Staphylococcus aureus bloodstream infection at a referral children's hospital in Cape Town, South Africa, 2018-2022.

BACKGROUND: In the absence of a vaccine, Staphylococcus aureus (S. aureus) remains a major cause of morbidity and mortality in children. Studies describing S. aureus bloodstream infection (BSI) in children in Africa remain few. METHODS: This was a retrospective descriptive study conducted at Red Cross War Memorial Children's Hospital (RCWMCH) in Cape Town, South Africa. All S. aureus positive blood culture results for the period 2018-2022 on the National Health Laboratory Service microbiology database were identified. Case records of participants with S. aureus positive blood cultures were retrieved. Demographic, clinical, antimicrobial management and participant outcome data were extracted from the case records. Continuous variables were presented as means and standard deviations or medians and interquartile ranges as appropriate. Parametric (Student t-test) and non-parametric (Wilcoxon rank sum test for independent variables) methods were used to compare continuous variables. Pearson's Chi-square or Fisher's exact tests were used to compare categorical variables. RStudio statistical computing software version 4.4.2 was used for the logistic regression analysis. Our findings were also compared with a similar study at our institution which analysed the epidemiology of Staphylococcus bloodstream infections between 2007 and 2011 (Naidoo et al. PLoS ONE 8(10):e78396, 2013). RESULTS: During the 5-year study period there were 245 S. aureus BSIs identified, with an overall annual incidence risk of 2.7 cases per 1000 admissions. Methicillin-resistant S. aureus (MRSA) caused 10% of the BSIs and was predominantly healthcare-associated (79.2%). Only 2.1% of participants were living with human immunodeficiency virus (HIV). However, underweight was present in 29.2% of children with S. aureus BSI. Healthcare-associated infection comprised 33.3% of all S. aureus BSIs. 70% of BSIs had an identifiable site of infection with skin and soft tissue (25%) being the commonest. All MRSA isolates were susceptible to linezolid, teicoplanin, tigecycline and vancomycin. The mortality rate within 28 days of S. aureus BSI diagnosis was 8.3%. The commonest attributable cause of death was septic shock (25%). Requiring intensive care unit (ICU) admission during a BSI, (adjusted Odds ratio (aOR): 5.4, 95% CI: 1.89-15.21) was associated with increased 28-day mortality rate. When compared to the 2007-2011 study, there has been a decline in the incidence risk of S. aureus BSI and the prevalence of healthcare-associated MRSA BSI. CONCLUSIONS: S. aureus BSI remains a major cause of morbidity and mortality in children. Admission to ICU during a BSI episode was significantly associated with 28-day mortality. Furthermore, over time changes to the epidemiology of S. aureus BSI have occurred at RCWMCH. CLINICAL TRIAL NUMBER: Not applicable (study is not a clinical trial).