Caffeine Use for Bronchiolitis-Related Apnea: A Propensity-Matched Multicenter Database Analysis

Tuesday, February 25, 2025

8:55am – 9:55am Eastern Time

Location: Connections Central Theater 1

Gabriella D. Dyke, DO (she/her/hers)

Akron Childrens Hospital

Disclosure(s): No relevant financial relationship(s) to disclose.
CP

Christopher Page-Goertz, MD

Pediatric Intensivist
Akron Children's Hospital

Disclosure information not submitted.
Danielle Maholtz, DO, FAAP (she/her/hers)

Cincinnati Children's Hospital

Disclosure information not submitted.
CH

Claire Hanson, MD

Pediatric Intensivist
Akron Children's Hospital

Disclosure information not submitted.
RN

Ryan Nofziger, MD, MBA,

Pediatric Intensivist
Akron Childrens Hospital

Disclosure information not submitted.
CH

Christopher Horvat, MD, MHA

Childrens Hospital of Pittsburgh of UPMC

Disclosure information not submitted.
Jonathan Pelletier, MD, FAAP (he/him/his)

Akron Children's Hospital

Disclosure(s): No relevant financial relationship(s) to disclose.

Disclosure(s):

Gabriella D. Dyke, DO: No relevant financial relationship(s) to disclose.

Jonathan Pelletier, MD, FAAP: No relevant financial relationship(s) to disclose.

Introduction: Caffeine has been shown to reduce neonatal apnea in some settings. Prior studies of caffeine to treat bronchiolitis-related apnea (BRA) have been underpowered and shown conflicting results. We aimed to analyze the effect of caffeine on BRA in a multicenter database.

Methods: This was a multi-center retrospective review of the Pediatric Health Information Systems database. We included patients under two years admitted between 1/1/2010-12/31/2023 with ICD codes for bronchiolitis and apnea. Daily charges were used to identify patients receiving caffeine, non-invasive ventilation (NIV), and invasive mechanical ventilation (IMV). To determine the effect of caffeine use on IMV, encounters receiving caffeine before IMV were 4:1 nearest-neighbor propensity matched for caffeine prescription. The marginal risk ratio for IMV was calculated in the matched cohort. A nested sensitivity analysis excluded patients with neonatal or cardiovascular complex chronic conditions.

Results: There were 7,655 admissions among 7,562 patients across 48 hospitals for BRA during the study period. 9.9% (756/7,655) received caffeine, and 82.3% of those (622/756) received caffeine before IMV. Caffeine use decreased over time from 12.4% (80/647) in 2010 to 4.3% (17/396) in 2023 (p < 0.001). Before propensity matching, patients prescribed caffeine were younger, had a higher proportion of complex chronic conditions, were more likely to be admitted to an ICU, and more likely to receive NIV. Matching performance was excellent with an average pair distance of 0.0038. After matching on demographics, ICU admission, and NIV use, patients prescribed caffeine were less likely to undergo IMV (marginal risk ratio = 0.861 [95% CI 0.762-0.972], NNT = 21.3, 95% CI: 12.4-105.7). The sensitivity analysis excluding patients with neonatal or cardiovascular conditions included 68.9% (5,274/7,655) of the original cohort. Matching had similar results, but with wide confidence intervals (MRR = 0.824 [95% CI 0.681-0.997], NNT = 19.4, 95% CI: 10.7-1,137.7).

Conclusions: Caffeine use was associated with a reduction in IMV in patients with BRA in this retrospective database analysis. However, wide confidence intervals in the effect size estimate suggest that further work is needed.