Abstract
Background: Numerous interventions for coronavirus disease 2019 (COVID-19) have been investigated by randomized controlled trials (RCTs). This systematic review and Bayesian network meta-analysis (NMA) aim to provide a comprehensive evaluation of efficacy of available treatments for COVID-19.
Methods: We searched for candidate COVID-19 studies in WHO COVID-19 Global Research Database, PubMed, PubMed Central, LitCovid, Proquest Central and Ovid up to December 19, 2020. RCTs for suspected or confirmed COVID-19 patients were included, regardless of publication status or demographic characteristics. Bayesian NMA with fixed effects was conducted to estimate the effect sizes using posterior means and 95% equal-tailed credible intervals (CrIs), while that with random effects was carried out as well for sensitivity analysis. Bayesian hierarchical models were used to estimate effect sizes of treatments grouped by their drug classifications.
Results: We identified 96 eligible RCTs with a total of 51,187 patients. Compared with the standard of care (SOC), this NMA showed that dexamethasone led to lower risk of mortality with an odds ratio (OR) of 0.85 (95% CrI [0.76, 0.95]; moderate certainty) and lower risk of mechanical ventilation (MV) with an OR of 0.68 (95% CrI [0.56, 0.83]; low certainty). For hospital discharge, remdesivir (OR 1.37, 95% CrI [1.15, 1.64]; moderate certainty), dexamethasone (OR 1.20, 95% CrI [1.08, 1.34]; low certainty), interferon beta (OR 2.15, 95% CrI [1.26, 3.74]; moderate certainty), tocilizumab (OR 1.40, 95% CrI [1.05, 1.89]; moderate certainty) and baricitinib plus remdesivir (OR 1.75, 95% CrI [1.28, 2.39]; moderate certainty) could all increase the discharge rate respectively. Recombinant human granulocyte colony-stimulating factor indicated lower risk of MV (OR 0.20, 95% CrI [0.10, 0.40]; moderate certainty); and patients receiving convalescent plasma resulted in better viral clearance (OR 2.28, 95% CrI [1.57, 3.34]; low certainty). About two-thirds of the studies included in this NMA were rated as high risk of bias, and the certainty of evidence was either low or very low for most of the comparisons.
Conclusion: The Bayesian NMA identified superiority of several COVID-19 treatments over SOC in terms of mortality, requirement of MV, hospital discharge and viral clearance. These results provide a comprehensive comparison of current COVID-19 treatments and shed new light on further research and discovery of potential COVID-19 treatments.
Introduction
In the first year of the COVID-19 pandemic, the global research community produced a massive volume of clinical data at an unprecedented pace. However, much of this research consisted of small-scale, rapid randomized controlled trials (RCTs) that compared a single drug against the "Standard of Care" (SOC). This created two major challenges for clinicians and policymakers: the Comparison Gap, where active treatments were rarely compared head-to-head, and the Granularity Problem, where it was difficult to determine if benefits were specific to a drug or shared across a therapeutic class.
This study addresses these gaps by using a Bayesian Network Meta-Analysis (NMA) to synthesize both direct and indirect evidence from 96 RCTs involving over 51,000 patients, providing a cohesive framework for evaluating the efficacy of early-pandemic interventions.
Methodology
The researchers employed a Bayesian Network Meta-Analysis using a hierarchical framework. Unlike standard meta-analysis, NMA allows for indirect comparisons; if Drug A and Drug B are both compared to SOC, the NMA can statistically estimate the relative efficacy of Drug A versus Drug B.
Key methodological features included:
- Hierarchical Modeling: Individual drugs were nested within broader classifications (based on the WHO’s ATC/DDD system). This allowed the model to "borrow" information across similar drugs to estimate class-level effects, which is particularly useful when data for a specific drug is sparse.
- Statistical Estimation: Markov chain Monte Carlo (MCMC) algorithms were used to estimate Odds Ratios (OR) and 95% equal-tailed Credible Intervals (CrIs).
- Evidence Grading: The researchers applied the GRADE approach to rate the certainty of evidence (High, Moderate, Low, or Very Low), ensuring that the strength of clinical recommendations matched the quality of the underlying data.
Data Synthesis
The study performed an exhaustive search across six major databases for studies published between January 1 and December 19, 2020. The final analysis included 96 eligible RCTs (93 unique trials) covering 51,187 patients. The scope included both peer-reviewed papers and preprints, encompassing various disease severities and randomization methods.
Key Findings
Mortality and Mechanical Ventilation
Dexamethasone emerged as the primary "anchor" for treatment. It was the only intervention to show statistically significant reductions in both mortality (OR 0.85) and the need for mechanical ventilation (OR 0.68) with a moderate level of certainty. Additionally, Lopinavir/ritonavir showed a significant reduction in mortality (OR 0.88), while Recombinant human GCSF showed a substantial reduction in the risk of requiring mechanical ventilation (OR 0.20).
Hospital Discharge and Recovery
Several treatments significantly improved hospital discharge rates compared to SOC:
- Interferon beta: OR 2.15
- Baricitinib plus Remdesivir: OR 1.75
- Tocilizumab: OR 1.40
- Remdesivir: OR 1.37
The success of the baricitinib and remdesivir combination highlights the potential power of multi-pathway therapeutic approaches over monotherapies.
Viral Clearance vs. Clinical Outcomes
Convalescent plasma (OR 2.28) and Nitazoxanide (OR 1.72) were the top performers for viral clearance. However, the study noted a critical distinction: strong performance in biological markers like viral clearance did not always translate to significant reductions in mortality, emphasizing the need to prioritize clinical recovery metrics in trial design.
Discussion and Limitations
While the NMA provides a comprehensive map of treatment efficacy, several limitations persist:
- Risk of Bias: Approximately two-thirds of the included studies were rated as having a high risk of bias, often due to a lack of blinding in urgent pandemic conditions.
- Network Sparsity: Most trials compared a drug to SOC rather than to another active drug, which can limit the precision of indirect estimates.
- Intransitivity: Differences in patient baselines—such as varying levels of disease severity across different trials—can complicate indirect comparisons.
Conclusion
The Bayesian NMA confirms the superiority of several treatments, most notably dexamethasone and certain antiviral/immunomodulator combinations, over the standard of care. As more double-blind RCTs are published, these models should be updated to refine estimates and resolve discrepancies found in early-pandemic data. This systematic framework provides a vital tool for navigating the complex landscape of COVID-19 therapeutics.
Summary updated