Device-Patient Interface Air Leaks During Noninvasive Ventilation: Single-Center Pediatric Experience and Outcomes, 2018-2023.

Objectives

Air leaks are common during noninvasive ventilation (NIV) of children in the PICU. We aimed to describe our experience with NIV and study the association between the occurrence of air leaks and clinical parameters, and the median (95% CI) of ventilator-free days (VFDs).

Design

Single-center retrospective cohort, 2018-2023, identified in local health databases.

Setting

A quaternary children's center PICU in Montréal, Quebec, Canada. PATIENTS All pediatric patients admitted to the PICU and ventilated with a dual-limb NIV circuit (2018-2023). Only the first 6 hours of NIV per admission were analyzed; repeated admissions were included as separate encounters.

Interventions

None.

Measurements

AND MAIN

Results

Among 1460 encounters (1292 distinct patients), the median (interquartile range [IQR]) of air leak occurrence was 74% (IQR, 62-83%). Younger age was associated with a higher occurrence of air leaks (p < 0.001): 79% (IQR, 71-86%) in 681 infants 0-6 months old vs. 70% (IQR, 56-82%) in 328 older infants and toddlers 7-24 months old (n = 328, 22%) vs. 66% (IQR, 49-78%) in 451 preschool and school-aged children older than 24 months. A 10% increase in median air leak occurrence during the first 6 hours of NIV was associated with 0.6 fewer VFDs (slope estimate, -0.06 [95% CI, -0.10 to -0.01]). Lower air leaks were associated with higher peak airway pressures (slope estimate, -0.79 [95% CI, -1.44 to -0.15]), higher respiratory rate (slope estimate, -0.19 [95% CI, -0.35 to -0.03]), and greater weight (slope estimate, -0.40 [95% CI, -0.54 to -0.26]).

Conclusions

In our center, 2018-2023, high air leaks were frequent during NIV and associated with fewer VFDs. As a potentially modifiable factor, further prospective research is needed to assess the impact of air leaks on outcomes.