Work

Early inflammation-driven damage and delayed viral clearance are associated with prolonged biliary perturbation in a Cameroonian cohort of COVID-19 patients

Year: 2026

Type: article

Abstract: Although COVID-19-associated liver abnormalities are documented, the link between delayed viral clearance, IgG seroconversion, and inflammation remains poorly defined, particularly in Sub-Saharan African populations, where baseline liver enzymes differ from Western standards. In 38 SARS-CoV-2-positive individuals monitored at D0, D14, and D30, we quantified liver biomarkers (ALT, AST, BILT, BILD) by spectrophotometry (COBAS), plasma IL-6 by Luminex, viral load (N and ORF1ab genes) by RT-qPCR, and IgG/IgM serology. Analyses were performed in GraphPad Prism v9.0. ALT and AST rose transiently at D14, returning to baseline by D30. IL-6 peaked at D14 and correlated positively with BILT (r = 0.64, p = 0.001) and BILD (r = 0.54, p = 0.006). ORF1ab Ct at D14 correlated negatively with ALT at D0 (r = -0.67, p = 0.004) and D14 (r = -0.71, p = 0.006); N gene Ct at D14 correlated with ALT at D0 (r = -0.61, p = 0.011). Mixed-effects models identified IL-6 as the sole independent predictor of AST (β = 3.04, p = 0.005) and Ct N gene as an independent predictor of direct bilirubin (β = -0.003, p = 0.025), supporting a dual-pathway model of COVID-19-associated cholestasis. Paradoxically, IgG seropositivity independently predicted higher direct bilirubin (β = 0.06, p = 0.030) but did not mark biliary normalisation. Elevated ALT and BILT at D0 were associated with detectable viral RNA beyond D14 (P = 0.03). Participants aged >30 years drove rising BILD, with significant increases at D14 (P = 0.03) and D30 (P = 0.04) versus D0. COVID-19 liver injury follows a biphasic course: D14 alterations are inflammation-driven (IL-6), while D30 abnormalities track with delayed viral RNA clearance independent of IgG seroconversion. Extended hepatic monitoring is warranted, especially in older patients with delayed viral clearance in resource-limited settings.

Source: PLoS ONE

DOI: 10.1371/journal.pone.0357503