Infection characteristics and treatment of Staphylococcus aureus bacteraemia at a tertiary children’s hospital

Alasdair P. S. Munro, Christopher C. Blyth, Anita J. Campbell, Asha C. Bowen

Staphylococcus aureus bacteraemia (SAB) causes considerable morbidity and mortality in children. Despite this, its epidemiology and risk factors are poorly understood, with minimal paediatric clinical trial data available to guide clinicians in management. We conducted a pilot study to characterise SAB and validate a severity classification for use in future clinical trials. Patients with SAB were prospectively identified at Princess Margaret Hospital for Children (Perth, Western Australia) from May 2011 to December 2013. Retrospective data were collected from clinical and laboratory records. Cases were classified based on a priori defined criteria as simple (single or contiguous, peripheral site focus) or complex (multi-site, deep tissue, no focus or sepsis) and tested against risk factors and markers of severity of infection. There were 49 cases of SAB (median age 7.7 years), with classification as simple (n = 30, 61%) and complex (n = 19, 39%) respectively. There were no deaths or relapses in our cohort. Only 10% of isolates were methicillin resistant S. aureus (MRSA), and none of these were healthcare-associated. Age, gender, Indigenous status, MRSA and healthcare-associated infections were not predictive of complex infection. Pre-existing malignancy was a risk factor for complex infection (p = 0.02). Complex infections were associated with a higher median maximum C reactive protein (216 mg/L vs 50 mg/L, p = < 0.001), longer median length of stay (42 vs 10 days, p = < 0.001) and longer duration of antibiotic therapy (43 vs 34 days, p = 0.03). This is the first attempt to categorise paediatric SAB as simple versus complex, to guide clinicians in decision making. There is a wide spectrum of disease severity in paediatric SAB, with maximum CRP, length of stay, and duration of therapy greater in those with complex disease. Distinct cohorts with simple and complex courses which may be a target for future clinical trials have been described.