Impact of different frequencies of controlled breath and pressure-support levels during biphasic positive airway pressure ventilation on the lung and diaphragm in experimental mild acute respiratory distress syndrome

BACKGROUND: We hypothesized that a decrease in frequency of controlled breaths during biphasic positive airway pressure (BIVENT), associated with an increase in spontaneous breaths, whether pressure support (PSV)-assisted or not, would mitigate lung and diaphragm damage in mild experimental acute respiratory distress syndrome (ARDS). MATERIALS AND METHODS: Wistar rats received Escherichia coli lipopolysaccharide intratracheally. After 24 hours, animals were randomly assigned to: 1) BIVENT-100+PSV(0%): airway pressure (P(high)) adjusted to V(T) = 6 mL/kg and frequency of controlled breaths (f) = 100 bpm; 2) BIVENT-50+PSV(0%): P(high) adjusted to V(T) = 6 mL/kg and f = 50 bpm; 3) BIVENT-50+PSV(50%) (PSV set to half the P(high) reference value, i.e., PSV(50%)); or 4) BIVENT-50+PSV(100%) (PSV equal to P(high) reference value, i.e., PSV(100%)). Positive end-expiratory pressure (P(low)) was equal to 5 cmH(2)O. Nonventilated animals were used for lung and diaphragm histology and molecular biology analysis. RESULTS: BIVENT-50+PSV(0%), compared to BIVENT-100+PSV(0%), reduced the diffuse alveolar damage (DAD) score, the expression of amphiregulin (marker of alveolar stretch) and muscle atrophy F-box (marker of diaphragm atrophy). In BIVENT-50 groups, the increase in PSV (BIVENT-50+PSV(50%) versus BIVENT-50+PSV(100%)) yielded better lung mechanics and less alveolar collapse, interstitial edema, cumulative DAD score, as well as gene expressions associated with lung inflammation, epithelial and endothelial cell damage in lung tissue, and muscle ring finger protein 1 (marker of muscle proteolysis) in diaphragm. Transpulmonary peak pressure (Ppeak,L) and pressure–time product per minute (PTP(min)) at P(high) were associated with lung damage, while increased spontaneous breathing at P(low) did not promote lung injury. CONCLUSION: In the ARDS model used herein, during BIVENT, the level of PSV and the phase of the respiratory cycle in which the inspiratory effort occurs affected lung and diaphragm damage. Partitioning of inspiratory effort and transpulmonary pressure in spontaneous breaths at P(low) and P(high) is required to minimize VILI.