and factors associated with infection are important to combat the drug resistance and treatment. This study aimed to assess the bacterial etiologies, antimicrobial susceptibility pattern, and possible associated factors among patients suspected of sterile body sites. Methods A hospital-based cross-sectional study was conducted from June 2022 to August 2022 at Debre Markos Comprehensive Specialized Hospital in Amhara regional state, Ethiopia. One hundred seven study participants were selected using consecutive convenient sampling techniques. A structured questionnaire was used to collect socio-demographic and clinical data. Gram stain was done for a preliminary report and inoculated into blood agar, MacConkey agar, and chocolate agar and incubated aerobically and micro aerobically at 37°C for 24 h. Antimicrobial susceptibility testing was done by the modified Kirby Bauer’s disk diffusion method. Data were analyzed using bivariate and multivariate logistic regression was used. A p-value less than 0.05 is considered as statistically significant. Results The overall magnitude of sterile body site infection among study participants was 7.5% (14/187). The majority of the isolates were Gram-negative bacteria with the predominant species Enterobacter cloacae accounting for 28.57% (4/14). Among isolates 78.57%(11/14) of them were multidrug-resistant isolates. Being inpatient, co-morbidity, and alcohol consumption were significantly associated with sterile body site infection. Conclusion In our study, Gram-negative bacteria were the predominant bacteria that infects sterile body fluid. The prevalence of multi-drug resistance bacteria isolates was significantly high. Therefore, before prescribing an empirical treatment, a medical professional should identify the bacterial etiology of sterile body fluids and the susceptibility of microbes to the drug.


Introduction
Sterile body fluids (SBF), which include cerebrospinal fluid (CSF), peritoneal fluid, pleural fluid, and synovial fluid, are fluids that have no normal flora of bacteria.Sterile body infection and drug resistance bacteria is life threatening ceases.One of the most urgent public health problems facing the world today is antibiotic resistance.It is typically associated with higher medical costs, longer hospital stays, and more mortality (1,2).Because of insufficient policy implementation for the prevention and control of drug resistance and laboratory infrastructure inadequacy for obtaining samples and identifying the particular agent, the situation simply gets worse.It was more challenging in those settings since it was necessary to collect samples from invasive locations and confirm the agents for diseases including meningitis, peritonitis, septic arthritis, and pleural empyema (3-8).These circumstances force medical professional to treat patients empirically, which exacerbates the problem of drug resistance already facing public health (1,2).
Meningitis, peritonitis, septic arthritis, and pleural empyema are only a few of the potentially fatal diseases caused by bacterial infections in sterile body sites (3)(4)(5)(6)(7)(8).A worldwide study found that there were 2.5 million cases of meningitis and 236,000 deaths from the disease (9).Also Wunrow et al. (9), bacterial meningitis affects approximately 1.2 million people each year and causes almost 170,000 deaths globally (10).In developing nations, bacterial meningitis continues to pose a serious threat to public health (11).The highest burden of bacterial meningitis occurs in sub-Saharan Africa, known as the meningitis belt (12).Despite the availability of effective antimicrobials, mortality rates in developing nations are higher, ranging from 16 to 32% (13).
Furthermore, spontaneous bacterial peritonitis (SBP) contributes 10 to 30% prevalence and its prevalence even increased among hospitalized patients (25-30%), of which 15% of the infection rate is caused by post-operative procedures (4,14,15).This disease is significantly associated with a high level of mortality (90%) (4).Gramnegative flora is the major cause of SBP and mortality rates reach up to 10 to 46% (16).Moreover, bacterial-related pleural infection has increased from 7.6 to 14.9% globally with a 20% mortality rate (15,17).Also, Septic arthritis is a serious orthopedic issue and the incidence of septic arthritis of native joints has been reported to be around two cases per 100,000 people per year (18,19).In this regard, the prevalence of bacterial isolates from sterile body fluids in Ethiopia is 11.5-20.2%.Of this, 10.6-40.7% of the isolates were Gram-positive bacteria whereas the rest 9.6-74.6%were Gram-negative bacteria (5)(6)(7)(8).
Currently, the emergency of multidrug-resistant (MDR) bacterial infection challenges treatment success and leads to increased mortality (20,21).MDR bacteria are bacteria that are resistant to three or more classes of antimicrobial drugs.Ethiopia, a developing country in Sub-Saharan Africa, further aggravates the problem with its inadequate laboratory infrastructure, poor antimicrobial surveillance and control measures, and high rates of antimicrobial misuse from drug stores even without a medical prescription (22,23).
For instance, a study conducted by Frehiwot and colleagues in Ethiopia found a high level of MDR (75.9%) in the bacterial isolates.According to this study 100% and 90% of MDR was from CoNS isolates and K. pneumoniae isolates, respectively (6).The main reason for the high prevalence of drug resistance in this setting could be related to a habit of self-medication of antimicrobials from drug stores, poor microbiology facilities, and diagnostic capacity (24).Studies have also shown that surgical procedures, trauma, age, hospitalization, chronic illnesses (HIV/AIDS, hepatitis, cirrhosis, cancer, and diabetes mellitus), as well as other behavioral factors (alcohol misuse and cigarette smoking), are significantly associated to bacterial invasion into the sterile body site (25-27).
Therefore, assessing the prevalence of bacterial etiology, and antimicrobial susceptibility patterns, and addressing potential associated factors of bacterial infection of sterile body sites in settings where poor surveillance and control measures, inadequate laboratory infrastructure, medical professionals treat patients empirically, as well as a lack of published data on this matter, could contribute to better management of bacterial diseases and address factors for the development of drug resistance.

Study setting, design, and population
A hospital-based cross-sectional study was conducted among patients suspected of bacterial sterile body site infections at Debre Markos Comprehensive Specialized Hospital (DMCSH), Northwest Ethiopia, from June 2022 to August 2022 from patients suspected of sterile body fluid infections and had no a history of antimicrobials intake within the last two weeks.Socio-demographic (age, gender, marital status, residence, occupation, income, and educational status), clinical data (history of hospitalization, co-morbidities, cigarette smoking), and other possible associated factors were collected by using a pretested structured questionnaire.DMCSH is found in Debre Markos town and provides specialized health care services through its medical, clinical and diagnostic departments for approximately 255,248 people per year.In the present study, the prevalence of bacterial growth was dependent variable whereas the Sociodemographic data (age, gender, residence, occupation, educational status, and monthly income), Clinical-related data (Patient setting (inpatient, outpatient), history of Hospital admission, history of invasive procedures& surgery, and clinical features), and Co-morbidity (DM, HIV, TB, chronic liver disease, cancer &others) and Behavioral factors (Alcoholism, Smoking cigarettes) were independent variables.

Sample size and sampling technique
The sample size was determined using a single population proportion formula by considering a prevalence (p) of 14.1% from the Tikur Anbesa Hospital, a 95% confidence interval of Z = (1.96), and a margin of error (d) of 5% (6).

Specimen collection, transportation and bacteriological analysis
One to five mL of sterile body fluid samples (lumbar puncture, paracentesis, thoracentesis, and arthrocentesis) were collected aseptically in sterile glass tubes from each participant by experienced physicians.The sample collection tube was labeled and closed tightly.Then specimens were transported to the hospital laboratory within 10 min and laboratory analyses were started within 30 min of specimen delivery.

Inoculation, incubation and isolation of bacterial agent
A drop of the suspension was inoculated into blood agar, chocolate agar, and MacConkey agar plates (Oxoid Ltd., Basingstoke, and Hampshire, United Kingdom).Blood and chocolate agar plates were incubated at 35-37°C in a candle jar with a capnophilic (5-10% CO 2 ) environment and MacConkey agar was incubated aerobically for the isolation of Gram-negative aerobic bacteria and examined after 24 h.Mannitol salt agar (Oxoid Ltd., Basingstoke, and Hampshire, United Kingdom) was used to isolate S. aureus when beta-hemolysis was observed from the blood agar plate.For mixed colonies, a sub-culture of blood agar and chocolate agar was performed to get pure colonies (28).

Identification of bacterial isolates
Isolates were identified by colonial morphology such as hemolytic reaction on blood agar plate, pigment production or color changes surrounding carbohydrate fermenting colonies on MacConkey agar plate.After obtaining pure colonies, further identification was conducted using standard bacteriological techniques including Gram reaction, colony morphology, and biochemical tests.A battery of biochemical test such as catalase test, coagulase test, Optochin sensitivity test, bacitracin sensitivity test, mannitol test, carbohydrate utilization, indole production, mannitol fermentation, citrate utilization, lysine decarboxylation, H2S production, triple sugar iron utilization, and motility testing used to identify Gram positive and Gram negative bacteria responsible for sterile body fluid infections (28).

Data analysis and interpretation
Data were entered into Epi-data version 4.6 software and exported to SPSS version 25 for analysis.Antimicrobial susceptibility pattern was analyzed by World Health Organization (WHO) NET 2022 software.Binary logistic regression models were used to predict the relationship between dependent and independent variables.Variables with a p-value <0.25 in the bivariate logistic regression analysis were moved to a multivariate logistic regression analysis.Adjusted odds ratios with 95% confidence interval (CI) were used and variables with a p-value less than 0.05 in multivariate analysis were considered as statistically significant.

Ethical considerations
Ethical clearance was obtained from the Ethical Review Committee of the College of Health Sciences, Debre Markos University (Hsc/R/C/ Ser/PG/Co/197/11/14), and Amhara Public Health Institute (APHI).Written informed consent and assent were obtained from each participant and parents or legal guardians.Information obtained from this study used only for the purpose of the study.

Sociodemographic characteristics
In this study, a total of 187 sterile body fluid samples were collected from patients suspected of bacterial infections.Among enrolled participants, the majority 103/187 (55.1%) of study participants were males, 72/187 (38.5%) were unable to read and write, and 105/187 (56.1%) were rural residents.The median and  1).

Factors associated with bacterial isolates of sterile body site infections
Patients in the inpatient department were 10.267 times more likely to be infected than outpatients [AOR =10.267, 95% CI: (2.37, 44.50), p value =0.002].Similarly, patients with comorbidity were   4).
In this study, E. cloacae was the most frequent bacterial etiologies causing sterile bodies infection, in contrast.E. coli was the most prevalent isolated bacteria in a previous study conducted in EPHI, Ethiopia (36), Harar, Ethiopia (8), India (33), and Nepal (32).The detection of E. cloacae may be an indication of an opportunistic pathogen associated with hospital-acquired infections, due to poor infection prevention practices such as overcrowding, lack of standard facilities with controlled ventilation, and poor sterilization of all gowns and surgical equipment.
In our study, all Gram-positive isolates were resistant to penicillin and cefoxitin, but most of the isolates were susceptible to erythromycins, doxycycline, and tetracycline.Similarly, high levels of resistance to beta-lactam agents were reported in previous studies in Ethiopia (8,36) and India (39).This may be due to the presence of mobile-resistant factors, widely usage of these drugs in the treatment of different infections, and irrational use of antimicrobials.
All S. aureus isolates were methicillin-resistant (MRSA), similar to the study conducted in Harar, Ethiopia (8).But this finding is higher than study conducted in India, 36.6% (34), Nepal (30%) (32), and Addis Ababa, Ethiopia (50%) (6).This may be due to geographic variation, differences in infection control practice, and differences in treatments used.This may be related to various factors.As evidenced, for example, by this study, patients in the inpatient department had an infection rate that was 10.267 times higher than that of outpatients [AOR =10.267, 95% CI: (2.37, 44.50), p value =0.002].Similarly, individuals with co-morbidity had a 6.31-fold increased risk of bacterial infections [AOR = 6.31, 95% CI: (1.487-26.771),p value =0.012] compared to those without co-morbidity.
In this study, high rates of multi-drug resistance (MDR) (78.57%) (95% CI: 56.30-99.70%)were reported, which is comparable to the findings in Addis Ababa, Ethiopia (75%) (6,36), Harar, Ethiopia (76.4%) (8), Mekelle, Ethiopia (90%) (5) and India (62.9%) (40).But higher than study conducted in Nepal (40%) (32).These variations may be due to different bacterial isolates, patients' awareness of the use   of the antimicrobials, the difference in antibiotic prescribing policies, easy availability of some drugs without a prescription, isolated strain difference, and indiscriminate use of common antimicrobials.In addition, poor infection control practice (the use of alcohol, bleach, disposable gloves, sanitizer), along with poor hospital sanitation conditions, waste management guidelines, and inadequate infrastructure, raise serious concerns in Ethiopia since they may contribute to the emergence of drug resistance.Being an inpatient, comorbidity, and the habit of drinking alcohol had a statistically significant association with bacterial infection of sterile body fluids.This finding is consistent with a study from Harar, Ethiopia (8).This could be due to the hospital's level of environmental hygiene and surgical invasive procedures.Another significant factor associated with bacterial infection of sterile body fluids was having comorbidities.This might be due to low immune status due to any underlying medical conditions.
Drinking alcohol was also statistically significant to bacterial infection of sterile body fluids.This result agrees with studies in Arba Minch, Ethiopia (25).The reason might be due to common alcohol consumption influences the contribution of the intestinal microbial populations by way of disturbing the stability of intestinal homeostasis and impairs the phagocytic function of macrophages and neutrophils recruitment, as well as minimizes bacterial clearance inside the lungs.
In the present study, the main shortcomings were the limited sample size and inadequate hospital infrastructure.The microbiology laboratory, for example, lacks anaerobic culture, which is expensive and requires specialized equipment.Hence, the identification was limited to anaerobic bacteria because only aerobic cultures were carried out.This might be one of the reasons for the low culturepositive yield in this study.

Conclusion
E. cloacae, P. aeruginosa, and S. pneumoniae were the most frequently isolated bacteria from various sterile bodily fluids.Antibiotics resistance was high among Gram-negative isolates.The majority of the isolates have shown multidrug resistance (MDR.Being an inpatient, comorbidity, and drinking alcohol were significant predictors for bacterial infection of sterile body fluids.Thus, there should be; regular monitoring of antimicrobial resistance patterns of infected body fluids, periodic antimicrobial surveillance, effective antibiotic policy with good medical practices, and proper infection prevention practices to control the disease transmission.Furthermore, as this study was done in a single study area, a multicenter study should be done to improve the clinical utility of the study.

FIGURE 1
FIGURE 1Bacterial isolates from patients diagnosed with sterile body site infections at DMCSH, from June 2022 to August 2022.

TABLE 1
Socio-demographic characteristics of patients suspected of sterile body site infections at DMCSH, from June 2022 to August 2022.

TABLE 3
Antimicrobial susceptibility patterns of Gram-positive bacterial isolate from patients suspected of sterile body site infections at DMCSH, from June 2022 to August 2022.

TABLE 2
Antimicrobial susceptibility patterns of Gram-negative isolates from patients suspected of sterile body site infections at DMCSH, from June 2022 to August 2022.

TABLE 4
Bivariate and multivariate analysis of factors associated with patients suspected of sterile body site infections at DMCSH, from June 2022 to August 2022.