Comparative Effectiveness and Safety of 32 Pharmacological Interventions for COVID-19

0
0
0

Comparative Effectiveness and Safety of 32 Pharmacological Interventions Recommended by Guidelines for Coronavirus Disease 2019: A Systematic Review and Network Meta-Analysis Combining 66 Trials

The global pandemic of coronavirus disease 2019 (COVID-19) has posed significant challenges to healthcare systems worldwide, with over 175.8 million confirmed cases and 859,130 deaths reported by June 2021. Despite urgent efforts to identify effective treatments, most pharmacological interventions recommended by clinical guidelines lack robust evidence. This study conducted a comprehensive network meta-analysis (NMA) to evaluate the comparative efficacy and safety of 32 pharmacological interventions across 66 randomized controlled trials (RCTs), aiming to inform clinical decision-making and guideline development.

Study Design and Methodology

The analysis included RCTs from Medline, Embase, Cochrane Library, clinicaltrials.gov, and three Chinese databases (SinoMed, CNKI, WanFang) up to July 20, 2020. Eligible trials involved patients with confirmed or suspected SARS-CoV-2 or SARS-CoV infections and compared pharmacological interventions such as antivirals, antibiotics, corticosteroids (COR), chloroquine (CQ), traditional Chinese medicine (TCM), and others against standard of care (SOC) or placebo. Primary outcomes included mortality, cure rate, viral negative conversion (VNC), and overall adverse events (OAEs). Secondary outcomes covered specific adverse events (e.g., diarrhea, acute kidney injury) and clinical parameters like hospitalization duration.

A Bayesian random-effects NMA was performed to synthesize direct and indirect evidence, with odds ratios (ORs) and weighted mean differences (WMDs) calculated for dichotomous and continuous outcomes, respectively. The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework was used to assess evidence quality.

Key Findings on Efficacy

Mortality

Among 18,881 patients evaluated, TCM and COR demonstrated significant mortality reduction compared to SOC. TCM showed the strongest effect (OR = 0.34, 95% CI: 0.20–0.56; moderate-quality evidence), followed by COR (OR = 0.84, 95% CI: 0.75–0.96; low-quality evidence). In contrast, high-dose chloroquine (CQ_HD) increased mortality risk (OR = 3.20 vs. SOC; 95% CI: 1.18–8.73; low-quality evidence). Remdesivir and other antivirals did not significantly reduce mortality.

Cure Rate

TCM and COR also improved cure rates. TCM nearly doubled the likelihood of cure compared to SOC (OR = 2.16, 95% CI: 1.60–2.91; low-quality evidence), while COR showed a modest benefit (OR = 1.17, 95% CI: 1.05–1.30; low-quality evidence). Antivirals, including lopinavir/ritonavir and remdesivir, did not significantly enhance cure rates.

Viral Negative Conversion

No intervention significantly improved VNC, highlighting a critical gap in therapies that accelerate viral clearance.

Key Findings on Safety

Overall Adverse Events

TCM and remdesivir at 10 mg/day (REM_10) were associated with fewer OAEs compared to SOC (TCM: OR = 0.52, 95% CI: 0.38–0.70; REM_10: OR = 0.31, 95% CI: 0.19–0.52). Conversely, CQ_HD (OR = 2.51 vs. SOC), interferons (IFN: OR = 2.69), and colchicine (COL: OR = 3.81) increased OAE risks.

Specific Adverse Events

  • Diarrhea: Increased risks were observed with COL (OR = 3.80 vs. SOC) and lopinavir/ritonavir (OR = 9.62). TCM reduced diarrhea incidence (OR = 0.43 vs. SOC).
  • Secondary Infections: TCM lowered the risk (OR = 0.33 vs. SOC), while COR showed no significant impact.
  • Hospitalization Duration: TCM, convalescent plasma (CON_PLA), and ribavirin/interferon combinations shortened hospitalization by 2.53–17.80 days compared to SOC.
  • Fever Resolution: TCM reduced fever duration by 1.03 days (WMD = −1.03, 95% CI: −1.09 to −0.97).

Evidence Quality and Methodological Considerations

The GRADE assessment revealed moderate-quality evidence for TCM’s mortality benefit but low or very low quality for most comparisons due to small sample sizes, methodological limitations (e.g., lack of blinding in 31.8% of trials), and heterogeneity. Transitivity and consistency assumptions were generally met, with no significant global inconsistency detected. Funnel plots indicated low publication bias risk for outcomes with ≥10 studies.

Clinical and Research Implications

The findings support using TCM and COR as adjunct therapies to reduce mortality and improve recovery in COVID-19 patients, particularly those with severe disease. However, the benefits of COR may vary by disease severity, warranting cautious use in mild cases. The harmful effects of CQ_HD emphasize the need for dose optimization and monitoring.

Despite the inclusion of 19,095 patients, limitations include sparse data for certain interventions, inability to perform subgroup analyses by disease severity, and reliance on low-quality RCTs. Future high-quality trials should prioritize head-to-head comparisons, individual patient data meta-analyses, and evaluation of long-term outcomes.

Conclusion

This NMA provides the most comprehensive evidence to date on COVID-19 pharmacological interventions. TCM and COR emerge as promising options for reducing mortality, while CQ_HD and some antivirals pose significant safety risks. Clinicians should weigh these findings against evidence quality and patient-specific factors. Urgent investment in large, rigorous RCTs is needed to address remaining uncertainties and optimize therapeutic strategies.

doi.org/10.1097/CM9.0000000000001672

Was this helpful?

0 / 0