‘Treatment of COVID-19 with convalescent plasma’ – Author’s reply

To the Editor,

We would like to thank Gazzaruso and colleagues for their interest in our review of convalescent plasma (CP) for treating coronavirus disease 19 (COVID-19) [1]. The authors describe an intriguing potential link between CP and thromboembolic therapy, suggesting that CP may exert its therapeutic effects through anticoagulant properties, specifically via antithrombin replacement. This new theory provides an interesting angle on the mechanism of CP as a therapeutic agent and may lend insight into future research. However, the gaps in the literature are too broad to make clinical recommendations.

The authors state that “mortality in patients with COVID-19 is mainly attributed to thromboembolic complications” and that “despite the anticoagulation, usually with heparin, mortality due to thromboembolic events is high”. While such complications have indeed been described in COVID-19 patients, the cited articles are not so definitive as to say that they are a primary cause of death. The first article cited describes only a few studies that have investigated the burden of thromboembolic complications in COVID-19 patients, with an incidence of venous thromboembolism (VTE) between 1% in ward patients and 35% in intensive care patients [2]. In addition, most of these studies had small sample sizes, and the clinical significance of such thromboembolic events, such as whether they increase mortality, remains uncertain. The second article demonstrates a relationship between low antithrombin and mortality, but this cannot be assumed to be a result of fewer thromboembolic events [3].

In addition, the authors describe that antithrombin deficiency may explain why some patients do not respond to heparin therapy. The authors cite that approximately 25% of patients with severe COVID-19 have antithrombin deficiency [4]. In addition to the fact that even the highest estimates of VTE incidence in COVID-19 are only 35%, this leaves us with a small proportion of patients who could theoretically benefit from antithrombin replacement, and all this is in the absence of clinical data to demonstrate that VTE causes a significant burden of mortality.

There are few data to either support or refute this hypothesis at present. The lack of certainty in clinical data is no fault of the authors and simply reflects the nature of this ever-evolving pandemic. Similar to the studies cited in our review, most of the data available to us are derived from observational studies with small sample sizes. The suggestion that CP could provide antithrombotic effects for those patients who are antithrombin deficient and thus potentially refractory to heparin is indeed intriguing. As suggested, it does seem theoretically reasonable for future studies to assess the antithrombin status in CP recipients. However, given that CP therapy is already fraught with practical challenges, prioritizing improved knowledge of the chain of assumptions underlying this theory may prove more beneficial, including the assumptions that (a) thromboembolic events contribute significantly to COVID-19 mortality, (b) heparin reduces this risk, and (c) antithrombin deficiency is responsible for the lack of efficacy in those patients for whom heparin is ineffective.

Transparency declaration

This work was supported by the Canadian Institutes of Health Research (CIHR) grant Canadian 2019 Novel Coronavirus (COVID-19) Rapid Research Funding Opportunity (No. OV3-170627) Therapeutics to HB. DJW has nothing to disclose.

Editor: L. Leibovici