Background: Flow asynchrony, a patient–ventilator asynchrony (PVA) related to insufficient support or high inspiratory effort, has been poorly described in children. We aimed to assess whether the visual patterns of flow asynchrony differ across pediatric ICU ventilators and how they are influenced by support level and inspiratory effort. Methods: We conducted a bench study using an ASL5000 test lung programmed with 2 models generated from actual patients: restrictive model (9-year-old, compliance 18 mL/cm H 2 O, resistance 48 cm H 2 O/mL/s, P mus 7 cm H 2 O) and obstructive model (15-mo-old, compliance 8 mL/cm H 2 O, resistance 179 cm H 2 O/mL/s, P mus 28 cm H 2 O). Four ventilators (AVEA, NKV, C6, ServoU) were tested using pressure-support ventilation (PSV), with 6 conditions per model combining 3 pressure support levels and 2 effort levels. Pressure and flow waveforms were analyzed offline by a blinded investigator to classify each breath as synchronous, flow asynchrony, double cycling, or ineffective effort. Results: Flow asynchrony was present in all ventilators at baseline (pressure support 5 cm H 2 O for restrictive model and 6 cm H 2 O for obstructive model) (100% of breaths), with double cycling breaths in one ventilator (C6). Flow asynchrony incidence decreased with higher support or reduced effort but was replaced by ineffective efforts in 3 ventilators (ServoU, C6, NKV). AVEA showed the most rapid flow asynchrony reduction. Mean peak flow varied from 0.121 ± 0.001 L/s to 0.137 ± 0.002 L/s in restrictive model and from 0.091 ± 0.001 L/s to 0.102 ± 0.001 L/s in obstructive model. Visual flow asynchrony patterns and peak flow varied depending on ventilators. Conclusions: Flow asynchrony occurred during simulated pediatric PSV and was influenced by both effort and ventilator algorithms. Adjusting support may abolish flow asynchrony but can also trigger other asynchronies, underscoring the need to consider ventilator-specific performance in managing PVA.
Vedrenne‐Cloquet et al. (Wed,) studied this question.