Antimicrobial Resistance in Enteric Bacteria Isolated from Kenyan Children at Discharge from Hospital

Introduction: Antimicrobial resistance (AMR) is associated with millions of deaths per year worldwide. The global burden of AMR is disproportionately high in Sub-Saharan Africa, primarily as a result of the high incidence of infectious diseases. Children who have been hospitalized and subsequently discharged in this region are at high risk for morbidity and mortality. It is possible that AMR plays a role in these poor outcomes through treatment failure due to antibiotic resistant pathogens or through the acquisition of new resistant infections either during hospitalization or the post-discharge period. In addition, children carrying resistance determinants may serve as an important reservoir of AMR that can be transferred to other household and community members. Conditions favorable to the spread of bacteria, and therefore AMR, such as crowded living conditions, insufficient sanitation and hygiene facilities and contaminated water sources, may exacerbate community transmission from these children to others. Methods: This dissertation used bacterial isolates from fecal samples from a recently completed clinical trial examining whether azithromycin reduced death and rehospitalization during the discharge period in children aged 6 – 59 months in western Kenya. Fecal samples were collected at the time of enrollment, prior to the administration of the first dose of the study drug or placebo. Study staff performed medical record extractions, conducted interviews with caregivers, and study clinicians performed a medical examination. Fecal samples were also collected at follow-up visits 3-months and 6-months post-enrollment. Bacterial isolates, including Escherichia coli (E. coli), Salmonella, and Shigella were isolated and underwent antimicrobial susceptibility testing (AST). Additionally, a random subset of caregivers were enrolled and provided fecal samples for the isolation of E. coli and subsequent AST. The proportion of E. coli isolated from children at enrollment resistant to a panel of antibiotics and extended spectrum beta-lactamase (ESBL) production and risk factors for carriage of ESBL-producing E. coli was determined (Chapter 1). Patterns of AMR in E. coli isolated from children and their caregivers at the time of hospital discharge were examined and correlates of concordance assessed (Chapter 2). Agreement of susceptibility to antibiotics in Salmonella or Shigella with E. coli isolated from the same fecal sample were also examined (Chapter 3). Results: The proportion of E. coli isolated from children resistant to ampicillin (95%), gentamicin (44%), ceftriaxone (46%), and the presence of ESBL (44%) was high. Use of antibiotics during the hospitalization (adjusted prevalence ration [aPR] = 2.23; 95% CI: 1.29 – 3.83) and being hospitalized within the prior year (aPR = 1.32 [1.07 – 1.69]) were associated with the presence of ESBL producing E. coli. Additionally, being female (aPR = 1.42; 95% CI: 1.15 – 1.76), practicing open defecation (aPR = 2.02; 95% CI: 1.39 – 2.94), and having a toilet shared with other households (aPR = 1.49; 95% CI: 1.17 – 1.89) were also associated with carriage of ESBL E. coli. Across all antibiotics tested, caregivers had less AMR than children. Concordance of AMR was more common when resistance to a particular antibiotic was particularly high (cefoxitin, chloramphenicol, and cotrimoxazole) or low (imipenem). ESBL concordance was less common in child-caregiver pairs where there was more than one child in the household compared to those without additional children (55.0% vs. 79.9% [p = 0.02]) and with caregivers who reported being employed or a student than those who were not employed or a student (45.0% vs. 55.0%) [p = 0.05]). E. coli had high concordance for susceptibility and was a reliable predictor of the outcome to cephalosporins and gentamicin but was a poor predictor of susceptibility to azithromycin and ciprofloxacin. Conclusion: Children who have been discharged from hospital have a high burden of AMR, which may contribute to poor outcomes in the post-discharge period. Given the frequency of AMR in this high risk population, new guidelines for the treatment of common infectious syndromes in children with a recent hospitalization should be considered. Additionally, because these children are likely to carry AMR from the hospital back into the community, they may pass AMR to family and other community members. Interventions which reduce the transmission of infectious agents may also reduce the transmission of AMR.