ObjectiveTo systematically review the efficacy of convalescent plasma (CP) in the treatment of coronavirus disease 2019 (COVID-19). MethodsPubMed, EMbase, The Cochrane Library, VIP, WanFang Data and CNKI databases were electronically searched to collect randomized controlled trials (RCTs) on the efficacy of CP in the treatment of COVID-19 from inception to September 15th, 2021. Two reviewers independently screened literature, extracted data and assessed the risk of bias of included studies; then, meta-analysis was performed using RevMan 5.4.1 software. ResultsA total of 16 RCTs involving 15 301 patients were included. The results of meta-analysis showed that CP treatment did not reduce mortality compared with no-placebo (RR=0.99, 95%CI 0.93 to 1.05, P=0.63) or normal saline (RR=1.06, 95%CI 0.60 to 1.86, P=0.84). However, compared with standard plasma, the mortality of CP group was lower (RR=0.59, 95%CI 0.37 to 0.95, P=0.03). In addition, compared with no-placebo or normal saline, CP treatment could not improve the clinical condition at 28-30 days, reduce mortality at early treatment and in patients without invasive mechanical ventilation when randomized. ConclusionCurrent evidence shows that compared with no-placebo or normal saline, CP does not reduce mortality in patients with COVID-19. However, when the disease progresses to the point where standard plasma is required, CP may reduce mortality. In addition, use of CP in patients with early or non-critical COVID-19 failed to reduce mortality. Due to limited quality and quantity of the included studies, more high-quality studies are required to verify the above conclusion.