To document the prevalence of asynchrony events during noninvasive ventilation in pressure support in infants and in children and to compare the results with neurally adjusted ventilatory assist.
Prospective randomized cross-over study in children undergoing noninvasive ventilation.
The study was performed in a PICU.
From 4 weeks to 5 years.
Two consecutive ventilation periods (pressure support and neurally adjusted ventilatory assist) were applied in random order. During pressure support (PS), three levels of expiratory trigger (ETS) setting were compared: initial ETS (PSinit), and ETS value decreased and increased by 15%. Of the three sessions, the period allowing for the lowest number of asynchrony events was defined as PSbest. Neurally adjusted ventilator assist level was adjusted to match the maximum airway pressure during PSinit. Positive end-expiratory pressure was the same during pressure support and neurally adjusted ventilator assist. Asynchrony events, trigger delay, and cycling-off delay were quantified for each period.
Six infants and children were studied. Trigger delay was lower with neurally adjusted ventilator assist versus PSinit and PSbest (61 ms [56–79] vs 149 ms [134–180] and 146 ms [101–162]; p = 0.001 and 0.02, respectively). Inspiratory time in excess showed a trend to be shorter during pressure support versus neurally adjusted ventilator assist. Main asynchrony events during PSinit were autotriggering (4.8/min [1.7–12]), ineffective efforts (9.9/min [1.7–18]), and premature cycling (6.3/min [3.2–18.7]). Premature cycling (3.4/min [1.1–7.7]) was less frequent during PSbest versus PSinit (p = 0.059). The asynchrony index was significantly lower during PSbest versus PSinit (40% [28–65] vs 65.5% [42–76], p < 0.001). With neurally adjusted ventilator assist, all types of asynchronies except double triggering were reduced. The asynchrony index was lower with neurally adjusted ventilator assist (2.3% [0.7–5] vs PSinit and PSbest, p < 0.05 for both comparisons).
Asynchrony events are frequent during noninvasive ventilation with pressure support in infants and in children despite adjusting the cycling-off criterion. Compared with pressure support, neurally adjusted ventilator assist allows improving patient–ventilator synchrony by reducing trigger delay and the number of asynchrony events. Further studies should determine the clinical impact of these findings.
1Adult Cardio-respiratory Physiotherapy, La Tour Hospital, Geneva, Switzerland.
2Medicine UFR, University of Picardie Jules Verne, Peritox EA 4284–UMI01 INERIS, Amiens, France.
3Neonatal and Pediatric Intensive Care Unit, University Hospital, Geneva, Switzerland.
4Intensive Care and Burns Unit, University Hospital, Lausanne, Switzerland.
5Pediatric Intensive Care Unit, University Hospital North, Amiens, France.
This work was performed at Neonatology and Pediatric Intensive Care Unit, University Hospital, Geneva.
Part of this study has been previously reported as an abstract (ATS meeting, May 2011, Denver, CO, A3191)
At the time of the study, Laurence Vignaux was working in the lab run by Philippe Jolliet. The remaining authors have disclosed that they do not have any potential conflicts of interest.
For information regarding this article, E-mail: firstname.lastname@example.org