Table 4.
Further analysis on mortality
| Studies number | Patient number | Event in NIV group | Event in control group | Risk ratio (95% CI) | I 2 | P | |
|---|---|---|---|---|---|---|---|
| Short-term mortality | 4 | 506 | 82 of 257 (31.9%) | 110 of 249 (44.2%) | 0.73 (0.58, 0.91) | 64% | 0.04 |
| ICU mortality | 3 | 466 | 54 of 237 (22.8%) | 73 of 229 (31.9%) | 0.72 (0.53, 0.97) | 33% | 0.03 |
| Hospital mortality | 4 | 506 | 82 of 257 (31.9%) | 110 of 249 (44.2%) | 0.73 (0.58, 0.91) | 64% | 0.04 |
| Patients with PaO2/FiO2 < 200 | 3 | 541 | 92 of 228 (40.4%) | 114 of 227 (50.2%) | 0.77 (0.61, 0.98) | 13% | 0.03 |
| Patients with PaO2/FiO2 > 200 | 2 | 126 | 29 of 62 (46.8%) | 45 of 64 (70.3%) | 0.46 (0.09, 2.34) | 88% | 0.35 |
| Patients with diagnosis of cancer and transplantation | 5 | 507 | 85 of 258 (32.9%) | 113 of 249 (45.4%) | 0.68 (0.48, 0.97) | 58% | 0.03 |
| Patients with diagnosis of drug-related immunosuppression | 2 | 81 | 7 of 33 (21.2%) | 5 of 30 (16.7%) | 0.95 (0.48, 1.87) | 0% | 0.37 |
ICU intensive care unit, NIV noninvasive ventilation, PaO 2 /FiO 2 ratio of arterial pressure of oxygen/fraction of inspired oxygen