ABSTRACT
INTRODUCTION: The spread of coronavirus disease 2019 (COVID-19) is having a profound effect on global health. In this study, we investigated early predictors of severe prognosis from the perspective of liver injury and risk factors for severe liver injury in patients with COVID-19. METHODS: We examined prognostic markers and risk factors for severe liver injury by analyzing clinical data measured throughout the course of the illness and the disease severity of 273 patients hospitalized for COVID-19. We assessed liver injury on the basis of aminotransferase concentrations and fibrosis-4 (FIB-4) index on admission, peak aminotransferase concentration during hospitalization, aminotransferase peak-to-average ratio, and albumin and total bilirubin concentrations. Furthermore, we analyzed age, aspartate aminotransferase (AST) concentrations, FIB-4 index on admission, hypertension, diabetes mellitus (DM), dyslipidemia, cerebral infarction, myocardial infarction, and body mass index as mortality risk factors. RESULTS: We identified advanced age as a risk factor. Among biochemical variables, AST concentration and FIB-4 index on admission were associated with high mortality. AST on admission and peak AST during hospitalization were significantly higher in the non-surviving (n = 45) than the discharged group (n = 228). Multivariable Cox hazards analyses for mortality showed significant hazard ratios for age, peak AST, and FIB-4 index on admission (p = 0.0001 and 0.0108, respectively), but not in a model including AST and FIB-4 index on admission. Furthermore, the AST peak was significantly higher among non-surviving patients with DM than in those without DM. CONCLUSIONS: We found that advanced age, high AST, and FIB-4 index on admission and a higher peak AST during hospitalization are risk factors for poor COVID-19 prognosis. Furthermore, DM was a risk factor for exacerbation of liver injury among non-surviving patients. The AST concentration and FIB-4 index should be assessed periodically throughout hospitalization, especially in patients with high AST values on admission and those with DM.
ABSTRACT
The number of COVID-19 patients in Japan is considered low, compared with U.S. and European countries. However, recent serological survey reported that several percent of population showed IgG positive to SARS-CoV-2. Specificity in the assays might influence the estimate, and possibility of overdiagnosis should be investigated. Serological tests for SARS-CoV-2 coronavirus were performed in pre-COVID-19 sera in Japan (400 healthy subjects in 2012-2015). Lateral flow assay (LFA) and enzyme-linked immunosorbent assay (ELISA) showed 1.5% (6/400) and 1.75% (7/400) IgG positives, respectively. Among those false positive samples, only one sample was positive in both LFA and ELISA (0.25%; 95% CI: 0.006-1.39%). Possible bias from pooling method was examined by Monte Carlo method and the possibility was unlikely at low false positive rate. Previous surveys might overestimate COVID-19 seroprevalence in several populations of Japan. These false positives could be excluded by combination of different diagnostics. Nonetheless, the result of seroprevalence should be carefully interpreted in less prevalent areas.
ABSTRACT
The number of COVID-19 patients in Japan is considered low, compared with U.S. and European countries. However, recent serological survey reported that several percent of population showed IgG positive to SARS-CoV-2. Specificity in the assays might influence the estimate, and possibility of overdiagnosis should be investigated. Serological tests for SARS-CoV-2 coronavirus were performed in pre-COVID-19 sera in Japan (400 healthy subjects in 2012-2015). Lateral flow assay (LFA) and enzyme-linked immunosorbent assay (ELISA) showed 1.5% (6/400) and 1.75% (7/400) IgG positives, respectively. Among those false positive samples, only one sample was positive in both LFA and ELISA (0.25%;95% CI: 0.006-1.39%). Possible bias from pooling method was examined by Monte Carlo method and the possibility was unlikely at low false positive rate. Previous surveys might overestimate COVID-19 seroprevalence in several populations of Japan. These false positives could be excluded by combination of different diagnostics. Nonetheless, the result of seroprevalence should be carefully interpreted in less prevalent areas.