Influence of liver attenuation on the severity of course COVID-19: a retrospective cohort study

Abstract

Introduction . A high prevalence of decreased liver density has been shown among patients with COVID-19, but there are no convincing data on the cause of this phenomenon. It is still debatable whether decreased liver attenuation is an independent risk factor for the severe course of COVID-19. Aim . Assessment the prognostic value of liver attenuation on CT scan in patients with COVID-19. Materials and methods . Retrospective cohort study. Data of COVID-19 outpatients were analyzed. Inclusion criteria: two chest CT scans, alanine aminotransferase (ALT), aspartate aminotransferase (AST) blood values, polymerase chain reaction to verify SARS-CoV-2. Four comparison groups were assigned according to the severity of lung lesions. Liver attenuation was analyzed by automatic segmentation, with values less than 40 HU being considered pathologic. Results . Data from 499 patients was included. No correlation between ALT and AST and changes in liver attenuation was found. Groups differed in age and liver attenuation on both CT scans. On follow-up CT, low liver density was seen in males (odds ratio (OR) 2.79 (95% CI 1.42–5.47), p-value = 0.003) and in patients with a baseline reduced liver density (OR 60.59 (95% CI 30.51–120.33), p-value < 0.001). Age over 60 years was associated with the development of lung lesions (OR 1.04 (95% CI 1.02–1.06) for extent of lung injury < 25%, OR 1.08 (95% CI 1.05–1.11) for 25–50%, OR 1.1 (95% CI 1.06–1.15) for 25–50%, p-value < 0.001). Low liver attenuation on the primary CT scan increased the odds of severe lung injury (OR 6.9 (95% CI 2.06–23.07), p-value = 0.002). Conclusion . In COVID-19, patients with low liver attenuation are more likely to develop severe lung damage.