Impact of COVID-19 therapeutics on the development of post-infectious lung fibrosis

TL;DR

This study found that certain treatments for COVID-19 may influence the risk of developing lung fibrosis after recovery.

Contribution

The study identifies remdesivir and baricitinib as factors associated with reduced and increased risk of post-COVID-19 pulmonary fibrosis.

Findings

Pre-existing renal disease and elevated CRP levels were independently associated with increased risk of PCPF.

Remdesivir use was significantly linked to a reduced risk of PCPF, while baricitinib use was associated with increased risk.

Abstract

Post-COVID-19 pulmonary fibrosis (PCPF) is a significant long-term complication in survivors of COVID-19. In this study, we aimed to identify clinical risk factors for PCPF and evaluate the impact of COVID-19–related therapies.

We retrospectively studied hospitalized adults with confirmed COVID-19 across three hospitals in South Korea from 2020 to 2022. Inclusion required chest computed tomography (CT) imaging both before and after COVID-19 infection. PCPF was defined as fibrotic changes seen on follow-up CT performed at least one month after recovery.

Among 5,720 hospitalized adults with COVID-19, 688 met the inclusion criteria, and 87 (12.6%) developed PCPF based on follow-up CT. In the multivariate logistic regression, pre-existing renal disease (adjusted odds ratio [aOR] 3.287; 95% confidential interval [CI]: 1.260–8.580; p = 0.014), higher hemoglobin levels (aOR: 1.194; 95% CI: 1.032–1.387; p = 0.018) and elevated CRP (aOR: 1.005; 95% CI: 1.001–1.009; p = 0.022) were independently associated with increased risk of PCPF. Remdesivir use was significantly associated with a reduced risk of PCPF (aOR: 0.359; 95% CI: 0.176–0.734; p = 0.005), whereas baricitinib use was associated with an increased risk (aOR: 5.633; 95% CI: 1.642–19.548; p = 0.006).

PCPF remains a relevant sequela in COVID-19 survivors. Remdesivir and baricitinib use were associated with a reduced and increased risk of PCPF, respectively. Although adjusted for multiple confounders, residual indication bias of each treatment cannot be completely excluded. Therefore, prospective studies are needed to validate these associations.

Figures