ABSTRACT
Background: COVID-19 is characterized by a dysregulated inflammatory response associated with disease severity, poor prognosis and death. The aim of this study was to describe the real-life use of high-dose anakinra (ANK, a recombinant IL-1 receptor antagonist) among patients with COVID-19 who received remdesivir (REM). Methods: Cohort study including 277 patients with COVID-19 hospitalized at IRCCS San Raffaele Hospital between September 1st,2020 and February 28th, 2021;58 patients were treated with REM+ANK and 219 patients with REM only. ANK was administered intravenously at a dose of 5mg/kg every 12 hours. Patients were treated according to available local and international guidelines;corticosteroids and anticoagulation were administered when not contraindicated. Results are described by median (IQR) or frequency (%);P-values (P) were calculated by chi-square or Fishers' exact test and Wilcoxon rank-sum test, as appropriate. Survival estimates at 28 days were calculated using Kaplan-Meier curves. Results: At hospital admission (Table 1), patients treated with REM+ANK tended to be older [69 years (57-77) vs 62 years (53-75), P=0.06], had a significant lower PaO2/FiO2 [135 (91-220) vs 246 (172-299), P=0.0001], higher aspartate aminotransferase [51U/L (34-74) vs 40U/L (30-53), P=0.001], lactate dehydrogenase [405U/L (296-496) vs 334U/L (279-419), P=0.008], D-dimer [0.86mcg/mL (0.48-1.57) vs 0.67mcg/mL (0.39-1.17), P=0.048], ferritin [1167ng/mL (804-1983) vs 683ng/mL (391-1153), P<0.0001] and C-reactive protein [82mg/L (38-136) vs 58 mg/L (27-96), P=0.004), and were more frequently admitted to the Intensive Care Unit within the first 48 hours [3 (1.1%) vs 0, P=0.007). REM and ANK were started early within a median of 0 (0-2) and 1.5 days (0-3) since hospitalization, respectively. The Kaplan-Meier estimate of mortality at 28 days was 17.2% (95%CI 8.8-32.1%) in the REM+ANK group (8 deaths) and 21.4% (95%CI 13.3-33.3%) in the REM group (18 deaths;log-rank test P=0.797). Median time to death was 14 days (9-29) in the REM+ANK group vs 19 days (12-27) in the REM group (P=0.523). Conclusion: Real-life use of high-dose ANK in COVID-19 patients treated with REM was reserved for subjects with severe respiratory failure and a more pronounced inflammatory status. Nevertheless, mortality at 28 days was not significantly different among patients treated with or without ANK. Further analyses are warranted to verify the impact of ANK addition to REM in patients with a hyperinflammatory profile.