Table 2.
Summary of findings of a systematic review of hydrocortisone and fludrocortisone for septic shock
| Outcomes | No. of studies | No. of patients | Effect estimates | Certainty in effect estimates † | Certainty assessment | ||
|---|---|---|---|---|---|---|---|
| HC + FC | Control | Relative effect | Absolute effect | ||||
| (95% CI) | (95% CI) | ||||||
| 28‐day mortality | 2 | 289/764 (37.8%) | 335/776 (43.2%) | RR 0.88 | 52 fewer per 1,000 | ⨁⨁⨁⨁ | |
| (0.78 to 0.99) | (4 fewer to 95 fewer) | High | |||||
| Long‐term mortality (90 days to 1 year) | 3 | 478/1,009 (47.4%) | 548/1,040 (52.7%) | RR 0.90 | 53 fewer per 1,000 | ⨁⨁⨁⨁ | |
| (0.83 to 0.98) | (11 fewer to 90 fewer) | High | |||||
| Shock reversal (at day 28) | 1 | 603/761 (79.2%) | 569/775 (73.4%) | RR 1.06 | 44 more per 1,000 | ⨁⨁⨁ | Imprecision |
| (1.01 to 1.12) | (7 more to 88 more) | Moderate | |||||
| Superinfection | 3 | 266/1,009 (26.4%) | 242/1,039 (23.3%) | RR 1.14 | 33 more per 1,000 | ⨁⨁ | Borderline inconsistency and imprecision |
| (0.85 to 1.51) | (35 fewer to 119 more) | Low | |||||
| GI bleeding | 2 | 50/764 (6.5%) | 53/775 (6.8%) | RR 0.96 | 3 fewer per 1,000 | ⨁⨁ | Serious imprecision |
| (0.66 to 1.39) | (23 fewer to 27 more) | Low | |||||
| Hyperglycemia | 3 | 547/614 (89.1%) | 520/626 (83.1%) | RR 1.07 | 58 more per 1,000 | ⨁⨁⨁⨁ | |
| (1.03 to 1.12) | (25 more to 100 more) | High | |||||
| Psychiatric disorders | 3 | 0/150 (0.0%) | 1/149 (0.7%) | RR 0.33 | 4 fewer per 1,000 | ⨁⨁ | Serious imprecision |
| (0.01 to 8.06) | (6 fewer to 47 more) | Low | |||||
CI, confidential interval; FC, fludrocortisone; GI, gastrointestinal; HC, hydrocortisone; RR, risk ratio.
†
Certainty in effect estimates was assessed with five domains: study limitations, inconsistency, indirectness, imprecision, and publication bias.