Although several liver- and inflammation-based scores to predict the clinical course of patients with coronavirus disease 2019 (COVID-19) have been evaluated, no direct comparison regarding their predictive ability has been performed.

1038 patients (608 males, age 63.5 ± 17 years) hospitalized with documented COVID-19 infection to the non-ICU ward, were included retrospectively. Clinical and laboratory characteristics on admission including evaluation of Fibrosis-4 (FIB-4) score and C-Reactive Protein (CRP) to albumin ratio (CAR) were recorded.

One hundred and twenty-four patients (11.9%) died during hospitalization after 8 (3–72) days. In multivariate analysis, FIB-4 (hazard ratio, 1.11; 95% confidence interval (CI), 1.034–1.19; P = 0.004), was independently associated with mortality, with very good discriminative ability (area under the receiver operating characteristic curve curve, 0.76). The patients with FIB-4 >2.67 (n = 377), compared to those with ≤2.67 (n = 661), had worse survival (log-rank 32.6; P < 0.001). Twenty-four (6.8%) of 352 patients with possible nonalcoholic fatty liver disease (NAFLD) (defined as Hepatic Steatosis Index >36) died during hospitalization. In multivariate analysis, CAR was an independent risk factor (1) for mortality (hazard ratio, 1.014; 95% CI, 1.002–1.025; P = 0.021), (2) the need for high-flow nasal cannula with or without intubation (hazard ratio, 1.016; 95% CI, 1.004–1.027; P = 0.007) and (3) development of acute kidney injury (hazard ratio, 1.017; 95% CI, 1.006–1.028; P = 0.002). In addition, the patients with possible NAFLD and CAR >12 (n = 154), compared to those with CAR ≤12 (n = 198), had worse survival (log-rank 5.1; P = 0.024).

FIB-4 was an independent factor for mortality with better performance compared to other liver function test- and inflammation-based scores in patients with COVID-19, while CAR was the only score independently associated with the clinical course in COVID-19 patients with possible NAFLD.