1. Amongst patients with severe COVID-19 pneumonia, there was no difference in 30-day all-cause mortality, ventilation or cure in patients receiving convalescent plasma (CP) plus standard therapy versus standard therapy alone.

2. CP treatment was associated with more adverse events, including 5 instances of respiratory failure requiring treatment interruption.

Patients who have recovered from SARS-CoV-2 infection produce convalescent plasma (CP). CP has been proposed as a therapeutic for COVID-19 patients to reduce disease progression and death. This study tests the efficacy of high-titre CP in treating patients infected with severe SARS-CoV-2 pneumonia.

This multi-site study, the Transfusion of Convalescent Plasma for the Early Treatment of Patients With COVID-19 (TSUNAMI) trial, randomized adult COVID-19 pneumonia patients to receive either standard therapy (ST) alone or CP+ST. Respiratory failure (partial pressure of oxygen–to–fraction of inspired oxygen (PaO₂/FiO₂) of <150mmHg), mortality, ventilation, viral cure, duration of hospital stay, and adverse events (AEs) were tracked for 30 days.

Patients in the CP group were treated an average of 7.7 days after symptom onset (n=241), most commonly with 2 doses of CP (mean CP titre = 226.3), but were otherwise comparable to the control group (n=246). Comparable proportions of patients experienced respiratory failure or death between those who did and did not receive CP (25.5% vs 28.0%, respectively) and 30-day all-cause mortality (6.1% vs 7.9%, respectively). One centre had worse respiratory outcomes in the CP group than the ST group.

The rate of need for ventilation, COVID-19 cure, and duration of hospitalization were the same in the CP and ST groups.

AEs were more common with CP treatment (5.0% vs 1.6%), including 5 instances of respiratory failure requiring CP to be stopped. One limitation of this study is that the subgroup analysis (pneumonia severity, disease onset, and serology) did not have a sufficient sample size. It may be that the 30-day follow up was insufficient to see the benefits of CP. Overall, this is a highly robust randomized controlled trial that suggests that patients with severe COVID-19 pneumonia treated with CP are no less likely to progress to respiratory failure or death than ST alone.

This study recruited adult patients with reverse transcriptase-polymerase chain reaction-confirmed COVID-19 and radiographically-confirmed pneumonia of moderate to severe level (baseline PaO₂/FiO₂ = 200-350mmHg) within the first 10 days of their symptom onset from 27 sites in Italy. Patients were excluded if they were pregnant or lactating, could not receive blood products, got immunoglobulins or interleukin 1, 6 or Janus kinase inhibitors within the month, or were non-invasively or invasively ventilated.

Computer-based randomization to either ST alone or 1 – 3 doses of 200 mL CP over 2 hours daily in addition to ST was stratified by site. The primary outcome was 30-day respiratory failure (PaO₂/FiO₂ <150mmHg) or mortality.

The 241 patients that received CP+ST (mean age = 65, 63.9% men) did not differ on baseline characteristics (age, sex, ethnicity, body mass index, time from initial symptoms to randomization, comorbidities, baseline PaO2/FiO2 ratio, Sequential Organ Failure Assessment score, and previous treatments) from the 246 patients that received ST alone (mean age = 63.5, 64.2% men). An average of 7.7 days passed between symptom onset and CP treatment. 15.1% of patients received only 1 dose of CP, 75.4% received 2, and 9.5% received 3 with a mean titre of CP of 226.3.

Respiratory failure or death occurred in 25.5% of patients who received CP+ST and 28.0% who received ST alone (OR = 0.88, 95% CI = 0.59-1.33, P = 0.54), suggesting that CP is ineffectual at decreasing the rate of respiratory failure and death. These results were consistent with subgroup analysis. In patients who had a PaO₂/FiO₂ of ≥300mmHg at baseline, respiratory failure or death occurred less in CP+ST patients than ST alone (11.6% vs 21.9%), but not statistically significantly less (OR = 0.47, 95% CI = 0.19-1.18). At one centre the patients that received CP had worse outcomes than those receiving only ST. The 30-day all-cause mortality rate was non-significantly lower in the CP group than the control (6.1% vs 7.9%, P = 0.43).

Rates of requiring mechanical ventilation or death, serological cure from COVID-19, and duration of hospitalization were the same in the CP and ST groups.

More AEs occurred in the CP group than the ST group (5.0% vs 1.6%, P = 0.04), including 5 AEs of worsening respiratory failure requiring treatment interruption, 1 with fever and another with rash, occurred.