Oscillation Transmission of Modern High-Frequency Neonatal Ventilators Under Different Lung Mechanics Conditions

BACKGROUND: High-frequency oscillatory ventilation (HFOV) is widely used in neonatal critical care, and several modern ventilators using different technologies are available to provide HFOV. These devices have different technical characteristics that might interact with patient lung mechanics to influence the effectiveness of ventilation. To verify this, we studied the oscillation transmission of 5 neonatal oscillators in a lung model mimicking the mechanical patterns commonly observed in neonatal practice.

METHODS: This was a benchtop, in vitro, physiological, pragmatic study using a model mimicking airways and lung of a 1-kg preterm neonate and the following patterns: normal (compliance: 1.0 mL/cm H₂O, resistance: 50 cm H₂O/L/s), restrictive (compliance: 0.3 mL/cm H₂O, resistance: 50 cm H₂O/L/s), and mixed mechanics (compliance: 0.3 mL/cm H₂O, resistance: 250 cm H₂O/L/s). Several permutations of HFOV parameters (variable mean airway pressure or amplitude or frequency protocols) were tested. Oscillations were measured with a dedicated pressure transducer validated for use during HFOV, and oscillatory pressure ratio (OPR) was calculated to estimate the oscillation transmission.

RESULTS: Overall OPR (calculated on all experiments) was significantly different between ventilators and the mechanical patterns (both P < .001). Different ventilators and patterns accounted for 35.6% and 20.6% of the variation in oscillation transmission, respectively. Sub-analyses per changing amplitude or frequency protocols and multivariate regressions showed that VN500 (standardized β coefficient [St.β]: 0.548, P < .001) and Fabian HFO (St.β: 0.421, P < .001; adjusted R²: 0.615) provided the best oscillation transmission. Fabian HFO also delivered oscillations with the lowest variability when increasing amplitude.

CONCLUSIONS: In an experimental setting mimicking typical neonatal lung disorders, the oscillation transmission was more dependent on the ventilator model than on the mechanical lung conditions at equal HFOV parameters. Fabian HFO and VN500 provided better oscillation transmission overall, and when increasing amplitude, Fabian HFO delivered oscillations with the lowest variability.