Positive end-expiratory pressure (PEEP) benefits in acute respiratory distress syndrome (ARDS) are driven by lung dynamic strain reduction: this depends on the variable extent of alveolar recruitment. The recruitment-to-inflation (R/I) ratio estimates recruitability across a 10-cmH2O PEEP range through a simplified maneuver. Whether recruitability is uniform or not across this range is unknown. The hypotheses of this study are that R/I represents an accurate estimate of PEEP-induced changes in dynamic strain, but may show a non-uniform behaviour across the conventionally tested PEEP range (15-5 cmH2O).
Twenty patients with moderate-to-severe COVID-19 ARDS underwent a decremental PEEP trial (PEEP 15-13-10-8-5 cmH2O). Respiratory mechanics and end-expiratory lung volume by nitrogen dilution were measured the end of each step. Gas exchange, recruited volume, R/I, changes in dynamic, static and total strain were computed between 15 and 5 cmH2O (global R/I) and within narrower PEEP ranges (granular R/I).
Between 15 and 5 cmH2O, median [Interquartile range] global R/I was 1.27 [0.40-1.69] and displayed a linear correlation with PEEP-induced dynamic strain reduction (r=-0.94, p<0.001). Intra-individual R/I variability within the narrower ranges was high (85% [70-109]). The relationship between granular R/I and PEEP was mathematically described by a non-linear, quadratic equation (r 2=0.96). Granular R/I across the narrower PEEP ranges had itself a linear correlation with PEEP-induced reduction in dynamic strain (r=-0.89, p<0.001).
Both global and granular R/I well estimate PEEP-induced changes in lung dynamic strain. However, the effect of 10-cmH2O of PEEP on lung strain may be non-uniform. Granular R/I assessment within narrower PEEP ranges guided by end-expiratory lung volume measurement may aid more precise PEEP selection, especially when R/I obtained with the simplified maneuver between PEEP 15 and 5 cmH2O yields intermediate values, that are difficult to interpret for a proper choice between a high vs. low PEEP strategy.