Under-ve children’s Bacterial Blood Stream Infection in Southern Ethiopia

Background Under-ve children’sblood stream infectionsmainly due to resistant pathogens found to be a higher risk of hospital mortality in low and middle income countries. This study intended to assess the bacterial prole and drug resistance of isolates of blood stream infection among under- 5 children in Southern Ethiopia. carried out microaerophilic by using Bacterial identication done primary on colony characteristics and Gram-stain reaction followed by proper biochemical tests. susceptibility determined by Kirby- Bauer method and to


Abstract Background
Under-ve children'sblood stream infectionsmainly due to resistant pathogens found to be a higher risk of hospital mortality in low and middle income countries. This study intended to assess the bacterial pro le and drug resistance of isolates of blood stream infection among under-5 children in Southern Ethiopia.

Methods
Retrospectively lab recordeddata's were retrieved to assess the bacterial isolates and rates antibiotic resistance at Hawassa University Comprehensive Specialized Hospital.

Results
Of 323bloods culture performed116 (35.9%) were positive for potential bacterial BSI. Positivity of blood culture decrease with age, hence neonatesblood stream infectionstream infection werehigher than other age groups (OR, 3; 95% CI 1.5-5.1; p = 0.001). The major gram positive isolateCoNSand S.aureushave shown highest level of resistance to penicillin 61.3.0% and tetracycline 78.8% whereas less level of resistance was reported to Meropenem 6.4%, Ceftriaxone 13% and doxycycline 13%. Although most of gram negatives isolates were resistant for tested antibiotics, K.pneumoniae has shown leaser resistance tocefotaxime and chloramphenicol.

Conclusion
A high level of blood stream infection accompanied by high drug resistant isolates was reported. The great burden in neonate than other children was reported. As a result to reduce the burden the hospital administrators need tofacilities effective infection prevention and control programmes, effective hygiene practices and well-functioning environment along with educating patient about proper hygiene practices.

Background
Blood stream infection (BSI) is the most common causes of morbidity and mortality in children across the world [1]. The burden of sepsis is most likely highest in low-and middle-income countries. It is estimated about 3 million newborn and 1.2 million children suffer from sepsis globally every year. Unlessdiagnosedon time and treated promptly, it can lead to septic shock, multiple organ failure and death. Antimicrobial resistance is a major factor determining clinical unresponsiveness to treatment and rapid evolution to sepsis and septic shock. Sepsis patients with resistant pathogens have been found to have a higher risk of hospital mortality [2].
Although, BSI can be preventable with timely attention, coherent antimicrobial therapy and supportive care, so far higher death rates reported in developing countries [3].The rate of BSI estimated to be 3-20 times higher in developing countries due to lack ofadequate and regular microbiology laboratory in these country [4].
Fevermay bethe only manifestation of childhood invasive infection. Nevertheless in Ethiopia with the added burden of malaria, HIV infection, tuberculosis and malnutrition, the treatment of fever is often syndromic, targeting multiple possible aetiologies, without necessarily considering BSI as a potential cause [5]. That is why in developing countrylike Ethiopia, where malaria is endemic, most febrile children misleadingly managed as malarial infection and only very little proportion of them empirically managed as bacterial infection. As timely treatment of bacteremia and appropriate use of antibiotics for children is crucial most recommended replacement of presumptive treatment of malaria by test before treat as strategy for management of other febrile illness [6].
In southern Ethiopia, there is no reported dataon blood stream bacterial infection among under-5 children. As a result this study provides information for local hospitals in the regions for empirical choice of antibiotics in the treatment of child bacteremia.

Methods
A retrospective study was conducted in Hawassa University comprehensive and specialized hospital (HUCSH). HUCSH is a teaching hospital providing both inpatient and outpatient services to more than 5 million people in the region. The microbiology lab is the only regional referral laboratory where patients from the region as well from parts of the region referred for bacteriological. All bacteremia-suspected patients were sent to microbiology lab for blood culture diagnosis of bacteremia and sensitivity tests.
We retrieved all microbiological reports on bacterial pathogens from January rst 2019 to March 30, 2020. This study was approved by IRB of Hawassa University College of Medicine and health sciences. Administrative permission to access data was requested through the hospital laboratory manager of Hawassa University College of Medicine and health sciences. All data obtained in the course of the study were reserved con dential and used only for this study. We have no inclusion criteria except we reject unsuitably recorded data's like illegible or incomplete in a sense of either age or nal culture results.

Bacteriological investigation
Our laboratory bacteriological analysis follows all standard operation procedures based on the Clinical and Laboratory Standards Institute's (CLSI).
The collected clinical samples were submitted to the laboratory and processed following standard procedures. Hence according to CLSI, a 1-2 mL blood sample was collected aseptically and added totryptic soya broth in a ratio of blood to broth 1:9 and incubated at 35-37°C. Visual inspection for possible microbial growth; like hemolysis, turbidity, gas production and pellicle formation was assessed until5 th days to report no bacterial growth.Turbid broth cultures were sub cultured into MacConkey agar, Blood agar plates and Chocolate agar plates (Oxoid™ Ltd, Thermo Fisher Scienti c, and Waltham, MA, USA) and incubated at 37°C for 24-48 hours. The Chocolate agar-incubated cultures were carried out in a microaerophilic atmosphere by using a candle jar. Bacterial identi cation was done primary based on colony characteristics and Gram-stain reaction followed by proper biochemical tests. Antimicrobial susceptibility pro le of isolates was determined by Kirby-Bauer disc diffusion method and the results were interpreted according to CLSI guidelines [7]. The following antibiotics were used; Pencillin (10 µg The data was entered and analysed using SPSS version 20(IBM Corporation, Armonk, NY, USA). Descriptive statistics like frequency and percentages of categorical variables was calculated then compared using Chi-square test. The odds ratio (OR) with 95% con dence interval (95% CI) were calculated to estimate the relationship of outcome with independent variables. The signi cance level was p<0.05.

Result
We processed a total of 323 blood culturesfrom child who were suspected for blood stream infection during the study period. The ratio of male were 172 (53.3%) and 151 (46.7%) females. From all processed clinical samples, 116/323 (35.9%) were culture positive for invasive bacterial infection. Among recorded 116 bacterial infections, 62 (53.4%) were male and 54 (46.6%) female.Dominant number of patients, 142 (44%) as well the highest positivity rate, 52.6% was recorded among the age group of 0-28 day neonates, p = 0.003 (Table 1).  7%) were the least isolate found (Fig. 1).From the isolated, Gram-positive bacteria were more prevalent than Gram-negative bacteria.

Antibiotic Resistance Pattern Of Gram Negatives
Yet again the most frequent isolate, K.pneumoniae recorded the high resistance almost for all common antibiotics mainly to ampicillin, Ceftazidime and Cipro oxacinhowever,least resistance documented to Chloramphenicol and Cefotaxime. In contrastP. aeruginosaandE.colicon rmed highest resistance to Chloramphenicol although still they showed high resistance for the rest of other antibiotics. Ceftriaxone seems the least resisted drugfor E. coli and Cefotaxime and Ceftriaxone appears least resisted by other isolates of gram negatives (Table 4).

Discussion
Bacterial blood stream infection of children is one of the most serious problems worldwide [2]. The lack of proper microbiological analysis in developing country like Ethiopia has challenged the process of reducing the burden.
While it was lower compared to the study done in Gondar 46.6% [17]among neonates suspected to sepsis.Possible reason for variation might be the variation in clinical indication for blood culture and patient age group. In our set up blood culture demanded after explicate diagnosis of other doubts due to the fact that blood culture is costly and take long time. Moreover poor effective hygiene practices and well-functioning environmentalso poor patient hygienic practices might increase the rate of infection.
In Ethiopia, bacterial sepsis was reported asthe leading cause of neonatal death [18].One of the systematic review study in Ethiopia has reported that a pooled prevalence of neonatal sepsis in Ethiopia was 49.98% [19]. Likewise in this study higher rate of blood stream infectionrecorded with the odds of neonates being more positive in blood stream infection than other age group of under 5 children (OR, 3; 95% CI 1.5-5.1; p = 0.001) ( Table 2). Most other study also reported neonates BSIhigher compared to [16,19].
Regarding isolate, the current studyhas foundthat almost 43% of the isolates were CoNS and S. aureusfollowed byKlebsiella 21.4% which is similarwith other studies in Ethiopia, in Arsi [20] and Gondar [16] also with most of other studies in Africa.When we look report from African study, BSI study on neonatal in Tanzania reported that S. aureus 36.5% and Klebsiella 29.7% dominate all isolates recognized [21]there is also agreementwith report from S. Africa wereS. aureus, Klebsiella and E. coliare predominant isolate BSI among children [5]. Type of isolates in this study is still similar to that reported in other African studies for instant report from S. Africa shows predominate isolates S. aureus, Klebsiella [14]amongblood cultured samples of children. Again In Zimbabwe, CoNS and S.aureus [22] were dominant.This might show sort of nosocomial infection due to procedural insertion of different medical device to child. At the mean time health workers pooreffective hygiene practices and well-functioning environmentthat can source birth canal contamination of Klebsiella to neonatal [21].likewise to this study a systematic review study done in resource limited countries had also reported that Gram-negativeorganisms,Klebsiella pneumoniae [23] was dominant isolate from child blood culture.In the other hand study from Ghanareported non-typhoidal salmonellae was dominant [24] As usual gram positive isolates shows highest level of resistance to penicillin 61.3.0% and tetracycline 78.8% which is similar with report in Jimma [8] and Gondar [16] also comparable with report of systematic review which covers most of the regions found in Ethiopia The overall BSI rates of gram negative isolates were 17% and almost all resisted Ampicillin (80%), Cipro oxacin (61.2%) and Ceftazidime (65.5%). As displayed in 4, most of the isolates are resistant for the tested antibiotics 36% -80% of resistance. The predominate isolate, K.pneumoniae have shown leaser resistance tocefotaxime and chloramphenicol while P. aeruginosa and E.coli resist more than 50%. Though comparison of our nding with others report is di cult due to the fact that antibiotic resistance vary place to place in addition to lack of consistency in the measurement and reporting of susceptibility data makes it di cult to compare ndings among different countries and laboratories, sometimes even within one country [28]. But most of the study revealed high resistance of gram negatives. In fact in resource limited countries large regional variations in resistance rates to locally prescribed antibiotics is common [16,23].Due to concerning other gram negative isolates we are unable to discuss on rate of resistance due to leaser number of observation.

Conclusion
In general the nding in this study indicates high level of BSI among under-ve children. A high drug resistance to locally used antibiotics also surge the challenge. Therefore we recommend for local health professionals to see choicesof best drug fromreported compiled date for their routine empirical treatment of blood stream bacterial infections. Furthermore the hospital administrators should facilities mainly on having functioning infection prevention and control programmes, effective hygiene practices and precautions, along with a clean, well-functioning environment and equipment also educating patient about effective hygiene practices.

Limitations
First as it is retrospective study from registered in laboratory log book we didn't got different documented independent variables for possible causal effect association. Hence we did not investigate risk factor for infection as well for drug resistance. Second, due to lack of appropriate facility anaerobicwere not included.Third we have excluded vancomycin resistance test due to lack ofuniformity. This study was approved by IRB of Hawassa University College of Medicine and health sciences. Administrative permission to access data was requested through the hospital laboratory manager of Hawassa University College of Medicine and health sciences. We have got waiver from hospital laboratory and all data obtained in the course of the study were reserved con dential and used only for this study.  Figure 1 The frequentisolates from under ve children with blood stream infection