Controversies’ clarification regarding ribavirin efficacy in measles and coronaviruses: Comprehensive therapeutic approach strictly tailored to COVID-19 disease stages

Table 5 Severe acute respiratory syndrome coronavirus, Middle East respiratory syndrome coronavirus and severe acute respiratory syndrome coronavirus-2 clinical studies focused on ribavirin treatment

Ref.	Total no patients/ Patients treated with RBV	Days from symptoms onset to RBV initiation, as mean	Dosing regimen/Duration	Other treatments	Outcome	Side effects	Authors’ conclusions	Comments
SARS-CoV clinical studies
Hsu et al[124], Singapore	20/14	10-14	20 mg/kg tid orally	Antibiotics, Oseltamivir	6 intubated, 3 died	No	No obvious response to RBV, some deteriorated in spite of its use	Too late RBV initiation when disease is already in Phase II (Pitfall 2)
Chiang et al[125], Taiwan	4	4-9	1 g/d orally	Levofloxacin,IVIG, If severe hypoxia developed → Mp 2 mg/kg/d	No mortality	No	Beneficial preliminary results. Despite early use of steroids in SARS may prolong its natural course, in rapid progression and severehypoxia it may prevent from further lung injury by cytokine storm	Despite the low administered RBV dosing (Pitfall 3), satisfactory outcome
Poutanen et al[126], Canada1	10/7	Unclear	2 g ld → 1 g qid × 4 d → 0.5 g tid × 4-6 d	Antibiotics, Oseltamivir No steroids	RBV → 1 died. 1 in ICU but improving and 5 recovered	No	Pts treated with RBV improved but due to an array of therapeutics. The effect of RBV is unclear	The time gap between illness onset and RBV initiation is not reported
Avendano et al[127], Canada	14	4.6 d stayed at home	2 g ld → 1 g qid × 4 d → 0.5 g tid × 4-6 d	Levofloxacin 8 pts received pulsed MP	All developed dyspnea, abnormal X-ray. None intubated. Full recovery	9 pts hemolysis (days 4-6), 2 pts transfuse. 8 pts discontinued RBV but 2 pts relapsed, restarted RBV→ Recovered	RBV was associated with hemolysis that might have increased morbidity in 9 pts. No death, No intubation. 3 pts with severe hypoxia treated with iv steroids	Very promising combination of RBV + Levofloxacin + Pulsed Mp when hypoxia occurred
Tsang et al[128], Hong Kong	10	9.6 ± 5.4 d	8 mg/kg tid iv or 1.2 g tid orally	Antibiotics Steroids iv in all	2 pts → died, 8 improved	No	Combination of RBV + high dose steroids coincided with clinical improvement	Late RBV administration (Pitfall 2)
Lee et al[129], Hong Kong	138	When fever persisted > 48 h or Leukopenia/ Thrombo- cytopenia occurred	1.2 g tid po. If worsening 0.4 g tid iv	Antibiotics, Oseltamivir, Ps 1 mg/kg. If worsening 2-3 Mp pulses 0.5 g iv daily	5 pts → died, 32 pts in the ICU. 19 pts intubated, 76 pts were discharged.	No	The similarity of disease imaging with BOOP and of histologic features with ARDS, prompted authors to use RBV + steroids. The majority of the cohort responded to the combination	CMR = 3.6%. The time-gap between the disease onset and the therapy initiation was not reported. Nevertheless, outcomes were satisfactory
Ho et al[130], Hong Kong	72	4d	8 mg/kg iv tid × 7 d → 1.2 g tid po, altogether 10-14 d	Antibiotics, Steroids in 3 different regimens: Hc or Mp at dosages similar to treatment of acute severe asthma or pulsed Mp as in ARDS	Day-21 as assessment for short-term outcome. 4/72 died, 12 admitted to ICU, 6 intubated	No	Initial use of pulsed Mp appears to be a more safe and efficacious steroid regimen when compared with regimens of lower dosages	CMR = 5.5% Satisfactory results for RBV + steroids when RBV early applied
Peiris et al[131], Hong Kong1	75	As soon as SARS diagnosis was established	8 mg/kg iv tid × 14 d	Antibiotics, Hc tailing regimen (200 mg iv tid × 10 d then tapered), Mp pulses if worsening 0.5 g iv/d for 2-3 doses	At day 21, 5 died (6.7%). Convalescence at home 27 pts, 43 pts remained in hospital of whom 13 in ICU (17%) and totally 19 pts intubated	No	Higher mortality than that reported from Lee et al[129] (6.7% vs 3.5%). The clinical progression, shifting radiological findings, and the inverted V viral-load profile suggest that worsening in week 2 is related not to uncontrolled viral replication but rather to immunopathological damage	The time-gap from symptoms onset to treatment initiation is unclear
Peiris et al[132], Hong Kong	50 monitored for 12 d	6.7 d	8 mg/kg tid iv 7-10 d	Antibiotics, Hc 200 mg tid tailed off	6 pts received treatment before ICU admission all recovered. 31 uncomplicated pts recovered. From 19 complicated pts 1 died	No	Complicated cases were associated with underlying diseases and delayed use of RBV and steroid treatment. CMR = 2%	Ok
Booth et al[133], Canada	144/126	First 48 h of hospitalization	2 g ld → 1 g qid × 4 d → 0.5 g tid × 3 d	Antibiotics Ster 40%, Hc 20-50 mg/d × 10 d	103 pts discharged. 8 pts died (6 with DM, 1 with cancer)	49% decrease in Hb > 2 g/dL. 40% transaminitis. 14% bradycardia	Poor outcome was associated with RBV treated pts but it was not significant	Despite unclear time gap between disease onset and RBV initiation, it seems that RBV alone (no Mp pulses, low steroid regimen in only 40% of pts), might not exert a clear benefit (Pitfall 4)
Zhao et al[134], China3	190/40. pts allocated to 4 groups	Not reported	group A: 0.4-0.6 g/d iv	Antibiotics	2 pts died. 3 intubated. The rest followed group D → improved	No	Early use of high- dose steroids with quinolone + azi gave the best outcome. No advantage from RBV	Unclear time-gap, too low RBV dosing (Pitfall 3). RBV treatment alone (Pitfall 4)
So et al[135], Hong Kong	31 pts → 1 recovered on antibiotics	5.5 d	RBV 400 mg iv tid × 3 d then 1200 mg bid orally × 10-14 d	Broad-spectrum antibiotics, Mp 1 mg/kg tid × 5 d then 1 mg/kg bid × 5 d. When worsening pulsed Mp 0.5 g iv. Then Ps 0.5 mg/kg bid × 5 d orally	17 pts showed rapid response. 13 achieved improvement with step-up or pulsed MP. None intubated. No mortality	No	Protocol provided satisfactory outcomes	No mortality reported
Lau et al[136], Hong Kong	88 pts → 3 recovered on antibiotics/ 68	5.8 d	So et al[135] treatment protocol applied	So et al[135] treatment protocol applied	18 pts required ventilation. 30 pts needed Mp pulses. All-cause mortality for pts aged < 60 was 0% (0/76) and 3/12 (25%) in aged > 60. CXRs of all survivors were significantly clearer in discharge	No	The standard treatment protocol of RBV + steroids and pulsed Mp resulted in satisfactory outcomes	Total CMR = 3.4% Ok
Dwosh et al[137], Canada	15 pts, treatment data only for 1 case	Post-intubation 9 d	2 g ld iv → 1 g qid × 4 d → 0.5 g tid × 6 d	Mp 40 mg × 2	Successfully extubated	No	No treatment conclusions	Late RBV initiation (Pitfall 2)
Sung et al[138], Hong Kong	138/94	3 d (0-11) to admission. RBV started after 48 h	2.4 g ld orally → 1.2 g tid. If dyspnea → 400 mg tid iv	Antibiotics Ps 0.5-1 mg/kg. If dyspnea → Hc 100 mg tid. Mp pulses for 3 d (up to 3 g)	25/94 pts responded toRBV. Mp in 107 non-resp. → 88.8% success. 15 pts died (mortality 10.9%)	Modest degree of anemia in 59%	RBV’s role is doubtful in treatment. Pulsed Mp associated with improvement	RBV alone or associated with low dose steroids seems insufficient for SARS Phase 2 (Pitfall 4). Possibly RBV is insufficient when applied in respiratory failure
Leong et al[139], Singapore	229/97 compared to a group of pts who did not receive RBV on day 6	6.4 d. Duration 5.6 d. Doctor- dependent RBV use	Oral 1.2 g tid iv 400 mg tid	Insufficient data	Mortality 10.3% vs 12.9% in control. HR of death for RBV 0.78 (P = 0.53). When adjusted for steroids HR = 1.03 (P = 0.93)	No difference in side effects	Use of RBV alone does not seem to confer any benefit	Late use of RBV (Pitfall 2), uneven groups, doctor-dependent use of RBV (Pitfall 5). RBV alone seems insufficient (Pitfall 4)
Leung et al[140], Hong Kong	1755/1467 met SARS criteria/ 1416 received RBV	On symptom onset: 25 pts. 1-3 d: 480 pts. 4-6 d: 499 pts. ≥ 7 d: 412 pts	Not reported	Not reported	302 died → mortality 17.2%. CFR of 25 pts: 4.0%, of 480 pts: 11.1%, of 499 pts: 10.0%, of 412: 12.5%, of 51 pts treatment not prescribed: 29.4%	No side-effects reported	The timing of RBV administration did not seem to statistically significantly influence outcome	Authors explain their finding that it possibly results from residual confounding or insufficient power to detect a difference given that most pts were treated (Pitfall 5)
Knowles et al[89], Canada	110 pts focused on RBV side-effects	Not reported	High-dose RBV(total > 20 g): 2 g ld → 1 g qid × 4 d → 0.5 g tid × 3 d; Low-dose RBV: 0.4 g iv tid × 4 d → 1.2 g po bid × 7 d	Antibiotics 50% steroids	61% hemolytic anemia. 28% transfused with ≥ 1 U of RBCs. A significant decrease (> 2 mg/dL) in Hb was seen at 6.8 d after RBV started, and reached a nadir at 13 d. Anemia associated with higher RBV doses (P = 0.005) and prolonged hospital stay (P = 0.001). 35/76 pts developed hypomagnesemia, 32/62 pts developed hypocalcemia. Teratogenic effect: it is recommended that 15 half-lives (6 mo) is required to complete washout after RBV discontinuation		In contrast to HK experience where RBV associated side effects have not been detailed, their comparable RBV doses suggest that associated side effects are frequent. The benefits of RBV use may not outweigh the risk of side effects with negative economic consequences on hospitals	No outcome results for the 110 pts were reported
Chan et al[141], Hong Kong4	75 pts compared with matched cohorts of 643 and 343 pts	As soon as SARS diagnosis established. Lop/r 5.5d and 1 d after RBV. Rescue therapy: 18 d	2.4 g oral ld → 1.2 po tid or 8 mg/kg tid × 10-14 d	Lop/r 400/100mg bid × 10-14 d. 1 group received it as initial treatment and a 2nd as rescue. In addition, tailing steroids regimen × 21 d and pulsed Mp	Lop/r as initial therapy CRF 2.3% vs 15.6%(P < 0.05), intubation rate 0% vs 11% (P < 0.05). As rescue no difference	No	Early Lop/r initiation in addition to standard treatment protocols (Ho, So) showed significantly beneficial outcomes	Combination of early RBV with Lop/r and steroid regimens with pulsed Mp when needed showed statistically significant results in intubation and mortalityrates. Ok
Chu et al[87], Hong Kong5	111pts historical controls compared to 41 pts treated with RBV + Lop/r	Once diagnosis was established for RBV. For Lop/r initial treatment group it was started at a median of 3.5 d while in the rescue group at 14 d	4 g oral ld → 1.2 g tid or 8 mg/kg iv tid × 14 d	Lop/r 400/100mg bid orally ´ 14 d. Tailing steroid regimen × 21 d and pulsed Mp	21-d adverse outcome (ARDS or death) was 28.8% for the historical control vs 2.4% in the initial treatment group (P < 0.001). No deaths in the treatment group	Mild gastrointestinal adverse-effects. Anemia (70%) → 2 pts transfused. 26.8% bradycardia	Apparent favorable clinical response to combination of Lop/r + RBV + steroids when needed	The second study showing statistically significant benefits from the combination of RBV+ Lop/r + ster when early applied in the disease course
Cheng et al[142], Hong Kong6	772	No data available	No data available	Steroids Lop/r	No data available	No	675 pts received RBV and 44 Lop/r. No obvious difference noted irrespective of treatment combination	In Table 2 of the article however, RBV + Lop/r + ster provided a CFR of 2.3%, IFN + ster 0%, RBV + pulsed Mp 5.9% and RBV + ster 7.7% compared to a 15.4% of supportive treatment
Lau et al[143], Hong Kong, Canada7	Integrated data base containing 1755 HK pts and 191 Toronto cases	Within 2 d from hospital admission	No data available	Data showed for HK pts crude CMR 23.3% in neither treatment, 29.4% in steroids, 8.9% in RBV and 12.6% for combination. For Toronto pts no treatment 20%, RBV 9.3% and RBV + ster 12.8%. Authors adjusted these results for propensity scores and balance was achieved among all pts characteristics. Side-effects not considered in this study		Estimated CFRs based on the generalized propensity score weighting, the model predicted that the overall CFR would have been highest if all pts in HK had been treated with RBV + steroids, whereas it would have been the lowest if none treated. Toronto results were consistent. The combination of RBV + ster has no therapeutic benefit		The generalized propensity score weighting model prediction reversed the initial finding for CFR 12.7% of the combination to 19.2% and of untreated from 23.3% to 15.4% (!!). Inconclusive study (Pitfall 5)
MERS clinical studies
Omrani et al[161], Saudi Arabia2	44 with severe pneumonia 20 treated 24 control. Scores APACHE II: 27, and SOFA: 11	3 d from diagnosis	2 g ld → 1.2 g tid 4 d → 600 mg tid × 4-6 d. Dosing adjusted to Crcl. Orally RBV	Antibiotics, Oseltamivir, PegIFN-α2α sc 180 μg/wk for 2 wk. Hc 200 mg/d in pts with refractory septic shock	41/44 intubated. 14-d mortality: treat 6/20 vs control 17/24 (P = 0.004). 28-d: treat: 14/20 vs control 20/24 (P = 0.054)	RBV well tolerated. Hb drop in treat > control (P = 0.002). No differences in transfusions, no treatment discontinuation	Significant benefit in 14-d survival. The loss of difference in 28-d might be explained by high initial APACHE II and SOFA scores and several comorbidities	Surprisingly, statistically significant results despite that eligible patients had initially severe pneumonia (Phase 2) (Pitfall 4) without high dose steroids applied. Long- lasting IFNs (peg) might not be the best form for acute infections
Shalhoub et al[162], Saudi, Arabia2	32 pts were already under MERS pneumonia and some with respiratory failure	For IFNs: 1 d after MERS diagnosis. For RBV not reported	2 g ld orally → 600 mg bid	Antibiotics, IFN-α2a sc 180 μg/wk × 2 wk. IFN-β1a sc44 μg × 3 times/wk	Overall mortality: 22/32 (69%). IFN-α2a + RBV: 11/13 (85%). IFN-β1a + RBV: 7/11 (64%). Hemodialysis pts: 14/14 (100%)	No	IFN-α2a or IFN-β1a + RBV were ineffective against MERS mortality	Unknown time-gap between symptom onset and treatment initiation. Very low RBV dose applied (Pitfall 3). In specific cases with severe pneumonitis high-dose steroids and Mp pulses should have been used for better outcomes (Pitfall 4)
Al Ghamdi et al[171], Saudi, Arabia2	51 pts	No data reported	No data reported	Antibiotics, IFN-α, IFN-β, MMF, Hc in 5 pts	31 pts received antivirals (IFNs,RBV) in several combinations, 8 pts MMF all survived. (IFN-β and MMF were given to less severely pts). CMR = 37%	No	IFN-β and MMF were predictors of increased survival	No time gap from symptom onset reported. No dosing reported. Inconclusive study for RBV treatment
Choi et al[172], Korea8	186 pts	6 d (1-20 d) 14% of pts within 48 h	81% IFN-α + RBV + Lop/r, 12.7% IFN-α + RBV, 5.0% RBV + Lop/r, No dosing regimens reported		CMR = 20.4% lower than others ranging 36.5%-65%	No	Unable to assess the clinical impact of therapies as most pts received antivirals	No dosing regimens, not duration reported
Arabi et al[166], Saudi, Arabia9	309/151 pts critically ill received steroids	3 d from ICU admission	Antivirals: RBV, IFN, RBV + IFN, oseltamivir. The median of the maximum daily Hc-equivalent was 300 mg with a median duration of 7 d		CMR 74.2% vs 57.6% (no steroids). After adjustment for baseline and time-varying confounders the use of steroids was not associated with increased 90-d mortality but with delayed RNA clearance	No	Steroids were commonly used in critically ill patients with MERS. Pts given steroids were more likely to have 1 or more comorbidities than those who did not (P = 0.001)	No Mp pulses were administered. Maximum Hc doses reported (300 mg) are equivalent to only 60 mg of Mp. In addition, authors do not comment about the impact of the co- administered antivirals (Pitfall 5)
Habib et al[163], Saudi Arabia2	63/61pts presented with severe illness (pneumonia 87.3% and septicemia 11%)	No data reported	No data reported	No data reported	Overall CMR 25.4%. Treated 22.9%. Survivors were more likely to have had received IFN + RBV than those who died (P = 0.01)	No	CMR 25% comparable to that of Omrani 30%, lower than AlMekhlafi (74.2%), Khalid (55%), and Al-Tawgiq (100%). Unable to determine the combination efficacy in the absence of a reference group	No dosing regimen, no time-gap from onset. The severity in admission probably implies an advanced disease phase, where antivirals are less effective (Pitfall 4)
Arabi et al[166], Saudi, Arabia2	349/144 critically ill all ICU pts	2d from ICU admission but 9 d (6-12) from symptom onset	RBV: 2 g ld po → 1.2 g po tid × 4 d → 600 mg tid po × 4-6 d	Peg-IFN-α2b → 1.5 mcg/kg sc × 2 wk Per-IFN-α2a → 180 μg/wk × 2 wk Peg-IFN-β1a → 44 mg sc × 3/wk	Crude CMR was higher in antiviral treated group 73.6% vs 61.5% (P = 0.02). However, with a marginal structural model there was no significant difference in 90-d mortality (aOR: 1.03; 95%CI: 0.73-1.44, P = 0.87). Also, no significant difference in RNA clearance (aOR: 0.65; 95%CI: 0.3-1.44, P = 0.29)		During ICU stay RBV/IFN treated pts were more likely to receive steroids (59.7% vs 44.9%P = 0.006). Future studies should test the efficacy of newer antiviral interventions	Very late antiviral initiation. Possible higher needs for steroids in antiviral – treated group could imply more severely ill pts (Pitfalls 2, 5)
AlMekhlafi et al[167], Saudi, Arabia2	31 pts in ICU. 13 pts received RBV+ IFN-α2α	ICU pts	Not reported	Not reported	CMR 74.2%. Among 13 pts who were given antivirals, 9 died	No	All pts who received either oseltamivir or RBV + IFN-α2a had no favorable outcomes	Antivirals may have no efficacy in Phase II-III of MERS (Pitfall 4)
Khalid et al[168], Saudi, Arabia2	14 pts intubated 11 pts received RBV	6 d	Not reported	Antibiotics RBV + Peg-IFN-α2a, Mp 1 mg/kg/d × 7 d	9 pts died in the ICU, 5 discharged	No	MERS with ARDS has high mortality rates. The role of RBV + IFN warrants further evaluation	Antivirals may have no effect in Phase II-III of MERS-infected pts under mechanical ventilation (Pitfall 4)
Khalid et al[164], Saudi, Arabia2	6 pts, 3 cases 74-84 yr, 3 cases 17-54 yr	1st group 12-19 d; 2nd group 1-2 d	2 g ld → 1.2 g tid × 4 d → 0.6 g tid × 4-6 d	IFN-α2b sc 180 μg/wk × 2 wk. 1 case received pulsed Mp and recovered	1st group pts all died. 2nd group all recovered	No	Combination of RBV and IFN-α2b have a role in treatment of MERS if started early in disease course	Very late (12-19 d) antiviral initiation in 1st group when disease is already in the ARDS phase (Pitfall 4). 1 case was helped by Mp pulses
Al-Tawfiq et al[169], Saudi, Arabia	5/5	11-21 d (after admission)	2 g ld → 400 mg po tid	Antibiotics Oseltamivir IFN-α2b Mp 40 mg tid or Ps 40 mg/d	All died	No	All pts were already intubated when treatment started	Antivirals in Phase 2, very low RBV dosing, low ster dosing for Phase 2-3 (Pitfalls 2, 3, 4)
Park et al[159], Korea2	43 HCW with high-risk exposure to MERS pneumonia pts. 21 HCW with more severe exposure received PEP. 22 HCW no PEP	Within 36 h after unprotected exposure	RBV 2.0 g ld orally → 1.2 g tid × 4 d → 600 mg tid × 6-8 d	Lop/r 400/100 mg bid × 11-13 d	6/43 HCW exposed developed MERS infection. The attack rate was lower in the PEP vs no-PEP (0% vs 28.6% OR: 0.405 P = 0.009). No MERS infection in PEP group. Only PEP therapy reduced significantly the risk of MERS infection (OR: 0.714; P = 0.009)	Mild: diarrhea, nausea, anemia, stomatitis, leucopenia, hyperbilirubinemia. No PEP discontinuation. All normalized after completion of PEP	PEP therapy was associated with a 40% decrease in the risk of infection	The only study reporting results of PEP prophylaxis with the combination of Lop/r + RBV. Ok
COVID-19 clinical studies
Tong et al[189], China2	115/44 pts Severe disease. 9 pts intubated 28 pts NINV	8 d from onset 4 d from diagnosis	500 mg iv bid	Antibiotics	Negative conversion time of SARS-CoV-2 test in RBV vs control (12.8 d vs 14.1 d, P = 0.314) CFR 17.1% vs 24.6% (P = 0.475)	No side effects. No difference in anemia	RBV administration was doctor-dependent and sometimes RBV was out of stock. In severe COVID-19 RBV is not associated with improved negative conversion time for SARS-CoV-2 test or improved mortality	Pitfall 2. Relatively moderate RBV dosing (Pitfall 3). Possibly not regular RBV administration (Pitfall 5)
Li et al[190], China2	151 pts, Number of pts treated with RBV was not specified. Moderate to critical disease	Not reported	500 mg iv bid or tid × 10 d	Umifenovir Lop/r, Traditional medicine, Peramivir, Oseltamivir, Penciclovir Ganciclovir	25 pts discharged 25 pts hospitalized 79 pts clinical improvement7 died (CFR = 4.6%)	The use of two-step clustering and subgroup analysis enabled an in-depth analysis of the effects of single or combined antiviral therapy. Following the antiviral therapy, there was indeed an improvement of severe patients' condition. Combination was superior to single or dual agents. A quadruple combinationof Umifenovir + RBV + Lop/r+ Lianhua Qingwen has been recommended for critically ill COVID-19 pts		Incomplete data (time-gap from symptom onset to treatment initiation) (Pitfall 5)
Yuan et al[191], China2	94 pts, 46 pts IFN-α + Lop/r.21 pts IFN-α + Lop/r + RBV. Median age 40 yr. 15 pts, 1 or 2 comorbiditie. Mild disease: 8 pts. Moderate: 75 pts. Critical: 11 pts	Hospitalized 7d after symptom onset	No data reported	No data reported	Significant correlation between the length of hospital stay and PCR negative conversion time in pts treated with IFN + Lop/r (P = 0.012) and with IFN + Lop/r + RBV (P = 0.0215). No death, no intubation, all recovered	No	These two regimens might be beneficial for COVID-19 treatment	Pitfall 2. No dosing regimens reported. Ok
Wu et al[192], China9	80/80 pts, 41 females, 46.1 yr. 77 pts mild to moderate symptoms. 3 pts severe. 38 pts chronic diseases	Not reported	Not reported. Duration 7 d	Moxifloxacin duration 7 d12 pts Mp to alleviate the shortness of breath	No death, no INV. 35 pts NINV. 55 pts abnormal chest CT. 3 pts transaminitis. 1 pt hemodialysis. As of writing, 21 pts discharged (stay 8 d)	No	Notably, infected patients may be falsely excluded based on 2 consecutively negative respiratory pathogenic PCR tests	Surprisingly, authors do not discuss at all the role of treatment administered (RBV + Mp + Moxi) (Pitfall 5)
Chen et al[193], China2	681 pts with severe disease/279 received RBV. 375 pts had comorbidities. Median 65 yr. 40-65 yr 46.1% of pts, > 65 yr 47.1% of pts	No time-gap between symptom onset and initiation of treatment, no dosing regimens reported, or drug combinations. 666 pts received antivirals, antibiotics (83.8%), IVIG (54.6%), and steroids (48.8%)			In a report from China overall mortality from COVID-19 was 2.3% while in critical cases 49%. In another from Italy CFR was 26% in ICU pts. Another study indicated a mortality of 15% while in ICU cases 38%. In this study CFR was 15.3%. 45.8% of the pts had preexisting cardiovascular disease, of which 23.4% died. In multivariate analysis, RBV and arbidol were positively associated with death, OR: 0.208 (95%CI: 0.07-0.618; P = 0.005). Of notice, RBV might have a beneficial effect in severe COVID-19 pts with cardiovascular diseases and cardiac injury by disease. Therefore, every drug regimen should include arbidol or RBV for severe cases			Impressive findings for both antivirals in reducing mortality in severe cases. The beneficial effect of RBV in cardiac injury is supported by another study which showed that RBV is mostly concentrated in heart and intestines
Peng et al[197], China2	75 pediatric pts. 8 most critical cases received RBV + IFN-α	4.9 d	10 mg/kg/d bid iv	IFN-α neb1-4 μg/kg/d bid. Antibiotics, arbidol 5 pts, oseltamivir 20 pts	All discharged. Length of hospital stay 10.6 d and SARS-CoV-2 clearance 6.4 d. The two most severe cases were treated with RBV	No	Severity in pediatric pts milder than adults. The efficacy of antiviral therapy in children remains to be evaluated	Ok
Huang et al[201], China10	101 pts, 33 pts RBV + IFN-α, 36 pts IFN-α + Lop/r, 32 pts RBV + Lop/r + IFN-α, Mild to moderate severity	4 d to enrollment	2.0 g ld iv → 400-600 mg tid depending on bw × 14 d	Lop/r 400/100 mg bid × 14 d IFN-α in h 5 MU bid × 14 d	SARS-CoV-2 time to negativity 12 d in group 2 vs 13 and 15 d in groups 1 and 3 (P = 0.23). Higher proportion of nucleic acid negativity in group 2 (61.1%) than (51.5% and 46.9%) in groups 1 and 3 in 14 d	GI side-effects mainly in the triple combination	No significant differences among the three regimens in terms of antiviral efficacy. Significant GI effects in the triple combination	Ok
Hung et al[202], China11, Open-label Phase 2 trial	127/81 pts 81 RBV + Lop/r + IFN-β1b. 41 Lop/r (control). Median age 52 yr. Men 54%, 51 pts had underlying diseases. Mild to moderate COVID-19	Triple combination: 5 d, control: 4 d	400 mg bid × 14 d	Oral Lop/r 400/100 mg IFN-β1b 8MU on alternate day sc up to 3 doses (within 1st wk). Hc 50 mg tid in oxygen desaturation	Abnormal chest X-ray in 96 pts. 17 pts oxygen desaturation → 6 in ICU, 1 intubated (96 yr) but extubated after 10 d. No one succumbed. Time to negative swab from treatment initiation in combo 7 d vs 12 d in control (P = 0.001)	Mild and self-limiting. Diarrhea, nausea, transaminitis, all resolved within 3 d from treatment initiation	Time to NEWS2 0 in combo 4 d vs 8 d (P = 0.0001) in control and time to SOFA 0 in combo 3 d vs 8 d in control (P = 0.041). Hospital stay duration: combo 9 d vs 14.5 d in control (P = 0.016). In subgroup analysis when authors compared pts with early (< 7 d) treatment initiation in both groups, all comparisons where statistically very significant (P < 0.0001) including improvement in NEWS2 and SOFA scores, and time to negative viral loads	Early antiviral triple therapy is superior to lop/r in shortening shedding, alleviating symptoms and facilitating discharge of pts with mild to moderate COVID 19. Ok
Eslami et al[199], Iran12	62/27pts All treated with SOC: Lop/r + HCQ	Not reported (at admission)	RBV 600 mg bid × 14 d	Sof/ daclatasvir 400/60 mg qd. All treated with Lop/r 400/100 mg bid × 5 d and HCQ 400 mg single dose	Median stay 5 d for Sof/d vs 9 d for RBV. CFR 6% in Sof/d vs 33% in RBV. Relative risk of death for those treated with Sof/d0.17 (95%CI: 0.04-0.73; P = 0.02)	Mild adverse effects reported but no discontinuation was demanded	Given these encouraging initial results, further investigation in larger-scale trials seems warranted	Unclear time-gap from disease onset to RBV initiation. Low RBV dosing. Confusing study as both arms were concurrently treated with other anti-coronaviruses agents (Pitfalls 3, 5)
Kasgari et al[200], Iran13	48 pts moderate disease, 24 pts → Sof/d + RBV24 pts → SOC: Lop/r + HCQ + RBV depending on recommendations at the time of the study	Not reported	RBV 600 mg bid	The median duration of hospital stay, number of ICU admissions, and number of deaths: no statistically significant differences between the two groups. Only trends for recovery and lower deaths in the Sof/d + RBV arm				Very small number of participants, Pt 2 unclear, fell under Pt3. Confounding results as both arms were concurrently being treated with antivirals, even with RBV (Pitfalls 2, 3, 5)
Liu et al[198], China14	Enrolled studies with COVID-19 (n = 12), MERS (n = 2), SARS (n = 4) and influenza (n = 1)	Interventions in the studies RBV (n = 3), HCQ (n = 5), favipinavir (n = 3), IFN (n = 3), Lop/r (n = 2), umifenovir (n = 1)			This review did not find persuasive evidence of benefit for treatment using RBV in a population of pts with COVID-19 and results from studies evaluating SARS or MERS provided no support for a reduction in mortality with RBV treatment[85,119,171]		Only treatment with Lop/r for which authors found low-quality evidence for a decrease in hospital stay in ICU. To date, persuasive evidence of important benefit in COVID-19 does not exist for any antiviral although for each treatment evidence has not excluded important benefit	Very controversial conclusion (Pitfall 5)
Zhong et al[88], China14	COVID-19 = 7 studies. SARS = 9, MERS = 2, RBV = 4 studies, RBV + Lop/r + ster = 2, RBV + IFNs = 3, RBV + ster = 1	Compared with comparators, interventions notably reduce mortality (RR: 0.65, 95%CI: 0.44-0.96, I2 = 81.3%). In subgroup analysis, the combination of RBV + ster remarkably decreased mortality (RR: 0.43, 95%CI: 0.27-0.68). Besides, Lop/r, RBV, RBV + IFN and combination of Lop/r + RBV + ster showed tendency of lower mortality. Interventions also remarkably ameliorated clinical and radiological improvement, without manifesting clear effect on virological eradication (except for Lop/r-based regimens), incidence of ARDS, intubation, and adverse effects				In conclusion, there was evidence of lower mortality, better clinical and radiological improvement in intervention group compared to control		A very large meta-analysis with remarkable conclusions for coronaviruses treatment. Ok

¹Prospective.

²Retrospective.

³Prospective randomized.

⁴Multicenter retrospective matched cohort.

⁵Open non-randomized prospective.

⁶Review.

⁷Systematic review.

⁸Retrospective observational.

⁹Multicenter retrospective.

¹⁰Randomized, open-label, prospective trial.

¹¹Multicenter randomized prospective.

¹²Open label parallel trial.

¹³Randomized controlled trial.

¹⁴Meta-analysis.

aOR: Adjusted odds ratio; ARDS: Acute respiratory distress syndrome; bid: Bis in die; bw: Body weight; CI: Confidence interval; COVID-19: Coronavirus disease 2019; CFR: Case fatality rate; Crcl: Creatinine clearance; GI: Gastrointestinal; Hb: Hemoglobin; HCQ: Hydroxychloroquine; HR: Hazard ratio; ICU: Intensive care unit; IFN: Interferon; IVIG: Intravenous immunoglobulin; neb: Nebulizer; Lop/r: Lopinavir/ritonavir; MERS: Middle East respiratory syndrome; MMF: Mycophenolate mofetil; OR: Odds ratio; pts: Patients; RR: Relative risk; SARS: Severe acute respiratory syndrome; RBV: Ribavirin; Sof: Sofosbuvir; ster: Steroids; tid: Ter in die.