Aims Following the introduction of conjugate childhood vaccines and declining trends in meningococcal disease over the past decade, community-acquired serious invasive bacterial infections in children have become less common but hospital admissions continue to rise. The objective of this study was to estimate the incidence of serious culture-confirmed invasive bacterial infections in children within a geographically-defined population over a three-year period.
Methods Blood and cerebrospinal fluid culture (CSF) results for all children aged 1 month to 16 years admitted to five South-West London (SWL) and Surrey hospitals during 2009–11 were obtained at regular intervals from each hospital microbiology department and a web-based questionnaire completed for all children with a positive culture. Hospital-acquired infection was defined as a significant pathogen isolated from a blood/CSF culture taken >48 hours after hospital admission.
Results During 2009–11, 44,118 children had 46,039 admissions to one of the five Hospitals, equivalent to 25.8 admissions per 1,000 children resident in SWL and Surrey. Blood/CSF cultures were obtained during 44.7% (n = 20,578) of admissions and 7.4% (n = 1,530) had a positive culture. Of these, only 571 were defined as a clinically significant serious bacterial infection (SBI), equivalent to 37.3% of positive blood/CSF cultures, 2.8% of all blood/CSF cultures taken and 1.2% of all hospital admissions. The population incidence of SBI in SWL and Surrey was, therefore, 31.9/100,000, with the highest incidence among 1–11 month-olds (172/100,000), followed by 1–4 year-olds (34.3/100,000) and 5–15 year-olds (16.4/100,000). A third of the SBI (208/571, 36.4%), however, were hospital-acquired and, of the community-acquired infections, more than two-thirds (252/363, 69.4%) occurred in children with co-morbidities. The incidence of community-acquired SBI in previously healthy children in SWL and Surrey was, therefore, only 6.2/100,000, with the highest incidence among 1–11 month-olds (33.9/100,000), followed by 1–4 year-olds (6.5/100,000) and 5–15 year-olds (3.2/100,000).
Conclusions Although a SBI was suspected in almost half of all childhood hospital admissions in SWL and Surrey, a significant pathogen was identified in only 3%, mainly among children with pre-existing co-morbidities. Improved targeting of children at very low risk of a SBI at presentation will facilitate increased management of unw.