Neurally-Adjusted Ventilatory Assist Versus Pressure Support Ventilation During Noninvasive Ventilation

BACKGROUND: Noninvasive ventilation (NIV) is increasingly used during ventilatory support. Neurally-adjusted ventilatory assist (NAVA) is a mode of mechanical ventilation that can improve patient-ventilator interaction. We conducted a meta-analysis to compare patient-ventilator interaction and clinical outcomes between NAVA and pressure support ventilation (PSV) in adult subjects during NIV. METHODS: The PubMed, Cochrane Library, Web of Science, OpenGrey, and Embase databases were searched for appropriate clinical trials comparing NIV-NAVA with NIV-PSV for adult subjects. Comparisons of asynchrony index (AI), types of asynchrony, and clinical outcomes were pooled. RESULTS: Fifteen studies were included involving 615 subjects. AI was significantly lower in NAVA than PSV group (mean difference [MD] −14.70 [95% CI −23.20 to −6.19], P < .001). Subgroup analysis grouped by exacerbation of COPD or non-COPD showed that the AI of NAVA was lower than PSV in COPD exacerbation (MD −14.56 [95% CI −21.04 to −8.09], P < .001) and non-COPD (MD −3.02 [95% CI −4.44 to −1.61], P < .001). Severe asynchrony was significantly lower in NAVA than in PSV (odds ratio 0.06 [95% CI 0.03–0.11], P < .001). Inspiratory trigger delay in NAVA was significantly lower than PSV (MD −129.60 [95% CI −148.43 to −110.78], P < .001). NAVA had longer ICU length of stay than PSV (MD 1.22 [95% CI 0.44–2.00], P = .002). Level of discomfort was significantly higher in NAVA group than PSV group (MD 0.62 [95% CI 0.02–1.21], P = .040). CONCLUSIONS: NAVA has advantages in ventilator-patient interaction compared to PSV in NIV. Further research is needed in order to estimate effects on clinical outcomes.