key: cord-0770946-xscsrexe
authors: Ding, Ze-yang; Li, Gan-xun; Shu, Chang; Yin, Ping; Zhang, Bixiang
title: Reply to: Comments on "Association of liver abnormalities with in-hospital mortality in patients with COVID-19”
date: 2021-06-06
journal: J Hepatol
DOI: 10.1016/j.jhep.2021.05.027
sha: e6faf5cc987bb776501612c197bdc7b5732e5218
doc_id: 770946
cord_uid: xscsrexe

nan

To the Editor:

We thank Singh et al, Horvath et al, and Luo et al for their comments on our recent study [1] . In this study, we focused on investigating the association of admission abnormal liver chemistries and in-hospital death, rather than the etiology of liver injury in COVID-19.We agree with comments from Singh et al that liver injury in COVID-19 may not attribute to COVID-19 infection alone, and associations of hypoxia, systemic inflammation, and hepatotoxic drugs with liver injury were explored in another study of our institute [2] . In our study, parameters of hypoxic Figure 1B ). We are expecting these predictive models to be validated in more cohorts in the future.

Luo et al raised concerns on the statistical analyses and suggested that it is better to use disease specific survival (DSS) instead of overall survival (OS) to build the nomogram. This suggestion is lack of feasibility, for that COVID-19 is an emerging infectious disease with its pathophysiology still exploring, and there is still no consensus on disease-specific death of COVID-19 [5] . In addition, Luo et al comments that based on the Riley's minimum sample size criteria [6] , a much larger sample size of 11200 is required for establishing a robust predict model in our study. The predict model of our study was used with an events per predictor parameter (EPP) of 20 (200 outcome events/10 parameters), which is compliant with the rule of thumb that a minimum of 10 EPPs is necessary for Cox models [7] . We noticed that when in introducing methods of calculating sample size in prediction models for a time-to-event outcome [6] . Whether the Riley's minimum sample size criteria is suitable for establishing predict models of acute diseases is needed to be confirmed and validated. In addition, the aim of the large sample size is to ensure the robust of the predict model, whereas this robust has been internal validated by setting the bootstrap resampling cohort in our study, and been external validated by Horvath et al in an Austrian cohort. 

Association of liver abnormalities with in-hospital mortality in patients with COVID-19

Clinical Features of Patients Infected With Coronavirus Disease 2019 With Elevated Liver Biochemistries: A Multicenter, Retrospective Study

High rates of 30-day mortality in patients with cirrhosis and COVID-19

COVID-19 and liver disease: mechanistic and clinical perspectives

COVID-19: Discovery, diagnostics and drug development

Calculating the sample size required for developing a clinical prediction model

The impact of COVID-19 on the clinical outcome of patients with cirrhosis deserves more attention and research