ARDS severity | p-value | |||
Mild | Moderate | Severe | ||
Subjects n | 5 | 12 | 3 | |
Age years | 64±19 | 57±15 | 76±3 | 0.17 |
Female sex | 0 | 4 (33) | 0 | 0.30 |
Body mass index kg·m−2 | 25±5 | 29±7 | 25±2 | 0.56 |
Time after ARDS diagnosis and length of mechanical ventilation before the study days | 5±2 | 6±2 | 5±3 | 0.61 |
[18F]FDG uptake rate mL blood·(mL lung)−1·min−1 | 42±13 | 50±27 | 116±52 | 0.07 |
Extent of lung inhomogeneities % lung volume | 15±1 | 22±6# | 24±4 | 0.02 |
Lung density g·mL−1 | 0.51±0.09 | 0.58±0.13 | 0.73±0.20 | 0.23 |
Total tissue g | 1402±186 | 1352±230 | 2001±93¶ | 0.02 |
Not inflated tissue % | 46±6 | 39±14 | 50±22 | 0.55 |
Poorly inflated tissue % | 23±3 | 35±7# | 37±10 | 0.01 |
Well-inflated tissue % | 32±6 | 25±13 | 12±15 | 0.15 |
Over-inflated tissue % | 0±0 | 0±1 | 0±0 | 0.17 |
Lung recruitability % lung tissue | 9±7 | 12±6 | 19±12 | 0.37 |
PaO2/FiO2 | 247±39 | 155±31# | 79±10¶ | <0.001 |
Shunt fraction % of cardiac output | 22±8 | 37±7# | 65±8¶ | <0.01 |
Minute ventilation L·min−1 | 8±1 | 9±2 | 10±2 | 0.27 |
Respiratory rate breaths·min−1 | 14±2 | 15±2 | 19±3 | 0.05 |
PaCO2 mmHg | 39±4 | 43±6 | 62±11 | 0.02 |
Physiological dead space % | 48±8 | 52±9 | 82±2¶ | 0.02 |
Respiratory system elastance cmH2O·L−1 | 21±3 | 27±9 | 39±21 | 0.22 |
Cause of ARDS n (%) | 0.72 | |||
Pneumonia | 2 (40) | 7 (58) | 2 (66) | |
Viral (H1N1) | 1 (20) | 3 (25) | 0 | |
Bacterial | 1 (20) | 4 (33) | 2 (66) | |
Sepsis | 2 (40) | 4 (33) | 0 | |
Trauma | 1 (20) | 1 (8) | 0 | |
Vasculitis | 0 | 0 | 1 (33) |
Data are presented as mean±sd or n (%), unless otherwise stated. ARDS: acute respiratory distress syndrome; [18F]FDG: [18F]2-fluoro-2-deoxy-d-glucose; PaO2: arterial oxygen tension; FiO2: inspiratory oxygen fraction; PaCO2: arterial carbon dioxide tension. Physiological data were collected at positive end-expiratory pressure 5 cmH2O. Physiological dead space was available in 19 patients (11 moderate ARDS). Data were compared with the Kruskal–Wallis test and, if the overall test was significant, multiple comparisons were performed with the Wilcoxon test and the p-value was corrected with Bonferroni's method. Categorical data were compared with Fisher's exact test. #: p<0.05 versus mild ARDS; ¶: p<0.05 versus moderate ARDS. For more details see online supplementary table S2.