No CrossRef data available.
Published online by Cambridge University Press: 02 May 2019
Introduction: Fever is a common presenting complaint in the emergency department (ED). Febrile infants are at particularly high risk of serious bacterial infection including bacterial meningitis. Unfortunately, recommendations as to when to perform a lumbar puncture in febrile infants older than 21 days remain conflicting. Our study seeks to establish the prevalence of bacterial meningitis in infants 22 to 60 days old and to evaluate the performance of our local fever without a source (FWS) workup protocol at identifying bacterial meningitis. Methods: This analysis represents the results of a retrospective cohort study which took place in an academic pediatric ED in Quebec City. Infants 22 to 60 days old investigated for FWS, were included in the study. Premature infants ( <37 weeks), as well as infants with chronic diseases, immunodeficiency, previous antimicrobial therapy, in-dwelling catheters, or septic shock were excluded. We evaluated the performance of our local FWS workup protocol which includes the Yale Scale, a complete blood count, blood culture, C-reactive protein, urinalysis and urine culture. The protocol recommends a lumbar puncture in all febrile infants <1 month old, and in all infants <3 months old with either leukocytes <5.0 or >15.0 X 10^9cells/L, petechia, or a Yale between 11 and 16. Results: We reviewed 1261 charts from 2012 to 2017, of which 920 met our inclusion criteria. In our cohort, 171 infants were 22 to 30 days old, 369 were 31 to 45 days old, and 380 were 46 to 60 days old. The proportion of infants with cerebrospinal fluid analysis in these 3 groups was 76% (n = 130), 25% (n = 98) and 12% (n = 46) respectively. In the entire cohort, two infants were diagnosed with bacterial meningitis resulting in a prevalence of 0.2% (95%CI: 0-0.5%); viral meningitis had a prevalence of 4.7% (95%CI: 3.3-6.1%). Sensitivity and specificity of the protocol were 100% and 52.8%; positive and negative predictive values were 0.4% and 100%, respectively. All charts were reviewed for 2 weeks following the index visit to screen for missed cases of bacterial meningitis. Conclusion: Systematically performing a lumbar puncture for workup of fever without a source in infants 22 to 60 days old appears unwarranted given the low prevalence of bacterial meningitis in this population. Our FWS workup protocol correctly identified the 2 cases of bacterial meningitis in our cohort. This is an ongoing study and more cases will be recruited to better evaluate the safety and performance of our protocol.