RESUMO
INTRODUCTION AND OBJECTIVES: The incidence of liver injury (LI) in hospitalized COVID-19 patients ranged from 14% to 53% based on sole or multiple elevated indexes for LI. The aims of our study were to investigate the changes of parameters (ALT, AST) in LI and determine the risk factors for LI in a cohort of 830 COVID-19 patients. METHODS: Demographic information, clinical features, and laboratory testing outcomes on admission were compared between patients with and without liver biochemistry abnormality (LBA). The same comparisons were performed between the LBA and LI groups. The updated RUCAM was used to determine the causality between drugs application and LI. Univariable and multivariable logistic regression analyses were used to explore the potential risk factors associated with LBA and LI. RESULTS: A total of 227 (27.3%) patients exhibited LBA and 32 (3.9%) patients were categorized as having LI based on the diagnostic criteria. 32.6% (74/227) of the LBA patients had RUCAM score >3, whereas the non-LBA patients had a slight lower at rate of 24.2% (146/603) (P?=?0.047). Multivariable regression showed that a higher incidence of LBA was associated with hepatic hypoattenuation on computed tomography (CT) (odds ratio: 2.243, 95% confidence interval: 1.410-3.592, p?=?0.001), lymphocyte proportion <20% (2.088, 1.476-2.954, p?1?mg/dL (2.650, 1.845-3.806, p?1 (2.558, 1.820-3.596, p?1.0?mg/dL, lymphocyte proportion <20%, AST/ALT ratio <1, and triglyceride levels >1.7?mol/L are potential risk factors for LI.