Recruitability and effect of PEEP in SARS-Cov-2-associated acute respiratory distress syndrome

Background A large proportion of patients with a SARS-Cov-2-associated respiratory failure develop an acute respiratory distress syndrome (ARDS). It has been recently suggested that SARS-Cov-2-associated ARDS may differ from usual non-SARS-Cov-2-associated ARDS by higher respiratory system compliance (C-RS), lower potential for recruitment with positive end-expiratory pressure (PEEP) contrasting with severe shunt fraction. The purpose of the study was to systematically assess respiratory mechanics and recruitability in SARS-Cov-2-associated ARDS. Methods Gas exchanges, C-RS and hemodynamics were assessed at 2 levels of PEEP (15 cmH(2)O and 5 cmH(2)O) within 36 h (day1) and from 4 to 6 days (day 5) after intubation. The recruited volume was computed as the difference between the volume expired from PEEP 15 to 5 cmH(2)O and the volume predicted by compliance at PEEP 5 cmH(2)O (or above airway opening pressure). The recruitment-to-inflation (R/I) ratio (i.e. the ratio between the recruited lung compliance and C-RS at PEEP 5 cmH(2)O) was used to assess lung recruitability. A R/I ratio value higher than or equal to 0.5 was used to define highly recruitable patients. Results The R/I ratio was calculated in 25 of the 26 enrolled patients at day 1 and in 15 patients at day 5. At day 1, 16 (64%) were considered as highly recruitable (R/I ratio median [interquartile range] 0.7 [0.55-0.94]) and 9 (36%) were considered as poorly recruitable (R/I ratio 0.41 [0.31-0.48]). The PaO2/FiO(2) ratio at PEEP 15 cmH(2)O was higher compared to PEEP 5 cmH(2)O only in highly recruitable patients (173 [139-236] vs 135 [89-167] mmHg; p < 0.01). Neither PaO2/FiO(2) or C-RS measured at PEEP 15 cmH(2)O or at PEEP 5 cmH(2)O nor changes in PaO2/FiO(2) or C-RS in response to PEEP changes allowed to identify highly or poorly recruitable patients. Conclusion In this series of 25 patients with SARS-Cov-2 associated ARDS, 64% were considered as highly recruitable and only 36% as poorly recruitable based on the R/I ratio performed on the day of intubation. This observation suggests that a systematic R/I ratio assessment may help to guide initial PEEP titration to limit harmful effect of unnecessary high PEEP in the context of Covid-19 crisis.