Spontaneous breathing with airway pressure release ventilation favors ventilation in dependent lung regions and counters cyclic alveolar collapse in oleic-acid-induced lung injury: a randomized controlled computed tomography trial

Abstract

Introduction: Experimental and clinical studies have shown a reduction in intrapulmonary shunt with spontaneous breathing during airway pressure release ventilation (APRV) in acute lung injury. This reduction was related to reduced atelectasis and increased aeration. We hypothesized that spontaneous breathing will result in better ventilation and aeration of dependent lung areas and in less cyclic collapse during the tidal breath.

Methods: In this randomized controlled experimental trial, 22 pigs with oleic-acid-induced lung injury were randomly assigned to receive APRV with or without spontaneous breathing at comparable airway pressures. Four hours after randomization, dynamic computed tomography scans of the lung were obtained in an apical slice and in a juxtadiaphragmatic transverse slice. Analyses of regional attenuation were performed separately in nondependent and dependent halves of the lungs on end-expiratory scans and end-inspiratory scans. Tidal changes were assessed as differences between inspiration and expiration of the mechanical breaths.

Results: Whereas no differences were observed in the apical slices, spontaneous breathing resulted in improved tidal ventilation of dependent lung regions (P < 0.05) and less cyclic collapse (P < 0.05) in the juxtadiaphragmatic slices. In addition, with spontaneous breathing, the end-expiratory aeration increased and nonaerated tissue decreased in dependent lung regions close to the diaphragm (P < 0.05 for the interaction ventilator mode and lung region).

Conclusion: Spontaneous breathing during APRV redistributes ventilation and aeration to dependent, usually well-perfused, lung regions close to the diaphragm, and may thereby contribute to improved arterial oxygenation. Spontaneous breathing also counters cyclic collapse, which is a risk factor for ventilation-associated lung injury.

Figure 1

Densitometries of dynamic lung computed tomography (CT). Center, example of an end-expiratory transverse CT scan of a slice close to the diaphragm. External boundaries of the lungs inside the ribs and the internal boundaries along the mediastinal organs were marked. The marked lung tissue was divided into a nondependent and a dependent region of interest bisecting a ventral to dorsal axis that represents the highest chest diameter in parallel to a reference line from sternum to spine. Upper and lower parts, densitometric analyses of all dependent and nondependent lung regions close to the diaphragm computed from dynamic CT at end-expiration and end-inspiration. Histograms show the normalized lung volume for Hounsfield units ranging from -1,000 to 100, plotted as means of all animals. Aeration categories (hyperinflated, normally aerated, poorly aerated, nonaerated; see text for details) are marked. Gray areas indicate increase during inspiration, black areas show decrease during inspiration.

Figure 2

Regional distribution of tidal gas volume. Tidal changes in gas distribution of diaphragmatic slices between a nondependent region of interest (ROI) and a dependent ROI during airway pressure release ventilation (APRV) with or without spontaneous breathing (+SB/-SB). Data presented as the percentage (mean ± standard error of the mean) of the total gas volume of voxels for each ROI. ^##P < 0.01, interaction of factor ROI with factor ventilatory mode, suggesting that the ventilatory mode has a significant influence on the regional distribution of tidal volume. *P < 0.05 between ventilatory groups (post hoc results given only if significant).

Figure 3

Regional distribution of tidal cyclic collapse. Tidal changes in nonaerated tissue (tidal recruitment) in diaphragmatic slices between a nondependent region of interest (ROI) and a dependent ROI during airway pressure release ventilation (APRV) with or without spontaneous breathing (+SB/-SB). Data presented as the percentage (mean ± standard error of the mean) of the total lung volume (gas and tissue) of voxels for each ROI. ^§§§P < 0.001, dependent ROI versus nondependent ROI. ^¶P < 0.05, APRV +SB versus APRV -SB. *P < 0.05 between ventilatory groups (post hoc results given only if significant).

Figure 4

Regional distribution of end-expiratory gas volume. End-expiratory distribution of gas in diaphragmatic slices between a nondependent region of interest (ROI) and a dependent ROI during airway pressure release ventilation (APRV) with or without spontaneous breathing (+SB/-SB). Data presented as the percentage (mean ± standard error of the mean) of the total gas volume of voxels for each ROI. n.s., not significant. ^#P < 0.05, interaction of factor ROI with factor ventilatory mode, suggesting that the ventilatory mode has a significant influence on the regional distribution of end-expiratory gas volume and nonaerated lung tissue.

Figure 5

Regional distribution of nonaerated lung tissue. End-expiratory distribution of nonaerated tissue in diaphragmatic slices between a nondependent region of interest (ROI) and a dependent ROI during airway pressure release ventilation (APRV) with or without spontaneous breathing (+SB/-SB). Data presented as the percentage (mean ± standard error of the mean) of the total lung volume (gas and tissue) of voxels for each ROI. n.s., not significant. ^#P < 0.05, interaction of factor ROI with factor ventilatory mode, suggesting that the ventilatory mode has a significant influence on the regional distribution of end-expiratory gas volume and nonaerated lung tissue. *P < 0.05 between ventilatory groups (post hoc results given only if significant).