Methods
Study design and patient population
This is a retrospective multi-site cohort study of adult patients during a one-year period from 11/6/2020 to 11/5/2021. All patients were admitted to one of the Mayo Clinic hospitals with COVID-19 diagnosis and received at least one dose of intravenous remdesivir. Per our standard practice, all patients hospitalized for COVID-19 were assessed for eligibility for a 5-day remdesivir course in the inpatient setting.
For this study, the patients were categorized into those who: 1) completed the 5-day remdesivir course entirely inpatient, 2) completed remdesivir course after transitioning to the outpatient setting, 3) did not complete remdesivir course but remained inpatient, and 4) did not complete remdesivir after discharge from the hospital. Determination of inpatient or outpatient status was based on patient classes, department, and manual review when unable to determine by other methods. Remdesivir treatment course was categorized as incomplete if <5 doses or <600 mg total was documented in the medication administration record.
Patients were eligible for this study if they were ≥18 years of age and received a first course of remdesivir. If there were >2 calendar days between doses of remdesivir, these were considered separate courses and the subsequent course(s) were excluded. Other exclusion criteria included receipt of an anti-spike neutralizing monoclonal antibody or convalescent plasma, use of remdesivir under another study protocol, and patients transitioning from an outside facility during remdesivir treatment. Minnesota residents who declined research authorization were excluded. The Mayo Clinic Institutional Review Board deemed this study to be exempt.
Clinical variables and outcomes
All-cause mortality was defined as death within 28 days of the first dose of remdesivir. Death related to COVID-19 was determined by chart review by study team physician or clinical pharmacist. Second review was performed by study physician if cause of death was unable to be determined by clinical pharmacist.
Liver function tests (LFT, alanine and aspartate aminotransferases) were reviewed for 10 days from the start of remdesivir treatment. Body mass index (BMI), World Health Organization (WHO) ordinal scale, Monoclonal Antibody Screening Score (MASS), and use of steroids, baricitinib, or tocilizumab were collected. MASS is a clinical risk prioritization score for the allocation of anti-spike monoclonal antibodies, and includes age, body mass index, chronic heart, kidney and lung diseases, diabetes mellitus, hypertension, pregnancy and an immunosuppressed condition or treatment.10 Primary diagnosis code was used to identify hospital and emergency care encounters related to COVID-19 within 30 days of the last administered dose of remdesivir and long COVID related encounters within 180 days.
Statistical analysis
Data are summarized using medians and interquartile ranges (IQR) for continuous data and frequencies and percentages for categorical data. Patient characteristics were compared between groups using Kruskal-Wallis tests for continuous data and either Chi-square or Fisher’s exact tests for categorical data. Multivariable logistic regression was used to assess associations between treatment group and the outcomes of having an abnormal LFT, all-cause mortality within 28 days, and death related to COVID-19 within 28 days of remdesivir treatment initiation after adjusting for age, sex, and MASS score. Rates of subsequent ED visit or hospitalization within 30 days and subsequent long COVID encounters within 180 of remdesivir initiation were estimated using the Aalen-Johansen method, where death was considered a competing risk. Cox proportional hazards regression was used to assess the associations between these outcomes and treatment groups. Pairwise comparisons between groups were adjusted for multiple comparisons using the false discovery rate method. All analyses were performed using SAS version 9.4 software (SAS Institute, Inc; Cary, NC).