333. Meningitis in Well-Appearing Febrile Infants Aged 1–90 Days

BACKGROUND: Fever in infants 1–90 days of age is common. Bacterial meningitis (BM) is a rare, potentially fatal infection that may occur in well-appearing febrile infants (FI). Our objectives were to identify infants with BM in a large population of well-appearing FI and evaluate factors associated with the diagnosis of BM in this population. METHODS: The Intermountain Healthcare System (IHS) is comprised of 22 hospitals across Utah and Idaho and includes Primary Children’s Hospital, the only pediatric hospital in a catchment of 400,000 miles(2). IHS has a care process model for the well-appearing FI. We queried the IHS EHR from July 1, 2004 to September 30, 2016 and captured data on age, laboratory testing, and outcomes. Diagnosis of BM required positive CSF culture. RESULTS: We identified 21,135 FI episodes; 54 infants (0.26%) had a diagnosis of BM. Gram-negative organisms predominated in FI 1–28 days [15/24 (63%)] and caused 28/54 (52%) cases overall (Figure 1). FI 1–28 days were significantly more likely to have BM than those 29–90 days (0.41% vs. 0.20%; RR 2.11, 95% CI 1.24–3.61). Laboratory screening showed abnormal white blood cell count in 63% of FI 1–28 days with BM and 50% of FI 29–90 days (P = 0.42); bands were abnormal in 33% and 47% respectively (P = 0.41); urinalysis was abnormal in 21% and 11% (P = 0.42). CSF profile was performed and interpretable in 48/54 (89%); CSF pleocytosis was present in 30/48 [(63%; 15/21 (71%) 1–28 days and 15/27 (56%) P = 0.34]. Nine of 54 (17%) FI with BM would not have been considered “high-risk” based on laboratory criteria alone. Of FI with BM, only 31/54 (57%) had bacteremia with the same organism [17/24 (71%) in those 1–28 days; and 14/30 (47%) in those 29–90 days; P = 0.099]. CONCLUSION: BM is rare and challenging to predict in well-appearing FI. Abnormal screening laboratory values identified 83% of FI with BM. Awaiting blood culture results before performing lumbar puncture would potentially miss 40%. Age was the only predictor for BM risk in our cohort. DISCLOSURES: A. J. Blaschke, BIoFire Diagnostics, LLC: I have intellectual property licensed to BioFire through the University of Utah, Independent Contractor and Investigator, Consulting fee and Licensing agreement or royalty. C. L. Byington, BIoFire Diagnostics, LLC: I have intellectual property licensed to BioFire through the University of Utah, Licensing agreement or royalty.