G345(P) Identifying children at risk of serious bacterial infection in the emergency department: risk scores and gut feeling. (12th March 2018)

• Record Type: Journal Article
• Authors: Snowden, C; Fidler, K; Bevan, C; Rabe, H; Llewelyn, M
• Is Part Of: Archives of disease in childhood. Volume 103(2018)Supplement 1
• Page Start: A140
• Page End: A140
• Publication Date: 2018-03-12
• Subjects: Children -- Diseases -- Periodicals; Infants -- Diseases -- Periodicals; 618.920005
• DOI: 10.1136/archdischild-2018-rcpch.335
• Languages: English
• ISSNs: 0003-9888

Abstract

Abstract : Aims: To assess the utility of a validated clinical risk score (Brent et al. 2012) to stratify febrile children presenting to the emergency department, as high or low risk of serious bacterial infection (SBI). To further assess clinician's gut feeling in identifying those children with SBI. Methods: A prospective cohort study of children aged <17 years presenting to a children's emergency department, with an actual or reported temperature >37.5°C and requiring blood tests as part of their medical management, were recruited between February 2014–January 2017. For each participant the validated clinical risk score was completed using triage and clinician documentation. Clinician's 'Gut feeling' was recorded prior to results review and management plan formation. SBI was defined according to criteria described by the original validation study. Final diagnoses were assigned as definite/probable bacterial or viral infection or indeterminate. Results: 200 children were recruited, 9 subsequently excluded, leaving 191 (median age 2.1 years (IQR 0.9–4.2) for analysis. 164 (85%) participants had scores≤5 suggesting low risk of SBI. Of these 46/164 (29%) had SBI. 8 participants had high risk scores>8 suggestive of an SBI and of these only 4/8 (50%) had an SBI. Where clinician's gut feeling was recorded as 'No SBI', 49% (34/69) of children had a bacterial infection. Where clinician's gut feeling was recoded as 'not sure/possibly' for SBI, 21% (22/104) had bacterial infection. Analysis of blood markers revealed total WCC and neutrophil count were not useful discriminators of SBI (AUC WCC 0.6, neutrophils AUC 0.7.) CRP was significantly higher in the groups definite and probable bacterial infection [(AUC CRP 0.86 (p<0.0001)]. Conclusions: The validated clinical risk score for SBI did not effectively 'rule out' or 'rule in' serious bacterial infection in this cohort. Clinician's gut feeling was not helpful in identifying definite bacterial or probable bacterial infection. As anticipated CRP was a useful indicator of bacterial infection but has limitations as it is used as part of the catergorisation of serious bacterial infectiion.