Proportional modes of ventilation: technology to assist physiology

Abstract

Proportional modes of ventilation assist the patient by adapting to his/her effort, which contrasts with all other modes. The two proportional modes are referred to as neurally adjusted ventilatory assist (NAVA) and proportional assist ventilation with load-adjustable gain factors (PAV+): they deliver inspiratory assist in proportion to the patient's effort, and hence directly respond to changes in ventilatory needs. Due to their working principles, NAVA and PAV+ have the ability to provide self-adjusted lung and diaphragm-protective ventilation. As these proportional modes differ from 'classical' modes such as pressure support ventilation (PSV), setting the inspiratory assist level is often puzzling for clinicians at the bedside as it is not based on usual parameters such as tidal volumes and PaCO2 targets. This paper provides an in-depth overview of the working principles of NAVA and PAV+ and the physiological differences with PSV. Understanding these differences is fundamental for applying any assisted mode at the bedside. We review different methods for setting inspiratory assist during NAVA and PAV+ , and (future) indices for monitoring of patient effort. Last, differences with automated modes are mentioned.

Keywords: Inspiratory assist; Mechanical ventilation; Proportional modes; Respiratory effort.

Conflict of interest statement

AHJ reports personal fees from Liberate Medical, outside the submitted work. GC reports personal fees from Air Liquide Medical System, personal fees from Löwenstein, outside the submitted work. LH has received grants from Orion Pharma and Liberate Medical and speakers fee from Getinge. LB conducts an investigator-initiated trial on PAV+ (NCT02447692) funded by the Canadian Institute for Health Research and a partnership with Medtronic Covidien; his laboratory also receives grants and non-financial support from Fisher & Paykel, non-financial support from Air Liquide Medical System, non-financial support from Philips, non-financial support from Sentec, other from General Electric. Other authors have nothing to declare.

Figures

Fig. 1

Example of the working principle of proportional assist ventilation with load-adjustable gain factors (PAV+). Short inspiratory occlusions are automatically performed (indicated by * in the flow signal) for the calculation of respiratory system resistance and compliance. Arrows indicate that airway pressure (Paw) is delivered proportional to the patient’s effort (esophageal pressure (Pes))

Fig. 2

a Schematic illustration of the relationship between patient effort (respiratory muscle pressure, Pmus) and tidal volume (VT) in unassisted spontaneous breathing (dashed line), during pressure support ventilation (PSV) and for proportional modes such as proportional assist ventilation with load-adjustable gain factors (PAV+) and neurally adjusted ventilatory assist (NAVA). b Patient-ventilator interaction during PSV. Increasing the pressure support level increases VT (blue line) and ventilator inspiratory time (Ti, green line), while patient effort (Pmus, grey dotted line) is downregulated. In addition, neural Ti (dark blue line) remains unaltered with increasing levels of assist which results in late cycling. c Patient-ventilator interaction during NAVA and PAV+. Ventilator assist is delivered proportional to the patient’s demand over the full inspiratory cycle (neural Ti = ventilator Ti, note that the dashed green and dark blue lines overlap). Increasing the inspiratory assist level (NAVA level or PAV+ gain) downregulates Pmus (grey dotted line). Because the patient’s brain controls mainly the desired VT, changing the level of assist often has only minimal effects on the VT, as shown by the horizontal blue line on the Volume vs. level of assist curve

Fig. 3

Representative example of over-assistance during pressure support ventilation (PSV). The patient was ventilated with an inspiratory pressure set at 10 cmH2O above a positive end-expiratory pressure of 8 cmH2O. A double-balloon nasogastric catheter was placed for measurements of esophageal pressure (Pes) and gastric pressure. Transdiaphragmatic pressure (Pdi) was calculated as gastric pressure minus Pes. As can be seen in the Pes waveform, the patient only triggers the ventilator (small drop in Pes) and relaxes inspiratory muscles thereafter, as demonstrated by the increase in Pes during the remaining of the inspiratory cycle and the absence of increases in Pdi

Fig. 4

Example of the neurally adjusted ventilatory assist (NAVA) preview during pressure support ventilation (inspiratory assist of 10 cmH2O above a positive end-expiratory pressure of 8 cmH2O). The grey curve shows a “preview” of the estimated airway pressure (Paw) that would exist if the patient was ventilated in NAVA mode. The shape of this Paw curve resembles the diaphragm electrical activity (EAdi) curve (i.e., proportionality). The amount of assist depends on the EAdi amplitude and the selected NAVA level (0.8 cmH2O/µV for this example)

Fig. 5

a Example of the calculations of the neuroventilatory efficiency index (NVE) and the patient-ventilator breath contribution index (PVBC). An unassisted breath is obtained by reducing the neurally adjusted ventilatory assist level to zero for one breath. NVE is calculated as the ratio of the tidal volume to peak diaphragm electrical activity (EAdi). When dividing this NVE by the ratio of tidal volume and EAdi of the previous assisted breath, a PVBC index is obtained. b Example of the calculation of the neuromechanical efficiency index (NME) during an end-expiratory hold manoeuvre. During the occlusion (zero flow), the ratio of delta airway pressure (Paw) and EAdi represents the NME