Table 2.
Effects of corticosteroids on mortality in different etiologies of ARDS
| Etiology of ARDS a | Number of studies | Number of patients | Relative risk (95% CI) | I 2 |
|---|---|---|---|---|
|
Unselected ARDS |
|
|
|
|
| Randomized controlled trials |
3 |
370 |
0.88 (0.65 to 1.18) |
28% |
| Cohort studies |
4 |
238 |
1.12 (0.78 to 1.60) |
40% |
|
Influenza-related ARDS
b
|
|
|
|
|
| Cohort studies |
3 |
283 |
2.45 (1.40 to 4.27) ** |
0% |
|
Sepsis-related ARDS |
|
|
|
|
| Randomized controlled trials |
2 |
201 |
0.50 (0.12 to 2.02) |
76% |
| Cohort studies |
2 |
208 |
1.04 (0.58 to 1.85) |
0% |
|
Post-operative ARDS
b
|
|
|
|
|
| Cohort studies | 1 | 20 | 0.10 (0.01 to 0.63) * | - |
aExcludes the studies of preventive use of corticosteroids; bno randomized controlled trials available. *P-value <0.05. ** P-value <0.005. ARDS, acute respiratory distress syndrome.