Efficacy and Safety of Novel Oral Antivirals in Hospitalized COVID-19 Patients: A Network Meta-Analysis of Randomized Clinical Trials

Figures & data

Figure 1 Flow diagram of the selection strategy.

Notes: The search flow diagram summarizes the search, screening, retrieval, and evaluation of articles finally included in the network meta-analysis.

Figure 2 Network of eligible comparisons for in-hospital mortality, adverse event, recovery time, and improvement in SpO2.

Notes: The figure plots the network of eligible direct comparisons for in-hospital mortality (36 studies) (A), adverse event (23 studies) (B), recovery time (8 studies) (C), and improvement in SpO2 (9 studies) (D). The width of the lines is proportional to the number of trials comparing each treatment pair, and the size of each node is proportional to the number of randomized participants.

Abbreviations: IVM, Ivermectin; FVP, favipiravir; CQ, chloroquine; LPV/RTV, lopinavir and ritonavir; HCQ, hydroxychloroquine; HCQ+ AZT, hydroxychloroquine plus azithromycin; SOC, standard of care; SpO2, peripheral capillary oxygen saturation.

Table 1 League Table of Network Meta-Analysis Results of Interventions for In-hospital Mortality(A), Adverse Event(B), Recovery Time(C), and Improvement in SpO2 (D)

A: In-hospital mortality
Baricitinib	1.33 (0.71,2.50)	1.42 (0.78,2.58)	1.44 (0.75,2.76)	1.51 (0.87,2.64)	1.59 (0.73,3.47)	1.64 (0.63,4.27)	1.65 (0.85,3.18)	1.74 (1.09,2.79)
0.75 (0.40,1.41)	HCQ+AZT	1.07 (0.66,1.71)	1.08 (0.64,1.83)	1.14 (0.75,1.71)	1.20 (0.61,2.33)	1.23 (0.51,3.01)	1.24 (0.69,2.20)	1.31 (0.87,1.96)
0.70 (0.39,1.28)	0.94 (0.58,1.50)	SOC	1.01 (0.75,1.36)	1.06 (0.84,1.35)	1.12 (0.67,1.86)	1.16 (0.51,2.63)	1.16 (0.71,1.88)	1.22 (0.85,1.76)
0.69 (0.36,1.33)	0.92 (0.55,1.56)	0.99 (0.73,1.32)	LPV/RTV	1.05 (0.75,1.47)	1.10 (0.67,1.82)	1.14 (0.48,2.68)	1.14 (0.66,1.97)	1.21 (0.78,1.88)
0.66 (0.38,1.15)	0.88 (0.58,1.33)	0.94 (0.74,1.19)	0.95 (0.68,1.34)	HCQ	1.05 (0.61,1.81)	1.09 (0.49,2.42)	1.09 (0.69,1.71)	1.15 (0.86,1.54)
0.63 (0.29,1.37)	0.84 (0.43,1.63)	0.89 (0.54,1.49)	0.91 (0.55,1.49)	0.95 (0.55,1.63)	FVP	1.03 (0.41,2.62)	1.03 (0.52,2.05)	1.09 (0.59,2.02)
0.61 (0.23,1.58)	0.81 (0.33,1.98)	0.86 (0.38,1.97)	0.88 (0.37,2.07)	0.92 (0.41,2.05)	0.97 (0.38,2.46)	CQ	1.00 (0.45,2.24)	1.06 (0.46,2.42)
0.61 (0.31,1.17)	0.81 (0.45,1.44)	0.86 (0.53,1.40)	0.88 (0.51,1.52)	0.92 (0.59,1.44)	0.97 (0.49,1.92)	1.00 (0.45,2.24)	IVM	1.06 (0.67,1.67)
0.57 (0.36,0.92)	0.76 (0.51,1.15)	0.82 (0.57,1.17)	0.83 (0.53,1.29)	0.87 (0.65,1.16)	0.91 (0.50,1.69)	0.94 (0.41,2.16)	0.94 (0.60,1.49)	Placebo
B: Adverse event
SOC	1.89 (0.63,5.67)	2.01 (0.75,5.40)	2.55 (0.91,7.11)	2.57 (0.86,7.69)	2.87 (0.93,8.87)
0.53 (0.18,1.58)	IVM	1.06 (0.30,3.81)	1.35 (0.56,3.21)	1.36 (0.40,4.61)	1.52 (0.61,3.78)
0.50 (0.19,1.34)	0.94 (0.26,3.38)	LPV/RTV	1.27 (0.40,3.97)	1.28 (0.40,4.12)	1.43 (0.41,4.94)
0.39 (0.14,1.10)	0.74 (0.31,1.77)	0.79 (0.25,2.47)	HCQ	1.01 (0.34,2.96)	1.13 (0.55,2.31)
0.39 (0.13,1.16)	0.74 (0.22,2.50)	0.78 (0.24,2.52)	0.99 (0.34,2.91)	FVP	1.12 (0.37,3.35)
0.35 (0.11,1.07)	0.66 (0.26,1.64)	0.70 (0.20,2.42)	0.89 (0.43,1.81)	0.89 (0.30,2.68)	Placebo
C: Recovery time (d)
IVM	0.22 (-1.36,1.80)	0.36 (-1.25,1.96)	1.36 (0.39,2.32)
-0.22 (-1.80,1.36)	HCQ	0.14 (-1.66,1.93)	1.14 (-0.12,2.39)
-0.36 (-1.96,1.25)	-0.14 (-1.93,1.66)	Baricitinib	1.00 (-0.29,2.29)
-1.36 (-2.32,-0.39)	-1.14 (-2.39,0.12)	-1.00 (-2.29,0.29)	Placebo
D: Improvement in SpO2(%)
FVP	-0.00 (-0.93,0.93)	-0.18 (-1.61,1.25)	-0.83 (-2.26,0.60)	-1.77 (-3.45,-0.09)
0.00 (-0.93,0.93)	LPV/RTV	-0.18 (-1.89,1.53)	-0.83 (-2.54,0.88)	-1.77 (-3.69,0.15)
0.18 (-1.25,1.61)	0.18 (-1.53,1.89)	HCQ	-0.65 (-2.52,1.21)	-1.59 (-3.41,0.22)
0.83 (-0.60,2.26)	0.83 (-0.88,2.54)	0.65 (-1.21,2.52)	SOC	-0.94 (-2.25,0.37)
1.77 (0.09,3.45)	1.77 (-0.15,3.69)	1.59 (-0.22,3.41)	0.94 (-0.37,2.25)	IVM

Notes: For in-hospital mortality and adverse event, estimates are presented as odds ratios (OR) with 95% confidence interval (CI) in parentheses (In-hospital mortality and adverse event are belongs to categorical variable). In other words, individuals receiving Baricitinib are estimated to have a 43% reduction in the odds of experiencing in-hospital mortality compared to those receiving placebo. OR <1 favour column intervention and OR >1 favour row intervention. Interventions in orange are significantly different since 95% CI does not include 1. For recovery time and improvement in SpO2, estimates are presented as mean differences(MD) with 95% confidence interval in parentheses (Recovery time and improvement in SpO2 are belongs to continuous variable). Compared to placebo, the IVM group experienced an average reduction of 1.36 days in recovery time. Compared to the IVM group, the FVP group had an average increase of 1.77 units in improvement in SpO2. MD <0 favour column intervention and MD >0 favour row intervention. Interventions in orange are significantly different since the 95% CI does not include 0. Figures should be read from left to right: A. In-hospital mortality (36 trials, 78 groups, 33,555 participants), B.Adverse event (23 trials, 49 groups, 11,530 participants), C. Recovery time (8 trials, 18 groups, 4,348 participants), D. Improvement in SpO2 (9 trials, 21 groups, 2,651 participants). Orange shaded results indicate statistical significance.

Abbreviations: IVM, Ivermectin; FVP, Favipiravir; CQ, Chloroquine; LPV/RTV, Lopinavir/Ritonavir; HCQ, Hydroxychloroquine; HCQ+AZT, Hydroxychloroquine + azithromycin; SOC, Standard of care; SpO2, Peripheral capillary oxygen saturation.

Figure 3 Ranking of treatment strategies based on probability of their protective effects on outcomes of in-hospital mortality (A), adverse event (B), recovery time (C), and improvement in SpO2 (D) by cumulative ranking area (SUCRA). Greater probability, stronger protective effects.

Notes: Graphs show the cumulative probability of each intervention ranking, from best (rank 1) to worst (rank 8, 6, 5 or 3 depending on the number of treatments) for each outcome. A ranking indicates the probability that an intervention is best, second best,etc. For example, baricitinib ranked best for reducing in-hospital mortality, while no intervention ranked worst.

Abbreviations: IVM, Ivermectin; FVP, favipiravir; CQ, chloroquine; LPV/RTV, lopinavir and ritonavir; HCQ, hydroxychloroquine; HCQ+ AZT, hydroxychloroquine plus azithromycin; SOC, standard of care; SpO2, peripheral capillary oxygen saturation.