Convalescent plasma therapy for COVID-19 patients: a protocol of a prospective meta-analysis of randomized controlled trials

Abstract

Background Coronavirus disease 2019 (COVID-19) is an infection with possible serious consequences. The plasma of recovered patients might serve as treatment, which we aim to assess in the form of a prospective meta-analysis focusing on mortality, multi-organ failure, duration of intensive care unit stay, and adverse events. Methods A systematic search was conducted to find relevant registered randomized controlled trials in five trial registries. A comprehensive search will be done continuously on a monthly basis in MEDLINE (via PubMed), Embase, Cochrane Central Register of Controlled Trials (CENTRAL), and Web of Science to find the results of previously registered randomized controlled trials. The selection will be done by two independent authors. Data extraction will be carried out by two other independent reviewers. Disagreements will be resolved by a third investigator. An update of the search of the registries and the first search of the databases will be done on the 21st of July. Data synthesis will be performed following the recommendations of the Cochrane Collaboration. In the case of dichotomous outcomes (mortality and organ failure), we will calculate pooled risk ratios with a 95% confidence interval (CI) from two-by-two tables (treatment Y/N, outcome Y/N). Data from models with multivariate adjustment (hazard ratios, odds ratio, risk ratio) will be preferred for the analysis. P less than 0.05 will be considered statistically significant. In the case of ICU stay, weighted mean difference with a 95% confidence interval will be calculated. Heterogeneity will be tested with I 2 , and χ 2 tests. Meta-analysis will be performed if at least 3 studies report on the same outcome and population. Discussion Convalescent plasma therapy is a considerable alternative in COVID-19, which we aim to investigate in a prospective meta-analysis.