(874) Which Objective Measurements Are Most Associated With HFNC Failure in Bronchiolitis?

Introduction: Introduction
HFNC therapy is the most commonly used mode of respiratory support for infants admitted to PICUs with bronchiolitis. PALICC-2 recently proposed criteria to diagnose possible PARDS or at risk for PARDS (ARF-PARDS) for patients on HFNC based on flow rate and S02/Fi02 ratio. We investigated whether these criteria were associated with the need for escalation from HFNC to NIV or EI (endotracheal intubation) in patients with bronchiolitis.

Methods: Methods
This is a retrospective cohort study including patients that were ≤ 1 year of age with bronchiolitis who were admitted to the PICU and treated with HFNC for > 24 hours during 2022. The primary outcome measure was escalation from HFNC to either NIV or EI. A database query was performed to identify the patients and then the charts were reviewed, and clinical data was extracted. Statistical analysis including unadjusted, adjusted and ROC analysis was performed. P values < 0.05 were considered significant.

Results: Results
A total of 60 patients were admitted to the PICU and treated with HFNC and 17 (28%) were escalated to NIV or ETT. All 17 were first tried on NIV and 5 of these patients were subsequently intubated. Patients that were escalated were younger (2 vs 4 months), had higher flow rates (1.84 vs 1.14 L/k/min) and lower S02/Fi02 ratios (211 vs 271) during the first 12 hours of admission. More of the escalated patients met criteria for ARF-PARDS (64.7 vs 14%) and their PICU LOS was longer ( 6.5 vs 2.8 days) [all P < 0.05]. There was no difference in the worst clinical respiratory score (CRS) in the first 12 hours, PRISM 3, number with a PMH diagnosis or diagnosis of RSV bronchiolitis. When adjusted for age, ARF-PARDs diagnosis had an OR 5.87 (95% CI 1.43, 24.18, p=0.014) for escalation to NIV/EI. ROC analysis for HFNC flow rate maximum had AUC of 0.759 and S02/Fi02 ratio had AUC 0.729.

Conclusions: Conclusion
Younger patients with bronchiolitis were more likely to be escalated from HFNC to NIV or EI in our PICU. When controlling for age, maximum flow rate and lowest S02/Fi02 rate during the first 12 hours were associated with escalation of therapy, while the CRS was not. Monitoring patients by utilizing PALICC-2 criteria for ARF-PARDS may be helpful in identifying those likely to fail HFNC.