125 - Investigating the prevalence of acute chest syndrome in patients with sickle cell disease presenting to a pediatric emergency department with fever
Fellow Physician Children's Hospital of Wisconsin Wauwatosa, Wisconsin, United States
Background: The National Heart, Lung, and Blood Institute (NHLBI) defines acute chest syndrome (ACS) as a new infiltrate on chest x-ray (CXR) and at least 1 of the following: fever (>=38.50 C), hypoxia, or respiratory symptoms. NHLBI expert consensus recommends a CXR in patients with sickle cell disease (SCD) who have fever and respiratory symptoms. However, evidence suggests that physicians fail to recognize ACS in children with SCD who present with fever, leading to broader CXR criteria and variation in CXR rates across sites. Providers in our emergency department (ED) record fever and the presence of each NHLBI-defined respiratory symptom at any visit for a child with SCD; a CXR is obtained unless a clear cause for fever exists even if there are no respiratory symptoms. Objective: Determine the prevalence of ACS in patients with SCD presenting to our ED with fever with or without respiratory symptoms. We hypothesized that the prevalence of ACS in children with SCD with fever and no respiratory symptoms is less than 5%. Design/Methods: A retrospective analysis of ED visits by patients with SCD (ages 0 to 20 years) who present with fever (>/= 38.5o C) within the past 24 hours between 1/2019 to 12/2022 was done. The presence or absence of NHLBI-defined respiratory symptoms was extracted as were CXR results from the index visit and any visit within 72 hours before or after. The primary outcome was ACS, defined as a new infiltrate on CXR. If no CXR was obtained, the chart was reviewed for any return visit within 1 week for a diagnosis of ACS. We determined the 95% confidence interval (CI) for ACS in children with no respiratory symptoms and used Fisher’s exact test to compare the ACS rates between children with and without respiratory symptoms. Results: There were 137 children (288 unique ED visits) with fever. Of the 288 visits, 38.5% had no respiratory symptoms. Overall, 13.9% of visits had a diagnosis of ACS. Of the 288 visits, 91% had a CXR, of which 14.8% had ACS. The prevalence of ACS in those with no respiratory symptoms was 0/111 (0%, 95% CI 0-3.3%), significantly less than 5% (supportive of our hypothesis) and lower than the prevalence of ACS in those with at least 1 respiratory symptom: 36/173 (20.8%, 95% CI 15.0-27.6%) (p < 0.001).
Conclusion(s): Our findings provide strong evidence that a CXR is not necessary in a child with SCD presenting to the ED with fever and no NHLBI-defined respiratory symptoms. This evidence could improve quality of care and decrease unnecessary CXR use in the SCD population.