Prognostic performance of an index based on lactic dehydrogenase and transaminases for patients with liver steatosis and COVID-19

Abstract

BACKGROUND Metabolic associated fatty liver disease (MAFLD) is associated with complications and mortality in patients with coronavirus disease 2019 (COVID-19). However, there are no prognostic scores aimed to evaluate the risk of severe disease specifically in patients with MAFLD, despite its high prevalence. Lactate dehydrogenase, aspartate aminotransferase and alanine aminotransferase have been used as markers of liver damage. Therefore, we propose an index based on lactate dehydrogenase, aspartate aminotransferase and alanine aminotransferase for the prediction of complications and mortality in patients with MAFLD and COVID-19. AIM To evaluate the prognostic performance of an index based on lactate dehydrogenase and transaminases (aspartate aminotransferase/alanine aminotransferase) in patients with COVID-19 and MAFLD [liver fibrosis and nutrition (LNF)-COVID-19 index]. METHODS In this retrospective cohort study, two cohorts from two different tertiary centers were included. The first was the derivation cohort to obtain the score cutoffs, and the second was the validation cohort. We included hospitalized patients with severe COVID-19 and MAFLD. Liver steatosis was evaluated by computed tomography scan. Area under the receiver operating characteristic (ROC) curve analysis and survival analysis were used. RESULTS In the derivation cohort, 44.6% had MAFLD; ROC curve analysis yielded a LFN-COVID-19 index > 1.67 as the best cutoff, with a sensitivity of 78%, specificity of 63%, negative predictive value of 91% and an area under the ROC curve of 0.77. In the multivariate analysis, the LFN-COVID-19 index > 1.67 was independently associated with the development of acute kidney injury (odds ratio: 1.8, 95% confidence interval: 1.3-2.5, P < 0.001), orotracheal intubation (odds ratio: 1.9, 95% confidence interval: 1.4-2.4, P < 0.001), and death (odds ratio: 2.86, 95% confidence interval: 1.6-4.5, P < 0.001) in both cohorts. CONCLUSION LFN-COVID-19 index has a good performance to predict prognosis in patients with MAFLD and COVID-19, which could be useful for the MAFLD population.