ABSTRACT
Introduction: Trials of remdesivir (RDV) for COVID-19 have provided evidence for regulatory approval. This is the first meta-analysis (MA) to evaluate the real-world effectiveness of RDV in patients hospitalized with COVID-19. Objective(s): To synthesize RDV observational data. Method(s): A systematic literature review identified observational studies of RDV. Outcomes were all-cause mortality and progression to invasive mechanical ventilation (IMV) assessed at early (day 14/15) and late (day 28/29/30) timepoints. MAs were conducted using standard random effects models;analyses were performed with R statistical software. Result(s): Of 1,069 studies identified, 29 met inclusion criteria for mortality data, 18 were excluded for low quality based on the ROBINS-I tool;11 studies from the United Kingdom, European Union, United States and Japan were included in the MA (N=166,399 patients). RDV was associated with a significant improvement in mortality at early (5 studies;risk ratio [RR] 0.71, 95% confidence interval [CI] 0.64-0.79) and late (10 studies, RR 0.82, 95%CI 0.71-0.95;Figure) timepoints. No significant effect was shown on the proportion of patients requiring IMV (evaluable only in the 3 studies denoted by asterisk in Figure, RR 1.07, 95%CI 0.84-1.34). Results were robust to scenario analyses. Conclusion(s): In a real-world setting, RDV is effective in reducing mortality in hospitalized COVID-19 patients.
ABSTRACT
Background: SARS-CoV-2 (i.e., COVID-19) has brought extracorporeal membrane oxygenation (ECMO) into the forefront of critical care. Its unique pathophysiology has added a level of complexity to ECMO therapy, particularly, the hematologic manifestations. Here we detail the spectrum and outcomes of bleeding complications in ECMO for COVID-19 and identify potential contributing factors. Methods: All patients who received ECMO for SARS-CoV-2 pneumonia severe acute respiratory distress syndrome at our institution between March 1, 2020 and April 12, 2021 were reviewed. Patient characteristics, laboratory results, and overall outcomes were recorded. Bleeding events were reviewed with regard to the type/location and intervention required. Severity was graded according to the degree of intervention for treatment (1 [conservative or minor] - 3 [major, life-threatening, or operative]). Laboratory results and patient characteristics were compared between patients with bleeding events and those without to identify factors associated with bleeding risk. Results: Fifty-four patients (mean age 53.2 years, 61.1% female, 51.9% Caucasian) underwent ECMO cannulation for SARSCoV-2 pneumonia at our institution. Thirty-eight (70.4%) received veno-pulmonary artery ECMO. The mean duration of support was 33.2 days with an in-hospital mortality of 42.6%. 68.5% of patients experienced at least one bleeding event during their ECMO course with 92 bleeding events (n=23 [grade 1], n=31 [grade 2], n=38 [grade 3]) over 1804 cannulation days. The most common types of bleeding types were nasal/oropharyngeal (n=30, 32.6%), pulmonary (n=18, 19.6%), and gastrointestinal (n=11, 12.0%). Eight (16.0%) patients required operative intervention and 11 (20.3%) died as a result of a bleeding event, mainly due to intracranial hemorrhages (n=5, 9.3%). There was no difference in the mean cumulative function for bleeding events between different ECMO support modalities (p=0.85) which demonstrated a linear pattern over time. Factors that increased the risk of bleeding included patient cumulative volume balance (OR 1.22 per 1000 mL increase from admission, p<0.001) while higher platelet count (OR 0.83 per 50x103/uL increase, p=0.03) was protective. Conclusion: ECMO for SARS-CoV-2 pneumonia is associated with a diverse and unique profile of bleeding complications. The incidence of bleeding complications is linearly related to cannulation duration. Certain patient factors may affect the risk of bleeding while on ECMO.