Association between convalescent plasma and the risk of mortality among patients with COVID-19: a meta-analysis

Background: Convalescent plasma (CCP) has been used for treating some infectious diseases; however, the efficacy of CCP in coronavirus disease 2019 (COVID-19) remains controversial. The aim of this research was to assess the efficacy of CCP as an adjunctive treatment in COVID-19 patients. Methods: Embase, PubMed, Web of Science, Cochrane and MedRix were searched for potentially relevant articles. All included papers were assessed for the quality using modified Jadad scale and Newcastle-Ottawa scale for randomized controlled trial (RCT) and non – RCT, respectively. We used a Q test and Egger test to assess the heterogeneity and publication bias among studies, respectively. Mortality rates between patients treated with standard treatment and standard treatment with CCP were compared using a Z test. Results: A total of 12 papers consisting of three cross-sectional studies, one prospective study, five retrospective studies, and three RCT studies were included in our analysis. Of them, a total of 1,937 patients treated with CCP and 3,405 patients without CCP were included. The risk of mortality was 1.92-fold higher in patients without CCP compared to patients treated with CCP (OR: 1.92; 95%CI: 1.33, 2.77; p=0.0005). In severe COVID-19 sub-group analysis, we found that patients without CCP had a 1.32 times higher risk of mortality than those treated with CCP (OR: 1.32; 95%CI: 1.09, 1.60; p=0.0040). Conclusions: CCP, as adjunctive therapy, could reduce the mortality rate among COVID-19 patients.

Any reports and responses or comments on the article can be found at the end of the article.

Introduction
The management of coronavirus disease 2019 (COVID-19) remains challenging.While the guideline for the management of COVID-19 has been established, [1][2][3] some reports still reported high mortality rate among COVID-19 patients. 4,57][8] Moreover, the findings from the World Health Organization (WHO) solidarity trials also failed to clarify the benefits of hydroxychloroquine, remdesivir, interferon, and lopinavir in the management of COVID-19. 8Therefore, new approaches to COVID-19 management are required.
Convalescent plasma (CCP), an immunological therapy, is suggested to have promising efficacy for managing several infectious diseases. 9CCP, a strategy of passive immunization, was first introduced by von Behring and Kitasato in 1890.
Initially, it was used to treat diphtheria and other infectious diseases such as scarlet fever and pertussis. 10Moreover, due to its good efficacy, this therapy was also used for the management of Ebola, severe acute respiratory syndrome (SARS), and Middle East respiratory syndrome (MERS). 11In patients with MERS, SARS, and Ebola, the clinical improvement and reduced mortality rate were observed in patients receiving CCP compared to patients without CPP. 12However, the efficacy of CCP against COVID-19 is conflicting.Furthermore, the previous meta-analyses resulted in inconclusive findings due to the lack of structured methodology.Therefore, a holistic meta-analysis is needed to provide insight into the clinical efficacy of CCP for the management of COVID-19.

Study design
A systematic review and meta-analysis covering the period July 2020 -December 2020 was conducted to assess the efficacy of CCP as an adjunctive treatment in COVID-19 patients.Studies from prominent bibliographic databases were searched, and the protocols followed the guideline from Preferred Reporting Items for Systematic Review and Metaanalysis (PRISMA). 13

Eligibility criteria
Relevant articles were assessed for inclusion and exclusion criteria before the final analysis.Our analysis included articles with the following criteria: (1) observational or randomized controlled trial studies; (2) providing sufficient data of COVID-19 diagnosis methods; and (3) well-identified methodologies represented with Newcastle-Ottawa Scale (NOS).Case reports, case series, letters to the editor, reviews, commentaries, low method quality, and those with pre-post test comparison were excluded.

Search strategy and data extraction
Relevant studies in four bibliographic databases (Embase, PubMed, Web of Science, and Cochrane) and a preprint database MedRix were searched as of 2 December 2020.The searches limited to English only using Medical Subjects Heading: ("COVID-19" OR "SARS-CoV-2") AND ("convalescent plasma" OR "serotherapy" OR "hyperimmune globulin therapy" OR "convalescent plasma treatment").A reference list of the relevant articles was also retrieved for additional references.If a duplicate publication was found, the article with the larger sample size was included.Information of: (1) name of the first author; (2) year of publication; (3) country of origin; (4) sample size of cases and controls, (5) CCP administration, and (6) mortality rate were collected from each article.Search strategy and data extraction were conducted by three independent investigators (MI, AAA & YP) using a pilot form.Disagreements were resolved in group discussions through a consensus approach.Before collecting the data, the investigators performed a discussion to define the study variables and the study protocols, and the understanding among the investigators was assessed using kappa test.

Assessment of the methodology quality
All included papers were assessed for the quality using modified jadad scale for randomized controlled trial (RCT) and Newcaste-Ottawa scale for non-RCT. 14The quality of the articles was classified as low, moderate, and high quality.
Articles with low quality were excluded from our analysis.The assessment was carried out by three independent investigators (MI, AAA & YP), and when there was a discrepancy among the investigators, a discussion was performed with a senior researcher (JKF).

Outcome measure
The primary outcome measure was all causes of mortality among COVID-19 patients treated with and without CCP.The predictor variable was COVID-19 patients treated with CCP.A sub-group analysis was conducted based on the severity of COVID-19 patients treated with CCP (i.e.mild and severe).

Statistical analysis
The association between CCP and the reduction of the risk of mortality among COVID-19 patients was assessed using a Z test.Before assessing the association, the potency of bias and heterogeneity was assessed.To assess the risk of bias, an Egger test was employed to calculate tau-squared, and a p-value of less than 0.05 indicates that the potency of bias was found.A Q test was used to assess the heterogeneity among the included papers.The p-value of less than 0.10 was considered that heterogeneity across the studies was found, and the correlation was therefore determined using a randomeffect model; otherwise, a fixed-effect model was employed.All analyses were carried out using Review Manager (Revman Cochrane, London, UK) version 5.3, and the cumulative calculation was presented using a forest plot.

Studies selection and baseline characteristics of the studies
A total of 1,143 papers were identified, and 1,105 papers were excluded because they had irrelevant topics.A total of 38 papers were included for review in full-text, and 26 additional papers were excluded because of review, pre-post test model, commentary, and low-quality papers.6][17][18][19][20][21][22][23][24][25][26] The article selection flowchart is depicted in Figure 1, and the study characteristics are presented in Table 1.

CCP efficacy against COVID-19
A total of 1,937 patients treated with CCP and 3,405 patients without CCP, collected from 12 papers, were included in our analysis.Data suggest that COVID-19 patients without the CCP had a 1.92-fold higher risk of mortality than patients treated with the CCP (OR: 1.92; 95%CI: 1.33, 2.77; p = 0.0005) (Figure 2A).A sub-group analysis among severe COVID-19 patients who were treated with CCP was conducted.This sub-group consisted of nine papers with 1,458 patients treated with CCP and 2,706 patients without CCP.The data revealed a 1.32-fold higher risk of mortality in COVID-19 patients without CCP compared to patients treated with CCP (OR: 1.32; 95%CI: 1.09, 1.60; p=0.0040) (Figure 2B).

Heterogeneity and potency of bias across the studies
The analysis revealed evidence of heterogeneity in total case of COVID-19.Therefore, a random-effect model was applied to assess the association between CCP and the risk of mortality among COVID-19 patients.In the severe COVID-19 sub-group, we found no heterogeneity, and we used a fixed-effect model to evaluate the correlation.Our analysis using an Egger test found no publication bias in both the total and the severe COVID-19 sub-group (Funnelplot is provided in supplementary file).

Discussion
Our data suggest that CCP treatment associated with a reduction of mortality both in all cases and severe COVID-19 patients.
8][29][30][31][32] The theory underlying the mechanism of CCP in COVID-19 patients remains open to controversy.Briefly, plasma transfer is the potential aspect that bridges the CCP and the reduced risk of mortality in COVID-19 patients.Plasma consists of various immunity components, including antibodies, anti-inflammatory cytokines, clotting and or anti-clotting factors, albumin, and protein C and S. 33,34 It is believed that CCP in COVID-19 may modulate the immune response through antiviral effects and has immunomodulatory effects. 35ntiviral effects of CCP may occur through neutralizing antibodies, and it was reported that IgG and IgM anti-severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) were the primary isotype antibodies identified from COVID-19 patients treated with CCP. 36This humoral immune response may inhibit protein S of SARS-CoV-2. 37Thereafter, they may exert the protective effects against COVID-19.The immunomodulatory effects of CCP may occur through the neutralization of cytokines and complements. 35,38These effects may inhibit the overactive immune system, including cytokine storm, complement activation, and hypercoagulable state regulation. 39These mechanisms may be responsible for causing clinical improvement of COVID-19 patients.Of them, it was considered that immunoglobulin transfer is the essential factor in modulating the protective effect of CCP. 40In SARS and influenza, it was reported that immunoglobulin transfer plays a vital role in governing clinical improvement. 9,11Moreover, in MERS, the CCP administration with the titers of antibodies 1:80 provided a significant immune response, and the titers of antibodies 1:40 did not provide a similar response. 41Additionally, in Ebola, MERS, and SARS, the antibodies from the CCP may bind to the CD4 binding site on the viral envelope, and therefore may reduce the viral load and the risk of infection of the new cells. 42It was also supported by previous studies that antibody titers from CCP donors also governed the clinical improvement of COVID-19 patients treated with CCP, 43,44 suggesting that antibody transfer might influence the outcomes of clinical improvement.
8][29][30][31][32] However, they had some significant limitations: (a) the systematic reviews involved had a small sample size while in our current study, we had a relatively larger sample size; (b) some studies did not perform meta-analysis calculations to synthesize the data 27,29 ; (c) some studies included several case reports and case series 28,29 in which should be excluded in the meta-analysis 13 ; (d) previous meta-analyses assessed the role of CCP in similar infectious diseases (SARS and influenza), and the results were implemented to the case of COVID-19 30,31 ; and (e) previous meta-analyses performed a mixed calculation where the data of the case vs. control model were combined with the data of pre-post intervention models, which might provide a high risk of bias due to the final effect that might be caused by other interventions. 29,32In the present meta-analysis, we only calculated the model of the case (standard treatment and CCP) vs. control (standard treatment only) and therefore might provide a better correlation.
In the present study, we emphasized that CCP provided good efficacy to reduce the risk of mortality among COVID-19 patients.Our findings might contribute to better management of COVID-19 patients, particularly to prevent the risk of mortality.It is expected that a medical council should elaborate on the standard procedures of CCP, including the dosage, donor criteria, side effects management, and post-intervention management.Since early administration of CCP provided  better clinical outcomes than those with later intervention, 45 the appropriate time of CCP administration should be determined, and further studies are warranted.
Several important limitations of this study should be discussed.Some confounding factors that might govern the final outcomes were not controlled, including the immunological status, the dosage of CCP, time of intervention, donor criteria, the titers of antibodies, comorbidities, and transmission area.The majority of the included papers were retrospective studies, and therefore a further meta-analysis of randomized-controlled trials with a bigger sample size might provide a better conclusion.

Conclusion
Administration of the CCP is associated with a lower risk of mortality among COVID-19 patients compared to those without CCP, and this highlights its potency to be used for the treatment of COVID-19.However, studies are warranted to formulate the dosage, time of intervention, donor criteria, and the titers of antibodies to optimize the effects.
ottawa; Jadad scale, and not jadad) in the whole text, including the abstract.
There are two dots in the end of the second sentence of the results section in the abstract.2.
In the introduction, the following sentence could be expanded: "the mortality rate of COVID-19 remains increased over the periods".Please, focus on the literature you cited and be specific as possible.For instance, some reports show the opposite, that is, that treatment has improved so the mortality rate has decreased.The authors' very own findings point to lower OR for death among those treated with CCP.

3.
Albeit it is true that the solidarity trial showed little efficacy of the drugs mentioned in the introduction, a robust finding was found in severe patients that were treated with dexamethasone.

4.
Would you please give more information on clinical improvement of the diseases treated with CCP cited in the second paragraph (introduction, ref. n. 12)?

5.
The following sentence needs a few supporting references: "However, the efficacy of CCP against COVID-19 is conflicting".

6.
I would suggest the authors to remove the word "holistic" when presenting their goal.7.
When presenting the eligibility criteria, I was wondering why pre-post comparisons were excluded.Would be worth it to justify this choice.

8.
I would change this sentence "If the disagreement was found, we performed a discussion to resolve the disagreement" into "Disagreements were resolved in group discussions".Also, did you ask for an external judge to assist in the discussion of disagreements?Or a consensus approach was used?This is unclear.

9.
There is a typo in the section describing the outcome variable (e.i.mild and severe).Please, correct to "i.e."; Also, the outcome variables could be more clear.Do you mean "the number or COVID-19 patients treated with CCP"? 10.
In the section "Heterogeneity and potency of bias across the studies", please be clear/complete in the following sentence "a random-effect model was applied to assess the Association…".Association between what?

11.
This sentence needs more information: "In the present meta-analysis, we only calculated the model of the case (standard treatment and CCP) vs. control (standard treatment only) and therefore might provide a better correlation".Would you please justify why your approach provides better 'correlation'?12.
Please, double check the comprehensiveness of Table 2.The table is supposed to present previous meta-analyses and some of the studies included there were judged to not calculate data synthesis.However, I see that some studies are only systematic reviews, such as Rajendran and collaborators.Reviewer Expertise: Health-related issues and systematic review and meta-analysis methodology.
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Iran
This meta-analysis aimed to investigate the association between convalescent plasma and the risk of mortality among patients with COVID-19.This is a great area of research but, in my opinion, the manuscript needs some major revisions as follows: Please indicate the name of the searched databases in the abstract section.

○
Please indicate the quality appraisal checklist in the abstract section.

○
Please indicate the method for investigating the heterogeneity and publication bias in the abstract section.

○
Please indicate the type of the included studies in the abstract, results, and table 1.

○
What is your reason for selecting this period for your search?○ Search strategy is not complete.

○
Please restructure the method section following the PRISMA item as you claim.

○
There are some problems regarding figure 1. Was there no duplicate record?It does not make much sense.

○
It needs to mention the type of the included studies and then we can speak about the quality appraisal checklist.it seems that NOS is not sufficient.NOS is for nonrandomized studies.

○
Please indicate inter-rater reliability between three raters.

○
Result section, please add a subheading for "study characteristics" based on PRISMA and first write a brief and then refer to table 1.Also, add the type of the control in column control.
○ Figure 1 has some problems.Your study is a meta-analysis.How were 11 studies included in qualitative synthesis?Which qualitative synthesis?○ There is a 6 % difference between I2 for A and B, what is your rationale for selecting the fixed or random-effect model?Please provide a reference for your claim.Please add the details in the method section.

○
Please add the funnel plot as a supplement.

○
Please remove table 2 from the discussion and also discuss the added value of your study regarding the existing meta-analysis.What is the novelty of your work?Reviewer Expertise: Health-related issues and systematic review and meta-analysis methodology.
I confirm that I have read this submission and believe that I have an appropriate level of expertise to state that I do not consider it to be of an acceptable scientific standard, for reasons outlined above.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com

Figure 1 .
Figure 1.A flowchart of study selection in our meta-analysis.

13 .
Morteza Arab-ZozaniSocial Determinants of Health Research Center, Birjand University of Medical Sciences, Birjand, Iran Thank you for clearly addressing my previous comments.In my opinion, the manuscript is acceptable in this way.Is the work clearly and accurately presented and does it cite the current literature?YesIs the study design appropriate and is the work technically sound?YesAre sufficient details of methods and analysis provided to allow replication by others?YesIf applicable, is the statistical analysis and its interpretation appropriate?YesAre all the source data underlying the results available to ensure full reproducibility?YesAre the conclusions drawn adequately supported by the results?YesCompeting Interests: No competing interests were disclosed.

○○
The conclusion is very optimistic.How did you come to that conclusion based solely on mortality?Is the work clearly and accurately presented and does it cite the current literature?PartlyIs the study design appropriate and is the work technically sound?PartlyAre sufficient details of methods and analysis provided to allow replication by others?PartlyIf applicable, is the statistical analysis and its interpretation appropriate?PartlyAre all the source data underlying the results available to ensure full reproducibility?YesAre the conclusions drawn adequately supported by the results?NoCompeting Interests: No competing interests were disclosed.