LAUSR

One of the limited number of choices for the treatment of coronavirus disease 2019 (COVID-19) is to administer plasma, containing neutralizing anti-viral antibodies, from donors who have recovered from the disease. This so-called convalescent plasma (CP) is a form of passive immunotherapy that has been used for the treatment and prevention of various infectious diseases for more than a century. Recently published reports indicate that CP provides a clinical benefit when it is given early in the course of COVID-19 and has a high titer of neutralizing antibodies. Despite the only recent use of CP in the treatment of COVID-19, several systematic reviews and meta-analyses have already been published. These publications have reached conflicting conclusions, probably because of the heterogeneity of the design of the studies included (e.g., peer-reviewed publications, preprints, randomized controlled trials [RCTs], and non-randomized studies), the patients' baseline characteristics (e.g., severity of COVID-19, time since symptom onset), number of doses transfused, and plasma titers. Taking this heterogeneity into consideration and incorporating data from a newly published systematic review, we reappraised the mortality outcome according to the amount of antibody in the therapeutic CP units. We were not able to undertake further subgroup analyses because the data reported from primary studies were limited and not stratified uniformly or clearly for other variable of interest (e.g., baseline characteristics of patients). For the mortality subgroup analysis, we selected three RCTs reporting the use of high-titer CP. The study by Simonovich et al. was not included because not reporting outcomes stratified by antibody titer. The study weight was calculated using the Mantel– Haenszel method and statistical heterogeneity was assessed using the I statistic. Measures of treatment effect were risk difference (RD) together with 95% confidence interval (CI). All calculations were conducted using Review Manager, version 5.4 software. The results of the analysis are shown in Figure 1. Treatment with high-titer CP decreased all-cause mortality significantly (RD, -0.06; 95 % CI, -0.12/0.00; p=0.05); the assumed risk of mortality in the control group was 11.5%, and the corresponding risk in the CP group was 5.8%. These data are in contrast to the results of the more recent systematic review, in which the analysis was performed not considering the CP titer in the primary studies, and included also data from six studies in which variable CP titers (low, unclear, or with no minimum titer cutoff) were given. In the overall analysis of the 10 RCTs included in the systematic review, CP transfusion was not associated with a decrease in all-cause mortality (RR, 1.02; 95% CI, 0.92/1.12; p = 0.68). We used the principles of the Grading of Recommendations Assessment, Development and Evaluation system to assess the quality of the body of evidence on the outcome analyzed (all-cause mortality), and constructed a ‘Summary of findings’ table (Table 1) using the Review Manager software. The certainty of a body of evidence involves consideration of within-trial risk of bias, directness of evidence, heterogeneity, precision