TABLE 1.
Descriptive baseline and outcome data
| Variable | 12 mg (n = 497) | 6 mg (n = 485) |
|---|---|---|
| Baseline variables | ||
| Country of inclusion | ||
| Denmark | 251 (50.5%) | 234 (48.2%) |
| India | 182 (36.6%) | 187 (38.6%) |
| Sweden | 40 (8.0%) | 39 (8.0%) |
| Switzerland | 24 (4.8%) | 25 (5.2%) |
| Age (years) | 65 (56–74) [22–88] | 64 (54–72) [22–90] |
| Weight (kg) | 80 (68–96) [45–198] | 80 (68–95) [42–164] |
| Type of respiratory support | ||
| Open system | 272 (54.7%) | 258 (53.2%) |
| NIV/CPAP | 118 (23.7%) | 128 (26.4%) |
| Invasive mechanical ventilation | 107 (21.5%) | 99 (20.4%) |
| Respiratory failure category (presented as numerical values) a | 2 (1–4) [1–6] | 2 (1–4) [1–6] |
| Respiratory failure category a | ||
| 1: Open system, low flow b | 141 (28.7%) | 126 (26.4%) |
| 2: Open system, high flow b | 131 (26.6%) | 132 (27.7%) |
| 3: NIV/CPAP, low FiO2 c | 69 (14.0%) | 73 (15.3%) |
| 4: NIV/CPAP, high FiO2 c | 45 (9.1%) | 47 (9.9%) |
| 5: Invasive mechanical ventilation, low FiO2 c | 64 (13.0%) | 51 (10.7%) |
| 6: Invasive mechanical ventilation, high FiO2 c | 42 (8.5%) | 48 (10.1%) |
| Oxygen flow rate (L/min, in patients on open system only) | 22 (15–40) [10–61] | 24 (15–40) [10–70] |
| FiO2 (%, in patients on closed system only) a | 60 (50–75) [25–100] | 60 (45–80) [25–100] |
| Number of comorbidities | 0 (0–1) [0–3] | 1 (0–1) [0–4] |
| Number of comorbidities (categorical) | ||
| 0 | 270 (54.3%) | 240 (49.5%) |
| 1 | 164 (33.0%) | 174 (35.9%) |
| 2 | 54 (10.9%) | 57 (11.8%) |
| 3 | 9 (1.8%) | 13 (2.7%) |
| 4 | 0 (0.0%) | 1 (0.2%) |
| Diabetes mellitus | 135 (27.2%) | 163 (33.6%) |
| Ischemic heart disease or heart failure | 67 (13.5%) | 69 (14.2%) |
| Chronic obstructive pulmonary disease | 57 (11.5%) | 56 (11.5%) |
| Immunosuppression within 3 months prior to randomisation | 40 (8.0%) | 43 (8.9%) |
| Lactate (mmol/L) d | 1.6 (1.1–2.3) [0.3–16.7] | 1.7 (1.2–2.3) [0.2–13.8] |
| Use of vasopressors or inotropes | 81 (16.3%) | 68 (14.0%) |
| Limitations in care (cardio‐pulmonary resuscitation or life‐support) | 30 (6.0%) | 25 (5.2%) |
| Predictions | ||
| Predicted risk of mortality at day 90 e (%) | 35.5 (22.2–52.9) [2.6–99.7] | 34.6 (22.6–49.4) [3.1–98.3] |
| Outcome variables | ||
| Days alive without life support at day 90 f | 84 (9–90) [0–90] | 80 (6–90) [0–90] |
| Mortality at day 90 g | 157 (32.0%) | 180 (37.7%) |
Note: Descriptive baseline, outcome and predicted data according to the internal prediction model for all variables assessed. Some baseline and the outcome data have been presented previously elsewhere. 4 Numeric data are presented as medians (interquartile ranges) [full ranges], whereas binary/categorical data are presented as numbers (%).
Abbreviations: CPAP, continuous positive airway pressure; NIV, non‐invasive ventilation.
A total of 13 patients on closed systems (1.3% of the full intention‐to‐treat population; 5 patients in the 12‐mg group and 8 patients in the 6‐mg group) could not be classified due to missing FiO2 values.
Low flow includes patients with oxygen flow values ≤ the median value in all patients on open systems (22 L/min), whereas high flow includes patients with values > this value.
Low FiO2 includes patients with values ≤ the median value in all patients on closed systems (60%), whereas high FiO2 includes patients with values > this value.
Lactate values were missing in 106 patients (10.8% of the intention‐to‐treat population; 57 patients in the 12‐mg group and 49 patients in the 6‐mg group).
Calculated using the stacked multiply imputed datasets; the mean predicted risk was 38.5% in the 12‐mg group and 37.6% in the 6‐mg group.
Days alive without life support at Day 90 were missing in 15 patients (1.5% of the intention‐to‐treat population; 8 patients in the 12 mg group and 7 patients in the 6 mg group).
Mortality at Day 90 values were missing in 14 patients (1.4% of the intention‐to‐treat population; 7 patients in the 12‐mg group and 7 patients in the 6‐mg group).