Effects of Variable Pressure Support Ventilation on Regional Homogeneity and Aeration

Tommaso Mauri, Marta Lazzeri, Alfio Bronco, et al.

Am J Respir Crit Care Med 2017; 195: e27-e28

Figure 1.Physiologic effects of variable pressure support assessed by electrical impedance tomography. After application of variable pressure support, distribution of tidal ventilation within the patient’s chest became more homogeneous (A and B) and was associated with improved lung aeration (C). Most of the increase in end-expiratory lung volume was in region of interest (ROI) 2 (E), whereas in the other regions changes in lung aeration were not evident (D, F, and G). ∆EELV = change in end-expiratory lung volume; PSV = pressure support ventilation; vPSV = variable pressure support ventilation.

A 70-year-old woman was admitted to the intensive care unit for out-of-hospital cardiac arrest. After 34 days, the patient fully recovered from a cardiological and neurological point of view, but prolonged ventilation dramatically altered her lung mechanics. We decided to switch the patient to variable pressure support ventilation (vPSV), a relatively new assisted ventilation mode that exerted salutary effects on alveolar recruitment, ventilation/perfusion matching, and systemic oxygenation in different lung injury models (1). During vPSV, the support delivered by the ventilator in each breath [Psupp(i)] is automatically generated as Psupp(i) = Psupp ± percentage-of-pressure-variability (i) × Psupp, where the percentage-of-pressure-variability (i) is randomly selected from the set variability range and follows a Gaussian curve (2). To assess the physiologic effects of vPSV, we used electrical impedance tomography, a noninvasive, radiation-free bedside lung imaging method that allows continuous monitoring of the regional distribution of tidal ventilation and of changes in end expiratory lung volume (∆EELV) (3). Alternative bedside methods to assess changes in lung aeration might have been lung ultrasounds (4) or the nitrogen washin–washout technique (5), although, to our knowledge, no other bedside monitor can measure regional ventilation. We compared electrical impedance tomography data recorded during clinical pressure support ventilation (PSV) (Psupp = 10 cm H2O; positive end-expiratory pressure = 5 cm H2O) with those obtained after 20 minutes on vPSV (Psupp = 10 cm H2O; positive end-expiratory pressure = 5 cm H2O; set variability range, 40%). As a reference, during vPSV, the average peak inspiratory pressure (Ppeak) did not change (16 vs. 16 cm H2O), whereas Ppeak range was much wider (12–19 vs. 16–17 cm H2O) in comparison to conventional PSV.

Figures 1A and 1B show the distribution of tidal ventilation within the patient’s chest during PSV and vPSV: the amount of tidal volume distending each region of interest (ROI1–4) became more homogeneously distributed (end-inspiratory ventilation homogeneity index, 0.82 vs. 0.74) (6), especially by relative increase in ROI2. Moreover, during vPSV, ∆EELV increased (Figure 1C), mostly in ROI2 (Figures 1D–1G). Improved ventilation homogeneity and lung aeration during vPSV were also associated with increased oxygenation and CO2 clearance. Indeed, peripheral oxygen saturation increased (100 vs. 97%), despite decreased FiO2 (0.45 vs. 0.40), and the oxygen saturation as measured by pulse oximetry/FiO2 ratio increased to 250 from 216%. At the end of the vPSV phase, end-tidal CO2decreased to 41 from 43 mm Hg, even in the presence of slightly decreased minute ventilation (7.8 vs. 8.1 L/min). After the trial, the patient was switched back to conventional PSV as per clinical decision and eventually died 1 week later as a result of septic shock, while still on mechanical ventilation.

The present case generates the hypothesis that vPSV might improve regional lung mechanics, increasing aeration and ventilation homogeneity. However, the mechanisms of improved ventilation during vPSV need to be studied further as it is unclear whether improved physiology followed the variable nature of vPSV or simply the occasionally high Ppeak values that recruited atelectatic lung parenchyma. If the results of the present report are confirmed, vPSV might impact weaning of patients undergoing prolonged mechanical ventilation by, for example, decreasing ventilation pressures and/or preventing fatigue of respiratory muscles.