Reclassifying acute respiratory distress syndrome

Rationale: The ratio of PaO2to FIO2(P/F) defines acute respiratory distress syndrome (ARDS) severity and suggests appropriate therapies. Objectives: We investigated 1) whether a 150-mm-Hg P/F threshold within the range of moderate ARDS (100-200 mm Hg) would define two subgroups that were more homogeneous; and 2) which criteria led the clinicians to apply extracorporeal membrane oxygenation (ECMO) in severe ARDS. Methods: At the 150-mm-Hg P/F threshold, moderate patients were split into mild-moderate (n = 50) and moderate-severe (n = 55) groups. Patients with severe ARDS (FIO2not available in three patients) were split into higher (n = 63) and lower (n = 18) FIO2groups at an 80% FIO2threshold. Measurements and Main Results: Compared with mild- moderate ARDS, patients with moderate-severe ARDS had higher peak pressures, PaCO2, and pH. They also had heavier lungs, greater inhomogeneity, more noninflated tissue, and greater lung recruitability. Within 84 patients with severe ARDS (P/F,100 mm Hg), 75% belonged to the higher FIO2subgroup. They differed from the patients with severe ARDS with lower FIO2only in PaCO2and lung weight. Forty-one of 46 patients treated with ECMO belonged to the higher FIO2group. Within this group, the patients receiving ECMO had higher PaCO2than the 22 non-ECMO patients. The inhomogeneity ratio, total lung weight, and noninflated tissue were also significantly higher. Conclusions: Using the 150-mm-Hg P/F threshold gave a more homogeneous distribution of patients with ARDS across the severity subgroups and identified two populations that differed in their anatomical and physiological characteristics. The patients treated with ECMO belonged to the severe ARDS group, and almost 90% of them belonged to the higher FIO2subgroup.