318 - Viral Bronchiolitis: Association Between Etiology, Length of Stay and Disease Severity

Bronchiolitis is the most common reason for hospital admission in infants and toddlers. No therapies have been shown to prevent admission or decrease length of stay (LOS). Patients with severe symptoms are treated in a pediatric intensive care unit (PICU) and exposed to diverse therapies depending on disease severity and physician preference. Increasing use of respiratory viral PCR panels have aided in diagnosing the viral etiology of bronchiolitis.

Objective:

Describe the association between the hospital length of stay (H-LOS), PICU LOS (P-LOS), severity of illness, x-ray findings and treatments in relation to the viral etiology in patients with bronchiolitis in the PICU.

Design/Methods:

We performed a multicenter, retrospective chart review from 1/2020 to 3/2022 of patients aged 1 month to 3 years admitted to the PICU at Wolfson Children’s Hospital, Jacksonville and Shands Children’s Hospital, Gainesville with the clinical diagnosis of bronchiolitis. Demographics, respiratory viral PCR results, respiratory support, x-ray findings and treatments were retrieved. The pediatric logistic organ dysfunction-2 score (PELOD-2) was calculated. H-LOS and P-LOS were compared among viral groups (and among PELOD-2 scores, and for race and ethnicity) using Kruskal-Wallis. The association among viral groups with all nominal outcomes and therapies were compared using Fisher’s exact test. The association between PELOD-2 scores and respiratory support and treatments was analyzed using Kendall’s Tau-b test. The association between single/multiple virus and H-LOS/P-LOS, and the association between gender and H-LOS/P-LOS was analyzed using Mann Whitney U test.

Results:

117 patients were included and divided into five groups (see Table 1). A significant difference was found in H-LOS among the viral subtypes (p = 0.015) with group 2 having shorter H-LOS than groups 1 and 3. No statistical difference was found in P-LOS between viral subtypes (p = 0.054). Higher PELOD-2 scores were found to be associated with longer H-LOS (p = 0.003), maximum respiratory support (p = 0.000), pneumothorax (p = 0.009), paralysis (p = 0.000) and antibiotics, but not prone positioning, steroids, mucolytics, pneumonia, pleural effusion, or lobar atelectasis. Steroids were given less frequently to viral group 1 (p = .001). There was no difference in H-LOS/P-LOS for race/ethnicity/gender or single versus multiple viral infection.

Conclusion(s):

The presence of an RSV infection, isolated or as a co-infection, is found to be associated with a longer H-LOS and more severe disease with higher PELOD-2 scores requiring higher level of respiratory support.