A monthly roundup of key articles in other journals

Abstract

Convalescent plasma in patients admitted to hospital with COVID-19 (RECOVERY): a randomised controlled, open-label, platform trial. RECOVERY Collaborative Group*, Lancet 2021; 397:2049–59. Convalescent plasma transfusion does not improve mortality in patients hospitalized with COVID-19. The search for treatments that reduce mortality in COVID-19 has proven challenging. Convalescent plasma (CP) transfusion was granted emergency use authorization and widely implemented into clinical practice, but efficacy has not been established. The RECOVERY trial—the largest clinical trial for any indication of convalescent plasma—sought to address this need. RECOVERY was a randomized, controlled, openlabel, adaptive platform trial of convalescent plasma (qualified by EUROIMMUN ELISA) in patients hospitalized with COVID-19, conducted across 177 National Health Service hospitals in the United Kingdom. Patients were assigned to usual care (n = 5763) or to transfusion with two units of CP (n = 5795). Median time from symptom onset to assignment was 9 days. Outcomes were measured at 28 days with additional analyses planned for 6 months. The primary outcome, all-cause mortality at 28 days, was not different between groups: 1399 (24%) of the CP group and 1408 (24%) of the usual care group died (rate ratio 1.00, 95% confidence interval [CI] 0.93–1.07; p = .95). Combining the findings with data from previously reported trials, the mortality rate ratio was 0.98 (95% CI 0.91–1.06; p = .63). Mortality was higher in patients who were seronegative at baseline in both groups. Median time to discharge and risk of progressing to the prespecified composite outcome of invasive mechanical ventilation or death were also comparable between groups, but there was some indication that patients who were seronegative at baseline and randomized to the CP group had slightly better outcomes. No differences in use of ventilation, successful stopping of mechanical ventilation, progression to use of renal replacement therapy, or cause-specific mortality were detected. Safety outcomes were comparable across groups. Adherence to the protocol was relatively good. Eighty percent of patients assigned to the CP group received the full two units, 11% received one unit and 9% received none; 17 patients in the usual care group received CP. Slightly fewer patients in the CP group received tocilizumab or sarilumab but use of remdesivir was comparable between groups. There were also fewer men randomly assigned to the CP group but regression analyses adjusted for sex did not suggest any impact on outcomes.