The most important feature when using NIV-NAVA pertains to the Edi catheter (a routinely used naso- or orogastric feeding tube with miniaturized sensors embedded to record Edi), available in sizes suitable for adults and children (Fig. 15.2). The Edi catheter is connected to a SERVO-i or SERVO-U or SERVO-n ventilator (Maquet Critical Care, AB, Solna, Sweden). The catheter is well tolerated and easy to place, as described in the literature. During NIV-NAVA, any interface can be used (e.g., face mask, helmet, nasal prongs) with maintained synchrony, because the control of the ventilator is not affected by leaks.

To initiate a ventilator breath, the Edi signal triggers inspiration once a threshold change in Edi has been exceeded. The pressure delivered after triggering increases during inspiration in proportion to the Edi, until neural exhalation begins, and the ventilator cycles off (70 % of the peak Edi). The NIV-NAVA level determines the proportionality between the Edi and the ventilator pressure. After cycling off, the assist returns to a user-defined positive end-expiratory pressure (PEEP) level. In this fashion, the patient is in control of their own ventilator rate and level of assist, which can vary on a breath-by-breath basis. Figure 15.3 demonstrates Edi and ventilator pressure tracings for an infant patient breathing on NIV-NAVA, and demonstrates the synchrony between the Edi (patient) and airway pressure (ventilator), both in terms of timing and proportionality. As in any another mode of mechanical ventilation, upper pressure limits can be set. Backup ventilation is provided in the case of apnea or accidental catheter removal.

The neural drive to breathe during NIV-NAVA is controlled by multiple afferent inputs originating in the lungs (stretch receptors), the central and peripheral chemoreceptors (CO₂ and pH sensitive), the respiratory muscles (muscle tension receptors), and the upper airway muscles (chemical and pressure receptors). Depending on clinical practice, a further influence on respiratory drive is sedation and analgesia. These multiple afferent inputs are continuously being “centralized” and “integrated” in the respiratory centers of the brainstem, with the resultant package of respiratory-related information being sent out via the phrenic nerves to electrically activate the diaphragm (i.e., the Edi) and other respiratory muscles (Fig. 15.1). Therefore, the physiological responses driving the patient’s diaphragm are also simultaneously driving the ventilator throughout each breath during NIV-NAVA.

A total of 21 papers appear on PUBMED with the topic of NIV-NAVA since its release in 2008. Since the update in Minerva in 2013 [3], five clinical NIV-NAVA studies and one experimental study have been published [4–9]. Three of these articles were accompanied by editorials [10–12].

In 12 adult patients with COPD, Doorduin et al. [4] performed automated patient-ventilator interaction analysis during NIV-pressure support ventilation (PSV) (delivered with two separate ventilators, Maquet’s SERVO-i and Respironics’ Vision) and NIV-NAVA. Synchrony was superior (the Neurosync index was low, 5 %) during NIV-NAVA compared with NIV-PSV (24 % Vision, 21 % SERVO-i). The improvement in synchrony was mainly due to reduced triggering and cycling-off delays. The authors also found that there was a progressive number of wasted efforts as the triggering and cycling-off delays got worse.

In pediatrics, three studies have all shown feasibility and tolerance of NIV-NAVA, as well as insertion of the Edi catheter [5–8].