DISCUSSION
The study of the therapeutic benefits of plasma transfusion of a cured person from infectious diseases began in the 20th century.2 As new antibiotics, antiviral agents, and vaccines are developed, administration of convalescent plasma is not a common treatment, but it can be still an important treatment in the absence of specific treatment of new infectious diseases.3 Over the decades, convalescent plasma has proved its effectiveness as a potential treatment in patients with MERS-CoV,4 H1N15 and H5N1 avian flu,5 and SARS-CoV.6 A systemic review and meta-analysis to evaluate the clinical effects of convalescent plasma shows a statistically significant reduction of mortality.1 In this context, convalescent plasma can be a promising treatment option for severe COVID-19 patients.
As can be seen in the two cases, both received lopinavir/ritonavir and hydroxychloroquine but showed persistent fever, rapidly aggravated hypoxemia and progressive bilateral infiltrations in accordance with the criteria of severe ARDS. After convalescent plasma infusion, the patients showed improved oxygenation and chest X-rays with decreased inflammatory markers and viral loads.
Intravenous methylprednisolone was started just before the convalescent plasma infusion in both cases. We did not use corticosteroids from the beginning as a routine treatment. Current guidelines recommend that systemic corticosteroids should not be given routinely for the treatment of COVID-19 due to the lack of evidence of its clinical efficacy on mortality reduction.78 However, we decided to start corticosteroids when the patients' condition rapidly deteriorated to ARDS. Methylprednisolone was administered one day and two days before the plasma infusion in case 1 and case 2, respectively. Serial laboratory and oxygenation parameters showed rapid improvement right after the corticosteroid administration even before the convalescent plasma infusion.
ARDS is partly caused by cytokine storm and host immune responses.7 Autopsy of patients dying from COVID-19 shows diffuse alveolar damage with exudate and inflammation very similar to those seen in SARS and MERS-CoV infections.9 Theoretically, systemic corticosteroids may have a role to dampen excessive lung damage due to inflammatory responses.10 The recent article about risk factors associated with ARDS and death among COVID-19 patients showed that treatment with methylprednisolone might be beneficial to reduce the risk of death for patients developing ARDS.11 However, corticosteroids are also thought to inhibit proper immune responses and viral clearance and delay antibody production.1213
Convalescent plasma infusion might play a role in the coexistence of benefits and concerns of corticosteroid use. Antibodies contained in the convalescent plasma will suppress viruses.14 In an animal study, passively transferred antibodies can provide total protection as well as the maintenance of high levels of antibody titer until the host's immune responses could be increased to clear the viral infection. Besides, in vivo studies showed that the effects of neutralizing antibodies were not only limited to viral clearance, but also included acceleration of infected cell clearance.15
In our cases, the viral load estimated by Ct values showed an increasing trend just before plasma infusion but began to decrease right after the use of convalescent plasma. Although improvement of inflammatory marker and oxygenation could be contributed to the combined use of corticosteroid, decreased viral load of SARS-CoV-2 might mean the effectiveness of convalescent plasma in the treatment of COVID-2.
Convalescent plasma was administered after 22 days from the onset of symptoms in Case 1, and 7 days in Case 2, respectively. Because these are not in the early phase of the disease, it is difficult to determine clearly that the decrease in the viral load shown in both cases is due to convalescent plasma or natural pathology of COVID-19. Other studies about viral kinetics of COVID-19 show naturally reducing viral titers after 7–10 days from onset in most patients.1617 However, Liu and colleagues reported that severe patients requiring intensive care unit admission due to COVID-19 had high viral load for a longer period than in mild patients.16 Both our cases presented severe ARDS and the viral loads were in increasing trend at the time of plasma infusion regardless of the date of onset.
In Case 2, the patient showed lymphopenia from day 1, and it persisted even after clinical improvement with corticosteroid use. When the convalescent plasma was administered on day 6, lymphocyte count immediately rose to normal level (from 0.52 × 103/µL to 1.21 × 103/µL) and then remained in the normal range. Patients with severe COVID-19 pneumonia and ARDS also presented with lymphopenia in other studies.11 Some authors hypothesized that continuous and gradual increases in lymphocyte count might be required for immunity against SARS-CoV-2 infection.11 In SARS patients, lymphopenia existed at the onset of illness and persisted until the recovery period.1819 These findings were consistent with the recovery of lymphocyte count with clinical improvement of Case 2 after the use of convalescent plasma.
We could not assess neutralizing antibody titers from the convalescent plasma. Plasma with high neutralizing antibody titers is likely to be available from the patients in the convalescent phase recovered from severe infection.20 To use plasma for treatment, a neutralization test is suggested as the optimal assay for assessing proper donor or plasma. However, some studies showed that ELISA IgG correlates well with neutralization titers in MERS cases so that it might be a suitable screening test for plasma donation.2021 In our cases, donors presented bilateral pneumonia in the course of COVID-19 and both showed positive results in the ELISA IgG test for SARS-CoV-2.
There are still limitations for the use of convalescent plasma. Scientific evidence is insufficient due to the lack of large-scale clinical trials that may be representative of the target populations. Second, the number of antibodies administered to each patient was not standardized. Finally, convalescent plasma usually proceeds with other treatments, such as antiviral agents and steroids, which can affect the relationship between convalescent plasma and antibody, confounding the results.
Despite the limitations, our cases suggest that convalescent plasma from patients who have recovered from COVID-19 infection might be an additional option to treat patients without causing any severe adverse effects. Also, when used with systemic corticosteroids, we might expect the possibility of reducing excessive inflammatory response by corticosteroids as well as promoting the reduction of viral loads by convalescent plasma simultaneously. Further well-designed studies are needed to demonstrate the efficacy and safety of convalescent plasma transfusion in COVID-19 patients.