Frontiers in Medicine (Nov 2020)
Pre-existing Liver Diseases and On-Admission Liver-Related Laboratory Tests in COVID-19: A Prognostic Accuracy Meta-Analysis With Systematic Review
Abstract
Background: We aimed to perform a systematic search and meta-analysis to evaluate the prognostic value of on-admission liver function tests and pre-existing liver diseases on the clinical course of coronavirus disease 2019 (COVID-19).Methods: The study was registered on PROSPERO (CRD42020182902). We searched five databases between 01/01/2020 and 04/23/2020. Studies that reported on liver-related comorbidities and/or laboratory parameters in patients with COVID-19 were included. The main outcomes were COVID-19 severity, intensive care unit (ICU) admission, and in-hospital mortality. Analysis of predictive models hierarchical summary receiver-operating characteristic (HSROC) was conducted with a 95% confidence interval (CI).Results: Fifty studies were included in the meta-analysis. High specificity was reached by acute liver failure associated by COVID-19 (0.94, 95% CI: 0.71–0.99) and platelet count (0.94, 95% CI: 0.71–0.99) in the case of mortality; chronic liver disease (CLD) (0.98, 95% CI: 0.96–0.99) and platelet count (0.82, 95% CI: 0.72–0.89) in the case of ICU requirement; and CLD (0.97, 95% CI: 0.95–0.98), chronic hepatitis B infection (0.97, 95% CI: 0.95–0.98), platelet count (0.86, 95% CI: 0.77–0.91), and alanine aminotransferase (ALT) (0.80, 95% CI: 0.66–0.89) and aspartate aminotransferase (AST) (0.84, 95% CI: 0.77–0.88) activities considering severe COVID-19. High sensitivity was found in the case of C-reactive protein (CRP) for ICU requirement (0.92, 95% CI: 0.80–0.97) and severe COVID-19 (0.91, 95% CI: 0.82–0.96).Conclusion: On-admission platelet count, ALT and AST activities, CRP concentration, and the presence of acute and CLDs predicted the severe course of COVID-19. To highlight, pre-existing liver diseases or acute liver injury associated by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection plays an important role in the prediction of mortality.
Keywords
