Assessment of the Safety and Therapeutic Benefits of Convalescent Plasma in COVID-19 Treatment: A Systematic Review and Meta-Analysis

Abstract

Background: The coronavirus disease (COVID-19), caused by the Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2), prompted a global health crisis, with no available specific treatments. Convalescent plasma (CP) with neutralizing antibodies could be a promising therapeutic approach to reduce mortality. Objectives: To evaluate the therapeutic potential of CP for COVID-19 and to assess its safety and efficacy in reducing the patients' mortality. Methods: We retrieved clinical trial references from multiple Databases (e.g., PubMed, B-On, SCOPUS), for complete studies until November 26th 2020. We included Randomized controlled trials (RCT) and controlled non-randomized trials (CNRT), that assessed the efficacy of CP to treat hospitalized COVID-19 patients. Trials were included regardless of concomitant medications in the intervention's arms. Eleven trials met our eligibility criteria. This study was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. We defined a methodological protocol to extract and evaluate all pertinent baseline demographics and interventions' characteristics from trials. The primary outcomes were the safety profile of CP, measured by the type, frequency and severity of adverse events, and CP effectiveness in reducing mortality, measured by the number of deaths registered for this therapy. Results: We assessed 11 trials (5 RCT and 6 CNRT) with 3,098 participants, of whom 923 patients were treated with CP. Only 32 (3.5%) of the treated patients suffered adverse events (from which 9.4% serious transfusion-related adverse events). The overall mortality rates were significantly decreased by CP administration {risk ratio (RR) 0.71, p = 0.005, 95% confidence interval (Cl) [0.57-0.90]}, with low heterogeneity. In the sub-analysis by period of transfusion, CP transfusion within a week of hospitalization contributed to diminished mortality rate (RR = 0.71, p = 0.03, 95%Cl [0.53-0.96]). CP therapy also led to significantly reduced viral loads at 72 h after transfusion (RR = 0.61, p = 0.04, 95%Cl [0.38-0.98]), despite high heterogeneity due to disease severity. Conclusion: This meta-analysis established CP as a safe and potentially effective therapy for COVID-19, decreasing the mortality rates and promoting a swift viral clearance. Further studies are necessary to provide stronger evidence.

Keywords: COVID-19; clinical trials; convalescent plasma; meta-analysis; mortality; safety; viral clearance.

Figure 1

PRISMA Flow diagram illustrating the study selection process. Schematic illustration indicating the various steps from the Database search until the inclusion of the selected trials for this meta-analysis. The number of trials excluded, and the reason for exclusion are also described in this scheme. RCT, Randomized Clinical trials; CNRT, Controlled Non-Randomized Trials.

Figure 2

Forest plots for mortality, and the respective sub-analysis, following CP treatment. (A) For the overall mortality, the results showed a significantly decreased mortality rate in the CP intervention group in comparison to standard treatment. The overall heterogeneity levels are low; therefore, the results for this outcome reported are consistent. Nonetheless, a sensitivity analysis was performed by multiple sub-analysis. (B) Sub-analysis by study design (RCTs vs. CNRTs). (C) Sub-analysis by disease severity (critically, severely, moderately ill and miscallaneous of critical and severe patients). (D) Sub-analysis by administration period of CP (before or after 7 days of hospitalization). (E) Sub-analysis by antibody titers administered (<1:297 or ≥1:297). Different sizes of data markers correspond to the relative weight assigned in the pooled analysis. Diamond marker indicates the overall result.

Figure 3

Forest plots depicting viral clearance at 72 h after CP transfusion, and respective sub-analysis. (A) At 72 h after CP administration, the results showed a significantly reduced number of positive laboratory-confirmed SARS-CoV-2 infection in the CP intervention group. Nevertheless, the overall heterogeneity levels are high; therefore, we conducted a sub-analysis. (B) Sub-analysis by disease severity (Critical, Severe, and Moderate clinical status). Different sizes of data markers correspond to the relative weight assigned in the pooled analysis. Diamond marker indicates the overall result.

Figure 4

Forest plots for length of hospitalization analysis and sub-analysis assessing CP intervention and standard treatment. (A) Beginning at hospital admission, the hospitalization length results showed a significantly decreased period of hospital stay for the standard treatment. However, the overall heterogeneity levels are high; therefore, we conducted a sub-analysis. (B) Sub-analysis by disease severity (miscellaneous of critical and severe patients, and moderately ill patients). (C) Sub-analysis by study design (RCTs vs. CNRTs). (D) Sub-analysis by administration period of CP (before or after 7 days of hospitalization). The results are presented in mean and standard deviation (SD). Different sizes of data markers correspond to the relative weight assigned in the pooled analysis. Diamond marker indicates the overall result.

Figure 5

Forest plots assessing the need for invasive (A–C) and non-invasive (D,E) mechanical ventilation in CP intervention and standard treatment. (A) Concerning the need of invasive mechanical ventilation both groups showed no difference although a tendency favoring the CP treatment can be observed. The results also indicate moderate heterogeneity levels; therefore we conducted a sub-analysis. (B) Sub-analysis by disease severity (miscellaneous of critical and severe patients, and moderately ill patients). (C) Sub-analysis by study design (RCT vs. CNRT). (D) The analysis regarding the need of non-invasive mechanical ventilation showed that both groups presented no significant difference, although patients in the CP group had lower necessity of supplemental oxygen. The heterogeneity levels were null, thus a sensibility analysis by disease severity was performed (E). Different sizes of data markers correspond to the relative weight assigned in the pooled analysis. Diamond marker indicates the overall result.