Patient-Ventilator Synchrony in Neurally-Adjusted Ventilatory Assist and Variable Pressure Support Ventilation

Background: Neurally-adjusted ventilatory assist (NAVA) improves patient-ventilator synchrony and reduces the risk of respiratory over-assistance. Variable pressure support ventilation (PSV) is a recently introduced mode of assisted ventilation that has shown reduction in patient-ventilator asynchronies. We hypothesized that NAVA would reduce patient-ventilator asynchronies and inspiratory effort compared to variable PSV because breathing variability was intrinsically determined by the patient and not by the ventilator. This study aimed to evaluate patient-ventilator asynchronies and inspiratory effort pressure-time product (PTP) between NAVA and variable PSV in subjects with mild ARDS. Methods: After 24 h of controlled mechanical ventilation, subjects (PaO2 /FIO2 200-300 and PEEP level < 10 cm H2O) were randomized in sequence 1:1 by using a web-based encrypted platform and assigned to NAVA or variable PSV groups. Both modes of ventilation were consecutively kept for 24 h unless there were clinical changes. The primary aim of this study was to evaluate differences in asynchrony index (AI) between variable PSV and NAVA. Our secondary aims were to evaluate the coefficient of variation (CV) of breathing patterns and inspiratory effort between the groups. Results: Thirteen subjects were randomized in the NAVA group and 13 subjects in the variable PSV group. AI over time and minute PTP (PTPmin) were not different between NAVA and variable PSV groups (AI t0 P = .52, AI t12 P = .27, AI t24 P = .12; and PTPmin-t0 P = .60, PTPmin-t12 P = .57, PTPmin-t24 P = .85, respectively). CV for tidal volume (VT) and pressure support (PS) was lower in variable PSV group over time compared with NAVA group (P < .05). Conclusions: In this randomized controlled trial including subjects with mild ARDS, NAVA and variable PSV had comparable effects on patient-ventilator synchronies and PTP. However, variable PSV reduced the variability of VT and PS when compared with NAVA