Table 4.
Trials using different corticosteroids (CS); design (Rs = retrospective and obs = observational and As = ambispective in yellow and MA = meta-analysis and SRMA = Systematic Review and Meta-analysis in orange), timing, duration, outcomes (focusing on mortality data), and viral shedding.
| Trial Name/Authors (Reference) | Steroid Used, Dosing | Trial Design, Population | Initiation Timing | Duration of Administration (Days) | Main Outcomes | Secondary Outcome(s) | Viral Clearance |
|---|---|---|---|---|---|---|---|
| Wu C et al. [14] | Max dose: 80 mg MP equiv (IQR = 40–80) mg | SC/Rs/Obs | NA | 7 (4–12) | D-60 in-hospital-mortality: Significant ↓ in CS (p = 0.0160) | NA | CS not associated with delayed viral clearance |
| Liu J et al. [18] | Not mentioned | Rs/Obs CS vs. no CS | CS D28 mortality 44.3% vs. no CS 31% (p < 0.001) Mortality in CS group linked to HD and early initiation (<3 d from hospitalization) |
Delayed (SHR 1.59 CI 95%, 1.17–2.15) p = 0.003 | |||
| Li Y et al. [36] | Median 200 mg/d HC equiv (100–320.9) | Rs/MC critically ill pts CS vs. no CS | 9 (5–14) | No significant difference in mortality | CS prolonged viral shedding | ||
| Lu X et al. [19] | Different CS, 100–800 mg/day HC equiv |
SC/Rs, critically ill pts, CS vs. no CS | Median 8 (4–12) | CS independent from overall mortality. HD CS:↑ mortality risk (p = 0.003) | NA | NA | |
| Li Q et al. [47] | 5 pts on pos prednisone (MP equiv dose 20 mg/d) and 50 pts on iv MP 20–40 mg/d | Rs nonsevere pts early LD CS vs. no CS | 1–5 d after hospital admission | 3 (Oral prednisone) 3–5 (iv MP) | CS in nonsevere COVID-19 pneumonia: ↑risk of progression to severe disease, ↑ use of antibiotics, ↑ duration of fever, ↑LOHS | CS in nonsevere COVID-19 pneumonia: prolongs virus clearance time | |
| Ma Y et al. [48] | MP equiv 56.6 mg median daily dose | Rs/MC, severe and nonsevere COVID-19 pts | 5 median duration | CS: ↑ LOS in nonsevere pts (p < 0.05) BUT no significant ↑ LOS of severe pts | CS:↑ viral shedding duration only in nonsevere pts (p < 0.05) | ||
| Li S et al. [49] | Not mentioned | Not mentioned | Not mentioned | Assessment of risk factors for long-term (>30 d) positive SARS-CoV-2 and viral shedding | NA | HD (80 mg/day; aHR, 0.67 (95%CI, 0.46–0.96) p = 0.031) but not LD CS (40 mg/day; aHR,0.72 (95%CI, 0.48–1.08), p = 0.11) potentially delayed viral shedding | |
| Shi D et al. [50] | Rs/SC | Factors associated with prolonged viral shedding | CS:not independent factor of prolonged viral shedding | ||||
| Hu Y et al. [37] | 0.75–1 mg/kg/d MP equiv | Rs CS vs. no CS | Median time since Sx onset: 7 | 6 (IQR, 4–8) | CS vs. no CS: no significant difference in: imaging progression, 90-day mortality. Significant difference in lymphocytes and CRP | CS vs. no CS: no significant difference in time to negative PCR | |
| Keller MJ et al. [33] | CS vs. no CS | Obs incl children | Within 48 h of admission | CS: no effect on mortality, BUT if CRP > 20; significant ↓ risk of mortality or MV [OR 0.23, 95% CI, 0.08–0.7]. If CRP < 10; significant ↑ risk of mortality or MV [OR 2.64, 95% CI, 1.39–5.03] | |||
| Wu J et al. [40] | MP equiv dose 40 mg/d | Rs severe and critically ill pts | 24 h after Dx | Median 6 (3–10) | CS independently associated with ↑ in-hospital mortality in severe COVID-19 (HR = 1.43, 95% CI: 0.82–2.49, p = 0.201 in IPTW; HR = 1.55, 95% CI: 0.83–2.87, p = 0.166 in PSM) and critically-ill pts (HR = 3.34, 95% CI: 1.84–6.05, p < 0.001 in IPTW; HR = 2.90, 95% CI: 1.17–7.16, p = 0.021 in PSM) | ||
| Albani F et al. [39] | Median dexa equiv dose 20 mg/day | Rs CS vs. no CS | CS not associated with in-hospital mortality | CS: ↓ ICU admission | |||
| Fang X et al. [51] | Median dose (HC equiv) pos 237.5 mg/iv 250 mg in general/severe pts | Rs, pts of different severity, CS vs. no CS | Not mentioned | Median duration: 7/4 in general/severe pts | CS:no significant difference in viral shedding irrespective of severity | ||
| Li X et al. [41] | Median cumulative dose equiv to 200 mg prednisone | As, CS vs. no CS | NA | Median duration: 4 | severity on admission, complications, Tx, and outcomes: HD CS during hospitalization (aHR, 3.5; 95%CI, 1.8–6.9] was significant risk factor associated with death in severe COVID-19 | NA | NA |
| Sarkar S et al. [20] | SRMA (2 RCTs and 10 cohorts) | CS: ↑ mortality (OR = 1.94, 95%CI: 1.11–3.4, I2 = 96%), irrespective of severity or dosage and ↑ LOHS (MD = 1.18 d, 95%CI: −1.28 to 3.64, I2 = 93%) | CS may prolong viral shedding (MD = 1.42 d, 95%CI: −0.52 to 3.37, I2 = 0%) | ||||
| Sterne JAC et al. REACT [23] | Different CS in different schemes | Ps/MA (7 RCTs) CS vs. SOC | CS: ↓ all-cause D28-mortality: dexa vs. SOC OR 0.64 [95% CI, 0.5–0.82. p < 0.01], HC vs. SOC OR 0.69 [95% CI, 0.53–0.82, p = 0.13], MP vs. SOC OR 0.91 [0.29–2.87, p = 0.87] | No difference in serious AEs | |||
| Pasin L et al. [24] | SRMA (5 RCT)CS vs. any other Tx | CS: ↓ Mortality rate (26% in CS vs. 28% in no CS, RR = 0.89, CI 95%, 0.82–0.96, p = 0.003) CS: ↑ mortality in pts on no O2 | Significant ↓ of MV risk in CS group | ||||
| Van Paassen [25] | Diverse CS strategies | SRMA (RCTs and Obs) | CS are beneficial in short-term mortality | CS: ↓ MV need | CS: not clear effect on viral clearance | ||
| Cano EJ et al. [26] | Different CS in different doses | SRMA | Mortality benefit in CS group in severe COVID-19: OR, 0.65; 95% CI, 0.51–0.83, p = 0.0006. No benefit or harmful effect among HD and LD CS | LD CS no significant effect on viral shedding |
AE = Adverse Events, aHR = adjusted hazard ratio, ARF = acute respiratory failure, As = ambispective, CI = confidence interval, cont = continuous, CS = corticosteroids, DB = double-blind, dexa = dexamethasone, Dx = diagnosis, equiv = equivalent, HC = hydrocortisone, HD = high-dose, HR = Hazard Ratio, incl = including, IPTW = inverse probability of treatment weighting, IQR = interquartile range, ITT = Intention-to-treat, LD = low-dose, LO(H)S = Length of (hospital) stay, MA = meta-analysis, MC = Multicenter, MP = methylprednisolone, MV = mechanical ventilation, NA = not applicable, NIV = noninvasive ventilation, Obs = observational, O-L = Open-label, OR = odds ratio, pos = per os, PP = per protocol, Ps = prospective, PSM = propensity score matching, pts = patients, R = Randomized, RCT = Randomized-controlled trial, RR = relative risk, Rs = Retrospective, SC = single center, SHR = subhazard ratio, SI = systemic inflammation, SOC = standard of care, Sx = symptom(s), SRMA = Systematic Review and Meta-analysis, Tx = treatment, vs. = versus, yo = years old, and Δ = change.