Optimal timing of remdesivir initiation in hospitalized COVID-19 patients administered with dexamethasone

Link to article at PubMed

Clin Infect Dis. 2021 Aug 22:ciab728. doi: 10.1093/cid/ciab728. Online ahead of print.


BACKGROUND: Evidence is lacking about any additional benefits of introducing remdesivir on top of dexamethasone, and the optimal timing of initiation.

METHODS: In a territory-wide cohort of 10,445 COVID-19 patients from Hong Kong who were hospitalized between 21st January 2020 and 31st January 2021, 1544 patients had received dexamethasone during hospitalization. Exposure group consisted of patients who had initiated remdesivir prior to dexamethasone (n=93), or co-initiated the two drugs simultaneously (n=373); whereas non-exposure group included patients who were given remdesivir after dexamethasone (n=149), or those without remdesivir use (n=929). Multiple imputation and inverse probability of treatment weighting for propensity score were applied and hazard ratios (HR) of event outcomes were estimated using Cox regression models.

RESULTS: Time to clinical improvement (HR=1.23, 95%CI 1.02-1.49, p=0.032) and positive IgG antibody (HR=1.22, 95%CI 1.02-1.46, p=0.029) were significantly shorter in the exposure group than that of non-exposure. The exposure group had a shorter hospital length of stay by 2.65 days among survivors, lower WHO clinical progression scale scores from five days of follow-up onwards, lower risks of in-hospital death (HR=0.59, 95%CI 0.36-0.98, p=0.042) and composite outcomes; and without experiencing an increased risk of ARDS. Differences in the cumulative direct medical costs between groups were no longer significant from 17 days of follow-up onwards.

CONCLUSIONS: Initiation of remdesivir prior to or simultaneously with dexamethasone was associated with significantly shorter time to clinical improvement and positive IgG antibody, lower risk of in-hospital death, in addition to shorter length of hospital stay in patients with moderate COVID-19.

PMID:34420051 | DOI:10.1093/cid/ciab728

Leave a Reply

Your email address will not be published.