A meta-regression study of the clinical significance of serum aminotransferases in COVID-19.

Abstract

The aim of the study was to appraise the capacity of serum aminotransferases to discriminate between hepatic and other extra-pulmonary COVID-19-related manifestations and, potentially, to serve as predictors of poor clinical outcomes. Ninety-eight studies were identified (79% from China), including 43,554 patients (57% males), 9,983 (62% males) with poor outcomes and 33,571 (50% males) with favorable outcomes. After splitting studies depending on whether serum alanine aminotransferase (ALT) concentrations were statistically different between patients with poor vs. favorable outcomes, the 35 'hepatic involvement' articles (p<0.05) included 28,510 patients (51% males), 5,279 (66% males) and 23,231 subjects (48% males) with poor and favorable outcomes, respectively. The 63 'extra-hepatic involvement' studies (p>0.05) included 15,044 patients (54% males), 4,704 (60% males) with poor outcomes and 10,340 (51% males) with favorable outcomes. The meta-analysis shows that serum aspartate aminotransferase (AST) concentrations were significantly higher in patients with poor outcomes than those with favorable outcomes (WMD 12.5 UI/L, 95% CI 10.9 to 14.1 p<0.001). Similarly, AST concentrations were significantly higher in the 'hepatic involvement' studies (WMD 16.3 UI/L, 95% CI 13.4 to 19.2 p<0.001) and in the 'extra-hepatic involvement' studies (WMD 10.3 UI/L, 95% CI 8.6 to 12.0 p<0.001). The different association of serum AST concentrations with some clinical, demographic, and biochemical factors in the two clusters suggests that in COVID-19 patients, serum AST elevation is not necessarily linked to real liver damage.