Neurally Adjusted Ventilatory Assist vs Pressure Support Ventilation for Noninvasive Ventilation During Acute Respiratory Failure: A Crossover Physiologic Study

Abstract

Background Patient-ventilator asynchrony is common during noninvasive ventilation (NIV) with pressure support ventilation (PSV). We examined the effect of neurally adjusted ventilatory assist (NAVA) delivered through a facemask on synchronization in patients with acute respiratory failure (ARF). Methods This was a prospective, physiologic, crossover study of 13 patients with ARF (median PaO2/FIO2, 196 [interquartile range (IQR), 142-225]) given two 30-min trials of NIV with PSV and NAVA in random order. Diaphragm electrical activity (EAdi), neural inspiratory time (TIn), trigger delay (Td), asynchrony index (AI), arterial blood gas levels, and patient discomfort were recorded. Results There were significantly fewer asynchrony events during NAVA than during PSV (10 [IQR, 5-14] events vs 17 [IQR, 8-24] events, P = .017), and the occurrence of severe asynchrony (AI > 10%) was also less under NAVA (P = .027). Ineffective efforts and delayed cycling were significantly less with NAVA (P < .05 for both). NAVA was also associated with reduced Td (0 [IQR, 0-30] milliseconds vs 90 [IQR, 30-130] milliseconds, P < .001) and inspiratory time in excess (10 [IQR, 0-28] milliseconds vs 125 [IQR, 20-312] milliseconds, P < .001), but TIn was similar under PSV and NAVA. The EAdi signal to its maximal value was higher during NAVA than during PSV (P = .017). There were no significant differences in arterial blood gases or patient discomfort under PSV and NAVA. Conclusion In view of specific experimental conditions, our comparison of PSV and NAVA indicated that NAVA significantly reduced severe patient-ventilator asynchrony and resulted in similar improvements in gas exchange during NIV for ARF. Trial registry ClinicalTrials.gov; No.: NCT01426178; URL: www.clinicaltrials.gov. Patient-ventilator asynchrony is common during noninvasive ventilation (NIV) with pressure support ventilation (PSV). We examined the effect of neurally adjusted ventilatory assist (NAVA) delivered through a facemask on synchronization in patients with acute respiratory failure (ARF). This was a prospective, physiologic, crossover study of 13 patients with ARF (median PaO2/FIO2, 196 [interquartile range (IQR), 142-225]) given two 30-min trials of NIV with PSV and NAVA in random order. Diaphragm electrical activity (EAdi), neural inspiratory time (TIn), trigger delay (Td), asynchrony index (AI), arterial blood gas levels, and patient discomfort were recorded. There were significantly fewer asynchrony events during NAVA than during PSV (10 [IQR, 5-14] events vs 17 [IQR, 8-24] events, P = .017), and the occurrence of severe asynchrony (AI > 10%) was also less under NAVA (P = .027). Ineffective efforts and delayed cycling were significantly less with NAVA (P < .05 for both). NAVA was also associated with reduced Td (0 [IQR, 0-30] milliseconds vs 90 [IQR, 30-130] milliseconds, P < .001) and inspiratory time in excess (10 [IQR, 0-28] milliseconds vs 125 [IQR, 20-312] milliseconds, P < .001), but TIn was similar under PSV and NAVA. The EAdi signal to its maximal value was higher during NAVA than during PSV (P = .017). There were no significant differences in arterial blood gases or patient discomfort under PSV and NAVA. In view of specific experimental conditions, our comparison of PSV and NAVA indicated that NAVA significantly reduced severe patient-ventilator asynchrony and resulted in similar improvements in gas exchange during NIV for ARF.